‫هب انم خدا‬

‫‪www.iran-mavad.com‬‬
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
COMPREHENSIVE MATERIALS
PROCESSING
EDITOR-IN-CHIEF

Saleem Hashmi
Dublin City University, Dublin, Ireland

ASSOCIATE EDITORS-IN-CHIEF

Gilmar Ferreira Batalha

University of São Paulo, São Paulo, Brazil
Chester J. Van Tyne
Colorado School of Mines, Golden, CO, USA
Bekir Yilbas
King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia

VOLUME 11

ADVANCED MACHINING TECHNOLOGIES

VOLUME EDITOR

Mustafizur Rahman
National University of Singapore, Singapore

AMSTERDAM • BOSTON • HEIDELBERG • LONDON • NEW YORK • OXFORD

PARIS • SAN DIEGO • SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
Elsevier
Radarweg 29, PO Box 211, 1000 AE Amsterdam, Netherlands
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK
225 Wyman Street, Waltham, MA 02451, USA

Copyright  2014 Elsevier, Ltd. unless otherwise stated. All rights reserved.

No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic,
mechanical, photocopying, recording or otherwise without the prior written permission of the publisher.

Permissions may be sought from Elsevier’s Science & Technology Rights department in Oxford, UK: phone (+44) (0) 1865 843830;
fax (+44) (0) 1865 853333; email: [email protected]. Alternatively you can submit your request online by visiting the Elsevier
website at http://elsevier.com/locate/permissions and selecting Obtaining permission to use Elsevier material.

Website links cited in the Encyclopedia were accepted as accurate when first submitted to the publisher, however, the Publisher is unable
to ensure that they will continue to remain so.

Notice
No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability,
negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein,
Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress

ISBN: 978-0-08-096532-1

For information on all Elsevier publications

visit our website at store.elsevier.com

Printed and bound in Italy

14 15 16 17 18 19 10 9 8 7 6 5 4 3 2 1

Project Manager: Gemma Tomalin

Associate Project Manager: Joanne Williams

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
EDITOR-IN-CHIEF

Saleem Hashmi graduated from East Pakistan University of Engineering and Technology in 1967,
and in 1972 he completed his master’s and PhD degrees from the University of Manchester. He
was appointed as a postdoctoral fellow with UMIST, Manchester, in January 1973, and in October
1973, he joined Sheffield Hallam University as a lecturer. In 1976, he was promoted to senior
lecturer, then to principal lecturer in 1977, and to reader in 1984.
In January 1987, Professor Hashmi founded the School of Mechanical & Manufacturing Engi-
neering at Dublin City University. While in the process of establishing this school, he attracted
substantial external research funding to set up the Material Processing Research Centre (MPRC),
a research center of excellence in the field of material processing. To this day, he continues his
affiliation with Dublin City University as an emeritus professor in the school he founded 27 years ago.
In 1990, Professor Hashmi established Advances in Materials and Processing Technologies
(AMPT), a much needed international conference in materials processing. He continues to be the
Chairperson of the Steering Committee for this series of conferences. In 1998, he was appointed
as editor-in-chief of Elsevier Journal of Material Processing Technology and continued in this role
until 2008.
In 1995, Professor Hashmi was awarded the higher doctorate degree of DSc, by the University of
Manchester, the highest academic degree any university in the United Kingdom can award.
Professor Hashmi has supervised or co-supervised 110 PhD and 55 MEng research students to successful completion. Over the years,
he has also acted as an external examiner and expert assessor for PhD candidates and engineering departments with universities in Ireland,
the United Kingdom, India, Pakistan, Bangladesh, Hong Kong, Canada, Australia, and Malaysia. He is a Chartered Engineer (CEng), a Fellow
of the Institution of Mechanical Engineers, UK (FIMechE) and also a Fellow of the Institution of Engineers of Ireland (FIEI). He has
published over 430 papers and 12 books and is still publishing.

v
www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
This page intentionally left blank

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
EDITORIAL BOARD

Chester J. Van Tyne is the FIERF (Forging Industry Education and Research Foundation) professor
at Colorado School of Mines (CSM) in Golden, Colorado. He received his undergraduate and
graduate degrees from Lehigh University and worked for Bethlehem Steel as a research engineer for
a short period before joining Lafayette College in the Metallurgical Engineering Department as
a professor in 1980. In 1988, he joined the Metallurgical and Materials Engineering Department at
CSM. He is actively engaged in the research operations of the Advanced Steel Processing and
Products Research Center at CSM. His research interests are in metal deformation processes
especially forging, sheet stamping, and friction during metalworking operations.

Gilmar F. Batalha graduated as a mechanical engineer from the University of Brasilia in 1982. He
completed his MSc in Mechanical Engineering at the Federal University of Santa Catarina (1987),
his PhD in Mechanical Engineering (1995) and habilitation/livre-docencia in manufacturing
processes (2010) at the Department of Mechatronics and Mechanical Systems, Polytechnic School
of Engineering, University of Sao Paulo, where he has been a tenured professor since 1989. He has
had postdoctoral research experience at the Friedrich Alexander University Erlangen–Nuremberg,
Germany (1997–98). He was a visiting professor at Ecole Centrale de Lille, France (2004–06), and
at the Faculty of Mechanical Engineering of the Silesian University of Technology, Gliwice, Poland
(2012–13). He served as an adjunct professor at the Faculty of Engineering, UDESC, Joinville, Brazil
(1984–89). He has been an advisor on several PhD theses and master’s dissertations, coordinator
of the Laboratory of Manufacturing Engineering of the EPUSP, and is responsible for several
international and national cooperation agreements for exchange of students and researchers as well
as for consultancy with the industry. He is the volume editor for the Materials Modeling and
Characterization volume in Elsevier’s Comprehensive Materials Processing series.

Sergio T. Button received his PhD in Mechanical Engineering from School of Mechanical Engi-
neering, University of Campinas, Brazil, in 1990. He is currently full professor of Manufacturing
Processes at the University of Campinas. His main research fields include metal forming and
numerical simulation. Currently, his research is focused on cross wedge rolling and hot stamping.
He is the coauthor of 2 books and more than 100 papers in international journals and conferences.
He is the editor and reviewer of many international journals, and participates in the organization of
three national and international scientific conferences.

vii
www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
viii Editorial Board

David Cameron is currently scientific advisor to Miktech Oy, Finland, and adjunct professor in
Dublin City University, Ireland. From 2003 to 2013, he was professor of Material Technology in
Lappeenranta University of Technology, Finland, where he established the Advanced Surface
Technology Research Laboratory (ASTRaL). He obtained his doctorate from the University of
Glasgow in 1980 and has spent his research career working on thin-film technology. He was
a senior research fellow at the Royal Signals and Radar Establishment in the United Kingdom until
1982 when he joined the School of Electronic Engineering at Dublin City University, becoming
associate professor. His research focus is on atomic layer deposition and its industrialization.

Joseph A. McGeough is an honorary professorial fellow in the School of Engineering at the

University of Edinburgh. He is a graduate from the universities of Glasgow (BSc and PhD) and
Aberdeen (DSc). He held research appointments at the universities of Leicester, Queensland, and
Strathclyde. As a student he gained numerous county, university, and national championship
athletic awards. His industrial training began as an undergraduate vacation apprentice for a firm of
electrical contractors based at ICI ltd Nobel Division, and he subsequently undertook postgraduate
experience at International Research and Development, Newcastle upon Tyne, before returning to
university service. He was successively a lecturer, senior lecturer, and reader in Engineering at
Aberdeen University, and then moved to Edinburgh University to take up its Regius Chair of
Engineering. His research work has dealt mainly with manufacturing processes for which he has
been awarded Institution of Mechanical Engineers and the Society for Underwater Technology
prizes. He has been the recipient of a Royal Society/SERC (Science and Engineering (now Engi-
neering and Physical Sciences) Research Council) Industrial Fellowship in order to transfer the
results of his research to industry. His publications include Principles of Electrochemical Machining,
Advanced Methods of Machining, and Micromachining of Engineering Materials (editor). His latest book
is The Engineering of Human Joint Replacements (2013), which contains details of the materials and
manufacturing processes used in this field. He is a fellow of the Institution of Mechanical Engi-
neers, International Academy for Production Engineering, Royal Society of Edinburgh, and Royal
Academy of Engineering.

Emin Bayraktar is a full university professor and research director at Supmeca/Paris, School of
Mechanical and Manufacturing Engineering, Paris, France. He was awarded a PhD degree in
Mechanical Engineering by IST/OSU and a second PhD degree in Materials Science and Engi-
neering in Paris. Subsequently, Professor Bayraktar was awarded a higher doctorate degree of
habilitation (DSc) in Mechanics of Materials in recognition of his research contributions at
Paris 6, University Marie-Curie. His teaching and research expertise lies on the processing and
manufacturing of the composite materials and damage analysis. Recently he has been appointed
as an adjunct professor in UNICAMP-Campinas, Brazil, on the new design of composites for the
application in aeronautical engineering. He is on the editorial board of a number of scientific
journals. He is also a member of the Steering Committee for the International Conference Series
on Advances in Materials and Processing Technologies (AMPT).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Editorial Board ix

Bekir S. Yilbas received his PhD degree in Mechanical Engineering from Birmingham University.
He was awarded the Doctor of Engineering degree in 2005 by Birmingham University due to his
significant contribution to his field of study. He has published 375 journal papers in reputable
international journals and presented over 60 papers at international conferences. He has served
as an editorial board member of international journals, including International Journal of Machine
Tool and Manufacture Design, Research and Application, International Journal of Subsurface Sensing
Technologies and Applications, Journal of Materials Processing Technology, Journal of Achievements in
Materials and Manufacturing Engineering, International Journal of Nanomanufacturing, and Archives of
Materials Science and Engineering. He has received numerous awards in recognition of his research
work.

Nabil Bassim was born and raised in Cairo, Egypt. Dr. Bassim received a BSc in Chemical Engi-
neering from Cairo University in 1965. He went to the University of Virginia, United States, where
he received his MSc in Chemical Engineering (Nuclear Engineering option), and a PhD in Materials
Science in 1973. He moved to Canada where he is now professor of Materials Science and Engi-
neering at the University of Manitoba. He served for years as associate dean of research at the
Faculty of Engineering and was director of International Programs for the faculty. His areas of
research interest include mechanisms and mechanics of plastic deformation, dislocation theory,
fracture and fatigue, and nondestructive evaluations. He has patents on the use of acoustic emission
for monitoring of structures. He also has an interest in studying the behavior of materials at high
strain rates and large strains as they affect processes such as metal forming and machining. He has
published over 120 papers and supervised numerous MSc and PhD theses.

Syed H. Masood is professor of Mechanical and Manufacturing Engineering at Swinburne

University of Technology in Melbourne, Australia, where he is also the program coordinator of the
master’s program in advanced manufacturing technology. He obtained his PhD from the University
of Queensland, Australia; MEng from the University of New Brunswick, Canada; PG Dip from IIT
Delhi; and BSc Eng Hons from Aligarh Muslim University, India. Professor Masood has worked
previously at Central Queensland University, Rockhampton, and PNG University of Technology,
Papua New Guinea. He has published over 275 technical articles in international journals, books,
and conferences in different areas of advanced manufacturing. His current research interests include
additive manufacturing technologies, especially fused deposition modeling and direct metal
deposition, plastics manufacturing processes, material development, and material characterization.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
x Editorial Board

Mustafizur Rahman is currently a professor at the Department of Mechanical Engineering at

National University of Singapore (NUS), and has been with NUS since 1982. He obtained his BSc
in Engineering from Bangladesh University of Engineering and Technology (BUET), MEng and PhD
from Tokyo Institute of Technology, Japan. His research interests include micro-/nanomachining,
design and development of miniature machine tools and devices for micro-/nanomachining, and
high-speed machining of difficult-to-machine materials. He has received Outstanding Achievement
Awards from the Singapore Institute of Engineers and the Japan Society of Mechanical Engineers in
2003 and 2005, respectively.

George Krauss is currently university emeritus professor at the Colorado School of Mines and
a metallurgical consultant specializing in steel microstructural systems. Dr. Krauss received the BSc
in Metallurgical Engineering from Lehigh University in 1955 and the MSc and ScD degrees in
Metallurgy from the Massachusetts Institute of Technology in 1958 and 1961, respectively, after
working at the Superior Tube Company as a development engineer in 1956. In 1962–63, he was an
NSF postdoctoral fellow at the Max Planck Institut für Eisenforschung in Düsseldorf, Germany. He
served at Lehigh University as assistant professor, associate professor, and professor of Metallurgy
and Materials Science from 1963 to 1975, and in 1975 joined the faculty of the Colorado School of
Mines as the AMAX Foundation professor in Physical Metallurgy. He was the John Henry Moore
professor of Metallurgical and Materials Engineering at the time of his retirement from the Colo-
rado School of Mines in 1997.
In 1984, Dr. Krauss was a principal in the establishment of the Advanced Steel Processing and
Products Research Center, an NSF Industry-University cooperative research center at the Colorado
School of Mines, and served as its first director until 1993. He has authored the book Steels: Pro-
cessing, Structure, and Performance, (ASM International, 2005), coauthored the book Tool Steels, fifth
edition, (ASM International, 1998), and edited or coedited conference volumes on tempering of
steel, carburizing, zinc-based coatings on steel, and microalloyed forging steels. He has published
over 330 papers and lectured widely in technical conferences, universities, corporations, and ASM
International chapters, presenting a number of keynote, invited, and honorary lectures. He presented the Edward DeMille Campbell
Memorial Lecture of ASM International in 2000, the Howe Memorial Lecture of ISS in 2003, and the Adolf Martens Memorial Steel Lecture of
AIST in 2010.
Dr. Krauss has served as the president of the International Federation of Heat Treatment and Surface Engineering (IFHTSE), 1989–91,
and as President of ASM International, 1996–97. He is a fellow of ASM International, TMS, and IFHTSE. He has been awarded the Adolf
Martens Medal of the German Society for Heat Treatment and Materials, the Charles S. Barrett Silver Medal of the Rocky Mountain Chapter
of ASM, the George Brown Gold Medal of the Colorado School of Mines, and several other professional and teaching awards, including the
ASM Albert Easton White Distinguished Teacher Award in 1999. He is an honorary member of the Iron and Steel Institute of Japan,
a distinguished member of the Iron and Steel Society, an honorary member of ASM International, and an honorary member of the Japan
Institute of Metals.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
CONTRIBUTORS TO VOLUME 11

M.R. Aravind Raghavendra M.D. Nguyen

National University of Singapore, Singapore National University of Singapore, Singapore
M. Arif A.K.M. Nurul Amin
Swiss Federal Institute of Technology of Lausanne International Islamic University Malaysia, Kuala
(EPFL), Lausanne, Switzerland Lumpur, Malaysia
A.B.M.A. Asad T. Obikawa
National University of Singapore, Singapore The University of Tokyo, Meguro-ku, Tokyo, Japan
I.A. Choudhury M. Rahman
University of Malaya, Kuala Lumpur, Malaysia National University of Singapore, Singapore
T.L. Ginta T. Saleh
Universiti Teknologi, PETRONAS, Malaysia International Islamic University Malaysia, Kuala
Lumpur, Malaysia
M.A. Habib
Islamic University of Technology, Gazipur, Bangladesh A. Senthil Kumar
National University of Singapore, Singapore
M.P. Jahan
Western Kentucky University, Bowling Green, KY, USA E. Shamoto
Nagoya University, Nagoya, Japan
S.S. Joshi
Indian Institute of Technology Bombay, Mumbai, India H. Shinno
Tokyo Institute of Technology, Yokohama, Japan
S.A. Lawal
University of Malaya, Kuala Lumpur, Malaysia and N. Suzuki
Federal University of Technology, Minna, Nigeria Nagoya University, Nagoya, Japan
X. Li Z. Wang
National University of Singapore, Singapore Makino Inc., Mason, OH, USA
D. Marla Y.S. Wong
Indian Institute of Technology Bombay, Mumbai, India National University of Singapore, Singapore
T. Masaki K.S. Woon
Masaki Giken, Osaka, Japan Singapore Institute of Manufacturing Technology,
Singapore
T. Matsumura
Tokyo Denki University, Tokyo, Japan X. Zhang
National University of Singapore, Singapore and
C. Nath
Machining Technology Group, Singapore Institute of
National University of Singapore, Singapore and
Manufacturing Technology, Singapore
University of Illinois at Urbana-Champaign, Urbana,
IL, USA X.Q. Zhang
Singapore Institute of Manufacturing Technology,
W.K. Neo
Singapore
National University of Singapore, Singapore

xi
www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
This page intentionally left blank

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
CONTENTS OF VOLUME 11

Editor-in-Chief v
Editorial Board vii
Contributors to Volume 11 xi
Preface xv

11.01 Introduction to Advanced Machining Technologies 1

M Rahman, A B M A Asad and Y S Wong
11.02 Machine Tools for Micro- and Nanometer Scale Processing 15
H Shinno
11.03 Continuum Micromechanics and Molecular Dynamics Simulation 27
X Li
11.04 Fast and Fine Tool Servo for Ultraprecision Machining 61
A Senthil Kumar, M R Aravind Raghavendra, W K Neo and M Rahman
11.05 Compound and Hybrid Micromachining Processes 89
M Rahman, A B M A Asad, Y S Wong, M P Jahan and T Masaki
11.06 Compound and Hybrid Micromachining: Part II e Hybrid Micro-EDM and Micro-ECM 113
M Rahman, Y S Wong and M D Nguyen
11.07 Micromilling 151
T Matsumura
11.08 Fundamentals and Modeling of Micro-End Milling Operation 179
M Arif and M Rahman
11.09 Diamond Turning 201
X Q Zhang, K S Woon and M Rahman
11.10 High-Speed Machining 221
Z Wang and M Rahman
11.11 Machining with Least Quantity Lubrication 255
T Obikawa
11.12 Burr Formation in Machining Processes: A Review 283
I A Choudhury and S A Lawal
11.13 Heat-Assisted Machining 297
A K M Nurul Amin and T L Ginta
11.14 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications 333
M P Jahan, M Rahman and Y S Wong

xiii
www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
xiv Contents of Volume 11

11.15 Electrochemical Micromachining 373

S S Joshi and D Marla
11.16 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 405
E Shamoto and N Suzuki
11.17 Ultrasonic Vibration Cutting: Part II e Ductile Cutting and Analytical Force Models
for the Elliptical Vibration Cutting Process 455
C Nath, X Zhang, A Senthil Kumar and M Rahman
11.18 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation 483
T Saleh and M Rahman
11.19 Microelectrochemical Deposition 523
M A Habib

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
PREFACE

In recent years considerable advances have taken place in materials and their processing technologies, the pace
of which is progressing at an accelerated rate. It is therefore vitally important that researchers, academics, design
engineers, and manufacturing technologists be up to date with these new developments. Such developments
open up excellent opportunities to improve effective utilization of material resources, improve efficiency,
reliability, durability, and cost effectiveness of the products. This comprehensive series will serve researchers all
over the world in keeping them fully abreast of such new developments.
Comprehensive Materials Processing is a one-stop resource consolidating and enhancing the literature of the
materials processing and manufacturing universe. The work addresses recent progress in materials technologies
and science as well as covering recent developments in specific manufacturing processes involved with materials
for applications in engineering, biomedical, environmental protection, health and safety, and sensor materials
and technologies. The authoritative analysis of these processes will assist scientists and engineers in the selec-
tion, design, and usage of materials, whether required in the lab or in action across industry.
As the editor-in-chief of this comprehensive series, a truly collaborative work, I am greatly indebted to the
volume editors, internationally renowned experts in their fields, for guiding and selecting the topics for their
respective volumes, commissioning authors, and reviewing the content so meticulously. Their true dedication to
the scientific community is reflected in the time and energy they have given to this project. My sincere thanks are
also due to all of the authorseresearchers and materials practitioners who have contributed extensive coverage
of literature review as well as recent works of research to this huge project. Their insight and specialist knowledge
in their respective fields is reflected in the high quality of this work.
Myself, the three associate editors-in-chief, and all of the volume editors are greatly appreciative of all the
hard work undertaken to turn this idea into a publishable work. Our special thanks go to Gemma Tomalin, the
project manager, along with Joanne Williams, and the rest of the team at Elsevier who served successively to
keep the project on track through friendly nudges in order to ensure timely completion. We are also hugely
grateful to Mohamed Imrankhan for his dutiful coordination of the proofs.
The extensive academic discussion of core theories and applications, supplemented by applied case studies
and advanced multimedia features has drawn together numerous areas of research and I sincerely hope that this
work will prove to be of great help to both the young and experienced members of the international research
community, academics, and industrial practitioners for many years to come.
Saleem Hashmi
Editor-in-Chief

xv
www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
This page intentionally left blank

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 11.01 Introduction to Advanced Machining Technologies
M Rahman, ABMA Asad, and YS Wong, National University of Singapore, Singapore
 2014 Elsevier Ltd. All rights reserved.

11.01.1 Need for Tool-Based Micro-/Nanoscale Machining 2

11.01.2 Synergistic Development for Micro-/Nanoscale Machining 4
11.01.3 Tool-Based Machining Processes for Micro-/Nanoscale Applications 4
11.01.3.1 Micromilling 5
11.01.3.2 Microturning 6
11.01.3.3 Mechanical Microdrilling 7
11.01.3.4 Microgrinding 8
11.01.3.5 Micro-EDM 8
11.01.4 Compound Micromachining – A Recent Frontier in the Horizon 10
11.01.5 Summary 12
References 12

A very senior software engineer colleague of one of the authors from a semiconductor processing equipment company once
commented being disgruntled with some biased decisions made by the process engineering team – “the most influential man in
manufacturing is the process engineer; they are the only one and they are all in all.” Though it was commented in a way to vent out
frustration on the heated up argument on a technical issue that he had with a process engineer in the organization, there is an
implication in his comment. Albeit manufacturing is not just about process engineering, but is multidisciplinary in nature when
dealing with the design, implementation, operation, and optimization of the process concerned. An equipment with very good
mechanical structure and excellent control system but devoid of superior process capabilities will not be useful. But, at the same
time an identified innovative process supported by fundamental of physical process and driven by potential industrial application
will not be realized if a pertinently designed equipment is not available to enable its implementation and in-depth study.
The aforementioned case applies impeccably for micro-/nanoscale processing technologies. Primarily due to the similarity
between the fundamental concepts of conventional and micro-/nanoscale machining processes, one common misconception
about micro-/nanoscale machining is perception of only difference being the machining performed at a lower dimensional scale.
Practically, the process performance will be hindered by the machine tool employed for machining having a wrong setup or the
capability demonstrated at a dimensional range is not appropriate. Therefore, it is important to untwine the correct question and
for micro-/nanoscale machining it should be: whether or not it can provide solutions to manufacturing problems. And the solution
has to be at the required dimensional range and supported by underlying physical principles given that suitable machining
platforms exist for performing such machining processes. In particular it is very important to pay attention to the term ‘required
dimensional range’ as that is expected and therefore, the answer is required to match to that dimensional range which is often not
the case. While machining of a 500 mm diameter cylindrical rod is not particularly a problem with present day available tech-
nologies, this becomes a massive challenge when this is tried for a 50 mm cylindrical rod (Figure 1) (1). Therefore, demonstrations
of capability of micro-/nanomachining at upper or mid-boundary of the dimensional range for micro-/nanomachining domain are

Figure 1 Plastic flow and rough surface generated during microturning of brass below 100 mm. Reproduced from Rahman, M.; Asad, A. B. M. A.;
Masaki, Takeshi; Wong, Y. S.; Lim, H. S. A Multi-Process Machine Tool for Compound Micromachining. Int. J. Mach. Tools Manuf. 2010, 50/4, 344–356.

Comprehensive Materials Processing, Volume 11 http://dx.doi.org/10.1016/B978-0-08-096532-1.01101-8 1

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
2 Introduction to Advanced Machining Technologies

not acceptable as practically significant challenges are lying at the lower boundary of micro-/nanomachining domain below
w50 mm dimensional range.
In order to answer the question asked above it is also important to provide a very brief overview on the nature of challenges for
micro-/nanomachining to happen at the lower boundary of micro-/nanomachining domain. Micro-/nanomachining is a relatively
new arena of advanced manufacturing technology and there exists many magnificent process ideas for machining of parts and
components with micrometer range geometry which are considered as the art of micro-/nanomachining (2–8). At par, there exists
enormous difficulties in translating these process ideas into profitable manufacturing reality which are the engineering of micro-/
nanomachining and demands optimization for a very complex set of mutually inversely dependent output parameters. The
optimization process requires fragmenting the complex set of technical challenges into seemingly simple units following
a systematic and rigorous approach which loops through the conventional development life cycle recurrently but it is not confined
in a set of heuristic rules which could be applied in every recurrent cycle. It involves understanding the process requirements, setting
the criteria for mechanical system, mechanical design, fabrication, and assembly of the mechanical structure, developing electronic
circuits and control systems, developing intelligence in the system to control the physics behind the process, and finally observing
and empathizing to what the machine needs when it is not capable of controlling the underlying physics of the process. On the
other hand, the requirement of such processes is defined by the need of engineering and manufacturing practices.

11.01.1 Need for Tool-Based Micro-/Nanoscale Machining

The drive for miniaturization and fabrication of components with a wide selection of materials will allow microsystems technology
to enhance health care, quality of life, to attain new technological breakthrough, and to coat engineering applications with envi-
ronment friendly and energy saving practices. The pervasive demands of miniaturization across all engineering disciplines have
imparted the challenge of fabrication of such components to the manufacturing engineers. In present day, state-of-the-art fabri-
cation techniques refer to the fabrication of components and parts for microelectromechanical systems (MEMS), subminiature
actuators and sensors, components for biomedical devices, high precision equipment, components for advanced communication
technology, long microchannels for lab-on-chips, shape memory alloy ‘stents,’ fluidic graphite channels for fuel cell applications
and many more (2,9–11). The more recent trends indicate that the drive has gone beyond the little earlier challenge of precision and
minuteness in dimension to a new level where components of same precision and even less visible dimensions are produced by
machining on tough materials at lower cost.
Micro-/nanomachining is a general term used to indicate a collective form of all the process technologies used for machining of
microstructures and microsystems. The formal dimensional range of micromachining has been adopted as 1–500 mm by the
Scientific Technical Committee of the Physical and Chemical Machining Processes of The International Academy for Production
Engineering (CIRP) (5). Most of the technologies that have been utilized in micromachining are the existing technologies adapted
to operate in micrometer dimensions or adopted from microelectronic fabrication processes.
The past two decades have evidenced tremendous research and development specially focused on fabrication techniques for
MEMS. MEMS is the integration of mechanical elements, sensors, actuators, and electronics on a common silicon substrate through
the utilization of microfabrication technology. The electronic components in an MEMS package are mostly fabricated using inte-
grated circuit (IC) fabrication processes and the micromechanical components are fabricated using specialized technologies unique
to silicon micromachining processes that selectively etch away parts of the silicon wafer or add new structural layers to form the
mechanical and electromechanical devices (6,8,12). The manufacturing resources for silicon ICs are quite impressive, representing
decades of research and billions of dollars worth of investment in developing manufacturing techniques and equipment and the
existence of the very sizeable IC fabrication infrastructure. MEMS manufacturing leverages much of this extensive technological
base, but has developed a distinct set of fabrication technologies which when combined with conventional IC manufacturing
processes enable MEMS to be realized. Exciting MEMS actuators and devices capable of performing complicated mechanical
functions, such as moving, touching, and rubbing include gears and motors at the micro-/nanoscale (Figure 2) (13).
However, the majorities of these methods are limited to a few silicon-based materials, essentially planar geometries and therefore
significant amount of design constraints and thus broad commercialization of MEMS-based products has been hindered. Micro-
fabrication for aerospace, automotive, and precision engineering industries pose a very wide spectrum of challenge on micro-/
nanoscale machining but from an entirely different domain. Advent of miniaturization and new technologies have demanded
micromachining of any shape including true 3D structures on almost every material such as metals, plastics and semiconductors,
which are required for the moving parts and guiding structures (14). Applications related to microsurgery, biotechnology, fluidics, or
high-temperature environments (15) are examples of broadly emerging need for fabrication of microparts with exquisite structure-
and strength-dependent properties, on a par with size-dependent properties, in which material properties of silicon often do not meet.
Microstructures produced by photolithography have the limitations of low aspect ratio and quasi-3D structure (16,17). It is
possible to fabricate high-aspect-ratio components with submicron structure by Lithographie Galvanoformung und Abformung
(LIGA) process (from the German – a combination of lithography, electroplating, and molding process) using the synchrotron
radiation process and focused ion beam (FIB) machining process. However, present laboratory-scale and industrial fabrication
techniques using LIGA require special and extremely expensive facilities like a synchrotron system and require machining of
expensive masks which has imposed a hindrance on quick and economical fabrication of microparts. Furthermore, the dimensional
ranges that such processes cover are sometimes not required which is illustrated in Figure 3 (18). It can be observed from the

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Introduction to Advanced Machining Technologies 3

Figure 2 (a) Intermeshing gears and (b) mirror and drive systems for light steering produced by Si surface micromachining in Sandia National
Laboratories’ SUMMiT. Reproduced from Kim, Seong H., et al. Nanotribology and MEMS. Nano Today 2007, 2 (5), 22–29.

Figure 3 Illustration on dimensional range and aspect ratio of several machining processes. Reproduced from Asad, A. B. M. A. Micro-EDM Process
for Tool-based Compound Micromachining. Ph.D. Thesis, National University of Singapore, 2012.

illustration that tool-based micromachining has a unique place for performing micromachining operations at the lower boundary
of micromachining range (between 5 mm and 50 mm) to bridge the gap between mechanical machining, photolithography, and
LIGA process for dimensional range and aspect ratio (19,20).
Tool-based micromachining technologies, including microturning, microgrinding, microelectrical discharge machining (micro-
EDM) and microelectrochemical machining (micro-ECM), have many advantages in productivity, efficiency, flexibility, and cost
effectiveness and consequently have been applied to a variety of substrates and materials to fabricate microstructures and trans-
ducers in addition to etching and LIGA (4,7,8,21–23). In many applications the workpiece is a final product while in other
applications it might be the X-ray lithography mask or a mold for electroplating to create a microstructure. Among the tool-based
micromachining techniques, micro-EDM, a nonconventional machining process has been identified as a very promising technology
for the machining of microcomponents due to its noncontact machining capability. It involves almost negligible amount of force
interaction between the tool and workpiece and is capable of machining a wide range of conductive materials irrespective of
toughness. The other techniques like micromilling and microturning have the advantage of larger material removal rate (MRR) as
well as the capability to machine nonconductive materials.
Recently, a new concept of tool-based micromachining utilizing the complementary strengths of different material removal
processes (conventional and nonconventional) within the same setup, where possible, for the machining of microstructures with
high dimensional accuracy has been proposed (24). For example, material removal by micromilling process has many advantages
and it is fast as mentioned earlier. But it is limited by the lower limit of machinable feature size which is an order of magnitude larger
compared to micro-EDM due to the presence of larger cutting force. On the other hand, micro-EDM has the limitation of relatively
high tool wear to workpiece removal ratio due to high tool wear rate and low machining speed. Moreover, usually the electrode for
micro-EDM milling is prepared using other EDM-based techniques like electrodischarge grinding (EDG), which is also a slower

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
4 Introduction to Advanced Machining Technologies

• High accuracy motion • Understanding the

and guide system physics behind the
• Fine position control process
• Provides support for an • Identify controllable
array of factors for expected
micromachining outcome
Machine Process
processes
tool physics
research research

Innovative Intelligent
process control
technologies system

• Development of • On board Processing of

innovative process real-time data
• Employ compound and • Automated
hybrid machining measurement feedback
technologies and correction

Figure 4 Synergistic research and development areas for successful compound micromachining.

process. For instance, in the machining of a microplateau-shaped structure, the bulk amount of material can be removed using
conventional micromilling process followed by micro-EDM milling process which can achieve the dimensional accuracy and finer
range of feature size, and superior surface finish could be generated by applying micro-ECM process after micro-EDM. The electrode
too can be manufactured using microturning process which is also much faster compared to the EDG process.

11.01.2 Synergistic Development for Micro-/Nanoscale Machining

In order to achieve meaningful implementation of micro-/nanoscale machining techniques, primarily four important areas are
required to be addressed as has been illustrated in Figure 4. First, the challenges posed by the availability of machine tool suitable for
performing various tool-based machining processes at the dimensional range and accuracy required by micro-/nanoscale machining.
Second, research work needs to be undertaken towards the understanding of process physics to provide relevant background for
modeling, measurement, identification of control parameters, and application of feedback control in order to control compound
manufacturing processes. Third, intelligent system needs to be developed and implemented on board for processing real-time data,
which will enable the system to be capable of measurement, feedback, and correction of parameters without operator intervention. An
intelligent system from manufacturing perspective is defined as a system capable of taking in real-time data from process parameters/
environmental variables to differentiate between different scenarios/conditions; and is able to output necessary corrective signals
based on the input data. This may require learning process either supervised or not supervised; or could even be systems that are
dependent on models. Finally, the difficulties of microfabrication using available techniques need to be realized and then compound
processes need to be proposed to complement the weaknesses of different processes. Integrated research and development effort in
these areas are essential to continue supporting the proliferation of miniaturization and the quality of the technology.

11.01.3 Tool-Based Machining Processes for Micro-/Nanoscale Applications

Through the 1990s, Masuzawa highlighted the need for the development of micromachining processes (25) and drew attention
towards the need of tool-based micromachining. In his opinion, in precision machining it is important to recognize what directly
determines the shape of the products in the machining process and based on shape specification element (SSE) he categorized the
micromachining processes into two basic groups one having SSE as the tool and the other one having mask as the SSE. He drew
attention towards the fact that processes having mask as SSE are basically two-dimensional in nature and have limitations in
generating an actual three-dimensional shapes. He also emphasized that, even though very small unit removal (UR) requirement for
precision engineering is mainly catered by mask-based micromachining technology which could reach atomic level dimension, URs
of tool-based micromachining processes are acceptable in many practical applications; and they could even be preferred due to their
low setup and maintenance cost. Through his extraordinary research efforts, tool-based micromachining technologies received
significant attention and acceptance in the scientific community (5,22,25,26). His initial research work in collaboration with Masaki
et al. (27–29) and Sato et al. (30,31) from Matsushita Research Institute Tokyo, formed the foundation of micromachining with
a main focus on micro-EDM process. Table 1 shows some of Masuzawa’s pioneering research work and citation report according to

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Introduction to Advanced Machining Technologies 5

Table 1 Citation report of pioneering research work in tool-based micromachining

Title Year of publication Total citations Citations in the last 5 years

Wire electrodischarge grinding for micro-machining (Masuzawa – 1985) 1985 250 135
State-of-the-art of micromachining (Masuzawa – 2000) 2000 243 183
Micro-EDM for three-dimensional cavities – Development of uniform 1998 124 79
wear method (Yu – 1998)
Three-dimensional micromachining by machine tools (Masuzawa – 1997) 1997 113 62

Scopus – citation database (as of 2 June 2012) and growing research interest in this area is clearly visible through the number of
citations received in last five years.

11.01.3.1 Micromilling
Milling is one of the most universal operations of machining processes. Extensive research on physical characteristics of milling
process, tool failure analysis as well as process planning has been done (4). In the adoption of the milling process for micro-
machining applications, the amount of UR needs to be reduced (25). One issue in reducing UR for micromilling is that microcutting
occurs in a small region, which contains only a few layers of molecules and can be discrete in nature rather than continuous, as is
assumed in conventional continuum mechanics. Molecular dynamics-based simulation has been utilized to solve this problem and
several recent results demonstrate that the microcutting processes are capable of reducing the UR to as small as 1 nm (32–34). Recent
development in ultraprecision machine tool technology provided excellent positioning accuracy as well as repeatability, which are
significant steps forward for successful implementation of micromilling process in industrial applications for small UR.
Unfortunately, even though micromilling is a scale-down adoption of the milling process for micromachining applications and
has the theoretical capability of UR as small as 1 nm, practically there exist significant challenges in realizing micromilling, espe-
cially when the diameter of the end-mills are down to 0.1 mm or smaller. Owing to the comparatively smaller and weaker size of
micromilling cutters and the fact that only a tiny portion of the tool involved in cutting at a particular instance of the process, tool
failure due to tool run-out has been reported as one of the major problems in micro milling (5,31,35–39).
Recent research efforts include design of tool geometry to reduce cutting force in order to avoid tool breakage. Figure 5 shows
some of the tool shapes being researched (36) and an finite element method (FEM) on relative rigidity demonstrated that the (d)
and (e) types tools are 12 and 8 times, respectively, more rigid than the conventional (a) type tools of 100 mm diameter having two-
flutes. FEM studies demonstrated that three times larger cutting force is required for occurring chipping and breakage in (d) and (e)
type tools compared to (a) type tools. In microend-milling operation, the tool run-out to tool diameter ratio becomes very big
compared to conventional end-milling operation. As a result, only one side of the tool performs the machining operation at any
point of time while the other edge does not touch the workpiece at all and this generates significant amount of directional force
variation on the tool. With such a small tool dimension, even if the cutting force is as small as few tens of mN, chipping and
breakage of the tool tip occur.
Another major concern is the limited availability of micromilling tools below 50 mm diameter. Even though commercially
25 mm milling tools are available (40) at present, high tool cost, unreliable tool life, and early tool failure can damage the workpiece;
thus could be rendered unacceptable for precision applications. Schaller and colleagues (41) have demonstrated that self-made

Figure 5 Various types of end-mills in micromachining. (a) Two-flute end-mills, (b) D-type end-mills with a straight body, (c) D-type end-mills with
a straight body, (d) D-type end-mills with a tapered body, and (e) D-type end-mills with a tapered body. Reproduced from Fang, F. Z., et al. Tool
Geometry Study in Micromachining. J. Micromech. Microeng. 2003, 13, 726–731.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
6 Introduction to Advanced Machining Technologies

Figure 6 Microend-mills made by focused ion beam sputtering having 2 (a), 4 (b), and 6 (c) cutting edges. Reproduced from Adams, D. P.,
et al. Micromilling of Metal Alloys with Focused Ion Beam Fabricated Tools. Precis. Eng. 2001, 25, 107–113.

end-mills with diameters ranging from 35 to 120 mm could be ground to cut microstructure grooves. Another rather common
custom-fabricated micromilling tool machining process besides mechanical method is to employ focused ion beam FIB machining
technique, as this process has gained popularity for machining of very small probe tips for biomedical use and microscalpels with
extremely sharp cutting edges (37,42,43). It has been demonstrated that FIB could be used to shape a variety of cutting tools with
diameter ranging from 15 to 100 mm with 40 nm cutting edge radius using a wide range of tool materials like single crystal
diamond, tungsten carbide and high speed steel (44,45). Figure 6 (44) shows such tools with 2, 4, and 6 cutting facets fabricated
using FIB. However, a worthwhile point to note, FIB process is capable of fabricating very fine custom micromilling tool – the
utilization of FIB technology demands a vacuum background and thus involves more expensive and complicated equipment, setup,
and operating cost. In addition to that, FIB process could not be performed in situ and thus it does not eliminate clamping error.

11.01.3.2 Microturning
The present state of microturning process is very similar to conventional turning process on a lathe that has been extended to
provide better precision and accuracy in machining process. Similar to micromilling process, microturning has the capability to
produce 3D structures on microscale (19). There are two major types of microturning – cylindrical shaft turning for machining of
micropins and face turning for machining of microgrooves, that has been reported for micromachining applications (19,46).
Microturning for machining of micropins is possible but it is more difficult to realize due to the deformation of the fine workpiece
which is very similar to the deflection of microend-milling cutter as could be seen in Figure 7(a). However, the situation is even
more critical for microturning as often the microturning workpiece is much weaker than the tool in micromilling (5) and thus the
major drawback of microturning process is that the machining force influences machining accuracy and the limit of machinable
size. Significant work has been done to develop different cutting paths and schemes to reduce the effect of cutting force on the fine
shaft. A micropin of around 350 mm diameter (Figure 7(b)) with intricate shape and kinks has been fabricated (7,19). But, it is very
difficult to achieve a straight shaft below 100 mm diameter and in many cases, the workpiece is either broken, or starts to wobble due
to excessive radial cutting force on the microshaft. Figure 8 shows one such microshaft machined using the conventional

Figure 7 (a) Workpiece deflection in microturning, (b) scanning electron microscopy image of compound shaped micropin. Reproduced from
Rahman, A., et al. Development of Micropin Fabrication Process Using Tool Based Micromachining. Int. J. Adv. Manuf. Technol. 2006, 27 (9–10), 939–944.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Introduction to Advanced Machining Technologies 7

Figure 8 A microlapping tool made by microturning. Reproduced from Masuzawa, T. State of the Art of Micromachining. CIRP Ann. 2000, 49/2, 473–488.

microturning process. The shaft was deformed plastically with very rough surface finish from plastic side flow caused by the strain
gradient-induced strengthening due to the constant radial force during turning at a slower feed rate (47). When a faster feed rate is
applied, the shaft breaks easily as the radial force increases to an excessive level.
In face turning, research has been conducted for many years on diamond turning and this has found wide applications in
machining of various components such as microlenses, lens arrays, and parts for measurement references, for example, surface
encoder for multiaxis position accuracy measurement (46,48,49). Diamond turning has been generally coupled with the term
ultraprecision machining as single point diamond turning is probably one of the few processes achieving mirror surface finish.
Finish of less than 10 nm and form error of less than 1 mm can be obtained when machined using an ultraprecision machine
tool capable of moving in high accuracy at nanometeric precision (14). It is also possible to fabricate microparts using
conventional ultraprecision turning. After cutting microsteps on the surface of a plate, microparts can be cut out by other
methods such as wire electrical discharge machining (WEDM) as shown in Figure 8 (5,48). Another important area of
microturning is to support microgrooving and microthreading needs required for the fabrication of microfluidic sensors,
microinductors, and microactuators. However, the major difficulty is in availability of such tools. Literature suggests that FIB can
be a potential technique for machining of such tools (43,50).

11.01.3.3 Mechanical Microdrilling

Microdrilling has been widely used in various applications such as printed circuit boards (PCBs), ink-jet printer nozzles in semi-
conductor industry, orifices for biomedical devices, cooling vents for gas turbine blades, and diesel fuel injector spray holes. There
also exist few parallel machining techniques of microholes, for instance, micro-EDM, micro-ECM, laser ablation in addition to
mechanical microdrilling. Mechanical microdrilling has several advantages over other techniques (5,51). In mechanical micro-
drilling, the electrical properties of the workpiece do not influence the process and therefore, most metal and plastics, including their
composites, can be machined easily. One typical example is the drilling of holes in PCB, which is laminated with composite
material consisting of copper foils, resin, and glass fiber cloth (52). Another advantage is unlike stochastic processes such as micro-
EDM, machining time in mechanical microdrilling can be controlled easily because the process is stable when an appropriate feed
rate per rotation is set.
On the other hand, mechanical microdrilling faces a lot of challenges as the high-aspect-ratio microdrill bits are fairly weak and
should be used with care in order to drill holes accurately and to prevent drill breakage. The drill point is the most important part
of the drill, which penetrates into the material of the workpiece during the machining process. The geometry of the drill tip is such
that the normal rake and clearance angles and velocity of the cutting edge vary with the distance from the center of the drill. Even
small variations in the geometry of symmetry errors can have a very strong influence on the performance of the drill. For example,
if the tip relief angle is too small, excessive heat is generated resulting in an increased wear rate. Conversely, too large angle can
cause chipping or breaking of the cutting edge. Thus one very important requirement for successful microdrilling is to minimize
tool run-out and clamping error. Another important requirement is the straightness of the product and the axes of the machine
tool (5). Machined holes are often inclined because the already-machined part of the hole influences on the orientation of the
drill. In order to avoid inclination, correct positioning is necessary when the drill tip begins to cut the workpiece. If the tip position
shifts by even a small distance from the target center of the hole, the drill bends to follow a certain angle guided by the hole, which
the drill itself produces. Also there are difficulties in machining very hard or brittle materials as the drill bit may break or crack
might be generated in the workpiece. The above-mentioned issues seriously affect the reliability and accuracy of the drilled holes.
Fang et al. (4,53) have reported a microhole drilling on brass using high speed steel drill with a diameter of 50 mm and due to the
weak tool tip and off-set between the drill center and the rotational center, the fabricated hole has a diameter of 85 mm which is
much bigger than the drill diameter.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
8 Introduction to Advanced Machining Technologies

11.01.3.4 Microgrinding
Grinding has been widely applied for machining pins and grooves with small dimensions due to the fact that UR of grinding is small
and cutting is realized by the interaction of micrometer-sized abrasive grains with the micrograins of workpiece surface. With
appropriate microgrinding tools, machining of 2D or 3D microcavities in a system similar to mechanical- or EDM-milling is feasible
and the advantage of this process is the capability of machining brittle and nonconductive materials which the former processes are
not capable of. Another potential advantage microgrinding can have is the ability to produce mirror surface on microcomponents.
However, in the field of micromachining, one of the technological problems in trying to realize microgrinding is the fact that the
tool must be made up of an abrasive and a matrix. When the tool size is very small, the grain size cannot be ignored, and this leads to
certain difficulties in forming the precise shape of the grinding tool. Very small grain sizes are essential for micromachining and the
tool too needs to be small. On the other hand, another difficulty is posed by the fact that the cutting force in grinding process is
considerably high which can easily damage the microsized features in micromachining.
Onikura et al. (54) proposed ultrasonic vibration grinding to reduce this limitation in fabricating microcylindrical tools and flat
microdrills of ultrafine grain cemented carbides. By adding ultrasonic vibration to the grinding process, they were able to produce
high-aspect-ratio tools such as an 11 mm diameter with a length of 160 mm. Ohmori et al. (55) fabricated microtools using
microgrinding, and investigated the surface quality of the tools since the surface quality is closely related to machining performance,
part quality, and tool rupture strength. They claimed that surface characteristics of the prepared microtools were controllable at the
nanometer level by controlling and optimizing the machining process conditions. They produced a microtool having an extremely
precise shape, measuring 2 mm in diameter at the tip with an extremely large aspect ratio.

11.01.3.5 Micro-EDM
Micro-EDM is a material removal process employing discharges between a workpiece and a microscale electrode in a dielectric fluid.
Discharges occur when the electric field between the electrode and workpiece exceeds a critical value and the dielectric breaks down.
Either increasing the electric potential or reducing the separation distance between the electrode and workpiece may cause the field
to exceed the critical value. Energy from each discharge melts a microscopic amount of material, which is subsequently flushed away
after the voltage drops and the discharge collapses (56–58).
Even though micro-EDM is based on the same physical principle of spark erosion, it is not merely an adoption of EDM process
for machining at micron level. There are significant differences in the size of the tool used, machining method of microsized tools,
the power supply of discharge energy, movement resolution of machine tool’s axes, gap control, flushing techniques, and also in the
processing techniques (5,59). For example, terms such as micro-EDM milling, WEDG, and repetitive pattern transfer are unique to
micro-EDM process and commonly not associated.
The literature has shown significant contributions to micro-EDM going back 40 years. In 1968, Kurafuji and Masuzawa (60)
demonstrated the first application of micro-EDM. Through the years, micro-EDM has been developed into a versatile tool for
fabricating a variety of micromechanical components, molds for plastic injection molding, sensors, micropumps, micronozzles,
microgrippers (6,15,22,27,61–64). Micro-EDM is suitable for these and similar applications because of its remarkable advantage of
low machining force as molten or vaporized material can be removed without direct contact. This property provides advantages to
both the tool and the workpiece as probable deformation by machining force is avoided. Another very important advantage of the
micro-EDM process is the capability of repetitive pattern transfer, which is illustrated in Figure 9. The repetitive pattern transfer
process, which Masaki et al. (65) called as micro-EDMn, is capable of fabricating very complex microstructures by series of pattern
transfer cycles.
Current micro-EDM technologies used for manufacturing microfeatures can be categorized into four major types (58): (1)
microwire EDM, where a wire of diameter down to 0.02 mm is used to cut through a conductive workpiece, (2) die-sinking micro-
EDM, where an electrode with microfeatures is employed to produce its inverted image in the workpiece, (3) micro-EDM drilling,
where microelectrodes of diameters down to 5 mm are used to drill microholes in the workpiece, and (4) micro-EDM milling,
where microelectrodes are employed to produce 3D cavities by adopting a movement strategy similar to that in milling. There
exists another important variant of the micro-EDM process practically very similar to that of WEDM with apparent grindinglike
setup and is known as EDG. Masuzawa et al. (26) were the first to propose a variant of EDG using running wire (WEDG).
The workpiece electrode is machined by feeding downwards against a traveling sacrificial wire. This process has been extended
to the use of sacrificial block and sacrificial disk for EDG process (57,66,67) and has found extensive applications in tool
fabrication (68).
Despite the number of publications extolling the improved capabilities of micro-EDM, they are still not widely used and
industrial acceptance of micro-EDM is considerably slow given its immense potential. This is mainly due to the fact that available
machine tools and process characteristics are still not sufficiently reliable. Until recently, micro-EDM has tended to be performed
using conventional EDM machines modified to accommodate the micromachining requirements (58) and due to this lack of
focused development for micro-EDM process, in addition to its stochastic nature, there exist significant number of challenges which
are summarized in Figure 10. Among the many problematic areas, the major problem is posed by the micro-EDM process char-
acteristics and the interaction between the power supply and spark characteristics as this determines the process parameters.
Therefore, most of the micro-EDM process parameters are obtained by experimental methods and because of the stochastic thermal
nature of micro-EDM process it is difficult to explain the effects of process parameters well (58).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Introduction to Advanced Machining Technologies 9

Figure 9 (a) Procedure of micro-EDMn, (b) multielectrode, 14.3-mm diameter material – CuW, (c) micropin mold, material – WC, (d) microtaper pin
mold, material – STAVAX, (e) close-up of (d). Reproduced from Masaki, T., et al. Repetitive Pattern Transfer Process of Micro EDM. Int. J. Electro
Mach. 2006, 11, 33–34.

Electrodes
Handling
Parts

Drilling
threading
holes (WEDM)
Off-machine
preparation
Manufacturing
Electrode micro 3D
and electrode
workpiece
preparation Stationary
block

On-machine Rotating disk

preparation
Guided
Micro-EDM running wire
problematic
areas
Machine

Process Electrode
parameters Uniform
wear
Sources of Jigs and
method
erros fixtures
Multi
Electrode
Machining electrode
wear and
process machining
Z Level
strategies
compensation
Path
generation Wear
monitoing
CAM
Knowledge system
Support
based
Surface system
quality
Measurement
Dimensions

Figure 10 Problematic areas in micro-EDM. Adapted from Pham, D. T., et al. Micro-EDM – Recent Developments and Research Issues. J. Mater.
Process. Technol. 2004, 149, 50–57.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
10 Introduction to Advanced Machining Technologies

Among the problematic areas, micro-EDM process related issues are inherent to the process itself along with the advantages of
micro-EDM and thus are practically impossible to eliminate with the available technology and process knowledge. High electrode
wear and low MRR are two such major challenges. Electrode wear, which results from each discharge removing some material from
the tool electrode, degrades the geometric accuracy of machined features. The relative electrode wear ratio is not constant across
different workpiece materials and different geometric features. Due to this issue, it can be very difficult to provide computer-aided
manufacturing (CAM) support for micro-EDM process. However, this effect can be minimized when making micropockets with the
uniform wear method, presented by Yu et al. (22) but this method further compromise the MRR. On the other hand, due to rather
comparatively larger size of the smallest possible UR – even though there is almost no cutting force involved in machining there is
an early limit to minimally achievable feature size (69,70).

11.01.4 Compound Micromachining – A Recent Frontier in the Horizon

Even though compound micromachining could be traced back to 1985 when Masuzawa et al. (26) first proposed the WEDG
process, tool-based compound micromachining (TCMMP) did not receive much attention until recently other than in the form of
WEDG process and therefore TCMMP is considered rather new development in micromachining area. There exist many research
efforts in compound micromachining where two processes were combined to increase production efficiency or production
quality (71–74) but not much research efforts were taken to transfer those ideas into the micromachining domain. Compound
micromachining is yet to be formally defined by the scientific community due the process being rather new and search of databases
returns less than 30 unique articles in total (this is excluding compound process of WEDG followed by micro-EDM drilling/micro-
EDM milling). For defining the scope of this research work, discussion on TCMMP will be limited to processes which are synergism

Figure 11 (a) The concept of machining on XY and YZ plane using a spherical PCD tool on a three-axis machine, (b) concept of freeform surface
generation, (c) on-machine fabricated spherical PCD tool, and (d) machined sample of a shaped convex spherical surface on tungsten carbide.
Reproduced from Masaki, T., et al. Study on Shaping Spherical Poly Crystalline Diamond Tool by Micro-Electro-Discharge Machining and Micro-Grinding
with the Tool. Int. J. Surf. Sci. Eng. 2007, 1/4, 344–359.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Introduction to Advanced Machining Technologies 11

of at least more than one constituent processes where all the processes are involved in material removal either at the same time (e.g.,
EDG) or sequentially (e.g., micro-EDM followed by laser ablation). The material removal requires to be either from the final
workpiece or via an intermediate workpiece but are performed in a single setup for machining of feature size ranging from 1 mm to
500 mm. Following are some of the application areas of TCMMP that have recently started gaining extensive interest:
l Electrode machining for micro-EDM by TCMMP
l Microcutting tool machining using EDG
l TCMMP for microgrinding
l Surface improvement using TCMMP
l On-machine microassembly after micromachining

Even though TCMMP will be addressed in full depth in a separate chapter of this book, in order to provide the reader
with an overview of TCMMP two selected examples are presented below. The first application, proposed by Masaki et al. in
2007 (75), provided an innovative approach of using poly crystalline diamond (PCD) to accomplish microshape grinding of
microfreeform surfaces. They fabricated a spherical PCD tool by EDG with a pin gauge tool electrode made of tungsten
carbide that was manufactured precisely by controlling its diameter and straightness. Using this spherical PCD tool fabricated
on-machine they performed a series of micromachining of various shapes from flat, concave, convex to freeform machining
and achieved mirror surface finish on tungsten carbide with surface roughness of 5 nm Ra. On conventional machining center
– ball end-mills are used for milling of a variety of complex shapes. A ball end-mill has a normal hemisphere and therefore
four or five axis control is necessary to realize the high degree of freeform shaping and convex and concave shapes
machining. They demonstrated that the micro-EDMed spherical PCD tool, which has innumerable cutting edges uniformly
located along its entire surface, can be used for machining orthogonal microfreeform shapes on the XY and YZ planes using
a three-axis machining platform (Figure 11(a)). Figure 11(b) shows the concept of freeform machining using the on-
machine fabricated spherical PCD tool (Figure 11(c)) and image of a shaped convex spherical surface on tungsten
carbide is shown in Figure 11(d). The second application selected for presenting as an introduction has been proposed by
Masuzawa et al. (63) who proposed a compound process of micro-EDM, micro-ECM and electroforming for complex high-
aspect-ratio micronozzle production (Figure 12), and demonstrated that this compound process can be used to fabricate

Figure 12 Processes involved in the fabrication of micronozzle. Reproduced from Masuzawa, T., et al. A Combined Electrical Machining Process
for Micronozzle Fabrication. CIRP Ann. 1994, 43/1, 189–192.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
12 Introduction to Advanced Machining Technologies

a wider choice of complex shapes, instead of just straight rods, such as tapered and stepped shapes at various sections. In
their study, they also proposed another TCMMP process, wire electrochemical grinding, which does not use any abrasive, as
in the normal chemical grinding process, but is an electrochemical process with a setup similar to WEDG, by simply
replacing electric discharge with electrochemical dissolution to provide better surface quality.

11.01.5 Summary

The exuberant demand for miniaturization, machining accuracy, choice of materials and throughput posed a great challenge ahead
for manufacturing engineers. Tool-based micromachining techniques have been gaining popularity as an alternative machining
process parallel to processes based on semiconductor machining techniques primarily due to their ability of machining 3D
structures on wide range of hard materials. Besides the conventional micromachining processes, nonconventional processes such as
micro-EDM process has the advantage of machining with minimal cutting force but the main disadvantage is in the UR capability in
addition to electrode wear and low material removal rate.
Even though tool-based conventional micromachining processes have several advantages over the present day semiconductor
machining technologies for micromachining, they presently face significant challenges at such miniaturized dimensions. One of the
main challenges has been posed by the cutting forces involved and therefore the full potential of having the theoretical UR
capability of as small as 1 nm found from molecular dynamics simulation could not be practically realized (5). On the contrary, the
process characteristics of electrical machining processes are heavily stochastic in nature and the interaction between power supply
and the discharge characteristics has not been studied enough to elucidate the process parameter selection process. A compound
process involves a combination of processes of complementary strengths, which has a lot of potential for addressing present day
manufacturing need.

References

1. Rahman, M.; Asad, A. B. M. A.; Masaki, Takeshi; Wong, Y. S.; Lim, H. S. A Multi-Process Machine Tool for Compound Micromachining. Int. J. Mach. Tools Manuf. 2010, 50/4,
344–356.
2. Corbett, P. A., et al. Nanotechnology: International Developments and Emerging Products. CIRP Ann. 2000, 49, 523–546.
3. Dornfeld, D., et al. Recent Advances in Mechanical Micromachining. CIRP Ann. 2006, 55/2, 745–768.
4. Fang, F. Z., et al. Tool-based Micromachining and Applications in MEMS, MEMS/NEMS Handbook Techniques and Applications. In Manufacturing Methods; Springer: US,
2006, Vol. 3; pp 678–740.
5. Masuzawa, T. State of the Art of Micromachining. CIRP Ann. 2000, 49/2, 473–488.
6. Meeusen, W., et al. Micro-Electro-Discharge Machining as Microsensor Fabrication Technology. IEEE Sens. J. 2003, 3/5, 632–639.
7. Rahman, A., et al. Development of Micropin Fabrication Process Using Tool Based Micromachining. Int. J. Adv. Manuf. Technol. 2006, 27 (9–10), 939–944.
8. Schoth, A., et al. Micro Wire EDM for High Aspect Ratio 3D Microstructuring of Ceramics and Metals. Microsyst. Technol. 2005, 11, 250–253.
9. Lang, W. Reflexions on the Future of Microsystems. Sens. Actuators 1999, 72, 1–15.
10. Madou, M. J. Fundamentals of Microfabrication; CRC Press: Boca Raton, 1997.
11. Weck, M., et al. Fabrication of Micro Components Using Ultra Precision Machine Tools. Nanotechnology 1997, 8, 145–148.
12. Xuan, F.; Zha., et al. Web-Enabled Database System Development for Design and Manufacturing of Micro-Electro-Mechanical Systems (MEMS), MEMS/NEMS Handbook
Techniques and Applications. In Design Methods; Springer: US, 2006, Vol. 1; pp 73–109.
13. Kim, Seong H., et al. Nanotribology and MEMS. Nano Today 2007, 2 (5), 22–29.
14. Rahman, M., et al. Tool-based Nano Finishing and Micromachining. J. Mater. Process. Technol. 2007, 185 (1–3), 2–16.
15. Kuo, C. L., et al. Fabrication of 3D Metal Microstructures Using a Hybrid Process of Micro-EDM and Laser Assembly. Int. J. Adv. Manuf. Technol. 2003, 21, 796–800.
16. Okuyama, H., et al. Micromachining with SR and FEL. Nucl. Instrum. Meth. Phys. Res. B 1998, 144, 58–65.
17. Rajurkar., et al. 3D Micro-EDM Using CAD/CAM. CIRP Ann. 2000, 49/1, 127–130.
18. Asad, A. B. M. A. Micro-EDM Process for Tool-based Compound Micromachining. Ph.D. Thesis, National University of Singapore, 2012.
19. Rahman, A., et al. CNC Microturning: an Application to Miniaturization. Int. J. Mach. Tools Manuf. 2005, 45, 631–639.
20. Rajurkar, K. P., et al. Micro and Nano Machining by Electro-Physical and Chemical Processes. CIRP Ann. 2006, 55 (2), 643–666.
21. Gianchandani., et al. A Micromachining Process for Die-Scale Pattern Transfers in Ceramics and Its Application to Bulk Piezoelectric Actuators. J. Microelectromech. Syst.
2006, 15 (3), 605–612.
22. Yu, Z. Y., et al. Micro-EDM for Three-Dimensional Cavities – Development of Uniform Wear Method. CIRP Ann. 1998, 47/1, 169–172.
23. Zhao, W., et al. A CAD/CAM System for Micro-ED-Milling of Small 3D Freeform Cavity. J. Mater. Process. Technol. 2004, 149, 573–578.
24. Rahman, M., et al. Integrated Hybrid Micro/Nano-Machining. In ASME Conference on Proceedings of the 2007 International Manufacturing Science and Engineering Conference
(MSEC 2007 – 31009), Georgia, Atlanta, USA, 2007.
25. Masuzawa, T., et al. Three-Dimensional Micromachining by Machine Tools. CIRP Ann. 1997, 46/2, 621–628.
26. Masuzawa, T., et al. Wire Electro-Discharge Grinding for Micro-Machining. CIRP Ann. 1985, 34/1, 431–434.
27. Masaki, T., et al. Electric Discharge Machining Method and Apparatus for Machining a Microshaft. U.S. Patent Number – 4,900,890, 1990.
28. Masaki, T., et al. Micro Electro-Discharge Machining and Its Applications. Micro Electro Mech. Syst. IEEE; Napa Valley, CA, 1990, 21–26.
29. Masaki, T., et al. Micro Electro-Discharge Machining. In Proceedings of International Symposium for Electro-Machining; 1989, pp 26–29.
30. Sato, T., et al. Electro-Discharge Machine for Micro Hole Boring. Natl. Tech. Rep. (Japan) Oct 1985, 81, 105–113.
31. Sato, T., et al. Application of WEDG for Microdrilling and Microendmilling. In Proceedings of Annual Spring Assembly of JSPE; pp 1091–1092 (in Japanese).
32. Shimada, S. Molecular Dynamics Simulation of the Atomic Processes in Microcutting, Micromachining of Engineering Materials; CRC Press, 2001, 63–83.
33. Shimada, S., et al. Feasibility Study on Ultimate Accuracy in Microcutting Using Molecular Dynamics Simulation. CIRP Ann. 1993, 42/1, 91–94.
34. Ikawa, N., et al. Minimum Thickness of Cut in Micromachining. Nanotechnology 1992, 3, 6–9.
35. Bao, W. Y., et al. Modeling Micro-End-Milling Operations. Part – II: Tool Run-Out. Int. J. Mach. Tools Manuf. 2000, 40, 2175–2192.
36. Fang, F. Z., et al. Tool Geometry Study in Micromachining. J. Micromech. Microeng. 2003, 13, 726–731.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Introduction to Advanced Machining Technologies 13

37. Friedrich, C. R., et al. Micromilling Development and Applications for Microfabrication. Microelectron. Eng. 1997, 35, 367–372.
38. Tansel, I., et al. Micro-End-Milling – I. Int. J. Mach. Tools Manuf. 1998, 38, 1419–1436.
39. Zaman, M. T., et al. A Three-Dimensional Analytical Cutting Force Model for Micro End Milling Operation. Int. J. Mach. Tools Manuf. 2006, 46, 353–366.
40. Chern, G. L., et al. Study on Burr Formation in Micro-Machining Using Micro-Tools Fabricated by Micro-EDM. Precis. Eng. 2007, 31, 122–129.
41. Schaller, T., et al. Microstructure Grooves with a Width of Less Than 50m Cut with Ground Hard Metal Micro End Mills. Precis. Eng. 1999, 23, 229–235.
42. Friedrich, C. R. Micromechanical Machining of High Aspect Ratio Prototypes. Microsyst. Technol. 2002, 8, 343–347.
43. Vasile, M. J., et al. Micrometer-Scale Machining: Tool Fabrication and Initial Results. Precis. Eng. 1996, 19 (2–3), 180–186.
44. Adams, D. P., et al. Micromilling of Metal Alloys with Focused Ion Beam Fabricated Tools. Precis. Eng. 2001, 25, 107–113.
45. Picard, Y. N., et al. Focused Ion Beam-Shaped Microtools for Ultra-Precision Machining of Cylindrical Components. Precis. Eng. 2003, 27, 59–69.
46. Gao, W., et al. Precision Nano-fabrication and Evaluation of a Large Area Sinusoidal Grid Surface for a Surface Encoder. Precis. Eng. 2003, 27, 289–298.
47. Liu, K., et al. Effect of Plastic Side Flow on Surface Roughness in Micro-Turning Process. Int. J. Mach. Tools Manuf. 2006, 46, 1778–1785.
48. Brinksmeier, E., et al. Machining of Precision Parts and Microstructures. In Initiatives of Precision Engineering at the Beginning of a Millennium, 10th International Conference
on Precision Engineering (ICPE); July 18–20, 2001. Yokohama, Japan.
49. Pramanik, A., et al. Cutting Performance of Diamond Tools during Ultra-Precision Turning of Electroless-Nickel Plated Die Materials. J. Mater. Process. Technol. 2003, 140,
308–313.
50. Adams, D. P., et al. Microgrooving and Microthreading Tools for Fabricating Curvilinear Features. J. Int. Soc. Precis. Eng. Nanotechnol. 2000, 24, 347–356.
51. Chyan, H. C., et al. Development of Curved Helical Micro-Drill Point Technology for Micro-Hole Drilling. Mechatronics 1998, 8, 337–358.
52. Watanabe, H., et al. Microdrilling for Printed Circuit Boards (PCBs) - Influence of Radial Run-Out of Microdrills on Hole Quality. Precis. Eng. 2008, 32, 329–335.
53. Yang, Z., et al. Principle of Precision Micro-Drilling with Axial Vibration of Low Frequency. Int. J. Prod. Res. 2002, 40/6, 1421–1427.
54. Onikura, H., et al. Fabrication of Micro Carbide Tools by Ultrasonic Vibration Grinding. CIRP Ann. 2000, 49/1, 257–260.
55. Ohmori, H., et al. Improvement of Mechanical Strength of Micro Tools by Controlling Surface Characteristics. CIRP Ann. 2003, 52/1, 467–470.
56. Ho, K. H., et al. State of the Art Electrical Discharge Machining (EDM). Int. J. Mach. Tools Manuf. 2003, 43, 1287–1300.
57. Alting, L., et al. Micro Engineering. CIRP Ann. 2003, 52/2, 635–658.
58. Pham, D. T., et al. Micro-EDM - Recent Developments and Research Issues. J. Mater. Process. Technol. 2004, 149, 50–57.
59. Wong, Y. S. Investigation of Micro-EDM Material Removal Characteristics Using Single RC-Pulse Discharges. J. Mater. Process. Technol. 2003, 140, 303–307.
60. Kurafuji, H.; Masuzawa, T. Micro-EDM of Cemented Carbide Alloys. Jpn. Soc. Electr. Mach. Eng. 1968, 2/3, 1–16.
61. Ansel, Y., et al. Development of Tools for Handling and Assembling Microcomponents. J. Micromech. Microeng. 2002, 12, 430–437.
62. Kuo, C. L., et al. Fabrication of Series-Pattern Micro-Disk Electrode and Its Application in Machining Micro-Slit of Less Than 10 mm. Int. J. Mach. Tools Manuf. 2004, 44,
545–553.
63. Masuzawa, T., et al. A Combined Electrical Machining Process for Micronozzle Fabrication. CIRP Ann. 1994, 43/1, 189–192.
64. Michel, F., et al. EDM for Micro Fabrication-Technology and Applications. In Proceedings of International Seminar on Precision Engineering and Micro Technology (Aachen),
2000, pp 127–139.
65. Masaki, T., et al. Repetitive Pattern Transfer Process of Micro EDM. Int. J. Electro Mach. 2006, 11, 33–34.
66. Lim, H. S., et al. A Study on the Machining of High-Aspect-Ratio Micro-Structures Using Micro EDM. J. Mater. Process. Technol. 2003, 140, 318–325.
67. Ravi, N., et al. The Effects of Electro-Discharge Machining Block Electrode Method for Microelectrode Machining. J. Micromech. Microeng. 2002, 12, 532–535.
68. Morgan, J. C., et al. Micro-Machining and Micro-Grinding with Tools Fabricated by Micro Electro-Discharge Machining. Int. J. Nanomanuf. 2006, 1/2, 242–258.
69. Kawakami, T., et al. Study on Factors Determining Limits of Minimum Machinable Size in Micro EDM. CIRP Ann. 2005, 54/1, 167–170.
70. Kunieda, M., et al. Advancing EDM through Fundamental Insight into the Process. CIRP Ann. 2005, 54/2, 599–622.
71. Aspinwall, D. K. Hybrid High Speed Machining (HSM): System Design and Experimental Results for Grinding/HSM and EDM/HSM. CIRP Ann. 2001, 50/1, 145–148.
72. Kitagawa, T., et al. Plasma Hot Machining for New Engineering Materials. Wear 1990, 139, 251–267.
73. Jia, Z. X., et al. Study on a New Kind of Combined Machining Technology of Ultrasonic Machining and Electrical Discharge Machining. Int. J. Mach. Tools Manuf. 1997, 37/2,
193–199.
74. Koshy, P., et al. Mechanism of Material Removal in Electrical Discharge Diamond Grinding. Int. J. Mach. Tools Manuf. 1996, 36/10, 1173–1185.
75. Masaki, T., et al. Study on Shaping Spherical Poly Crystalline Diamond Tool by Micro-Electro-Discharge Machining and Micro-Grinding with the Tool. Int. J. Surf. Sci. Eng. 2007,
1/4, 344–359.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
This page intentionally left blank

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 11.02 Machine Tools for Micro- and Nanometer Scale Processing
H Shinno, Tokyo Institute of Technology, Yokohama, Japan
 2014 Elsevier Ltd. All rights reserved.

11.02.1 Introduction 15
11.02.2 Industrial Demands on Machine Tools for Micro- and Nanometer Scale Processing 15
11.02.3 Structural Design of Machine Tools for Micro- and Nanometer Scale Processing 16
11.02.4 Fundamental Structural Modules in Machine Tool for Micro- and Nanometer Scale Processing 19
11.02.4.1 Main Spindle System for Micro- and Nanometer Scale Processing 19
11.02.4.2 X–Y Planar Nanomotion System 21
11.02.4.3 Vertical Nanomotion System 22
11.02.4.4 Advanced Nanomotion Control System 22
11.02.5 Future Evolutionary Trends of Machine Tool Technologies for Micro- and Nanometer Scale Processing 24
11.02.5.1 Micro- and Nanometer Scale Machining Function 24
11.02.5.2 Establishment of New Structural Design Methodology 24
11.02.5.3 Machine Tool Equipped with Multiple System Functions 25
11.02.5.4 Intelligent and Autonomous Machine Tool 25
11.02.6 Conclusions 25
References 25

11.02.1 Introduction

Demands for micro- and nanometer scale processing have recently increased in a wide range of industries and sciences such as
aerospace, energy, electronics, automotive, and astronomy. Typical applications for micro- and nanometer scale processing include
various precision structured surfaces, i.e., sliding bearing surfaces with fine pattern, heat transfer devices with micro- and nanometer
scale channels, and radiation devices with micro- and nanometer scale three-dimensional (3D) free form surface (1). In order to
rationally meet such processing requirements, ultraprecision machine tools for micro- and nanometer scale processing have been
developed by the leading machine tool manufacturers and in the advanced research institutes over the world so far. Some
commercially available ultraprecision machine tools, i.e., ultraprecision diamond turning machines, ultraprecision grinding
machines, ultraprecision milling machines, and ultraprecision polishing machines, have already provided machining capability of
nanometer scale resolution.
Future machine tools for micro- and nanometer scale processing require machining capability of subnanometer order resolu-
tion (2). In order to surely realize such an extreme machining process, it is necessary and indispensable to remove the existing error
factors from the overall machine tool structure, and then to develop an ideal machine tool structure. In addition, such innovative
machine tools should be constructed by the innovative fundamental structural components and modules for micro- and nanometer
scale processing, i.e., an ultraprecision high-speed spindle system, a nanomotion table system, a vertical nanomotion mechanism,
and a precise tilting motion platform.
This chapter describes a desirable machine tool structure for micro- and nanometer scale processing and its fundamental
structural components and modules. The overall machine tool and its structural modules described here are fundamentally based
on a new design concept. In addition, the overall machine tool structure is designed so as to remove all existing error sources from
the objective machine tool structure and to minimize the influence of error factors and error sources. An actual machine tool
structure for micro- and nanometer scale processing and its structural modules are described in detail. Furthermore, future
evolutionary trends of machine tool technologies for micro- and nanometer scale processing are finally summarized and some
problems to be solved are described at the end of the chapter.

11.02.2 Industrial Demands on Machine Tools for Micro- and Nanometer Scale Processing

Demands for generating 3D micro- and nanometer scale geometries over a relatively large working space, as shown in Figure 1, have
remarkably increased in a wide range of industrial sectors, i.e., energy, environment, aerospace, semiconductor, optics, medical
instruments, etc. Actually, such applications of micro- and nanometer scale processing range from automotive and aerospace to
astronomy, medical, optics, and metrology. In order to meet such advanced and complicated industrial requirements, it is necessary
and indispensable to develop the machine tools for micro- and nanometer scale processing. Various kinds of ultraprecision machine
tools have been developed so far (3–5), while many ultraprecision machine tools cannot fully satisfy the present severe industrial
requirements due to the limited thermally and dynamically structural stability, the limited work space, etc. Machine tools for micro-
and nanometer scale processing can be defined as ultraprecision machine tool systems for performing micrometer and nanometer
scale machining processes.

Comprehensive Materials Processing, Volume 11 http://dx.doi.org/10.1016/B978-0-08-096532-1.01103-1 15

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
16 Machine Tools for Micro- and Nanometer Scale Processing

Figure 1 A machining space for micro- and nanometer scale processing.

Figure 2 Demands for micro- and nanometer scale processing and machine tool structure.

Figure 2 shows the coming future trends of micro- and nanometer scale processing procedures and the desirable machine tool
structures for meeting the future trends. In order to rationally meet such trends, it is necessary and indispensable to realize an
innovative machine tool structure based on a new design concept. In particular, it will be necessary to provide a thermally and
mechanically stable structure, large machining space, and motion error-minimized structure.
New structural materials used in such machine tools include advanced ceramics, advanced composite materials, engineering
plastics, fiber-reinforced plastics (FRPs), fiber-reinforced metals, etc. Table 1 shows some new structural materials and their
applications to machine tools. In particular, carbon FRPs have been recently applied to the key components of advanced machine
tool structures, i.e., a high-speed spindle system, a housing of spindle head.
In general, such machine tools use the linear motor of a noncontact nanomotion-controlled actuator and the aerostatic bearings
for supporting noncontact guideways. In consequence, future machine tools for micro- and nanometer scale processing will be
successfully able to generate micro- and nanometer scale geometries on the hard and brittle materials such as glass ceramics, silicon
carbide, silicon nitride, etc.

11.02.3 Structural Design of Machine Tools for Micro- and Nanometer Scale Processing

Demands for micro- and nanometer scale processing have recently increased in various industries. In order to rationally achieve
such nanomachining process, it is necessary to realize a thermally and dynamically stable machine tool structure. Figure 3 shows an
actual design concept of thermally and dynamically stable machine tool structure. The concept of stable machine tool structure
shown in Figure 3 are fundamentally realized by the following attributes: (1) desirable structural design of a machine tool for

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Machine Tools for Micro- and Nanometer Scale Processing 17

Table 1 Some structural materials and their applications

Materials Characteristics Applications

Concrete Portland cement High damping capacity Base

Resin concrete Thermal characteristics Bed
Polymer concrete Brittle material Column
Low cost
Ceramics Alumina ceramics Thermal characteristics Spindle
Zirconia ceramics Light weight Bearing
Silicon carbide Less wear Guideway
Silicon nitride Hard and brittle material Stage
High cost Column
Bed
Frame
Tool
Fiber-reinforced plastics (FRPs) Glass FRP High damping capacity Spindle
Carbon FRP Light weight Chuck
Aramido FRP High strength Tool holder
High cost Cover panel
Fiber-reinforced composite C–C composite Thermal stability Spindle
Carbon fiber-reinforced metal Light weight Ball screw
High cost Stage
Panel Steel honeycomb Light weight Bed
Aluminum honeycomb Damping capacity Column
Thermal characteristics Table
Plastics Polyamide, Light weight Gear
polytetrafluoroethylene, polyacetal Less friction Slideway
Light weight Guide bush
Glass ceramics Zerodur (Schott) Thermal stability Spindle
ULE (Coning) Low stiffness
Criston O (Hoya) Brittle material
High cost

Figure 3 Concept of stable machine tool structure for micro- and nanometer scale processing.

micro- and nanometer scale processing, (2) isolation of the error sources, (3) minimization of the error sources, and (4) control of
the error sources. Key design factors to be considered include independent metrology frame, symmetric machine structure, heat flow
control, minimization of Abbe offset, perfect noncontact structure, and active vibration control.
Based on such a design concept, an innovative machine tool for micro- and nanometer scale processing can be realized, as shown
in Figure 4. As can be seen in Figure 4, the machine tool structure is composed of the X–Y planar nanometer translational motion
table system, the vertical nanometer scale motion platform equipped with a noncontact gravity compensator, an air turbine–driven
aerostatic-spindle bearing system, a flat rectangular bed made of advanced ceramics, the columns and top beams made of advanced

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
18 Machine Tools for Micro- and Nanometer Scale Processing

Figure 4 Design concept of machine tool for micro- and nanometer scale processing.

ceramics, and active vibration isolation system. An intelligent performance controller equipped with sensorless monitoring and
compensating functions of machining force is also key component of the machine tool. The overall machine tool structure is
symmetrically designed with respect to the vertical nanomotion axis. In addition, all the structural modules with moving function,
i.e., an X–Y planar nanomotion table system (6) and a vertical nanomotion platform, are levitated by aerostatic bearings and driven
by the voice coil motor (VCM) in a perfect noncontact condition.
In consequence, various nonlinear phenomena can be removed from the machine tool structure and then the error factors in the
overall machine tool structure can be successfully minimized. Figure 5 shows an ideal structure of the table system for ultraprecision
machine tools. In order to realize an ideal machine tool structure, it is fundamentally important to provide the following design
attributes:
1. Symmetrically designed machine tool structure with respect to the vertical center line
In general, an overall machine tool structure is symmetrically designed with respect to the vertical center axis. Actually,
a symmetrically designed double column–type machine tool structure has been also applied for conventional machining centers.
Such a symmetrically designed machine frame can be applied to the machine tool structure for improving the structural stability. In
order to enhance the structural stability of the machine tool, it is an effective means to design such structural configuration.

Figure 5 An ideal table system for ultraprecision machine tools.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Machine Tools for Micro- and Nanometer Scale Processing 19

2. Abbe error offset–minimized machine tool structure and installation of the metrology frame structure
The principle of Abbe requires collinearity of the standard and the measurement. In order to achieve nanomotion control of the
table system, it is necessary to minimize the Abbe offset. In addition, the concept of the metrology frame should be considered for
achieving the accurate nanometer positioning. In the large optics diamond-turning machine designed and built by Lawrence
Livermore National Laboratory, the metrology frame is made from Super Invar and it is kinematically mounted to the structural
frame with three flexures, so that thermal and mechanical loads cannot be transferred between the two systems (7).
3. Motion error–minimized structure
It has been recognized that minimizing of motion errors of a moving table is an important issue for nanometer scale positioning. In
the case of the translational motion table system, it is important to symmetrically design the overall system with respect to the center
line. Effective means for realizing the motion error-minimized structure include the actuation of the moving table at the center of
gravity. The actuation of the moving table at the center of guideways is also one of the effective means.
4. Thermal deformation–minimized structure (8)
In general, thermal deformation of machine tool structures has been one of the most important issues to be considered in the
machine tool design. Thermal deformation–minimized structure should be designed based on the considerations from various
viewpoints. Many effective remedies of thermal deformation have been developed in machine tool manufacturers in the world so
far, but the structural design methodology for thermal deformation–minimized structure has not been established yet. Various
remedies of thermal deformation should be applied to the objective machine tool structure so as to minimize the thermal
deformation during nanomachining.
5. Long-stroke nanomotion system driven by hybrid actuator
Ultraprecision machine tools for nanometer scale processing require both long-range and nanometer positioning capabilities. Thus,
many actuators cannot simultaneously achieve such requirements. The achievement of such requirements can be effectively realized
by a concept of hybrid actuator. Hybrid actuator, which combined a coarse motion drive with a fine motion drive, has high
potentials for achieving new industrial requirements.
6. Planar nanomotion platform
Conventional two-dimensional (2D) X–Y translational motion table is composed of two-stacked single-axis motion tables. Such
table systems with stacked type cannot satisfy the requirements of micro- and nanometer scale positioning due to an unstable
structure with overhang, nonsymmetrical structure, and superposition. In order to rationally achieve such severe requirements, it is
necessary to realize an innovative table structure. A planar nanomotion table system is one of the most effective means for achieving
a wide-range simultaneous X–Y multiple-axis nanomotion control.
7. Vertical axis nanomotion platform with a gravity compensator
There have been few research papers on a vertical nanomotion mechanism so far. In order to provide accurate vertical nanomotion,
it is necessary to compensate the gravity load during nanomotion. Figures 4 and 14 show an example of vertical nanomotion
mechanisms. In this figure, the machine tool has the vertical nanomotion platform equipped with a gravity compensator with a pair
of vacuum cylinders. In addition, the vertical nanomotion platform is levitated by aerostatic bearing guideways and driven by
a VCM (a kind of linear motor) in a perfect noncontact condition.
8. Outer disturbance–isolated structure
In order to isolate the objective machine tool structure from various outer disturbances, such as temperature fluctuation, vibration
from the floor and thermal radiation, the overall machine tool structure should be constructed on the active vibration isolation
system installed inside a temperature-controlled enclosure. In the case of minimizing the thermal deformation, temperature
fluctuation in the enclosure should be controlled at a temperature of less than 0.1 K.
The above-mentioned design attributes should be successfully and simultaneously achieved by ideal hardware and suitable
software. Figure 6 shows an example of the machine tool structure for micro- and nanometer scale processing. This machine tool
structure is fundamentally satisfied with all the above-mentioned key attributes. As shown in this figure, this machine tool structure
is constructed from the dual frame structure on the large machine base made of ceramics, i.e., the metrology frame and the structural
frame. In order to isolate the structural deformation and vibration, a laser interferometer system is installed on the metrology frame
separated from the structural frame.

11.02.4 Fundamental Structural Modules in Machine Tool for Micro- and Nanometer Scale Processing
11.02.4.1 Main Spindle System for Micro- and Nanometer Scale Processing
A spindle system is one of the most important key components in an overall machine tool structure. A typical spindle system is
fundamentally constructed by a spindle, bearings for supporting the spindle, a driving system for driving the spindle rotation,
a cooling system for circulating fluids, and spindle housing for supporting the spindle and the bearings, etc. The rotational speed,

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
20 Machine Tools for Micro- and Nanometer Scale Processing

Figure 6 An example of machine tool for micro- and nanometer scale processing.

power, torque, dynamic stiffness, and thermal properties of the spindle system determine the machining accuracy and productivity
of the machining process. An air turbine–driven aerostatic spindle system provides high speed and precision spindle rotation in
a noncontact condition. The air turbine has 2D spiral grooves from the rotational center and provides the rotational torque without
ripple by supplying air from the spindle center. Therefore, the spindle system is one of the most effective spindle systems for micro-
and nanometer scale processing.
Figure 7 shows the structural configuration of the spindle system. In this figure, a main spindle is driven by an air turbine, which
has a cooling effect proportional to the spindle rotational speed. The spindle rotation causes an increase in heat generation at
bearing clearance, but the heat generation can be canceled out by the cooling effect of the air turbine, which is located near the axial
bearing. The main spindle made of ceramics is supported by aerostatic bearings in both the radial and axial directions. The aerostatic
bearings are made of porous materials.
The slot restrictor used for the radial bearings makes it possible to isotropically support the main spindle system, as shown in
Figure 8. Radial error motion of the spindle system can be minimized by the laminar air flow from the slot restrictor of radial
bearings.
Figure 9 shows an example of the air turbine–driven aerostatic spindle system (9). As can be seen, the spindle system has
a simple and compact structure. In consequence, the heat generated at bearings can be successfully cooled by the adiabatic
expansion in the air turbine. Actually, performance evaluation results of the spindle system confirm that thermal deformation of the
spindle system is negligibly small at high-speed rotation.

Figure 7 An air turbine–driven aerostatic spindle system.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Machine Tools for Micro- and Nanometer Scale Processing 21

Figure 8 Slot restrictor of radial bearing.

Figure 9 An example of the air turbine–driven aerostatic spindle system.

11.02.4.2 X–Y Planar Nanomotion System

The X–Y table systems of stacked type with servo motors and lead screws have been applied for conventional machine tools.
Although the table of this type has relatively X–Y long travel ranges, it has been widely known that they have nonlinear phenomena
such as friction and stick-slip. It is difficult to eliminate such nonlinear phenomena; therefore, various approaches have been
developed.
The X–Y planar nanomotion table is levitated by aerostatic bearings made of porous material and driven by hybrid linear motor
in a noncontact condition, so that it is easy to construct an ideal control model without nonlinear phenomena in the system and
then it is easy to construct the simplified model for control (10). In order to provide high motion accuracy and high structural
rigidity, the vacuum attraction force preloads the aerostatic bearings of the moving table.
Figure 10 shows the basic design concept of the hybrid linear motor (11), which is composed of two different kinds of linear
motors, i.e., a VCM and a shaft-type linear motor. Characteristic features of VCM include quick response, accurate force control
capability, and short stroke. Inversely, the shaft-type linear motor characterizes a long stroke. In order to enlarge both the X- and
Y-axes strokes of the X–Y planar nanomotion table system, the shaft-type linear motors independently drive the moving rectangular
frames, which have coils of VCMs so as to keep a constant relative distance within a stroke of the VCM. In consequence, the
nanomotion control system of the moving table drives the VCM without considering its stroke limit.
Figure 11 shows an example of the X–Y planar nanomotion system. The white rectangular moving table made of alumina
ceramics (Al2O3) can be freely driven and positioned on the large X–Y plane. The moving rectangular table levitated on the X–Y
plane is driven in a noncontact condition so as to eliminate nonlinear phenomena such as stick-slip, friction, etc.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
22 Machine Tools for Micro- and Nanometer Scale Processing

Figure 10 A concept of X–Y planar nanomotion system.

Figure 11 An X–Y planar nanomotion table system.

Figure 12 shows a 1-nm stepwise response of the X–Y planar nanomotion system. In consequence, a clear nanometer stepwise
positioning can be observed in each X- and Y-axis directions. Results of simultaneous X–Y-controlled stepwise responses confirmed
this table system has two-axes interference-free nanometer scale positioning capability. Figure 13 shows a typical example of the
circular motion test results in the X–Y plane with a diameter of 100 mm and a feed rate of 62.8 mm s1. A smooth nanometer scale
circular motion test results can be clearly observed without both the quadrant glitches and the inverse error.

11.02.4.3 Vertical Nanomotion System

Degree of freedom of shape-generating function for machining 3D geometries can be fundamentally realized by the translational
motion in the horizontal and vertical axes (X, Y, and Z axes) and the rotational motions (A, B, and C axes), as shown in Figure 1.
Realization of the vertical nanomotion system is one of the most important issues to be solved for developing the ultraprecision
machine tools with nanometer scale resolution. The mechanism of the vertical nanomotion system requires the compensating
function of the gravity load during nanomotion of the moving body.
The countermass and the counterbalancing mechanisms equipped with pneumatic cylinders are widely used for gravity
compensation in conventional large machine tools. However, the use of such mechanisms remarkably deteriorates the system
performance due to vibration, low response, or pressure fluctuation. Inversely, pressure fluctuation in the vacuum cylinders is
significantly lower than that in pneumatic cylinders because the absolute pressure in vacuum cylinders is negligibly small. A pair of
vacuum cylinders can be used as an effective gravity compensator. Figure 14 shows a typical example of the structural configuration
of the vertical nanomotion system equipped with a pair of vacuum cylinders (12). In this figure, the Z-axis nanometer scale
translational motion can be driven by a VCM installed at the center of parallel allocated guideways. This structural configuration can
achieve a long-stroke vertical motion of a subnanometer scale resolution.
Figure 15 shows a 0.6-nm stepwise response of the vertical nanomotion system equipped with gravity compensator.
As shown in this response, a clear stepwise nanomotion of sensor resolution level can be observed without any vibration in the
Z-axis direction.

11.02.4.4 Advanced Nanomotion Control System

The moving bodies on the X–Y planar nanomotion table system and the vertical nanomotion platform in the machine tool
mentioned above are levitated by five aerostatic bearings and then driven by eight linear motors (VCMs) in a perfect noncontact
condition. These moving bodies achieve accurate and smooth nanomotion, while the machining force causes the motion errors

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Machine Tools for Micro- and Nanometer Scale Processing 23

Figure 12 Nanometer stepwise responses in the X- and Y-axis directions.

Figure 13 A typical example of the circular motion test results.

during micro- and nanometer scale processing. Therefore, a sensorless monitoring and compensation function of machining force
(13) shown in Figure 4 is implemented on the nanomotion control systems. Therefore, the perfect noncontact machine structure
enables the making of an accurate physical model of the system and then the successful identification of system parameters in the
models.
The sensorless monitoring system described here is based on a disturbance observer and can eliminate the machining
force from both the input and output signals of a positioning controller without additional in-process sensors. Because the
driving at the center of moving body mass can be achieved in the nanomotion control system, machining force compen-
sation can be successfully achieved through the feedback of the estimated machining force without angular excitation of the
moving table. In addition, each nanomotion control system has a separate sensorless monitoring and compensating
functions.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
24 Machine Tools for Micro- and Nanometer Scale Processing

Figure 14 Vertical nanomotion systems with a pair of vacuum cylinders.

Figure 15 Nanometer stepwise responses in the Z direction.

11.02.5 Future Evolutionary Trends of Machine Tool Technologies for Micro- and Nanometer Scale Processing
11.02.5.1 Micro- and Nanometer Scale Machining Function
In order to achieve the effective generation of 3D micro- and nanometer scale geometries, it is necessary and indispensable to realize
the innovative machine tools equipped with simultaneous multiple axis–controlled nanomachining function. However, accurate
and stable rotary motion technologies for nanometer scale processing have not been established so far. Typical trunnion (cylindrical
protrusion used as a mounting and/or pivoting point)-type rotary motion table systems are widely used for the rotary table system
in conventional simultaneous five axes–controlled machining centers, but the performances, i.e., the position resolution and the
structural stability, are insufficient for micro- and nanometer scale processing. Thermal stability of the direct drive rotary motor used
for such driving systems is also a serious problem to be solved.
One of the most effective means is to develop the rotary motion platform driven by a hybrid actuator. The hybrid actuator can be
constructed by integrating a pneumatic rotary actuator with an electromagnetic actuator. The pneumatic actuator can generate high
torque without heat generation, while the electromagnetic actuator provides significant advantages such as quick response, accurate
torque control, and dead band free. Consequently, the resultant hybrid actuator will make it possible to achieve a remarkable
performance.
Intelligent machining function is also one of the most important issues for realizing a future machine tool for micro- and
nanometer scale processing. On-machine measuring function with optical method will be implemented in such machine tools.

11.02.5.2 Establishment of New Structural Design Methodology

It is important to establish a design methodology for realizing an innovative machine tool structure for micro- and nanometer scale
processing without following the conventional machine tool structure. A new design concept should be proposed for providing of
a thermally and dynamically stable machine structure. Future machine tool structure for micro- and nanometer scale processing
should be optimally designed from the viewpoint of both structural design and control design.
In order to realize the machine tool with subnanometer resolution, new structural materials will be applied to the machine tool
structure. Such new structural materials include a variety of advanced materials such as ceramics, engineering plastics, and

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Machine Tools for Micro- and Nanometer Scale Processing 25

composite materials. In such case, the most important issue is to put the right material in the right portion in the machine tool
structure.

11.02.5.3 Machine Tool Equipped with Multiple System Functions

Various system functions of machine tools have been increased to respond to the industrial demands for high accuracy, high
flexibility, and high productivity. To meet various industrial requirements, the machine tools equipped with multiple functions
have recently developed and have been commercialized (14).
Future machine tools for micro- and nanometer scale processing will perform the multiple processes such as the combination of
the machining processes and high-density energy beam processes. In such cases, on-machine measuring or in situ measuring
functions will be also required for performing additional corrective processing to achieve the required accuracy and surface
roughness.
In addition, physical cleaning function of workpiece or tool surface will be implemented with ultrasonic, energy beam, or dry ice
blasting on machine.

11.02.5.4 Intelligent and Autonomous Machine Tool

Future machine tools will realize intelligent and fully autonomous machining systems. The intelligent and autonomous machine
tool should support human activities and replace them to an unmanned machining system.
In order to realize the machine tool, it is necessary to energetically conduct the extensive studies on intelligent sensors, modular
autonomous devices, dexterous robots, intelligent monitoring systems, intelligent process control system, autonomous process
planning system, etc. Finally, implementation of such intelligent and autonomous machine tool will largely change the machining
environment.

11.02.6 Conclusions

Ultraprecision machine tools have been recently designed and developed for productive nanometer scale machining of 3D
nanogeometries made of hard and brittle materials. In addition, the micro- and nanometer scale processing technologies based on
mechanical material removal process has been widely recognized as a core production technology of the twenty-first century.
Specification and performances of products in the global market, some more and some less, depend on production technologies for
producing the high-precision components and parts with reasonable cost.
This chapter described primary concerns in future industrial demands and new development of machine tools. Specifically, the
contents include the structural design and the actual development examples of an innovative ultraprecision machine tool for
generating micro- and nanometer scale geometries in a large work space. The machine tool for micro- and nanometer scale pro-
cessing described here provides a dynamically and thermally stable machine tool structure.
The fundamental structural components and modules for the machine tools were also discussed in detail. The functions and
structure of the objective machine tool are of the utmost importance in realizing future innovative machine tools. The structural
design of the machine tool and the nanomotion control technologies related to the machine tool presented in this chapter will be
widely applied in a variety of industries.
Finally, future evolutionary trends of machine tools for micro- and nanometer scale processing were summarized from the
viewpoint of machine tool design.

References

1. Bruzzone, A. A. G.; Costa, H. L.; Lonardo, P. M.; Lucca, D. A. Advances in Engineered Surfaces for Functional Performances. CIRP Ann. 2008, 57 (2), 750–769.
2. Shinno, H.; Yoshioka, H.; Sawano, H. A Newly Developed Long Range Positioning Table System with a Sub-nanometer Resolution. CIRP Ann. Manuf. Technol. 2011, 60 (1),
403–406.
3. Takeuchi, Y.; Sakaida, Y.; Sawada, K.; Sata, T. Development of 5-Axis Control Ultraprecision Milling Machine for Micromachining Based on Non-Friction Servomechanism. CIRP
Ann. 2000, 49 (1), 295–298.
4. Sriyota, P.; Nakamoto, K.; Sugai, M.; Yamazaki, K. Development of 5-Axis Linear Motor Driven Super-Precision Machine. CIRP Ann. 2006, 55 (1), 381–384.
5. Shinno, H.; Hashizume, H.; Ito, Y.; Sato, C. Structural Configuration and Performances of Machining Environment-Controlled Ultraprecision Diamond Turning Machine ‘Capsule’.
CIRP Ann. 1992, 41 (1), 425–428.
6. Shinno, H.; Yoshioka, H.; Taniguchi, K. A Newly Developed Linear Motor-Driven Aerostatic X-Y Planar Motion Table System for Nano-Machining. CIRP Ann. 2007, 56 (1),
369–372.
7. Slocum, A. H. Metrology Frame, Precision Machine Design; Prentice Hall: Englewood Cliffs, NJ, 1992; pp 236–246.
8. Shinno, H. Various Remedies for Reduction of Thermal Deformation. In Thermal Deformation in Machine Tools; Ito, Y., Ed.; McGraw-Hill: USA, 2010; pp 117–141 (chapter 4).
9. Sawano, H.; Kobayashi, R.; Yoshioka, H.; Shinno, H. A Proposed Ultraprecision Machining Process Monitoring Method Using Causal Network Model of Air Spindle System. Int. J.
Automation Technol. 2011, 5 (3), 362–368.
10. Shinno, H.; Hashizume, H. Nanometer Positioning of a Linear Motor-driven Ultraprecision Aerostatic Table System with Electrorheological Fluid Dampers. CIRP Ann. 1999,
48 (1), 243–246.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
26 Machine Tools for Micro- and Nanometer Scale Processing

11. Kurisaki, K.; Sawano, H.; Yoshioka, H.; Shinno, H. A Newly Developed X-Y Planar Nano-Motion Table System with Large Travel Ranges. J. Adv. Mech. Des. Syst. Manuf. 2010,
4 (5), 976–984.
12. Takahashi, M.; Yoshioka, H.; Shinno, H. A Newly Developed Long-Stroke Vertical Nano-Motion Platform with Gravity Compensator. J. Adv. Mech. Des. Syst. Manuf. 2008, 2 (3),
356–365.
13. Shinno, H.; Hashizume, H.; Yoshioka, H. Sensor-Less Monitoring of Cutting Force during Ultraprecision Machining. CIRP Ann. 2003, 52 (1), 303–306.
14. Moriwaki, T. Multi-functional Machine Tool. CIRP Ann. 2008, 57 (2), 736–749.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 11.03 Continuum Micromechanics and Molecular Dynamics Simulation
X Li, National University of Singapore, Singapore
 2014 Elsevier Ltd. All rights reserved.

11.03.1 Continuum Mechanics Modeling of Machining Processes 27

11.03.1.1 Modeling of Workpiece Material Flow Stress Properties in Machining of Ductile Materials Based
on Continuum Mechanics 27
11.03.1.2 Predictive Modeling of Milling Geometry 29
11.03.1.3 Predictive Modeling of Face Milling Forces 31
11.03.1.4 Predictive Modeling of Helical End Milling Cutting Forces 32
11.03.1.5 Predictive Modeling of the Dynamics of Milling Process 34
11.03.1.6 Predictive Modeling of Dynamic-Regenerative Uncut Chip Thickness 34
11.03.1.7 Milling Process Simulation System Using the Predictive Models 35
11.03.1.8 Experimental Verification of the Simulation System 36
11.03.2 Micromechanics Modeling of Cutting of Composite Materials 38
11.03.2.1 Special Challenges in Cutting of Metal Matrix Composite 38
11.03.2.2 The Rule of Mixtures 39
11.03.2.3 Modeling of Tool Wear Rate in Relation to the Reinforcement in MMCs 40
11.03.2.3.1 MMC Workpiece Specimens 40
11.03.2.3.2 Special Tool Wear Mechanisms Found from Experimental Cutting Tests 41
11.03.2.3.3 Critical Percentage of Reinforcement Causing Enhanced Abrasive Wear 42
11.03.3 Molecular Dynamics Modeling and Simulation of Ductile Mode Cutting of Brittle Materials 46
11.03.3.1 Transition of Ductile Mode to Brittle Mode in Cutting of Brittle Materials from Nanoscale
to Micro/Millimeter-Scale 46
11.03.3.2 Molecular Dynamics Modeling of Cutting of Silicon 48
11.03.3.2.1 The Crystal Structure of a Silicon Workpiece 48
11.03.3.2.2 Molecular Dynamics Modeling of Nanoscale Cutting of Silicon with a Signal-Point Diamond Tool 49
11.03.3.3 The Simulation System Using the Molecular Dynamics Model 50
11.03.3.4 Validation of the MD Simulation System 50
11.03.3.5 Molecular Dynamics Simulation Study of the Characteristics in Nanoscale Cutting of Silicon 53
11.03.3.5.1 Effects of Tool Edge Radius on the Cutting Forces 53
11.03.3.5.2 The Upper Bound of Cutting Edge Radius for Ductile Chip Formation 54
11.03.3.5.3 Dynamic Hard Particles in the Chip Formation Zone 57
11.03.3.5.4 Groove Wear of Diamond Tool in Ductile Mode Cutting of Silicon 57
References 58

11.03.1 Continuum Mechanics Modeling of Machining Processes

Three typical materials representing the three categories of workpiece materials are metals, metal matrix composites (MMCs), and
silicon, as shown in Figure 1. In this section, how continuum mechanics is applied to modeling and simulation of machining of
metals is presented. Continuum mechanics applies to materials that follow the laws of conservation of mass, conservation of
momentum, and conservation of energy. In modeling of machining of more or less ductile and homogeneous workpiece materials
such as metals, continuum mechanics has been used and the models have been working well. This section shows, for example, in
milling of steels that is based on continuum mechanics, the workpiece flow stress properties can be modeled, and then predictive
modeling of milling geometry, cutting forces in face milling, cutting forces in helical end milling, dynamics of milling process, and
dynamic-regenerative uncut chip thickness, and development of a milling process simulation system using the predictive models,
with experimental verification.

11.03.1.1 Modeling of Workpiece Material Flow Stress Properties in Machining of Ductile Materials Based
on Continuum Mechanics
Based on continuum mechanics, the steel workpiece material flow stress properties can be modeled as functions of strain, strain rate,
and temperature, and can be obtained from the material deformation under high-frequency alternating-current induced heating
and compressive split Hopkinson bar tests, as shown in Figure 2.
With the models for the workpiece material flow stress as a function of strain, strain rate, and temperature, represented by the
flow stress, s, and work hardening index, n, a predictive model for machining characteristic factors in oblique cutting of steel
workpiece materials, including the cutting forces Fc, Ft, and Fr, the undeformed chip thickness t2, the cutting temperature T, and the

Comprehensive Materials Processing, Volume 11 http://dx.doi.org/10.1016/B978-0-08-096532-1.01102-X 27

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
28 Continuum Micromechanics and Molecular Dynamics Simulation

Figure 1 Microstructure of three categories of machining workpiece materials. (a) For metals; (b) for metal matrix composites; and (c) for silicon
wafers.

Pressure gun

Striker bar Strain gauge Output

bar
Incident bar Specimen

Signal conditioner

Strainmeter

Digital osilloscope

Figure 2 Ductile and homogeneous workpiece material flow stress properties can be obtained from material deformation under high-frequency
alternating-current induced heating and compressive split Hopkinson bar tests.

tool life L, can be developed using Oxley’s predictive machining theory (1) that is based on continuum mechanics, as shown in
Figure 3.
With the predictive model for machining characteristic factors, predictive models for various cutting processes can be developed.
As an example, a predictive model for milling processes is described.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Continuum Micromechanics and Molecular Dynamics Simulation 29

Figure 3 A predictive model for predicting the cutting forces Fc, Ft, and Fr, the undeformed chip thickness t2, the cutting temperature T, and the
tool life L, from strain and strain rate that can be calculated from cutting conditions, and workpiece material flow stress properties.

11.03.1.2 Predictive Modeling of Milling Geometry

The milling geometry is first modeled for the true uncut chip thickness (see Figure 4(a)), true entry and exit angles (see Figure 4(b)),
the effect of cutter runout (see Figure 4(c)), and the dynamic shear length based on true tooth trajectories (see Figure 4(d)). The true
uncut chip thickness htr can be expressed as a function of milling process parameters Nt – number of teeth, ft – feed per tooth per
revolution, and r – the radius of the cutter, and 4j – the instantaneous position angle (in rad) of the tooth tip analyzed clockwise
with reference to the positive Y-axis:
0 1
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
B u
u 2ft sin 4i 2
f cos 24i f sin 4i cos 4i C
3 2
htr ¼ jMCj  jNCjyr B
@1  u
t1   t  þ t 3 C A
Nt f t Nt ft 2 Nt ft
rþ cos 4i rþ cos 4i rþ cos 4i
2p 2p 2p

Y R tmt1 1 Y
1 Workpiece Workpiece
feed
feed
N M
htr entry

0 O C 2
0 O C 2 X
X
exit

3 R tmt2
3

(a) (b)

Shear length, l sd
Workpiece Y
Y
1 feed M Vf
M
h tr
N
Tooth 1 Tooth V
Q
Tooth 4 Vr
2 O C 4
X C
Tooth 3 X

Tooth 2

(c) (d)

Figure 4 Modeling of milling geometry. (a) True uncut chip thickness; (b) entry and exit angles; (c) effect of cutter runout; and (d) dynamic shear
length.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
30 Continuum Micromechanics and Molecular Dynamics Simulation

The true entry angle 4entry can be expressed as a function of milling process parameters Nt – number of teeth and ft – feed per
tooth per revolution:
ft
4entry y  
ft Nt 
[1]
2 rþ
2p
The true exit angle 4exit can be expressed as a function of milling process parameters Nt – number of teeth and ft – feed per tooth
per revolution:
ft
4exit yp þ 
ft Nt 
[2]
2 r
2p
The true ideal peripheral machined workpiece surface roughness can be represented by the magnitude of the height of tooth
mark above point of the lowest machined surface level, Rtmt (see Figure 4(b)), as a function of milling process parameters Nt –
number of teeth and ft – feed per tooth per revolution:
  4entry
Rtmt1 ¼ r 1  cos 4entry ¼ 2rsin2
2
rft2 [3]
y 
ft Nt 2
8 rþ
2p
ð4exit  pÞ
Rtmt2 ¼ rð1  cosð4exit  pÞÞ ¼ 2rsin2
2
rft2 [4]
y 
ft Nt 2
8 r
2p
The effect of cutter runout can be represented by an increase in the chip load derj:
derj ¼ rj0  rjþ1
0
[5]

rj0
where the radius of the cutter at the jth tooth and 0
rjþ1is the radius of the cutter at the (j þ 1)th tooth.
The dynamic shear length can be determined based on the true tooth trajectories, as shown in Figure 4(d). When the tip of the
tooth number i reaches the position of point M, where its angular position is 4i, the instantaneous cutting velocity V of the cutting
tooth relative to the workpiece is the vector-sum of the instantaneous rotational velocity Vr and the feeding velocity Vf of the cutter
relative to the workpiece. If the instantaneous position of the tool center is C, then MC must be perpendicular to Vr, and the
inclination angle between Vf and Vr is identical to 4i. Assuming the inclination angle of V to Vr is d, from the triangular relationship,
it can be derived that
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
v ¼ vr2 þ vf2 þ 2vr vf cos 4i [6]

vr þ vf cos 4i
d ¼ arccos [7]
v
where v, vr, and vf are modulus of the velocity vectors, V, Vr, and Vf, respectively, and vr ¼ 2prn, vf ¼ ftNtn.
For a given shear angle, say 4, the shear plane should intersect the free surface at a point Q, and the length of the line segment
MQ should be the required shear length. If the angle included between MQ and the positive X-axis is j, the following relationship
can be written as:
yM  yQ
¼ tan j [8]
xM  xQ

The coordinates of point M can be written according to the parametric equations representing the tooth path as described in
Chapter 3, i.e.,

xM ¼ rð4i  iD  q0 Þ þ r sin 4i
[9]
yM ¼ r cos 4i

Since the point Q is located on the free surface that is generated by the preceding cutting tooth number i þ 1, Q must be located
on the trajectories of the tooth number i þ 1. Assuming that when the tip of the cutter tooth number i þ 1 is at the position of point
Q, its angular position is 4i þ 1, then the coordinates of point Q are:

xQ ¼ rð4iþ1  ði þ 1ÞD  q0 Þ þ r sin 4iþ1
[10]
yQ ¼ r cos 4iþ1

From the geometry,

j ¼ p  4i  ðf  dÞ [11]

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Continuum Micromechanics and Molecular Dynamics Simulation 31

Substituting eqns [9]–[11] into eqn [8] gives

r cosð4iþ1  4i  ðf  dÞÞ  r cosðf  dÞ þ rð4iþ1  4i  DÞsinð4i þ ðf  dÞÞ ¼ 0 [12]

The only unknown in eqn [12] is 4iþ1. As it is a transcendental equation, numerical methods have to be used to solve it. Bisection
method is again used. Once 4iþ1 is solved, the coordinates of point Q can be determined from eqn [10]. Then the actual shear length
lsd can be subsequently determined using the equation:
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 2  2
lsd ¼ jMQj ¼ x M  x Q þ yM  yQ [13]

Equations [6]–[13] formulate the model governing the shear length. Since it covers the variable inclination of the work surface in
front of the tool, it is the dynamic shear length model.
According to Oxley’s theory (1), when the actual shear length is determined, an equivalent undeformed chip thickness h*, which
is actually the uncut chip thickness at the outer end of the shear zone relative to the current cutting position, can be found as
h
¼ lsd sin f [14]
and the instantaneous chip thickness t2 can be expressed as
t2 ¼ lsd cosðf  aÞ [15]
Then the cutting forces in face milling can be modeled.

11.03.1.3 Predictive Modeling of Face Milling Forces

The shear length and shear angle having been determined, a theoretical model for the cutting forces in face milling can be estab-
lished by using the predictive oblique cutting model as described in Section 11.03.1.1. A typical face milling system is considered as
shown in Figure 5, where the coordinate system XYZ is fixed with respect to the machine tool structure, its X-axis opposite to the
feeding direction of the cutting tool, its Z direction upward along the cutter axis, and its Y-axis normal to X and Z. The milling cutter
rotates clockwise and has Nt teeth uniformly distributed and labeled clockwise as number 0, 1, ., Nt  1. The tooth-working angle
is equal to the difference between exit angle and entry angle, j ¼ 4ex  4st .
For a tooth number (j) that is engaged in cutting at the instantaneous angular immersion 4j measured clockwise from the Y-axis
at a particular instant, a rotating coordinate system UjVjWj, which is fixed with this tooth, is used as shown in Figure 5. Referring to
this system, this cutting tooth can be viewed as a single-point cutting tool of a lathe, thus P1j, P2j, and P3j, which are the cutting force
components produced in this tooth in the tangential, radial, and axial direction, respectively, in this rotating coordinate system can
be calculated from the corresponding chip load, cutting condition, and tool geometry by using the predictive machining theory of
continuous cutting with a single-point cutting tool as described in Section 11.03.1.1. The instantaneous shear length can be
determined using the model described in this section, either the static shear length model or the dynamic shear length model. It
should be noted that regardless of the intermittent nature of milling, in calculations for the cutting forces of a cutter tooth during
milling, the cutting temperature which is determined from the work done in the chip formation and is used for determinations of
the workpiece flow stress properties and cutting forces, as described in Section 11.03.1.1, should be determined as if it is in
continuous cutting, as far as the cutting edge is engaged in cutting, because the temperature increases directly from the work done in
the chip formation zone and simultaneously reacts back on the workpiece material under deformation in the chip formation zone.
Then, P1j, P2j, and P3j can be expressed in terms of the fixed coordinate system XYZ as FXj, FYj, and FZj, force components in the X,
Y, and Z directions, respectively. To resolve and sum the force components P1j, P2j, and P3j into X, Y, Z directions, the following
equation is used:
8
< FXj ¼ P1j cos 4j þ P2j sin 4j
FYj ¼ P1j sin 4j þ P2j cos 4j [16]
:
FZj ¼ P3j

Work Uj

Wj
P2j
j Vj
P3j
FZ
Fx
P1j

Fy

Figure 5 A typical face milling system.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
32 Continuum Micromechanics and Molecular Dynamics Simulation

The total forces acting on the cutter are the sum of the forces on each tooth, which can be determined using the following
equation:
8 NPt 1 NPt 1 
>
>
> FX ðtÞ ¼
> FXj ¼ P1j cos 4j þ P2j sin 4j
>
>
>
>
j¼0 j¼0
< NPt 1 NPt 1 
FY ðtÞ ¼ FYj ¼  P1j sin 4j þ P2j cos 4j [17]
>
>
>
>
j¼0 j¼0
>
> NPt 1 NPt 1
>
>
: FZ ðtÞ ¼ FZj ¼ P3j
j¼0 j¼0

where the immersion angle 4j varies with time as 4j(t) ¼ 2pnt/60 þ 2pj/Nt, and n is the milling cutter rotating speed in
rpm. The milling force variation against cutter rotation can be predicted by calculating the FX , FY and FZ for different
values of t.

11.03.1.4 Predictive Modeling of Helical End Milling Cutting Forces

End milling is the most versatile form of milling that can be used to machine slots, shoulders, die cavities, contours, and profiles. An
end milling process consists of a cylindrical cutter that has multiple cutting edges on both its periphery and its tip, permitting end
cutting and peripheral cutting. These cutting edges or flutes are usually made helical to reduce the impact that occurs when each flute
engages the workpiece. In this section, a theoretical model for cutting forces in helical end milling is developed based on Oxley’s
predictive machining theory (1).
A typical right-hand helix, right-hand cut flat end milling is schematically depicted in Figure 6. There are Nt flutes (or teeth)
uniformly distributed. The helix angle for each flute is b. The flutes can be labeled clockwise using an index w ¼ 0, 1, ., Nt  1, and
the entering sequence of these flutes is Nt  1, Nt  2, ., 2, 1, 0. A coordinate system XYZ is used to describe the cutting system, and
the origin O is located in the point where the cutter axis intersects with its face plane. The X direction designates the negative cutter
feed direction with respect to the workpiece, Z direction points up along the cutter axis, and Y direction is the normal to the feed and
following the right-hand convection.
In order to account for the helix angle b, the helical end milling cutter within the axial depth of cut da can be discretized into Ns
number of small slices along its axis in the Z direction, as shown in Figure 6. Thus the thickness of each slice is Dd ¼ da/Ns. At each
slice, the cutting action of an individual flute segment can be modeled as oblique cutting, with the cutting edge having an incli-
nation angle equal to the helix angle b. These slices are labeled from bottom to top using an index of v ¼ 1, 2, ., Ns, and the first
slice is located in the cutting ends.

Ns-th slice

da

Δd
First slice

Fz

Fy

dr

X Fx

Figure 6 Schematic of a helical end milling process.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Continuum Micromechanics and Molecular Dynamics Simulation 33

If the feed per tooth is ft, the radius of the cutter is r, and the rotating speed of the tool is n (rpm), then 4v,w, the angular position
of the cutting point of the wth flute in the vth slice, with respect to the negative Y direction and measured clockwise at a given time t,
can be expressed from the geometrical relationship as,
2pnt 2pw ðv  1Þda tan b
4v;w ðtÞ ¼ þ  [18]
60 Nt Ns r
During the cutting process under ideal conditions, each axial cutting edge segment removes a continuously varying chip
thickness hv,w. This varying undeformed chip thickness can be determined by using the true undeformed chip thickness model,
 
hv;w ¼ f 4v;w ; ft ; r; Nt [19]

when r is far larger than ft, and hv,w can be approximated by using the traditional undeformed chip thickness model, which is
expressed as,
hv;w ¼ ft sin 4v;w [20]

To model the cutting forces in helical end milling, by discretizing the cutting part of a helical end milling tool into Ns number of
slices along its axis in the Z direction to account for the helix angle b, the cutting action of an individual flute segment within each
slice can be modeled as single point oblique cutting, which has an inclination angle of b. For a cutting edge segment at the vth slice
component of the wth helical flute, a rotating coordinate system UVW, which is fixed with this edge, is used as shown in Figure 7.
Referring to this system, this cutting edge segment can be viewed as a single-point cutting tool of a lathe, thus P1v,w, P2v,w and P3v,w,
which are the cutting force components produced in this segment in the tangential, radial, and axial direction, respectively, in this
rotating coordinate system, can be calculated from the corresponding chip load, cutting condition, and tool geometry by using the
respective oblique cutting models as described in Section 11.03.1.1.
However, there is a distinctive difference between the first cutting slice element and the rest of slices in their cutting actions. For
the first cutting slice, it has cutting edges on both its end face and its periphery; therefore both end cutting and peripheral cutting
exist simultaneously in this slice. For other slices, on the other hand, only peripheral cutting exists on each of the individual flute
segment within each slice, and there is no cutting edge on the end face.
Considering the distinctive characteristics of the first cutting slice and other slices, as well as considering the mechanics of
milling, a theoretical force model for helical end milling can be developed based on the predictive oblique cutting model as
described in Section 11.03.1.1. The cutting action of a flute segment in the first slice can be modeled as single point oblique cutting
with the effects of end cutting edge and tool nose radius. The cutting action of a flute segment in the other slices can be modeled as
simple oblique cutting with a single straight cutting edge without the effects of end cutting edge and tool nose radius. Using these
models, for the cutting point of the wth flute in the vth slice, the cutting force components P1v,w, P2v,w, and P3v,w can be determined
from the corresponding cutting condition, tool geometry, work material properties, and chip load (or width and length of the shear
plane) in a way similar to the force model for face milling as described in Section 11.03.1.3.
The total forces acting on the cutter are the sum of the forces on each tooth segment. With P1v,w, P2v,w, and P3v,w calculated, the
cutting force on this edge segment can be expressed in terms of the fixed coordinate system XYZ. Thus the total cutting forces of the
helical end milling cutter at the present instance, Fx, Fy, and Fz, as shown in Figure 6, can be determined using the following equations,
8
PNs NP t 1 
>
>
> Fx ðtÞ ¼
> P1v;w cos 4v;w þ P2v;w sin 4v;w
>
>
>
<
v¼1 w¼0
PNs NP t 1 
Fy ðtÞ ¼  P1v;w sin 4v;w þ P2v;w cos 4v;w [21]
>
>
>
>
v¼1 w¼0
>
> PNs NP t 1
>
: Fz ðtÞ ¼ P3v;w
v¼1 w¼0

U
work P1v,w W
V

v,w P3v,w

P2v,w
Fz
X Fx
v -th slice
of the milling
Fy cutter
Y

Figure 7 Coordinate systems for the vth slice of the milling cutter.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
34 Continuum Micromechanics and Molecular Dynamics Simulation

Uj

Vj
kx
ϕj
Work

cx Cutter

ky cy
Y

Figure 8 Model for milling system with two degrees of freedom.

11.03.1.5 Predictive Modeling of the Dynamics of Milling Process

In the dynamic model, the milling cutter system is assumed to be a system with two modes of vibration in two mutually
perpendicular directions, X and Y, and the feed direction of the workpiece is along the X-axis, as shown in Figure 8. The
coordinate system is fixed with respect to the machine tool structure; its axes are aligned with the principal modes of oscillation.
This is a common characteristic of the spindle–tool assembly, which is the most flexible part of a typical milling machine. The
milling cutter has Nt teeth, which are equally spaced. The dynamics of the milling system can be given by the differential
equations as:
8 NPt 1
>
>
> mx€x þ cx x_ þ kx x ¼
< Fxj ¼ Fx ðtÞ
j¼0
NP
[22]
> t 1
>
>
: my €y þ cy y_ þ ky y ¼ Fyj ¼ Fy ðtÞ
j¼0

where mx and my are the masses, cx and cy are the damping coefficients, and kx and ky are the stiffness of the machine tool
structure in modes X and Y, respectively. Fx and Fy are the components of the cutting force that are applied on tool in the
directions of X and Y.
The dynamic characteristics of the machine tool structure, including the masses, damping coefficients, and stiffness, or in another
form of the dynamic stiffness, natural frequency, and the damping ratio, can be obtained through modal testing by using the
impulse method. Once the instantaneous dynamic-regenerative undeformed chip thickness is determined, the cutting force
components can be predicted from the predictive force model as described in the previous sections. It is assumed here that dynamic
cutting forces in milling react instantaneously to the changes in undeformed chip geometry that are occurring due to the dynamic-
regenerative effects. Within two adjacent simulation steps in time domain, the undeformed chip thickness is considered to be
constant and the predictive oblique cutting model is applicable.

11.03.1.6 Predictive Modeling of Dynamic-Regenerative Uncut Chip Thickness

In a dynamic milling process, the cutting forces excite the structure in the feed and normal directions, causing relative vibration
between the tool and the workpiece. If the tool vibrates, each tooth leaves behind a wavy surface and the following tooth cuts into
it and regenerates undulations, as shown in Figure 9. The structure and the cutting process are involved in a feedback loop:
vibration causes a modulation of the cutting force, and the variable cutting force excites vibration in turn (2). As a result, the
undeformed chip thickness will be affected not only by the instantaneous vibration of the system, but also by the amount of
undulation left from the previous teeth. The resultant undeformed chip thickness therefore consists of static part h and dynamic
components caused by the vibrations of the tool at the present and previous tooth periods, which can be referred to as
modulation vj and vj0 , respectively. The instantaneous dynamic-regenerative undeformed chip thickness for the jth tooth at an
angular position of 4j can be expressed as:
   
hrg ¼ h  vj þ vj0 g 4j [23]

where h, the static part, is the nominal instantaneous undeformed chip thickness in steady-state cutting.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Continuum Micromechanics and Molecular Dynamics Simulation 35

Feed
Workpiece

uj
uj 0
ϕj
X

ft

Figure 9 Model for dynamic-regenerative uncut chip thickness.

The inner modulation vj is caused by the present tool–workpiece relative displacement. Since the chip thickness is measured in
the radial direction, if x and y are the instantaneous deflections in the system, vj is given by
vj ¼ x sin 4j þ y cos 4j [24]

The outer modulation vj0 is the possible undulation that could be left behind from the previous teeth. vj0 is usually
modulated by the previous tooth (j þ 1). However, under large amplitudes of oscillation, the (j þ 1) tooth may lose contact
with the surface being machined, then the outer modulation vj0 may be left by the (j þ 2) tooth, or by another more previous
tooth. This is the basic nonlinearity of the dynamic milling process. As an approximation, the outer modulation is generally
given by
 
vj0 ¼ min vjþ1 ðt  TÞ; vjþ2 ðt  2TÞ þ h; vjþ3 ðt  3TÞ þ 2h; / [25]

2p 60
where t is the current time, and T ¼ ¼ is the period between successive tooth engagements. vjþk(t  kT) is the possible
UNt nNt
modulation on the current surface by the tool–workpiece vibration at the time (t  kT) when the tooth (j þ k) was passing the
angular position 4j, which is
vjþk ðt  kTÞ ¼ xðt  kTÞsin 4j þ yðt  kTÞcos 4j [26]

If 4entry and 4exit represent the entry and exit immersion angles of the cutter to and from the cutting, respectively, which can be
determined from the cutting process geometric parameters, then a unit step function that determines whether the tooth is in or out
of cut. g(4j), can be determined as
(  
g 4j ¼ 1 if 4entry  4j  4exit
  [27]
g 4j ¼ 0 if 4j < 4entry or 4j > 4exit

When h(4j) > 0, it means that the cutting tooth number (j) is in contact with the arc being machined, the cutting force
components on it can be determined in terms of this chip load. If h(4j) < 0, it means that this tooth is out of cutting at the given
angular position, and the force on it must be set as zero. This also expresses the basic nonlinearity of the dynamic cutting process.

11.03.1.7 Milling Process Simulation System Using the Predictive Models

Using the predictive time domain model for dynamic cutting forces and chatter in milling, a windows-based simulation system
MillSimu has been developed, which reads the input data of cutting conditions, machine tool characteristics, work material, tool
geometry, and other related parameters, and outputs the cutting status and progress of chatter in milling. The flow chart of the
simulation system is shown in Figure 10.
The simulation is carried out for each step of the angular increment of cutter rotation. The angular increment can be adjusted to
ensure that the corresponding time step is less than 1/10 of the lowest natural period of the machine tool system. At every
simulation step, the angular position of each tooth is determined, then for each tooth which is currently engaged in cutting; the
corresponding dynamic chip load is calculated, taking into account the modulation effects of present and previous cutting tool–
workpiece vibration. The cutting force components for this tooth are determined using the predictive milling force model. These
forces are summed together with the cutting force components for other teeth engaged in cutting at the same time to give the total
forces for this simulation step. With the forces obtained, the instantaneous vibration displacements of the cutting tool in X and Y
direction can be determined by solving the differential equations of the system using numerical methods. The process continues
until the cutter complete one rotation, and the period of milling to be simulated is determined by the number of cutter rotations to
be simulated, which is user-definable. The simulation system outputs the predicted dynamic milling force components and tool

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
36 Continuum Micromechanics and Molecular Dynamics Simulation

Start

Input:: cutting conditions; machine

tool characteristic parameters; cutter
parameters; tool and work material

Setup initial values;

First simulation time step: i = 0

First tooth: k = 0

Calculate the current angular position of this

h

No This tooth in cutting?

Yes
Calculate the instantaneous chip thickness;
calculate the cutting force on this tooth;

Yes
Next tooth k+ < Teeth Number?

No

Calculate the current chatter vibration

di l t

No
i++;
Next

Yes
Yes
Next Continue?
No
Output: forces, vibration, etc.

Figure 10 Flow chart of the simulation system.

displacements in the feed and normal directions against the cutting time, in both graphic and numerical formats, and with both the
maximum and peak-to-peak values presented. A typical window for the simulation output is shown in Figure 11.

11.03.1.8 Experimental Verification of the Simulation System

The simulation system has been verified for both face milling and end milling operations, with the simulation results compared
with experimental results. For example, the end milling results are shown next.
The experimental end milling tests were performed on a Deckel Maho DMU-50V computer numerical control (CNC) machine.
The cutter used was a four-fluted, 30 helical carbide square end mill with 12 mm diameter, and the measured cutter runout values
were 0.032, 0.040, 0.038, and 0.029 mm. A series of cutting conditions were used, including slot milling, half immersion up
milling, half immersion down milling, as well as up milling and down milling with 2.5 mm radial depth of cut. Two types of feed
per tooth, 0.08 and 0.06 mm, respectively, and various axial depth of cut were used for all cases. The spindle speed included 800,
1600, 2400, 3200, 4000, and 4800 rpm.
The dynamic parameters of the end mill were determined from modal tests by using the impact method, which are listed in
Table 1.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Continuum Micromechanics and Molecular Dynamics Simulation 37

Figure 11 A typical output window from the simulation system.

Table 1 Dynamic parameters of the end mill

Direction Frequency (Hz) Damping (%) Stiffness (N mm1)

X 1024 7.3 35 448

Y 1040 5.6 25 774

The machine tool used in the experiment was a simultaneous five-axis CNC milling machine, where the worktable is capable of
rotating in three directions during cutting to enable the five-sided machining. This feature, however, decreases the stiffness and thus
stability of the worktable. During the cutting experiment under larger depth of cut, significant vibration between tool and workpiece
was observed. Therefore, the effects of workpiece vibration should be included in modeling the dynamic milling process. The
measured dynamic parameters of the workpiece are listed in Table 2.
For up milling with a radial depth of cut of 2.5 mm, a cutting speed of 4000 rpm, and a feed per tooth of 0.08 mm, the
predicted critical axial depth of cut was 13.1 mm using the proposed model. When the axial depth of cut was da ¼ 12 mm,
which was just below the critical axial depth of cut, no chatter occurred during the cutting test. Figure 12(a) shows the cor-
responding measured cutting forces Fx and Fy during two revolutions of the tool. The simulation results of the cutting forces and
tool vibration displacements in X and Y directions during two revolutions of the tool for this case are shown in Figure 12(b).
The calculated stability index was 1.02. It can be seen that both the predicted and experimental results were in a stable cutting
state, and the simulated cutting forces agreed well in amplitude and pattern with the measured forces. During the cutting test,
when the axial depth of cut was da ¼ 15 mm, chatter occurred, which was evident from the recorded acceleration signals as
shown in Figure 13.

Table 2 Dynamic parameters of the workpiece

Direction Frequency (Hz) Damping (%) Stiffness (N mm1)

X 180 14.1 205 908

Y 172 13.6 191 084

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
38 Continuum Micromechanics and Molecular Dynamics Simulation

Figure 12 The measured (a) and simulated (b) results for up milling, n ¼ 4000 rpm, da ¼ 12 mm.

Figure 13 Recorded acceleration signal for up milling, n ¼ 4000 rpm, da ¼ 15 mm.

Figure 14(a) shows the corresponding measured cutting forces Fx and Fy during two revolutions of the tool, indicating a fully
developed chatter. The simulation results of the cutting forces and tool vibration displacements are shown in Figure 14(b). The
calculated stability index was 1.80, which indicated the instability of the cutting condition. Hence the simulation results predicted
the occurrence of chatter.

11.03.2 Micromechanics Modeling of Cutting of Composite Materials

Micromechanics follows the rule of mixtures, applying continuum mechanics separately to the different phases of material in
a composite. This section reveals that when the workpiece material is made from two or more constituent materials with signifi-
cantly different physical or chemical properties, such as a composite material, the machining characteristics, such as the tool wear
rate, have to be modeled based on workpiece properties that have to be modeled using micromechanics.

11.03.2.1 Special Challenges in Cutting of Metal Matrix Composite

In cutting of composite materials, such as cutting of MMC materials, the workpiece material is made of two phases of material –
a soft metal material functioning as a matrix embedding a much harder material in segment or particulate as the reinforcement (see
Figure 15) for physical or mechanical properties such as wear resistance, friction coefficient, or thermal conductivity that are much
better than those of the individual materials, however, leaving special challenges to machining technology. One of the special

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Continuum Micromechanics and Molecular Dynamics Simulation 39

Figure 14 The measured (a) and simulated (b) results for up milling, n ¼ 4000 rpm, da ¼ 15 mm.

Figure 15 Microstructure of a metal matrix composite, with aluminum alloy as the matrix and silicon carbide particulates as the reinforcement.

challenges in cutting of MMCs is the significantly increased tool wear rate. The tool wear rates in cutting of MMCs have to be
modeled for tool life prediction so that machining automation can be established. In modeling of the tool wear rate in cutting of an
MMC, the two different phases of the material in the MMC have to be considered separately by following the rule of mixtures, and
then to each of which continuum mechanics can be applied.

11.03.2.2 The Rule of Mixtures

As a composite is a combined material with two phases of material, its properties are a combination of those of the material phases.
The rule of mixtures provides a theoretical upper and lower bound on the properties of the composite, including the physical and
mechanical properties such as electrical conductivity, mass or weight density, ultimate tensile strength, and hardness. As an example,
the rule of mixtures for an MMC is analyzed for the upper bound of the mass and weight densities, as shown.
For the mass density of an MMC, taking Vr as the volume fraction of the reinforcement and 1–Vr as the volume fraction of the
matrix, then the upper bound of the mass density of the MMC, rmmc, is:
rmmc ¼ Vr ðrr Þ þ ð1  Vr Þðrm Þ [28]
where rr and rm are the mass densities for the reinforcement and for the matrix, respectively.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
40 Continuum Micromechanics and Molecular Dynamics Simulation

Similarly for the weight density of an MMC, taking wr as the weight fraction of the reinforcement and 1 – wr as the volume
fraction of the matrix, then the upper bound of the weight density of the MMC, wrmmc, is:
wrmmc ¼ wr ðwrr Þ þ ð1  Vr Þðwrm Þ [29]
where wrr and wrm are the weight densities for the reinforcement and for the matrix, respectively.

11.03.2.3 Modeling of Tool Wear Rate in Relation to the Reinforcement in MMCs

Applications of composite materials are among the most important developments in materials engineering in recent years.
However, the applications of MMCs are limited by their poor machinability, which is a result of their highly abrasive nature that
causes excessive wear in carbide and even diamond cutting tools. Therefore, unless the search for an economical and efficient
machining method for MMCs is fruitful, the tremendous engineering potentials of MMCs are always overshadowed by their high
machining costs. To overcome the difficulties in machining of MMCs, the details of the tool wear mechanism in cutting of MMCs,
including correlation between the tool wear and the percentage of reinforcement in the MMCs have been studied. This section
shows the special tool wear mechanisms in cutting of aluminum–silicon carbide metal matrix composite (Al–SiC MMC) using
tungsten carbide tools, which include erosion, abrasion, and enhanced abrasion relating to a certain range of the percentage of
reinforcement in the MMCs, and a model developed for the percentage of reinforcement in the MMCs in relation to the enhanced
abrasive tool wear, based on micromechanics analyses.

11.03.2.3.1 MMC Workpiece Specimens

Al–SiC MMC workpiece specimens were cast using aluminum alloy AA2024 as the matrix and containing various percentages (2.5,
5.0, 7.5, 10.0, 12.5, and 15.0%) of silicon carbide particles of mean diameter 50 mm. The specimens were in the form of cylindrical
bars of length 178 mm and diameter 22 mm.
For the liquid phase casting, an economical method widely used in the industry was adopted. The equipment setup includes
a resistance furnace, a crucible, and a temperature controller. The casting followed the following procedure:
1. The corresponding mass of aluminum alloy was contained in a crucible. It was placed into a resistance furnace and heated to
800  C, which is above the aluminum alloy’s melting temperature of 660  C. It was then held at this temperature for 30 min to
degas and fully melt the aluminum alloy.
2. At the same time, the corresponding mass of SiC particulate of mean particle size 50 mm was poured into another crucible and
preheated at 800  C for 1 h. This was done to improve the surface wetting through desorption of the absorbed gases from the
ceramic surface, thus avoiding the formation of Al2O3, which prevents interfacial wetting (3).
3. The preheated SiC particulate was then added into the aluminum alloy matrix at 800  C. The mixture was stirred for 10 min, long
enough to obtain an even distribution of SiC in the aluminum alloy matrix but not so long for excessive formation of Al4C4 and
Al4SiC4, which could lower the weight percentage of SiC particulate reinforcement in the MMC (3).
4. After stirring, the melt was poured into the mild steel mold. The mold was preheated at 300  C to remove moisture. The pre-
heating of the mold also minimized dendritic separation and coring, which might occur as a result of different cooling rates from
core to surface when the MMC is cooled and solidified. This process ensured that the structure of the composite would be
homogeneous.
5. The casting was completed by removing the specimen from the mold on cooling.
6. After the casting was completed, the actual percentage of reinforcement in the cast Al–SiC MMC specimens was determined by
checking the SiC weight percentage in the two ends of each specimen using the acid dissolution method as shown in
Figure 16.

Al-SiC MMC HCL acid Remaining SiC

Weight Dissolution Filtering Weight

Figure 16 Schematic diagram for acid dissolution.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Continuum Micromechanics and Molecular Dynamics Simulation 41

11.03.2.3.2 Special Tool Wear Mechanisms Found from Experimental Cutting Tests
Experimental cutting tests were performed on a Leblond Makino 15 Inch Regal Precision Lathe. In the cutting tests for erosive and
abrasive tool wear mechanisms, coated carbide inserts (CNMG N308) with a tool holder of 6 rake angle and 95 approach angle
was used. In the cutting tests for correlation between tool wear and percentage of SiC in the MMCs, coated carbide inserts (SNMG
T260) with a tool holder of 8 rake angle, 45 approach angle, and 0 inclination angle were used. The tool wear measurements were
performed on an Olympus Measuring Microscope with a resolution up to 0.0001 mm. Tool wear surface and machined workpiece
surface were observed using a scanning electron microscope (SEM).

11.03.2.3.2.1 Erosive Wear of the Tools

One of the features in the cutting of Al–SiC MMCs is that the reinforcements in the MMCs work as hard particles impinging one after
another on the tool cutting edge during the cutting process. This raises a suspicion that one of the causes of the tool wear is erosion.
Studies in the field of erosion wear reported (4) that for a brittle surface impinged by solid particles, the erosion rate was found to be
small at low impingement angle q and increases to a maximum value when the impingement angle is 90 (see Figure 17). In the
context of turning of Al–SiC MMCs, the impinging particles are the SiC particles and the cutting edge forms the impinged surface. As
shown in Figure 17, the inclination angle setting l is actually the complement of the impingement angle q. Therefore, the presence
of erosion wear on the tool in cutting of Al–SiC MMCs can be verified if the increase in the cutting tool’s inclination angle setting l
(equivalent to the decrease in impingement angle) results in the decline of the tool wear.
Turning experiments were conducted to test the presence of erosion wear of the tools. Cutting tools of different inclination
angles, 0 , 10 , 20 , 30 , and 40 , were used in cutting with an equal workpiece material removal. An Al–SiC (15% SiC by weight)
MMC workpiece was turned under the cutting conditions of cutting speed 64.7 m min1, feed rate 0.1 mm rev1, and depth of cut
0.5 mm. The tool nose wears (representing the maximum flank wears) were measured plotted against inclination angle setting. Each
test was repeated once to ensure reproducibility of the results. The results as shown in Figure 18 indicate that the tool flank wear
decreases as the inclination angle increases. Therefore, it can be concluded that erosion caused by impingement of SiC particles on
the tool cutting edge is one of the mechanism of tool wear in the cutting of Al–SiC MMCs.

11.03.2.3.2.2 Abrasive Wear of the Tools

A microscopic analysis to verify the presence of abrasive wear on the tool in the cutting of Al–SiC MMCs was conducted. The
abrasive wear to be verified was of two types: two-body abrasive wear and three-body abrasive wear. Two-body abrasive wear is
caused by rubbing of a softer surface by a hard rough surface while three-body abrasive wear is caused by hard particles entrapped
between two sliding surfaces (4). The presence of excessive grooves that were in parallel to the cutting direction on the worn tool’s
flank face indicates that two-body abrasion was the mechanism dominating the tool wear. As reported by Fang et al. (5), a char-
acteristic of three-body abrasion is that it can create either scratches at low wear rate or craters at high wear rate. The scratches and
craters found on the flank wear surface confirmed the occurrence of three-body abrasion.

11.03.2.3.2.3 Enhanced Abrasive Wear of the Tools

As one of the important findings from the study, an enhanced abrasive wear of the tools in cutting of Al–SiC MMCs was found
through examining the sequence in which different mechanisms of the tool wear, erosion, two-body abrasion, and three-body
abrasion in turn dominate the tool wear process in the turning of an Al–SiC MMC.

θ
λ

Plan view A–A Plane

Figure 17 Erosion in context of turning of Al–SiC MMC.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
42 Continuum Micromechanics and Molecular Dynamics Simulation

Nose wear against inclination angle

1.2
1.1

Nose wear (mm)

1
0.9 Nose wear (mm)

0.8
0.7
0.6
0 5 10 15 25 40
Inclination angle

Figure 18 Verification chart for erosive wear in turning of an Al–SiC MMC.

The sequence of the tool wear dominated by different mechanisms can be examined by studying the path of a SiC particle in an
Al–SiC MMC workpiece during a turning process. When the workpiece is turned, the aluminum alloy matrix will be split and
sheared by the cutting edge to form a chip, as shown in Figure 19(a). When there is a SiC particle at the splitting point, the particle
will impinge onto the cutting edge. As the tool is brittle, the impingement will either initiate a crack or propagate an existing crack
without causing much plastic deformation on the cutting edge (4). Consequently, two possible cases can take place. In the first case,
the SiC particle’s centroid is above the line of the cutting force Fc exerted by the cutting edge as shown in Figure 19(b).
As the SiC particle is much harder than the tool, it is more likely to be moved than cut or broken (6). As a result of the anti-
clockwise moment, Ma is created by the cutting force, as shown in Figure 19(b), and the SiC particle is likely to roll toward the
rake face of the cutting tool and move in the direction of the chip flow. This is the case mentioned by Quigley et al. (7), that a SiC
particle was ‘plucked’ from the workpiece. During the dislodgement of the SiC particle from the workpiece, it plows at the workpiece
surface while decreasing its plowing depth and creates a groove with narrower and shallower ending on the surface, as observed such
grooves from SEM examination of the machined workpiece.
In the second case, the SiC particle’s centroid is below the line of the cutting force Fc exerted by the cutting edge, as shown in
Figure 19(c). In this case, after impinging onto the cutting edge, the SiC particle will be pressed by the tool’s flank face such that it is
embedded into the softer aluminum alloy matrix of the Al–SiC MMC. While it embedded into the matrix, it plows at the tool flank
face and creates a groove on the flank face as well as on the workpiece surface. In this way the groove created on the workpiece
surface should have a wider and deeper ending, because at the end of its sliding path the SiC particle will finally reside at the end of
the groove that it created on the workpiece surface. Such grooves were found on the machined workpiece surface from SEM
examination of the machined workpiece.
The two cases described concern two-body abrasion. If in the path of being dislodged or embedded, the SiC particle hits
a neighboring SiC particle, its movement will be impeded. The interfering SiC particle, which is harder than the carbide cutting tool,
will provide a very high normal stress for the impeded SiC particle to plow at the cutting tool’s surface. This interaction between the
SiC particles is thought to be a secondary mechanism that enhances the two-body abrasion and causes excessive wear to the cutting
tool. This hypothesis was proven by the findings from an analysis of the interference between the particulate particles in the MMCs
in relation to the percentage of reinforcement, together with machining experimental results showing that the tool wear becomes
very severe after the SiC weight percentage reaches a critical value and that the inter-particle spacing in MMC is no longer large
enough for the impacted SiC particle to be embedded or dislodged without interference by a neighboring particle. This
phenomenon is explained in the next section.

11.03.2.3.3 Critical Percentage of Reinforcement Causing Enhanced Abrasive Wear

In general, the tool wear increases with the increase in SiC weight percentage in Al–SiC MMC, because the hardness of MMC
increases as its SiC weight percentage increases. On top of this, according to the enhanced abrasion mechanism for tool wear as
described in Section 11.03.2.3.2.3, if in the path of being dislodged or embedded, a SiC particle hits a neighboring SiC particle, its
movement will be impeded and the interfering SiC particle, which is harder than the carbide cutting tool, will provide a very high
normal stress for the impeded SiC particle to plow at the cutting tool surface and causes excessive wear to the cutting tool. The
critical SiC weight percentage in Al–SiC is hence the value which the inter-particle spacing in MMC is no longer large enough for the
impacted SiC particle to be embedded or dislodged without interference of a neighboring SiC particle.
An analysis to derive the critical SiC weight percentage can be conducted as follows. In turning of an Al–SiC MMC workpiece,
a SiC particle (idealized as a sphere) may impinge on the major cutting edge, nose, or miner cutting edge of the tool, as shown in

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Continuum Micromechanics and Molecular Dynamics Simulation 43

Plan view Plane A–A view

(a)

Plan view A–A Plane view

(b)

Plan view A–A Plane view

(c)

Figure 19 (a) Splitting of workpiece in turning of Al–SiC MMC; (b) dislodgement of SiC particle from the machined MMC; and (c) embedment of
a SiC particle into the machined MMC.

Figure 20(a). It is assumed that the maximum particle displacements in the X and Z directions are r, the radius of the impacted SiC
particle that occurs when the impact point is near the SiC particle’s centroid. The particle displacement in the Y direction is
a translation of kr in which k is a constant. Figure 20(b) shows the assumed particle displacement. The typical path, s, taken by an
embedded SiC particle is shown in Figure 20(c). Imagine that the cast Al–SiC MMC workpiece is constituted by N number of evenly
distributed elementary units as shown Figure 20(d).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
44 Continuum Micromechanics and Molecular Dynamics Simulation

Plan view Side view

(a)

Y Y

kr kr

M Mc

Z r X r

Major cutting edge Minor cutting edge

Particle displacement in Y–Z plane Particle displacement in X–Y plane

(b)

s
Y
Y
Z
2r
X
SiC

SiC Z
(c)

(d)

Figure 20 (a) Possible locations where SiC particles impinge onto cutting tool; (b) particle displacement after impingement; (c) typical path taken
by an embedded SiC particle; and (d) elementary unit that constitutes the MMC.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Continuum Micromechanics and Molecular Dynamics Simulation 45

For the impacted SiC particle to be embedded without interference of a neighboring SiC particle, the dimensions of the
elementary unit, X, Y, Z, must be at least:
X ¼ 2r þ x 0 X ¼ 3r [30]

Y ¼ 2r þ y0Y ¼ ð2 þ kÞr [31]

Z ¼ 2r þ z 0 Z ¼ 3r [32]
The total volume of SiC in the MMC is, therefore,
4
VSiC ¼ N pr 3 [33]
3
The total volume of the MMC is
Vmmc ¼ NðX  Y  ZÞ0Vmmc ¼ ½9ð2 þ kÞNr 3 [34]
The volume ratio of SiC to MMC,
VSiC ð4=3ÞNpr 3 4p
VSiCmmc ¼ ¼ 0VSiCmmc ¼ [35]
Vmmc 9Nð2 þ kÞr 3 27ð2 þ kÞ
The SiC weight ratio,
WSiC 4prSiC
WSiCmmc ¼ 0WSiCmmc ¼ [36]
Wmmc 27ð2 þ kÞrmmc
where WSiC is the weight of SiC particles in the MMC, Wmmc is the weight of MMC, rSiC is the density of the SiC particles, rmmc is the
density of the MMC, and k is a constant.
Applying the law of mixtures, the theoretical density of the MMC, rmmc, is given by eqn [28], as
rmmc ¼ VSiCmmc ðrSiC Þ þ ð1  VSiCmmc ÞðrAl Þ [37]
and the equivalent expression in SiC weight ratio terms is given by eqn [29], as
   
rmmc 1 ¼ WSiCmmc rSiC 1 þ ð1  WSiCmmc Þ rAl 1 [38]

Substituting eqn [38] into eqn [36],

4prSiC
WSiCmmc ¼ [39]
ð54 þ 27k  4pÞrAl þ 4prSiC
The implication of this expression is that in turning of any particulate-reinforced MMC with cutting tool that is softer than the
reinforcement, there exists a critical reinforcement weight ratio, above which there will be a transition of the tool wear from a range
of relatively low tool wear to a range of very high tool wear. These critical reinforcement weight ratios, WRmmc, are functions of the
density of the matrix-reinforcement system, rR and rM, and a constant k that is independent of the reinforcement particle size. The
generalized expression for the critical reinforcement weight ratio is:
4prR
WRmmc ¼ [40]
ð54 þ 27k  4pÞrM þ 4prR
To verify the above analysis, turning tests on the effect of the Al–SiC MMC’s SiC weight percentage on the tool wear was per-
formed. Six Al–SiC MMC workpieces, each with SiC weight percentage of 2.5, 5.0, 7.5, 10.0, 12.5, and 15.0%, were turned with
equal material removal at cutting speed 65 m min1, feed rate 0.1 mm rev1, and depth of cut of 0.5 mm. The measured flank wears
were plotted against the SiC weight percentage of the respective workpiece. Keeping other machining parameters unchanged, the
turning tests were repeated at higher and lower cutting speeds of 88.61 and 24.26 m min1, as well as higher and lower feed rates of
0.2 and 0.05 mm rev1, to examine the sensitivity of the results obtained to the cutting speed and feed rate.
The results obtained as shown graphically in Figures 21 and 22 suggest that the tool wear increases with the increase in SiC
weight percentage. Detailed analysis of the graph reveals that there are two tool wear regions: one with relatively low tool wear when
SiC weight percentage is below 5% and another with very high tool wear when SiC weight percentage is 7.5% and above. This
phenomenon appears in all the cutting conditions investigated, thus supporting the analysis as shown in eqns [30]–[40].
The two ranges of tool wear values corresponding to the two ranges of percentage reinforcement in the MMCs as observed in the
experimental results in Figures 21 and 22 can be well explained by using the analysis shown in eqns [30]–[40]. By substituting
rSiC ¼ 3100 kg m3 and rAl ¼ 2700 kg m3 together with the value of k determined from SEM micrograph of the machined
workpiece surface on a typical SiC particle translation (SiC particle embedded into the machined workpiece surface along the
cutting velocity direction, indicating that k is approximately 6) into eqn [39], the critical weight ratio of SiC to MMC for the
workpiece material tested is:
WSiCmmc ¼ 0:0662

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
46 Continuum Micromechanics and Molecular Dynamics Simulation

Average flank wear against SiC%

2.1

Average flank wear

1.6 Avg Flk Wr

(mm)
v = 24.259m min–1 (mm)
1.1
Avg Flk Wr
0.6 v = 64.671m min–1 (mm)
2.5 7.5 12.5
Avg Flk Wr
SiC (%) v = 88.605m min–1 (mm)

Figure 21 Turning tests with cutting speed variation.

Average flank wear against SiC%

Average flank wear

2.1
Avg Flk
1.6 Wrs = 0.05mm rev–1 (mm)
(mm)

1.1 Avg Flk

Wrs= 0.1mm rev–1 (mm)
0.6
12.5
2.5

7.5

Avg Flk
Wrs=0.2mm rev–1 (mm)
SiC (%)

Figure 22 Turning tests with feed rate variation.

As this critical SiC weight percentage, 6.62% falls between the observed transition range of 5 and 7.5%, the hypothesis that
abrasive wear of the tools is enhanced by SiC particle interference in Al–SiC MMC when it is turned is therefore verified.

11.03.3 Molecular Dynamics Modeling and Simulation of Ductile Mode Cutting of Brittle Materials

When the workpiece materials are brittle rather than ductile, such as tungsten carbide, glass, and silicon wafer material, to achieve
ductile chip formation in cutting, the workpiece material in the cutting region has to be transformed to a metallic phase. The
material phase transformation is achieved in the cutting process within the cutting region, and is governed by molecular dynamics
(MD) that is based on interatomic energy potential functions. Therefore, ductile mode cutting of brittle materials has to be modeled
based on MD. As an example, MD modeling and simulation of ductile mode cutting of silicon wafers are briefly described in this
section.

11.03.3.1 Transition of Ductile Mode to Brittle Mode in Cutting of Brittle Materials from Nanoscale to Micro/Millimeter-Scale
Figure 23 shows the machined surfaces from groove cutting of tungsten carbide, indicating the transition of chip formation from
ductile mode to brittle mode as the undeformed chip thickness increased from zero to a critical value against the radius of the tool
cutting edge.
In cutting of glass, Figure 24 shows the machined workpiece surfaces from cutting of soda-lime glass, indicating that to achieve
ductile mode cutting of glass, the undeformed chip thickness has to be small, on nanometer scales, and smaller than that of tool
cutting edge radius.
In cutting of silicon wafer, Figure 25 shows that to achieve ductile mode cutting of silicon wafer, the undeformed chip thickness
has to be smaller than that of the tool cutting edge radius, in which the cutting edge radius was on nanoscale.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Continuum Micromechanics and Molecular Dynamics Simulation 47

A
300 μm

(a)

(b) (c)

Figure 23 SEM micrographs of the machined workpiece surfaces and chips from grooving of tungsten carbide with the depth of cut increased
from zero to a large value. (a) The micrographs for the machined surfaces before and after the depth of cut reached a critical value. (b) The contin-
uous chips generated from the grooving before the depth of cut reached a critical value. (c) The particle chips generated from the grooving after the
depth of cut reached a critical value.

Figure 26 shows that in cutting of silicon wafer, when the radius of the tool cutting edge was increased in exceeding a critical
value, about 800 nm, even the undeformed chip thickness was smaller than the cutting edge radius, and the chip formation mode
changed from ductile to brittle.
Two special conditions for ductile mode cutting can be observed from cutting test results as shown in Figures (23)–(26): the
higher the brittleness of the workpiece material, the smaller the tool cutting edge radius has to be for ductile mode chip formation,
on nanoscales; and the undeformed chip thickness must be smaller than the cutting edge radius, on nanoscales. These two special

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
48 Continuum Micromechanics and Molecular Dynamics Simulation

(a) Fracture free surface (b) Fractured surface

Figure 24 SEM micrographs of the machined workpiece surfaces in cutting of soda-lime glass with a polycrystalline diamond tool of cutting edge
radius 650 nm. (a) Fracture-free surface when the undeformed chip thickness was 63 nm. (b) Fractured surface when the undeformed chip thickness
was 319 nm.

Figure 25 Chips formed in cutting of Si wafer using a diamond tool (cutting edge radius ¼ 45 nm) at different undeformed chip thickness, ac;
continuous chips were from ductile chip formation and fractured particle chips were from brittle chip formation.

conditions provide an extremely large hydrostatic pressure in the chip formation region, which is so large that in the cutting region
the original phase of the workpiece material is transformed to a metallic phase so that the chip formation is in a ductile mode as in
cutting of ductile material. In modeling such a chip formation process, the volume of the involved workpiece material in the chip
formation zone is in only a few atomic layers governed by the interatomic energy potential, the continuum mechanics dose not
apply, and MD has to be used.

11.03.3.2 Molecular Dynamics Modeling of Cutting of Silicon

MD is based on material interatomic energy potential functions. In this section, the crystal structure of a silicon workpiece and
modeling of nanoscale cutting of silicon with a single-point cutting tool based on MD are briefly described.

11.03.3.2.1 The Crystal Structure of a Silicon Workpiece

The monocrystalline silicon workpiece is crystallized in the diamond lattice as shown in Figure 27, which is a special cubic crystal
structure. In the face-centered cubic crystal, atoms are located at the corners and at the face centers. Based on the face-centered cubic
crystal, the diamond lattice has four further atoms located inside the lattice. Their centers can be found on the body diagonals at
a height of 1/4 and 3/4 , respectively, measured from the base plane. In a diamond lattice, the interatomic forces are developed by the
covalent bond, and the lattice constant a ¼ 0.543 nm.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Continuum Micromechanics and Molecular Dynamics Simulation 49

Figure 26 The machined silicon workpiece surfaces from cutting using a range of tool edge radius, with the undeformed chip thickness kept
smaller than the tool edge radius, showing an upper bound value of 717 nm for the tool edge radius for ductile mode chip formation that generates
fracture-free surface.

Figure 27 The diamond crystal lattice of silicon. (a) Spatial illustration with covalent bonding and (b) projection view.

11.03.3.2.2 Molecular Dynamics Modeling of Nanoscale Cutting of Silicon with a Signal-Point Diamond Tool
As an example, a three-dimensional MD model was developed for nanoscale ductile mode cutting of silicon, as shown in Figure 28,
where (a) shows a schematic diagram of the MD model, and (b) shows an output of the MD simulation system.
In the model, the positions of silicon atoms are arranged according to the crystal structure of silicon. The dimension of the
workpiece is 30a  20a  4a. According to the solid-state physics, every atom is vibrating around the equilibrium position at some
temperature. The velocities of all the atoms should satisfy the Maxwell–Boltzmann distribution (8). The simulation is conducted at
an absolute temperature T ¼ 293 K. At this temperature, the initial velocities of all the atoms are randomly assigned according to the
Maxwell–Boltzmann distribution.
The workpiece atoms comprise three different atoms: boundary atoms, Newtonian atoms, and thermostat atoms. The boundary
atoms are fixed in positions to reduce the boundary effects, avoid rigid body motion of the workpiece, and maintain the proper
symmetry of the lattice. The motion of the Newtonian atoms is determined by Newton’s equation of motion. The thermostat atoms,
which are used to simulate the heat conduction, are arranged to surround the Newtonian atoms to make the boundary temperature
close to ambient temperature. The thermostat atoms’ movement also should comply with Newton’s equation of motion, but their
velocities need to be scaled at some integrating steps, such as 50 steps. To reduce the boundary effect, the model scale should be
large. As a result, the computing time would be enormous. To avoid this problem, a periodic boundary condition is maintained
along the direction perpendicular to the cross section, as shown in Figure 28(a).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
50 Continuum Micromechanics and Molecular Dynamics Simulation

Figure 28 A three-dimensional MD model for simulation of nanoscale ductile mode cutting of silicon with a signal-point diamond tool.
(a) A schematic of the MD model and (b) an output of the MD simulation system.

The diamond cutting tool is assumed to be rigid, in which the relative positions of tool atoms remain unchanged in the
simulation. As the tool cutting edge radius cannot be ignored, the carbon atoms are arranged along an arc with the radius of the
cutting edge. The cutting velocity is set as 20 m s1.

11.03.3.3 The Simulation System Using the Molecular Dynamics Model

Based on the MD model as described in Section 11.03.3.2.2, an MD simulation system was developed. In the system, the inputs are
given to initialize the parameters. After computing, the results can be output. The results can include displacement, velocity, force,
and other information of all the atoms. Before simulation, the model needs to be relaxed for about 100 time steps to reach
equilibrium. Figure 29 shows a flow chart of the MD simulation system.

11.03.3.4 Validation of the MD Simulation System

The MD simulation system is validated by using it to simulate nanoscale cutting of monocrystalline silicon under different
combinations of tool edge radius and undeformed chip thickness for the chip formation zone stress conditions in relation to
workpiece material phase transformation and the consequent the chip formation modes, ductile or brittle, and compared the results
with the experimental results as shown in Section 11.03.3.1.
First, cutting with under conditions that the tool cutting edge radius R is 3.5 nm and the undeformed chip thickness is 2.8 nm
(ac < R) is simulated. The workpiece material in the chip formation region is observed as shown in Figure 30, in which the tool
cutting edge radius was larger than the workpiece undeformed chip thickness, resulted in extremely large compressive stress in the
chip formation zone, transforming the monocrystalline silicon (see insert (a)) into amorphous (see insert (b)). The interatomic
bond length of the silicon workpiece material in the undeformed zone is 2.35 Å, which is consistent with the theoretical value for
covalent bonding length between the atoms in silicon, whereas in the chip formation zone, the atomic structure of the silicon
workpiece material has been changed such that the interatomic bond lengths vary in a wide range, as shown in Figure 31. The values
of the interatomic bond lengths in the chip formation zone had two peaks, at 2.455 and 2.61 Å, respectively, indicating the silicon
phase transformation from diamond cubic structure to b silicon in the chip formation zone. The b phase of silicon is a metallic
body-centered tetragonal structure and can lead to plastic deformation. Moreover, in both b silicon and amorphous phase silicon in
the chip formation zone, nearly all the bond lengths became longer than those of original monocrystalline silicon. According to Gao
et al. (9), for pure silicon, the hardness is inversely proportional to the interatomic bond length. Therefore, slight increases in the
bond lengths will greatly soften the silicon atom groups. Since nearly all the bond lengths of the silicon material in the chip
formation zone became longer than the normal bond length (see Figure 31), the material in the chip formation zone became softer
than the original silicon. As a result, the silicon workpiece material in the chip formation zone was deformed in a form of plastic
flow–ductile mode chip formation. The silicon phase transformation from crystalline to amorphous in the chip formation zone

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Continuum Micromechanics and Molecular Dynamics Simulation 51

Figure 29 Flow chart of the MD simulation system.

Figure 30 Under the cutting conditions of tool cutting edge radius R is 3.5 nm and the undeformed chip thickness is 3.2 nm (ac < R), the silicon
workpiece material phase is transformed from monocrystalline in the undeformed zone (see insert (a)) to amorphous in the chip formation zone (see
insert (b)).

could be caused by the extremely high hydrostatic pressure in the zone, given by the tool cutting edge. According to the simulation
results, the average hydrostatic pressure was around 12 GPa, which is high enough for the phase transformation (10,11). This
explains the observed phenomenon as shown in Figure 26, in which ductile mode cutting of silicon was achieved when the
undeformed chip thickness was smaller than the cutting edge radius thus resulted in extremely high hydrostatic pressure on the chip
formation zone.
Further, crack initiation in nanoscale cutting of monocrystalline silicon is simulated using the MD simulation system. The tool
cutting edge radius R is fixed at 4.0 nm, and three cuts are performed, at the undeformed chip thickness of ac < R (ac ¼ 3.2 nm),

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
52 Continuum Micromechanics and Molecular Dynamics Simulation

80

70 Undeformed workpiece material

60

Distribution frequency (%)

50

Chip formation zone

40
(at cutting distances:
6.0, 6.9,
30
7.8, 8.7,
9.6 nm)
20

10
2.35 2.455
2.61
0
2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9
Interatomic bond length ( Å)

Figure 31 Comparison of distribution frequency of interatomic bond length in the undeformed silicon workpiece material with that in the chip
formation zone at different cutting distances.

ac ¼ R (ac ¼ 4.0 nm), and ac > R (ac ¼ 4.5 nm), respectively. It can be seen from Figure 32(a) that when the undeformed chip
thickness was smaller than the cutting edge radius, the largest deformation took place at the surface of the workpiece. However,
when the undeformed chip thickness was equal to or larger than the cutting edge radius, there was a peak deformation zone in the
chip formation zone corresponding to the connecting point of tool edge arc and the rake face, as shown in Figure 32(b) and 32(c),
respectively. It can be seen that when the undeformed chip thickness was equal or larger than the cutting edge radius, there was
a workpiece material deformation peak in the chip formation zone, aligning along the direction of crack initiation. These simulation
results are consistent with the cutting experimental results as shown in Figure 25, and can be explained by considering the stress

Figure 32 In cutting of monocrystalline silicon, the workpiece material deformation in the chip formation zone when the cutting edge radius
R ¼ 4.0 nm and the undeformed chip thicknesses were (a) ac ¼ 3.2 nm, (b) ac ¼ 4.0 nm, and (c) ac ¼ 4.5 nm.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Continuum Micromechanics and Molecular Dynamics Simulation 53

Figure 33 Two different chip formation modes. (a) Ductile mode; (b) Brittle mode.

conditions in the workpiece material in front and next to the peak deformation zone in the chip formation zone, as shown in
Figure 33, where the sxx and syy represent the normal stresses in the X and Y directions. Figure 34 shows the variations of the
stresses, where the simulation steps indicate tool advancing against the workpiece. It can be seen that as the undeformed chip
thickness increased from smaller to larger than the tool edge radius, sxx was positive (compressive stress) for all cases of undeformed
chip thickness. The stress in the direction nearly perpendicular to the direction of crack initiation, syy, however, decreased as the
undeformed chip thickness increased. Especially, as the undeformed chip thickness was larger than the cutting edge radius, syy
decreased to negative (the stress state changed from compressive to tensile).
As shown in Figure 33(a), when the undeformed chip thickness was smaller than the tool cutting edge radius, the stress syy was
compressive and no peak deformation zone occurred in the chip formation zone. Under this condition, the workpiece material next
to the chip formation zone was purely compressive stressed. Therefore, there was no crack initiation.
In contrast, as shown in Figure 33(b), when the undeformed chip thickness was equal to or larger than the tool cutting
edge radius, there was a peak deformation zone generated in the chip formation zone. Unlike ductile materials, which have
much larger fracture strains, silicon is a kind of brittle material with extremely small fracture strain. As a result, with the
tensile stress in the Y direction and compressive stress in the X direction, the workpiece material will be split, initiating
a crack in the material next to the peak deformation zone, forming a crack initiation zone next to the peak deformation
zone. In this crack initiation zone, a crack would be formed that propagates along the direction of crack initiation – from the
connecting point of tool rake face and edge arc to the deformation peak of chip formation. This explains the phenomenon
that in nanoscale cutting of silicon when the undeformed chip thickness is larger than the tool cutting edge radius, crack
initiation and propagation occur and the chip formation is in a brittle mode, such as the cutting experimental results shown
in Figure 25.

11.03.3.5 Molecular Dynamics Simulation Study of the Characteristics in Nanoscale Cutting of Silicon
With the MD simulation system developed, the major characteristics of nanoscale cutting of silicon can be studied in detail by
simulation. A brief description of the MD simulation study of the effects of tool edge radius on the cutting forces, the upper bound
of cutting edge radius for ductile chip formation, the dynamic super-hard particles in the chip formation, and the groove wear of
diamond tool in ductile mode cutting of monocrystalline silicon is given in this section.

11.03.3.5.1 Effects of Tool Edge Radius on the Cutting Forces

An output of the MD simulation of nanoscale ductile mode cutting of monocrystalline silicon is shown in Figure 35, where ac is the
undeformed chip thickness, ds denotes the spring-back thickness of the finished surface, and da denotes the affected zone thickness
of the tool edge, respectively. The simulated results are shown in Table 3, where R denotes the radius of the tool cutting edge, and Fc
and Ft denote the cutting force in the cutting direction and the thrust force perpendicular to the cutting direction, respectively.
Figure 36 shows the MD simulated cutting forces versus cutting edge radius. Three obvious features can be observed:
1. The thrust force is much larger than the cutting force during cutting, which is different from that in conventional cutting, where
the cutting force is generally greater than the thrust force and the tool cutting edge radius is far smaller than the undeformed chip
thickness.
2. The cutting force exhibits no significant change as the tool edge radius increases with the undeformed chip thickness fixed.
3. Unlike the cutting force, the thrust force changes significantly. The thrust force obviously increases as the cutting tool edge radius
is increased.
Similar trends were observed in actual cutting experiments, in which four diamond tools of different cutting edge radii were
applied and cuts in ductile mode were conducted (feed rate was 300 nm rev1 and workpiece rotation rate was 1000 rpm). The
cutting edge radii were 52.7, 71.10, 97.00, and 110.10 nm, respectively, and the undeformed chip thickness was fixed at 9.587 nm.
The undeformed chip thickness was set to be smaller than the cutting edge radius to ensure ductile mode cutting conditions.
Figure 37 shows the mean cutting forces versus cutting edge radius.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
54 Continuum Micromechanics and Molecular Dynamics Simulation

12

10

(GPa)
8

xx

ac = 3.2 nm
ac = 4.0 nm
4 ac = 4.5 nm

2
350 000 400 000 450 000

Si m u la tio n step s

(a)

0
(GPa)

–2
yy

–4
ac = 3.2 nm
ac = 4.0 nm
–6 ac = 4.5 nm

–8
350 000 400 000 4500 000

S im u la tio n ste p s

(b)

Figure 34 The variations of normal stresses with varying the undeformed chip thickness against the tool cutting edge radius R ¼ 4.0 nm. (a) sxx;
(b) syy.

11.03.3.5.2 The Upper Bound of Cutting Edge Radius for Ductile Chip Formation
As shown in Figure 26, in cutting of silicon there is an upper bound for the cutting edge radius, above which the chip formation
mode is brittle even the undeformed chip thickness is smaller than the cutting edge radius. This can be explained through MD
simulation of the stress conditions in the chip formation.
A series of MD simulations have been carried out under different cutting edge radii, which were 2.5, 4.0, and 6.0 nm,
respectively. In the simulations, the ratio of undeformed chip thickness to the cutting edge radius was fixed at 0.8, so that the
corresponding undeformed chip thicknesses are 2, 3.2, and 4.8 nm, respectively. This cutting condition makes it possible that
the chip formation is in ductile mode. The working environment temperature was set at 293 K and the cutting speed was set
at 20 m s1.
Figure 38 shows a snapshot of the MD simulation of a nanoscale cutting process. The normal stress along the Y direction syy in
the stress-calculating zone as shown in Figure 38 has been calculated. Figures 39–41 show the normal stress syy distributions in the

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Continuum Micromechanics and Molecular Dynamics Simulation 55

ac Chip
Cutting tool
ds

da

Workpiece

Figure 35 An output of the MD simulation of nanoscale ductile cutting.

Table 3 The MD simulated results for R varying from 2.5 to 5 nm (ac ¼ 2 nm)

R (nm) Fc (109 N) Ft (109 N) ds (nm) da (nm)

2.5 70 130 0.6198 1.4394

3.0 83.1 188.8 0.6268 1.7990
4.0 102.4 226.1 0.5643 1.9773
5.0 100.1 260.5 0.5940 2.2899

260

240

220

200 Fc
Ft
N)

180
–9
F o rce (1 0

160

140

120

100

80

2 .5 3 .0 3 .5 4 .0 4 .5 5 .0

Radius (nm)

Figure 36 The MD simulated cutting forces varying with cutting edge radius.

stress-calculating zone under different cutting conditions, R ¼ 2.5 nm and ac ¼ 2.0 nm, R ¼ 4.0 nm and ac ¼ 3.2 nm, and R ¼ 6.0 nm
and ac ¼ 4.8 nm, respectively. From these figures, it can be seen that in most areas of the stress-calculating zone, syy is compressive
stress and only in a small area it is tensile stress (the positive value represents compressive stress and the negative value tensile
stress). It also can be seen that this small area with tensile stress borders on the interface of plastic and elastic deformation zones as
shown in Figure 32. Moreover, the maximum value of tensile stress syy will increase and the location of the maximum tensile stress
syy will be nearer to the interface of plastic and elastic deformation zones.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
56 Continuum Micromechanics and Molecular Dynamics Simulation

2 .0

1 .8

1 .6

1 .4

1 .2
F o rc e (N ) Fc
1 .0
Ft
0 .8

0 .6

0 .4

0 .2

0 .0
135 140 145 150 155 160 165 170

Radius (nm)

Figure 37 Experimental cutting forces varying with the cutting edge radius.

Figure 38 A snapshot of the MD simulation of nanoscale cutting process.

Figure 39 The normal stress syy distribution in the stress-calculating zone when R ¼ 2.5 nm and ac ¼ 2.0 nm.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Continuum Micromechanics and Molecular Dynamics Simulation 57

Figure 40 The normal stress syy distribution in the stress-calculating zone when R ¼ 4.0 nm and ac ¼ 3.2 nm.

Figure 41 The normal stress syy distribution in the stress-calculating zone when R ¼ 6.0 nm and ac ¼ 4.8 nm.

11.03.3.5.3 Dynamic Hard Particles in the Chip Formation Zone

As shown in Figure 31, in the chip formation zone of ductile mode cutting of silicon, while the majority of the interatomic
bond lengths are larger than the original bond length, 2.35 Å, a very small percentage of the bond lengths are shorter or much
shorter than 2.35 Å. It has been indicated that for pure silicon, the hardness is inversely proportional to the interatomic bond
length. Therefore, the shorter the bond lengths, the harder the silicon atom groups with these bond lengths. For example,
when the bond length is shortened from 2.35 Å to 2.05 Å, the hardness will increase by 1.85 times. As a result, the silicon
atom groups with the shorter bond lengths could have acted as ‘hard particles’ in the chip formation zone, which have
hardness of a few times harder than the original hardness of silicon. Since these silicon atom groups with shorter bond lengths
appear only in the chip formation zone under dynamic high hydrostatic pressure, they could be called ‘dynamic hard
particles.’ Such ‘hard particles’ could be observed in the chip formation zone from the MD simulation results, as shown in
Figure 42, where the ‘hard particles’ have atoms with bond lengths shorter than 2.30 Å (smaller than the normal bond length
of 2.35 Å) are marked. The actual existence of such dynamic hard particles and the level of the hardness can observed from
a special groove wear on the tool flank face in ductile mode cutting of silicon with single crystalline diamond tools, as
described in the next section.

11.03.3.5.4 Groove Wear of Diamond Tool in Ductile Mode Cutting of Silicon

Li et al. (12) studied the diamond tool wear characteristics in nanoscale ductile mode cutting of monocrystalline silicon and the
effects of tool wear on the chip formation mode. In that study, the cutting tool was natural diamond with a perfect lattice. It was
shown that in nanoscale ductile mode cutting of monocrystalline silicon wafers, micro/nanogroove wear could be observed on the
cutting tool flank, forming sub-cutting edges of much smaller radii on the main cutting edge, as shown in Figure 43. In conventional
cutting processes, such as cutting of metallic alloys, the groove wear of tool flank face is usually caused by abrasion, in which hard
particles (carbide, nitride, or oxide) in the workpiece material abrade the tool face and form grooves on it. However, since
a monocrystalline silicon body is supposedly free of hard particles, and the diamond cutting tool material is much harder than the

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
58 Continuum Micromechanics and Molecular Dynamics Simulation

Figure 42 Three-dimensional representation of the chip formation zone having atom groups with shortened interatomic bond lengths (the broken
lines between the atoms mark the bond lengths shorter than 2.30 Å), showing the ‘dynamic hard particles’ in the chip formation zone.

Main cutting edge

Nano/microgrooves

Flank face Sub-cutting edges

(a) (b)

Figure 43 (a) SEM photographs of the tool flank face after ductile mode cutting, showing micro/nanogrooves on the diamond tool flank face; (b)
Sub-cutting edges of much smaller edge radii formed on the main cutting edge by the micro/nanogrooves at the tool flank.

monocrystalline silicon at room temperature, the generation of grooves in nanoscale cutting of monocrystalline silicon with dia-
mond tools is beyond the understanding based on conventional cutting processes.
With the MD simulation finding of the ‘dynamic hard particles’ as described in Section 11.03.3.5.3, the mechanism of the tool
flank groove wear formation in cutting of silicon with a diamond tool becomes clear. In the cutting process, the temperature in the
chip formation zone would increase, forming carbon oxides on the diamond tool surface, which make the diamond softer than the
original. Further, the increase of temperature will directly result in the decrease of the diamond material hardness. On the other
hand, as shown in Section 11.03.3.5.3, in the workpiece material, the phase transformation caused by extremely high hydrostatic
pressure creates atom groups with shortened bond lengths, forming ‘dynamic hard particles’ that are a few times harder than the
original silicon. As a result of the ‘dynamic hard particles’ plowing on the softened diamond tool flank face, micro/nanogrooves are
formed on the tool flank face.

References

1. Oxley, P. L. B. The Mechanics of Machining: An Analytical Approach to Assessing Machinability; Ellis Horwood Ltd: Chichester, 1989.
2. Tlusty, J. Automatic Supervision in Milling. In Automatic Supervision in Manufacturing; Szafarczyk, M., Ed.; Springer-Verlag: London, New York, 1994; pp 35–65.
3. Ibrahim, A.; Mohamad, F. A.; Lavernia, E. J. Particulate Reinforced Metal Matrix Composites. J. Mater. Sci. 1991, 26, 1137–1156.
4. Suh, N. Tribophysics; Prentice Hall: Englewood Cliffs, NJ, 1986.
5. Fang, L.; Zhou, Q. D.; Li, Y. J. An Explanation of the Relation between Wear and Material Hardness in Three Body Abrasion. Wear 1991, 151, 313–321.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Continuum Micromechanics and Molecular Dynamics Simulation 59

6. Tomac, N.; Tonnessen, K. Machinability of Particulate Aluminium Matrix Composites, 41/1/1992. Ann. CIRP 1992, 41, 55–58.
7. Quigley, O.; Monaghan, J.; Reilly, P. O. Factors Affecting the Machinability of an Al/SiC Metal Matrix Composite. J. Mater. Process. Technol. 1994, 48, 21–36.
8. Haile, J. M. Molecular Dynamics Simulation: Elementary Methods; Wiley: New York, 1992; p. 65, p. 293.
9. Gao, F.; He, J.; Wu, E.; Liu, S.; Yu, D.; Liu, D.; Zhang, S.; Tian, Y. Hardness of Covalent Crystals. Phys. Rev. Lett. 2003, 91, 501–502.
10. Zhang, L. C.; Tanaka, H. On the Mechanics and Physics in the Nano-Indentation of Silicon Monocrystals. JSME Int. J. A. 1999, 42, 546–559.
11. Cheong, W. C. D.; Zhang, L. C. Molecular Dynamics Simulation of Phase Transformation in Silicon Monocrystals Due to Nano-Indentation. Nanotechnology 2000, 11, 173–180.
12. Li, X. P.; He, T.; Rahman, M. Tool Wear Characteristics and Their Effects on Nanoscale Ductile Mode. Wear 2005, 259, 1207–1214.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
This page intentionally left blank

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 11.04 Fast and Fine Tool Servo for Ultraprecision Machining
A Senthil Kumar, MR Aravind Raghavendra, WK Neo, and M Rahman, National University of Singapore, Singapore
 2014 Elsevier Ltd. All rights reserved.

11.04.1 Introduction 61
11.04.2 Design Challenges of an FTS System 62
11.04.2.1 Guiding Mechanism 62
11.04.2.2 Actuators for FTS 63
11.04.2.2.1 Piezoelectric FTS 63
11.04.2.2.2 Lorentz Force FTS 63
11.04.2.2.3 Magnetostrictive FTS 64
11.04.2.2.4 Normal-Stress Electromagnetically Driven FTS 65
11.04.2.2.5 Hydraulic FTS 66
11.04.3 Classification of Fast Tool Servo 66
11.04.3.1 Tool-Positioning System 67
11.04.3.2 Error Compensation Mechanism 71
11.04.4 Long-Stroke Fast Tool Servo 74
11.04.5 Hybrid FTS/STS 78
11.04.5.1 Experiments and Discussions 81
11.04.6 More Promising Challenges 84
11.04.7 Conclusions 86
References 86

11.04.1 Introduction

Rotationally symmetrical parts such as cylinders, spheres, and paraboloids are predominantly machined by the process of turning,
the most common practice in the manufacturing arena. Turning is performed using lathes in which the components rotate and the
tool axis translate along the component at a certain speed and feed rate, respectively. The tool mounted on to the tool-axis remains
stationary, maintaining a constant distance (z-height) throughout the travel. However, with the rapid pace of technological growth,
there is a great deal of demands for machining nonrotational symmetrical freeform surfaces with high degrees of complexity.
Basically, an increasing complexity is often associated with a loss of symmetry of the surface. With an increase in the number of
degrees of freedom needed for moving a tool to produce a surface, the number of controllable machine axes will also increase.
Therefore, these freeform surfaces can be turned on a lathe if one or more axes are capable of moving at frequencies that are several
multiples higher than the spindle’s rotational frequency. The high-bandwidth, high-acceleration axis required to turn rotationally
nonsymmetric parts and features is called a fast tool servo.
A fast tool servo (FTS) is an auxiliary servo axis that is predominantly used in conjunction with a diamond turning machine (DTM)
to generate complex freeform textured surfaces with nanometer-scale resolution. As the name clearly emphasizes, an FTS system
provides a rapid displacement to tool-tip for every rotation of the workpiece. The complexity and the types of surfaces that can be
machined greatly depend on, but are also limited by the achievable acceleration, accuracy, and bandwidth of FTS. Typically, the
range of bandwidth varies from a few hertz to tens of thousands hertz and amplitude from few microns to millimeter range. FTS
exploits the benefits of the single finishing operation of the DTM to generate microfeatures with an optical finish quality.
The important factor which makes FTS stand out is its low mass characteristics, high bandwidth operation, and extremely high
positioning resolution. Due to its reduced mass inertia, it compliments the generation of higher bandwidth features such as
complex freeform surfaces with a diamond tool to produce optical surface finishes. FTS provides the capability to rapidly machine
rotational nonaxis symmetric surfaces such as cam, toric, off-axis segment of a parabolic mirror, microstructures with sinusoidal
grids, microlens array (MLA), micropyramid array, etc., with a subnanometer surface finish. Figure 1 shows a schematic of the FTS
system mounted on a DTM.
The machine axes represent the slow tool servo (STS) which are conventionally used in machining both rotationally and
nonrotationally symmetric continuous surfaces with low frequency and high amplitude. The tool path is generated by synchro-
nizing the two machine slide axes with the spindle axes (C-axis). This process is effectively used in machining continuous surfaces
with a no inflection or steep surface slopes. But, all the above-mentioned microstructures, which have sudden change in their surface
profile, cannot be realized using the STS system. The increasing complexity of freeform surfaces requires shorter spatial wavelength
components, which drives the needs for higher bandwidth and acceleration, and better accuracy of the FTS. The FTS axis (Z0 ), as
mentioned in Figure 1, acts an individual servo axis which in conjunction with the machine axes and spindle axis (C-axis), facilitates
realizing discontinuous microstructures with high surface quality.
Over the past several decades, FTS technology has been evolved to fulfill the demands for ultraprecision accuracy and excellent
surface quality of freeform surfaces. This evolution in turn marks the tipping point on the evolution of metrology for freeform

Comprehensive Materials Processing, Volume 11 http://dx.doi.org/10.1016/B978-0-08-096532-1.01104-3 61

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
62 Fast and Fine Tool Servo for Ultraprecision Machining

Figure 1 Schematic of an FTS system.

surfaces. These new evolutions have not been fully explored to unleash the hidden potential of freeform surfaces. This new field also
brings us many new challenges for design, machining, and testing. The following sections shall mainly focus on the machining
challenges and barriers in the FTS diamond turning process.

11.04.2 Design Challenges of an FTS System

A significant amount of research effort has been devoted to FTS design over the last several decades. It is essential to design a good
FTS system for providing nanometric accuracy position. There are three main components to be taken into design consideration for
a good FTS system, which are guiding mechanisms, FTS actuator, and its corresponding control system depending on the type of
actuator.

11.04.2.1 Guiding Mechanism

Mechanical guide plays an important role in the FTS system since it provides the necessary rectilinear path to the FTS. Most of the
FTSs are normally placed into the workpiece either radially or axially. The important characteristic of the guiding mechanism is to
provide the normal displacement with any parasitic erroneous motion to the tool. In order to prevent motion loss and backlash
during high frequency actuation, it is necessary to provide a preloading to the actuator. Preloading also prevents the actuator
damage due to impact loading during machining.
Flexures are the most commonly used element in the design of precision components such as nanopositioners and microgrippers
for Hi-Tech applications including scanning probe microscopy, lithography, nanometrology, beam steering for optical commu-
nication systems, fabrication and assembly of nanostructures, handling and manipulation of micro-objects, microassembling,
machining, and in manipulation of flexible and fragile biological micro-objects such as cells and bacteria in biotechnology (1–3).
Flexure-based mechanisms are the most commonly used mechanical guiding mechanism in an FTS system. Flexure mechanisms
obtain all or most of its motion by elastic deformation of the flexures about its slender cross-section and store energy in the form of
strain energy. Flexure has various advantages which are discussed in detail in most of the literature (4–7). The main advantages due
to which flexure hinges are used in the design of FTS guiding mechanism are
l friction-free motion and theoretically no wear;
l reduced part count making the manufacturing process simpler and avoidance of assembly-prone errors;
l reduced cost, but increased precision and reliability; and
l higher off-axis stiffness to withstand tangential cutting forces.

Though the flexure-based mechanisms are delimited by their output displacement, they are extensively incorporated in appli-
cations such as FTSs, due to their high quality motion attributes which are otherwise impossible or difficult to achieve using
conventional rigid-body mechanisms (4). Flexure-based design has traditionally been evolved on creative thinking and engineering
intuitions which is considered a designer’s delight (8). In the design of the guiding mechanism, systematic and symmetric
arrangement of the flexure hinges provide constrained motion along a particular motion without any parasitic motion errors. Also,
symmetric arrangement of flexure hinges helps in avoidance of thermal expansion of the mechanism, thus maintaining the axis-
normal assembly and actuation of the FTS system. Mathematical approaches such as the elliptical integrals has been used to

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fast and Fine Tool Servo for Ultraprecision Machining 63

analyze end-loaded large deflection cantilever beams in order to obtain closed-form solutions (4). However, this mathematical
approach is difficult to use and only provides little insight about the motion or stiffness of the beam. Alternatively, a parametric
approximation model called the Pseudo-Rigid-Body Model (PRBM) introduced by Howell et al. (8) becomes a bridge between the
classical rigid-body mechanism theory and compliant mechanism analysis. It acts as a visualization tool for designers to analyze the
deflection of flexure mechanisms which is more simplified and accurate. The abundant literature and the ease of design and
manufacturing makes the flexure hinges an inevitable choice for the FTS system.

11.04.2.2 Actuators for FTS

FTS are broadly classified based on their actuation type, achievable range of motion and frequency response, and topology.
Researchers have developed an FTS with a wide range of actuators based on the requirements of a particular application. Different
types of FTS based on actuators and their characteristics discussed by various authors are presented below. FTS actuators can be
divided into the following categories:
l Piezoelectric
l Lorentz force or shear-stress electromagnetic
l Magnetostrictive
l Normal-stress electromagnetic
l Hydraulic

11.04.2.2.1 Piezoelectric FTS

Piezoelectric-based FTSs are short-stroke FTSs which are similar to micropositioning devices (9). The advantage of piezoelectric
materials is its high dynamic stiffness (greater than 50 N mm1), force generation, and bandwidth in the order of kHz with
nanometric positioning accuracy and compact size. However, piezoelectric FTSs exhibit the following demerits due to the piezo-
electric ceramic’s properties:
l Energy losses and heat generation in the material when undergoing deformation with applied voltage affects the tracking
performance due to the nonlinear behavior and hysteresis. So, the need for liquid cooling of the piezoelectric ceramic becomes
a mandate.
l The effective stroke significantly reduces during high frequency and acceleration operation.
l Care is needed while designing piezoelectric FTSs to provide proper preload and motion guidance in order to avoid tangential
loads and compressive stresses, failing which could lead to cracking of the actuator ceramic.
l The advantage of achievable bandwidth of the standalone piezoelectric actuator is reduced to that of the lesser resonant
frequency of the guiding structure depending on the movable mass of the FTS.
The above mentioned demerits can be eliminated with much ease with the availability of a range of commercially available
piezoelectric actuators. These piezoelectric actuators are preloaded and enveloped in water/dust-proof casing. The availability of the
high power amplifier and servo-controllers based on an in-built strain gage sensor provides a better closed-loop performance with
the applied voltage. Continuous development in the field piezoelectric ceramic and electronics eliminates the need for liquid
cooling during high frequency actuation by proper selection of piezoelectric ceramic and the complimentary electronic circuitry.
Proper care in design of the guiding mechanism, mostly based on flexure hinges, prevents the piezoelectric actuator from parasitic
tangential forces and helps in improving the resonant frequency of the system along the direction of motion.
Nevertheless, piezoelectric-actuated FTSs have seen significant research contributions and are widely used. In summary,
piezoelectric actuators have significant applicability in FTSs for microstructure machining and machine tool error compensation in
order to improve the accuracy.

11.04.2.2.2 Lorentz Force FTS

Moving coil motors work on Lorentz forces generated by an external flux field on current carrying coils, and is also sometimes
referred as a voice coil motor (VCM) or shear stress actuation motor. They provide much higher stroke than those of the piezoelectric
FTSs, but with a limitation in the acceleration due to the heat generated in the coil and the achievable force density due to magnetic
flux density.
Douglass (10) developed a linear-motor-driven slide for the FTS to perform the on-axis turning of an off-axis paraboloid.
A capacitance probe based position feedback system was provided with bandwidth of about 100 Hz. Wang et al. (11) described
a voice-coil-based actuator to achieve 0.4 mm stroke and 0.1 mm resolution. At Fraunhofer Institute for Production Technology in
Aachen, Germany, Prof Manfred Weck and his research group contributed toward the development of linear-motor-based FTSs.
They developed a hybrid long-stroke FTS system with a piezoelectric actuator having a stroke of 40 mm at 1 kHz bandwidth with
a maximum force of 2400 N and stiffness of 50 N mm1 and a linear motor having a stroke of 2 mm at 40 Hz (12). This system
achieved a closed-loop position bandwidth of 84 Hz. The peak thrust force was 900 N with a time constant of 5 ms. The power loss
of the linear motor and the piezoelectric translator was dissipated by a water-cooling system. The moving mass is reduced using
a parallel leaf spring mechanism giving an overall system resonant frequency of 68 Hz. In another design, Prof Weck replaced the
leaf springs with air bearings, and eliminated the piezoelectric element (13). This system uses a linear motor with a peak force of

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
64 Fast and Fine Tool Servo for Ultraprecision Machining

Figure 2 CAD and photograph of ultrafast tool servo. Reproduced from Lu, X. Electromagnetically-driven Ultra-fast Tool Servos for Diamond
Turning. Ph.D. Thesis, Massachusetts Institute of Technology, 2005.

300 N to accelerate a 0.65 kg moving mass at up to 45 g. The moving mass in this design is made of synthetic fibers, and has a first
bending mode at 400 Hz. It has a maximum travel of 16 mm, and can follow 100 Hz trajectories with a travel of 1 mm. A note on
the inertial reaction forces while implementing linear FTS, and their tendency to excite structural resonances were dealt in their
design.
Greene and Shinstock developed a linear voice-coil-based FTS (14) with 6 mm stroke and 100 Hz bandwidth. Because of the low
mass of the actuator-tool assembly and the low bandwidth, the dynamic stiffness at 15 Hz was so low that it was considered not
feasible for FTS application due to chatter instability. Todd and Cuttino built a long-range traction-type FTS for diamond turning
applications (15). The tool shuttle was driven by a rotary motor through a steel ribbon. The achieved performance was 1 mm stroke
at 20 Hz.
Lu (16) worked on the development of two variants of FTS prototypes with powder iron core (Figure 2) and Ni–Fe tapecore.
Considerable work has been presented in development of rotary FTS with a novel ultrafast electromagnetic actuator, power
amplifier system, and control system with real-time PC control. A 23 kHz closed-loop bandwidth, as low as 1.7 nm RMS error, 30
micron stroke, 500 G peak acceleration at 10 kHz open-loop operation, and 2.1 nm (0.04%) error tracking at 3 kHz sinusoid of 16
micron p-v was achieved with the first prototype FTS using powder iron cores. Using this developed rotary FTS, two-dimensional
sinusoidal surfaces in copper and aluminum with 0.5 azimuthal spatial period, 160 micron radial spatial period, and 2 micron
peak-to-valley amplitude, at 500 rpm spindle speed were achieved. The results developed with a Ni–Fe tapecore system to achieve
1200 G peak acceleration were not presented.
Another active research based on Lorentz force FTS is in the development of high amplitude, high acceleration, and high
bandwidth rotary FTS. It is another class of FTS where the actuation of FTS produces a rotational torque at the tool-tip which is
amplified due to the offset of tool from the axis of rotation. An advantage of this design is that the rotary axis can carry two or more
tools which can be rotated into position, for example, for roughing and finishing operations that use two different types of tools.
A detailed description of the need, advantages, and the disadvantages of the rotary FTS is dealt under the Section 11.04.4.

11.04.2.2.3 Magnetostrictive FTS

Magnetostrictive materials behave similar to a piezoelectric ceramics, but in the presence of a magnetic field. In contrast to
piezoelectric materials, which exhibit a coupling between the mechanical and the electrical domain, magnetostrictive materials
respond with strain to an applied magnetic field (17–21). They exhibit a better performance than piezoelectric ceramics with the
following merits: higher strain output with no requirement for high voltage supply, and are less temperature dependent (22). The
magnetic field can be applied by using either permanent magnets (for a steady bias field) or magnet coils surrounding the materials
(for time-varying dynamic magnetic field). A common magnetostrictive material is called Terfenol-D (Terbium-Ferrous-Naval
Ordnance Laboratory-Dysprosium), which also suffer from hysteresis and heating problems similar to that of piezoelectric actu-
ators. The performance characteristics reported in the literature appears to significantly fall behind that achieved by piezoelectric
devices. There are some research efforts on such devices in the following literature.
Researchers at the Michigan Technological University have developed a magnetostrictively actuated FTS to investigate vibration
abatement in turning processes (23–25). Michler et al. (23) presented a design of a magnetostrictive FTS. This original design was
further developed by Liu et al. (25) with the implementation of control algorithms and is illustrated in Figure 3. FTS with a 50 mm
stroke with a resonance frequency at 1.5 kHz and a maximum output force of 490 N was developed. A velocity feedback controller
was used to augment damping, and a PID controller was implemented to machine noncircular mechanical parts. However, no
closed-loop bandwidth specification was reported for this device.
Eda et al. (26) developed a giant magnetostrictive material, and incorporated it into a FTS with 1 mm stoke with 10 nm
resolution. The design was improved by compensating for the thermal expansion caused by the heat generation of the actuator (27)

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fast and Fine Tool Servo for Ultraprecision Machining 65

Figure 3 A schematic of magnetostrictive-based FTS. Reproduced from Liu, D.; Sutherland, J. W.; Moon, K. S.; Sturos, T. J.; Kashani, A. R.
Surface Texture Improvement in the Turning Process via Application of a Magnetostrictively Actuated Tool Holder. ASME J. Dyn. Syst., Meas., Control
1998, 120, 193–199.

and by implementing a hybrid actuator consisting of piezoelectric and magnetostrictive actuators. The magnetostrictive actuator
and the piezoelectric actuator covered the lower and the higher bandwidth requirements, respectively (28). They use water
cooling and permanent magnets to establish a bias field which results in a very linear relationship between the actuator
displacement and the applied magnetic field. A capacitance sensor is used to measure tool displacement, but it appears as though it
is only used for testing purposes and the actuator position is set open-loop in operation. The machining result that indicates
a surface finish of 7 nm Rq is achieved with the FTS. Tang et al. (29) developed a magnetostrictive FTS with a variable structural
controller and a tool positioning precision of 2 mm could be achieved. In this study, however, the actuator was used at lower
frequency ranges and at limited output strokes to avoid the hysteretic nonlinearity of the actuator. El-Sinawi and Kashani (30)
designed an active magnetostrictive FTS with a Kalman estimator-based feed forward control scheme to suppress the vibration of the
tool. The FTS improves the surface roughness of the workpiece by an average of 25% over a conventional tool holder. Al-Zaharnah
(31) used two identical magnetostrictive FTS orthogonally located in both radial and feed directions in order to apply forces on the
cutting tool in both directions, independently. With the implementation of a Kalman estimator-based controller, vibration
suppression of up to 30% was achieved.
The magnetostrictive FTSs pose some advantages in machining processes compared to piezoelectric materials, including larger
achievable strain, higher Curie temperature, resistant to fatigue failure, low voltage requirement, and greater stiffness. However,
their performance falls behind that achieved by piezoelectric devices due to their low-energy efficiencies, bulky sizes, and high
electrical loss resulting in heat generation (32).

11.04.2.2.4 Normal-Stress Electromagnetically Driven FTS

Shear stress actuation motors such as the VCM and linear motor possess the advantage of higher stroke and acceleration, but a lower
force density. On the contrary, normal stress solenoid-type actuators (also referred as normal-stress variable reluctance actuator) can
also be used for FTSs, due to potentially higher force density and thus possibly higher acceleration than their shear-stress coun-
terparts. The main issue with the development of normal-stress FTSs is the difficulty of controlling these devices in the presence of
inherent nonlinearities. The nonlinearity is inherent because the actuating force is proportional to current squared and inversely
proportional to air gap squared. This leads to positioning error while a high frequency operation of the actuator is required. Certain
research works in which normal-stress electromagnetic actuators are used are presented below.
Higuchi et al. describes an electromagnetic-actuated FTS mechanism for turning noncircular pistons (33). The pistons have
nonrotational surface feature heights of 0.1–0.5 mm. They used four variable reluctance actuators (arranged in two opposing pairs)
with flexure guides, to drive the cutting tool. The authors described two generations of designs and in the second design a mass of
7.8 kg is actuated up to an accuracy of 2 mm.
Professor Paul Ro’s research group at North Carolina State University has also developed an FTS design based upon variable
reluctance actuators (34–36). The FTS developed has a range of 800 mm, a positioning accuracy of 100 nm, and a bandwidth of 100 Hz
and the tracking performance at only 30 Hz (36). Though the stroke length was larger, the design did not take advantage of the high
acceleration potential of the variable reluctance actuator, probably due to lager air gaps which reduces achievable flux density. Also,
the reaction forces resulting from the tool acceleration have a very strong likelihood to create error motions in the other axes. An
improved tracking performance of resonance frequency 200 Hz was achieved by incorporating a sliding mode control algorithm (37).
Lu and Trumper presented an electromagnetically actuated FTS with a 50 mm stroke, 160 G acceleration to frequencies of 3 kHz,
10 kHz bandwidth, and 1.4 nm resolution (38,39). The compliant flexure beams provide the necessary single axis motion to the tool

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
66 Fast and Fine Tool Servo for Ultraprecision Machining

Figure 4 CAD model of configurable variable reluctance FTS. Reproduced from Lu, Xiaodong; Trumper, David L. Electromagnetically Driven Fast
Tool Servo. In Proceedings from ASPE 2003 Annual Meeting, 2003; pp 103–106.

tip. The motion of the armature is measured by the capacitance probe which passes through the center of the rear solenoid. Figure 4
shows the CAD model of the developed FTS.
Montesanti designed a hybrid rotary/linear FTS with a stroke of 70 mm and 10 kHz closed-loop bandwidth (40). This FTS
achieves 400 G acceleration at 5 kHz and 870 G acceleration at a mechanical resonance frequency of 10 kHz.
Normal-stress electromagnetically driven FTS thus provides the necessary stroke, acceleration, and force density which is
mandatory while using FTS to machine harder material and overcome reaction force. They are prone to nonlinearity and the control
of the system becomes a challenge to the user. Flux biasing can be used to eliminate this nonlinearity (41).

11.04.2.2.5 Hydraulic FTS

Hydraulic FTS compared to other actuator-based FTSs have limited application in machining of microfeatures and machine
performance enhancement. Yet, there is some good literature of hydraulic FTS applied to machining and particularly in FTS
development and control are presented. Hydraulic FTS seems to have limited application due to its complexity in development and
control of the device.
Jran and DeBra (42) developed a fast, short stroke hydraulic actuator, suitable for spindle motion error correction or for
noncircular. The hydraulic FTS achieves a 180 mm stroke with a frequency response which is limited to 3 kHz. The actuator flow
saturation limits the range and frequency of hydraulic actuators (17). The thermal stability of the FTS is greatly improved due to
interaction between the hydraulic oil and the mechanical assembly. Tsao and his associates have developed the associated control
algorithms and choice of servo-pack which determines the dynamic response of the FTS (43–48). Tsao also cites that the allowable
finish errors turned camshaft is in the range of 25–75 mm. In a facing operation, surface finishes of better than 25 nm Ra were
achieved, and for noncircular cutting the surface finish was about 40 nm Ra.

11.04.3 Classification of Fast Tool Servo

FTS can be classified into two major categories: Application and Topology as shown in Figure 5. FTSs are applied in manufacturing
arena for optical surface generation and microfeature creation. Many researchers have developed error correction system in order to

Figure 5 Classification of FTS.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fast and Fine Tool Servo for Ultraprecision Machining 67

Figure 6 Freeform optical surfaces by FTS process. Reproduced from http://www.iiviinfrared.com.

provide a subnanometer surface accuracy to the components in which microfeatures are produced. Hence, the applications of FTS
are further classified as tool-positioning and error compensation mechanisms. Tool-positioning deals with synchronized tool movement to
generate microstructures. Some researchers have incorporated a precompensated tool-path in order to overcome the form error of
the machine axes. Another major area of FTS applications is in the dedicated error compensation system, which is used to overcome
the inherent error of slower machine axes. Another important classification of FTS is based on the axis of FTS movement into linear
and rotary FTS. Linear FTS follows the same Cartesian coordinate as that of the machine axes while the rotary FTS follows a polar
coordinate and generally have their rotation axis aligned to that of machine tool’s rotary B-axis. The second classification mainly
focused on the stroke length of FTS and shall be further discussed in the Section 11.04.4.
Some of the works on FTS were dated back as early as the 1980s, Meinel et al. (49) has successfully produced phase corrector
plates for wavefront correction, and Luttrell et al. (50) has been able to fabricate off-axis conic surfaces and tilted flats with the FTS.
Some of the freeform optical surfaces manufactured by FTS diamond turning are illustrated in Figure 6.
FTS have seen a wide scope of application in the ultraprecision manufacturing regime. FTSs development has benefited
a spectrum of fields such as optics, photonics, opto-mechanics-electronics, telecommunications, biomedical product, semi-
conductor, etc., (52) as microstructured surfaces exhibiting excellent functionalities compared to a smooth surface. Micropyramid
structures of different heights acts as a microfin significantly improve the cooling performance of a superheated wall (53). These
microfins and microchannels pose a greater potential to improve the efficiencies of semiconductors and microprocessors with their
effective heat-exchange capabilities. Sinusoidal grids of micrometer wavelengths are used as surface encoders for detecting multi-
degree-of-freedom translational and tilt motions of precision stages (54). Microstructured surfaces created on the tool face lead to
a high cutting performance (55,56). Microporosity and morphology patterned on crystalline silicon solar cells can reduce surface
reflectance and improve the cells’ efficiency (57). Microlens array has been used in fiber coupling and optical switching, collimation
of lasers diodes, imaging systems and sensors, beam homogenizers for lasers and illumination systems, array optics featuring high
precision, aspherical lenses for creating the best imaging characteristics, etc. (58).
So, the need for an FTS system which actively changes the height of the tool along z-axis of the lathe has increased manifolds.
Numerous research groups have dedicated their efforts toward the research and development of an FTS system, the most common
application of FTS is in the direct fabrication of lenses, lens molds for biomedical applications which correct astigmatism and in the
field of optics where feature arrays such as MLA and MLP. Another broad area of research where FTSs are incorporated is in machine
tool error compensation. The advantages of FTS are directly utilized to overcome the normal surface errors of the conventional machine
slides. Due to the larger range of frequency response difference between the machine slides and FTS, the latter effectively
compensates the geometric error both offline and in real time. This methodology proves to be a cost-effective way to machine
precise components using any low precision conventional machine tools.
FTSs are divided based on the application of the FTS as a tool-positioning system for high quality surface generation and error
compensation system in machine tools. The following contributions from the literature have shed some light on the design
consideration and application of FTSs.

11.04.3.1 Tool-Positioning System

Patterson and Magrab (59) designed a piezo-based FTS for diamond turning at Lawrence Livermore National Laboratory. This device
(Figure 7) was not specifically designed to cut asymmetric parts, but instead to compensate for error motions (spindle and axes) at
frequencies beyond the bandwidth of the large axes. FTS with a stack type PZT (6.3 mm diameter (approx.) and 12.7 mm length)
provided 1.27 mm stroke at 100 Hz bandwidth and dynamic repeatability of 1.3 nm. The stiffness sufficient enough to place the
resonance above 1 kHz when supporting the maximum weight of the movable tool holder with the tool is limited to 380 g.
The two parallel and symmetric diaphragm (leaf) flexures fixed to a cylindrical support provided the rectilinear motion and greater
reduction in mass of the moving unit. An integrated capacitance sensor provides position feedback with a closed-loop position
bandwidth of 640 Hz. However, the existing large axes on LODTM are so accurate that the piezo servo was not used in its operation (60).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
68 Fast and Fine Tool Servo for Ultraprecision Machining

Figure 7 Cross-section of the FTS developed by Patterson and Magrab. Reproduced from Patterson, S. R.; Magrab, E. B. Design and Testing of
a Fast Tool Servo for Diamond Turning. Precis. Eng. 1985, 7 (3), 123–128.

Another group of researchers from Precision Engineering Center at North Carolina State University has also been active in the
area of asymmetric turning and FTS design since 1988. Falter et al. (61) have developed an FTS of 20 mm stroke and 2 kHz
bandwidth (Figure 8). The heart of the servo was a hollow piezoelectric actuator (25 mm OD and 18 mm long) with resonance
frequency of approximately 10 kHz. But, at 1 kHz, the FTS has a maximum stroke which was limited to 5 mm and heat generation in
piezoelectric actuators interrupted continuous operation of the FTS. The developed FTS has been applied in several investigations
such as compensation of inherent vibration during cutting (62).
From the same group, Dow et al. (63) designed the piezoelectric FTS for the machining of nonsymmetrical surfaces. FTS has
a maximum range of 10 mm and a usable bandwidth of nearly 2 kHz at reduced extension. This device was attached to the tool arm
of their parallel axis ultraprecision lathe (64). The doctoral thesis of Falter (65) describes in detail the tool arm which is similar to
a phonograph stylus replaced by a diamond tool.

Figure 8 Designed FTS. Reproduced from Fawcett, S. C. Small Amplitude Vibration Compensation for Precision Diamond Turning. Precis. Eng.
1990, 12, 91–96.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fast and Fine Tool Servo for Ultraprecision Machining 69

Cuttino et al. (66) reported a novel FTS by employing a long piezoelectric stack with 100 mm stroke and 100 Hz bandwidth.
Generally, the long piezoelectric actuator has the severe hysteresis problem. This study has proposed that by adding a hysteresis
module can successfully compensate the error by 43% for full-range travel and by 80% for a travel range of 70 mm.
In 1990, Okazaki (67) proposed a piezo tool servo by employing a stacked ring piezoelectric actuator (25 mm OD, 14 mm ID,
and 19 mm long, 15 mm stroke). The piezoelectric actuator was fixed inside a steel block with N-shaped slit from its slide (Figure 9).
The effective stroke of the FTS had reduced to 7 mm because of the stiffness of the flexure. Hara et al. (68) developed a microcutting
device which consists of a parallel flexure-based tool holder actuated using a preloaded piezoelectric actuator and a piezoelectric
contact sensor for real-time measurement. The active stroke of the device is 3.7 mm with 80 N mm1 stiffness and is used in
machining constant micrometer depth of cut groove. This study mainly focused on the investigation of the initial contact point and
the capability to adjust the depth of cut in real time using the contact sensor output. The repeatability of the device is within 1 mm.
No machining result was reported.
Horiuchi et al. (69) also used a piezoelectric-driven diamond tool to compensate for error motions in the work spindle and
linear cross slide of a diamond turning machine. A pair of capacitance sensors was used as a feedback unit to measure the error along
the surface of a face-plate mounted on to the spindle. The characteristic of the developed device is not stated in this publication, but
typical finish cuts on that particular machine are at a depth of 0.5 mm.
Rasmussen et al. (70,71) developed a system for asymmetric turning which used a piezoelectric actuator to drive a tool through
a lever assembly. A schematic drawing of this device is shown in Figure 10. The tool-tip had a travel of 50 mm and a closed-loop

Figure 9 Schematic of FTS system with closed-loop feedback. Reproduced from Okazaki, Y. A Micro-positioning Tool Post Using a Piezoelectric
Actuator for Diamond Turning Machines. Precis. Eng. 1990, 12 (3), 151–156.

Figure 10 Schematic of FTS system for asymmetric turning. Reproduced from Rasmussen, J. D.; Tsao, T. C.; Hanson, R. D.; Kapoor, S. G. A
Piezoelectric Tool Servo System for Variable Depth of Cut Machining. In Precision Machining: Technology and, Machine Development and Improvement;
ASME, 1992; Vol. PED-58, pp 119–130.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
70 Fast and Fine Tool Servo for Ultraprecision Machining

position bandwidth of about 200 Hz. They turned steel and aluminum parts with rotational asymmetries of 21 mm at spindle speeds
of up to 1200 rpm, with form errors of less than 5 mm. These larger form errors due to workpiece deflection could not be detected
using the strain gage attached to the piezoelectric actuator.
Kouno et al. (72) designed a piezoelectric-actuated FTS system with 6.5 mm stroke, 10 nm resolution, 70 Hz bandwidth, and
300 N mm1 stiffness. A linear variable differential transformer is integrated for close-loop position feedback. The device was
proposed to be used in a precision machine for nanometric precision tool actuation in order to overcome the inherent systematic
errors of the machine tool. This reference gives an exceptionally well-written account of the design and testing of this FTS. However,
neither error compensation techniques nor machining accuracy had been presented in these studies.
While the traditional application of piezoelectric FTS has focused on diamond turning applications, which requires relatively small
chip loads and small cutting force disturbances. An active research group under Prof Yusuf Altintas at University of British addressed
the use of a piezoelectric FTS for precision shaft machining in conventional CNC turning machines. Zhu et al. (73) and Woronko et al.
(74) employed an adaptive sliding-mode controller to compensate for cutting force-based disturbances and hysteresis of the actuator.
Zhu et al. (73) performed both rough and semifinish operations which were performed on a tool with a conventional CNC
machine and the ultraprecision cutting were accomplished by the same machine using the developed FTS system (Figure 11).
A significant improvement on the surface quality was obtained which could not be achieved with a traditional CNC machine.

Figure 11 (a) Piezo-based FTS attached to conventional CNC machine. (b) Schematic of FTS with clamping unit. Reproduced from Zhu, W. H.; Jun,
M. B.; Altintas, Y. A. Fast Tool Servo Design for Precision Turning of Shafts on Conventional CNC Lathes. Int. J. Mach. Tool. Manuf. 2001, 41, 953–965.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fast and Fine Tool Servo for Ultraprecision Machining 71

Woronko et al. (74) further improved the performance by developing a piezoelectric-actuated FTS with 38 micron stroke, 370 N
per micron stiffness, 3200 Hz natural frequency, and 10 nm positioning during finish hard turning operation. The final finishing
depth of cut is solely executed by the actuator within the actuator stroke with no change of the CNC radial position. Analytical and
finite element models of the flexures are presented, and the design process is outlined. A sliding mode controller is implemented
which provides 200 Hz bandwidth and 20 nm tool-positioning resolution by rejecting cutting force disturbances (average radial
cutting force of 6 N) and piezo-ceramic nonlinearity during precision machining. Furthermore, in order to increase the stiffness the
FTS system would need in order to overcome the vibration during turning harder material, two piezo-actuated clamping units were
incorporated in the design. Stepped, tapered, and convex microprofiles are machined to verify the control command is accurately
reflected in the part form. The results demonstrate that the actuator is capable of precise and rapid tool positioning during shaft
machining when mounted to a conventional machine tool. Maximum surface roughness of 2.0 mm was achieved in machining
4340 steel with 35–40 HRC hardness, and 0.85 mm for 4320 steel with 58–62 HRC hardness.

11.04.3.2 Error Compensation Mechanism

Since approximately 1975, the term ‘nano-technology’ came into existence. This term was first coined by Norio Taniguchi (75), who
used it to describe “manufacturing to finishes and tolerances in the nano-metric regime.” So, during the mid-1980s, many researchers
dedicated their efforts to the improvement of the existing machine tools from the precision zone to the ultraprecision regime.
Researcher from Hitachi Ltd., Japan, had studied the effect of residual small vibration on the sophisticated ultraprecision DTM
which inevitably deteriorates the surface roughness machined components. Takasu et al. (76) presented an analytical study and an
experimental validation on the effect of small amplitude and low frequency vibration on the DTM. Based on these findings, Fawcett
(62) developed a Fast Low Amplitude Tool Servo to compensate the inherent workpiece–tool surface normal vibration between the
tool and the Z-axis in real time. In situ measurement of the surface ahead of the tool, along the tool feed direction, is measured using
a capacitance sensor. Waviness was significantly reduced to achieve a P–V of 32 nm and Ra of 3.8 nm. However, the system is prone
to inherent noise due to the external sensor and other quasi-static error, such as the axes straightness was not considered.
Similarly, Kim et al. (77) developed a microcutting device with piezo-electric actuator (Figure 12) to control the tool to provide
a precise depth of cut and simultaneously compensate the waviness on the surface of the workpiece. Eddy current type noncontact
sensor with resolution of 0.5 mm is used to measure the table waviness. The waviness profile of the machined Al workpiece is
reduced from 1.3 to 0.7 mm. Description of the waviness measurement or real-time compensation was not mentioned in their paper.
No insight about surface roughness studies or flexure mechanism modeling was presented.
In addition, they proposed the same concept by compensating the waviness that is found on the machined surface with the
developed microcutting device. The error was determined from the machined surface, and subsequently, set as the compensation
reference for the microcutting device. Kim and Kim (78) developed an FTS with a piezoelectric actuator (45 mm length, 18 mm
OD). Same mechanism design which was employing the parallel spring principle was reported. The results indicated that the FTS
can successfully machine surfaces with 0.1 mm flatness after implementing feed forward and PID controller.
Kohno et al. (79) proposed an idea to improve the form accuracy by in-process measuring the relative movement between the
tool and workpiece and compensating it with the FTS. Miller et al. (80) developed a piezo-actuated FTS and the necessary digital
signal processor-based control interface to synchronize the slides servo and FTS servo. The slide servo forms the fundamental closed-
loop for the axes error compensation and FTS servo for small perturbation z-axis motions synchronized to the rotation of the
workpiece spindle (q) and the radius (X) during machining nonrotationally symmetric surfaces. Straightness and the yaw error of
X-slide are compensated using the Z-slide and vice versa. X and Z slides straightness error is measured using laser interferometer and
stored as a look-up table in the controller memory. A controller algorithm is developed in order to compute the next move of each
slide based on the error table stored, thus actively compensating the slide error before the particular position is physically reached.
A synchronized FTS and axes servo is incorporated while machining nonaxis symmetric profile machining. FTS servo computes the

Figure 12 Developed micropositioner and the error compensation mechanism. Reproduced from Kim, J. D.; Kim, D. S. Waviness Compensation of
Precision Machining by Piezoelectric Micro Cutting Device. Int. J. Mach. Tool. Manuf. 1998, 38, 1305–1322.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
72 Fast and Fine Tool Servo for Ultraprecision Machining

current position of the axes from the laser interferometer feedback, spindle index, and FTS’s own position feedback from the high
resolution capacitance sensor. This method of real-time error compensation is promising, but it is limited by the mechanical
system’s resolution and data transferring time between the two controllers. The slide error is compensated to 40 nm in both X and Z
axes and a PV value of 0.7 mm while machining parabolic profile.
Pahk et al. (81) reported a real-time dual servo mechanism for positional accuracy improvement by employing a flexure-based
PZT-actuated fine motion stage to improve the positioning accuracy of the coarser slide (global stage). A laser interferometer with
a 10 KHz transfer rate (maximum controller interrupt) is used to form a closed-loop feedback system. The movement until 10 mm is
achieved using the global stage and further movement is controlled by actuating the microstage. Position accuracy of 10 nm is
achieved with this dual servo mechanism due to the low inertia, higher accuracy, and response of the flexure-based microstage. Still,
the need to use bulk and expensive laser interferometer, this method will not be suitable for real-time machine tool error
compensation at the cutting location. Also, alignment of the optics will be of greater concern since it will lead to Abbe and cosine
error during measurement.
Gao et al. (82) had presented a detailed study and compensation mechanism of the slide error of X-axis and the spindle errors
along the axial Z-direction in a diamond turning machine. Though the design of the FTS was not mentioned in their work, they used
an FTS unit with a bandwidth of 2.5 kHz and a stroke of 10 mm. The closed-loop schematic is shown in Figure 13. The error along
Z-axis at various X-slide locations was stored and a signal sent to FTS for every pulse from the rotary encoder of the spindle. The error
compensation signal to the FTS has three components, that is, the Z-directional out-of-straightness of the X-slide, angular, and axial
motion of the spindle. A PV value of 0.22 mm is achieved with the compensation of the X-slide error along Z-axis alone. While the
spindle axial and angular errors were compensated along with the X-slide waviness, a PV value of 0.12 mm is achieved compared to
the initial noncompensated surface PV of 0.27 mm.
Xu et al. (83) had described their efforts to refine an existing cutting tool system for active error compensation with the aid of an
ultrasonic vibration cutting technique. They used two piezoelectric actuators which were integrated in the cutting tool, one of which
is for active error compensation and the other for ultrasonic vibration cutting (Figure 14). The active compensation is achieved
using two sets of differential capacitance sensors to measure the spindle run-out error in real time, whose output actuates the
compensator using their custom built stack type piezo-actuator. The profile (roundness) and the surface roughness were studied
with and without compensation and ultrasonic vibration. The profile error is effectively compensated in their study with an
improvement by 26.7% (12–8.8 mm) with no sign of surface roughness improvement. With both the error compensation and
nonconventional ultrasonic vibrational cutting in place, the roundness profile is improved by 41.7% (to 7.0 mm) and roughness is
improved by 16.2% (1.36–1.14 mm). Though the active error compensation is employed in their study, both the profile and surface
roughness values are in the micron range which are far away from the current industry needs. Also, a detailed study on the existing
machine tools’ performance and slide-ways accuracy which plays a vital is not presented.
Gan (84) developed a fine tool servo system for real-time error compensation in a miniature ultraprecision lathe. In this study,
two piezoelectric-actuated flexure-based FTS systems had been developed. The objective of the work is to compensate for the X-axis
slide error, due to which the profile accuracy of the machined components is affected. An external cost-effective position sensitivity
detector of 0.1 mm resolution is used to measure the slide error in real time and is compensated using the developed closed-loop
system comprising of piezo-actuated flexure based tool-post (Figure 15). The surface waviness is reduced from 277 to 31 nm for
aluminum workpieces and from 123 to 64 nm for brass workpieces. The surface roughness value was 18 nm for both the workpiece
material with and without the FTS implementation.
In the second design, a hybrid system was developed which had incorporated force and a capacitance sensor to monitor the
machining force and FTS displacement in real time during machining, respectively, (Figure 16). The surface roughness is found to
be between 90 and 27 nm for various cutting conditions. At a particular cutting condition, the surface roughness Ra is found to be

Figure 13 Schematic of the closed-loop control of FTS. Reproduced from Gao, W.; Tano, M.; Araki, T.; Kiyono, S.; Park, C. H. Measurement and
Compensation of Error Motions of a Diamond Turning Machine. Precis. Eng. 2007, 31 (3), 310–316.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fast and Fine Tool Servo for Ultraprecision Machining 73

Figure 14 Schematic of experimental cutting system. Reproduced from Xu, W. L.; Han, L. Piezoelectric Actuator Based Active Error Compensation
of Precision Machining. Meas. Sci. Technol. 1999, 10 (2), 106.

Figure 15 Schematic of the flexure-based FTS and sensor assembly. Reproduced from Gan, S. W. Development of Hybrid Fine Tool Servo System
for Nano-machining. Doctoral Dissertation, National University of Singapore, 2009.

164 nm and 35 nm without and with the compensation using AFM measurements. The surface waviness is found to be in the range
of 0.12–0.2 mm for feed rate of 0.5–2.0 mm rev1 at 500 and 1000 rpm, respectively. But, at a higher feed rate of around 5 mm rev1,
the surface waviness is 0.1 mm for both 500 and 1000 rpm. This shows that the error compensation of the FTS is only effective at
lower rpm and higher feed rates.
Therefore, a detailed discussion of the available literature on FTS was discussed above. So, for easier and better understanding for
the reader, the FTS was classified based on the application. The effectiveness of the FTS in surface generation and real-time error
compensation makes the FTS an inevitable component in the ultraprecision machining.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
74 Fast and Fine Tool Servo for Ultraprecision Machining

Figure 16 CAD and photograph of the hybrid FTS system. Reproduced from Xu, W. L.; Han, L. Piezoelectric Actuator Based Active Error Compen-
sation of Precision Machining. Meas. Sci. Technol. 1999, 10 (2), 106.

11.04.4 Long-Stroke Fast Tool Servo

Another major classification of the FTSs is based on the topology of FTS. Generally, two types of FTSs are used in both research
and industry: Linear and Rotary FTS. This section mainly focuses on the effective stroke length of FTS. However, most FTS systems
are having an effective stroke length not more than 1 mm. This is the main reason that FTS is not suitable for machining surfaces
with a high profile. Thus, there are numerous works to address this major setback by having a longer stroke length to fulfill the
sag height requirement. Common methods for having FTS with longer stroke are by using rotary FTS (60) and designing flexure
having a higher displacement amplification mechanism (85,86), which are incorporated with voice coil and/or piezoelectric
actuators.
At Precision Motion Control Laboratory, MIT, under Professor David L. Trumper, an active group of researchers work on fairly
a new class of FTS – Rotary FTS. Rotary FTSs are synonymous to linear FTSs, which produce high acceleration and bandwidth system
with comparatively higher amplitude than its linear counterparts. The tradeoff between simultaneous bandwidth and amplitude as
mentioned in previous sections are overcome with rotary FTS. Rotary FTS use a rotary table (B-axis) instead of the traditional linear
machine tool axis. Work presented in the PhD thesis of Ludwick (60) provides an excellent overview and comparison of a rotary
topology and a linear topology for FTSs. Figure 17 shows the schematic of the cross-section of the developed rotary FTS mounted on
a diamond turning machine having peak accelerations of 500 m s2. This rotary FTS is also capable to machine a surface feature
having amplitudes of up to 10 mm at 50 Hz. However, there is a tool position error of 0.63 mm reported due to the higher harmonic
frequency error unable to filter-off during cutting.
Kim et al. (8) has developed a long-stroke FTS having a maximum stroke of 432 mm. It incorporates a piezoelectric actuator with
a displacement amplification mechanism composed of several levers and hinges (see Figure 18). Rakuff et al. (86) utilizes a voice coil
actuator and a pair of flexure hinges in a long-stroke FTS system (see Figure 19) having maximum accelerations of 260 m s2 and
bandwidths of up to 140 Hz. The maximum displacement range of the cutting tool was 2 mm. In contrast to piezoelectric actuators,
permanent magnet voice coil actuators are free of hysteresis with a nearly linear current versus force relationship for smaller strokes.
However, the flexure structure in this FTS has a low resonant frequency which can cause resonance and its low stiffness is liable to
generate vibration in vertical direction. Both of these effects have an adverse impact on the quality of machined surface.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fast and Fine Tool Servo for Ultraprecision Machining 75

Figure 17 Schematic diagram of long-stroke rotary FTS system by Ludwick. Reproduced from Ludwick, S. J., Jr. A Rotary Fast Tool Servo for
Diamond Turning of Asymmetric Optics. Ph.D. Thesis, Massachusetts Institute of Technology, June 1999.

Figure 18 Displacement amplification mechanism of LFTS. Reproduced from Kim, H. S.; Lee, K. I.; Lee, K. M.; Bang, Y. B. Fabrication of Free-form
Surfaces Using a Long-stroke Fast Tool Servo and Corrective Figuring with On-machine Measurement. Int. J. Mach. Tool. Manuf. 2009, 49 (12–13),
991–997.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
76 Fast and Fine Tool Servo for Ultraprecision Machining

Figure 19 (a) FTS system with voice coil motor; (b) Its flexure mechanisms. Reproduced from Rakuff, S.; Cuttino, J. F. Design and Testing of a
Long-range, Precision Fast Tool Servo System for Diamond Turning. Precis. Eng. 2009, 33, 18–25.

Sosnicki et al. (87) also developed a long-stroke FTS using an amplified piezo actuator (APA) from Cedrat Technology (88) with
an elliptic shell producing a stroke of 400 mm at above 600 Hz. The principle for this APA actuator (see Figure 20) is based on the
deformation of an elliptic shell to amplify the ceramic strain. The PZT ceramic stack is aligned with the great axis of the ellipse.
A small deformation of the great axis creates a large displacement of the small axis. The amplification ratio can typically reach
20 times which means such actuators can reach strokes of 1 mm. Although this APA actuator provides a high stiffness and relatively
large displacements, their mass is not optimized and the metallic shell implies a large penalty on the mass.
Recently, there is an emerging trend on utilizing air bearing guides in the development of long-stroke FTSs. Buescher et al.
(89,90) proposed a live-axis turning technique which utilized an air-bearing slider and linear motors to increase the stroke length up
to 4 mm, but at a relatively low bandwidth of 20 Hz. However, there was vertical vibration with amplitude of 0.2 mm. Zdanowicz
et al. (91) further developed a new system to address the limitation of Buescher’s FTS with VCM and porous air bearing. Marten et al.
(92) developed a long-stroke FTS which consisted of an air bearing stage and a three-phase oil-cooled linear motor. This FTS had

Figure 20 Cedrat APA400MML. Reproduced from Amplified Piezo Actuator, http://www.cedrat-technologies.com.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fast and Fine Tool Servo for Ultraprecision Machining 77

Figure 21 Schematic diagram of the hybrid macro- and micro-range FTS. Reproduced from Liu, Q.; Zhou, X.; Wang, L. A New Hybrid Macro- and
Micro-range Fast Tool Servo. In 2010 International Conference on Mechanic Automation and Control Engineering (MACE), 2010, http://dx.doi.org/10.1109/
MACE.2010.5535544.

a stroke of 25 mm at 20 Hz and was capable of accelerations of 100 m s2. Precitech (93) and Moore Nanotechnology (94) have also
developed several long-stroke FTSs with air bearing guides. Despite numerous advantages for utilizing air bearing guides, which
have been discussed, air bearing has a low damping stiffness as compared to rigid flexure hinge. Thus, a mass damper must be
designed for these air-bearing FTS to cancel dynamic loads due to mass inertia.
Hybrid methods were also employed to increase the stroke length of FTS. Weck et al. (95) developed a hybrid long-stroke FTS,
which integrated a finer piezoelectric system of 40 mm stroke length at 1 kHz bandwidth and a linear motor of 2 mm at 40 Hz. The
piezoelectric translator had a maximum force of 2400 N and peak thrust force of 900 N with an axial stiffness of 50 N mm1 and
a resonance frequency of around 2 kHz. The power loss of the linear motor and the piezoelectric translator was dissipated by a
water-cooling system. A compensation of these forces to reduce excitation of the supporting machine was achieved by the use of
a balance mass. Liu et al. (96) has introduced a hybrid macro- and micro-range FTS system (see Figure 21) that enables diamond
turning of optical freeform surfaces. The macro-range FTS is driven by a voice coil motor (VCM), and a PZT actuator is used to
drive the micro-range FTS, both of which are guided by a flexure hinge. The output force of the VCM is enlarged by a lever. The
macro-range FTS can be used to machine large asymmetry surfaces, and the small asymmetry surfaces are machined by the
micro-range FTS.
A novel long-stroke FTS utilizing a VCM, a piezoelectric actuator, and two flexure hinges has been developed with a displacement
range 0.5 mm and a bandwidth of 109.6 Hz, as illustrated in Figure 22 (97). The long-stroke flexure hinge is actuated by a high
resonance frequency in the z-axis direction which is approximately 109.6 Hz. The short-stroke flexure hinge actuated by piezoelectric
actuator can improve the closed-loop stiffness in the y-axis direction. By exploiting both advantages of VCM and piezoelectric
methods, the proposed FTS has high stiffness in the y-axis direction and high resonance frequency in the z-axis direction.
From the literature review, it can be concluded that long-stroke FTSs are usually actuated by piezoelectric and voice coil actuators.
Piezoelectric FTSs are usually guided by flexure hinge structures which are more suitable for error compensation. However,
piezoelectric FTSs often have a low resonance frequency because of the lever mechanism. The lever mechanisms also bring hysteresis
and tracking error because of the lever bending. Voice coil FTSs have longer strokes, but lower bandwidths than other FTSs. Hence,
the stroke and the bandwidth are two separate performance parameters which cannot be simultaneously optimized for most cases.
Although the rotary FTS have the advantage of high acceleration with a minimal reaction force system, the rotary FTS have
a reduced structural stiffness compared to its linear counterpart due to the cantilever arm which holds the tool. Higher acceleration
required the arm to be of dense material, this leads to a reduced structural stability of the FTS. Also, the force achievable at the tool-
tip in a rotary FTS is smaller compared to linear FTS due the conversation of force to velocity at the cantilever arm end. Thus,
machining of harder materials which requires higher actuation force becomes difficult while using a rotary FTS. However, linear FTS
transmits the force achieved at the actuator directly to the tool-tip with negligible loss due to the structural stiffness of the FTS’s
guiding mechanism. So, while using a rotary FTS, the designer needs to consider the cutting force generation before designing the
system. This constraints the FTS application to selected soft material such as plastics lens machining. Another factor which barri-
cades the use of rotary FTS is the tool-path generation for large stroke rotary FTS as the FTS creates parasitic lateral displacement of
considerable magnitude. If this lateral displacement is not taken in account during tool-path generation, the required surface feature
such as diffracted lens cannot be achieved. Calibration of rotary FTS needs to be based on slightly complicated polar coordinates,
but the linear FTS are straight-forward measurements since they follow the same Cartesian coordinates as machine axes. The
motions of the rotary FTS are also more likely to lead to an interrupted cut.
Although there are numerous works to increase the limiting stroke length, it may take a lot of efforts and long duration time to
modify an existing FTS system for the sake of fulfilling the surface height requirement. There are other promising solutions to
address the development of long stroke length in FTS. First, STS diamond turning is a well-established technology which is engi-
neered to address the travel limitation by FTS systems. STS diamond turning made its debut appearance in 2003 (98) and exhibited
its distinguished performance to fabricate freeform surfaces exceeding 1 mm sag height with excellent surface quality and accuracy.
This marks the tipping point for the growing interest of this novel ultraprecision machining technique to fabricate freeform optical
surfaces with a larger sag height. STS technology utilizes the existing diamond turning machine Z-slide for the tool motion by

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
78 Fast and Fine Tool Servo for Ultraprecision Machining

Figure 22 FTS with hybrid flexure design: (a) Long and short flexures; (b) Working principle. Reproduced from Liu, Q.; Zhou, X.; Xu, P.; Zou, Q.;
Lin, C. A Flexure-based Long Stroke Fast Tool Servo for Diamond Turning. Int. J. Adv. Manuf. Technol. 2012, 59, 859–867.

adopting a linear motor to replace ball screws. This allows more flexibility in the motion of the slide without damaging the ball
screw. It has advantages of fabricating parts with much larger deviation than the short-stroke FTSs. Although several advantages have
been highlighted, there is a major setback in this STS technology in which high inertia forces slow acceleration in the movement
speeds due to heavy weight linear axes. Hence, it is not suitable for machining freeform surfaces with higher frequency asymmetries
and lower machining speed as compared with FTS systems. Slow tool servo is also plagued by thermal drift during the extremely
long fabrication time, similarly to those traditional fabrication methods such as grinding, polishing, or flycutting.
In order to overcome this problem, a second method has been developed for long-stroke FTSs by employing a hybrid FTS/STS
method (99) to speed up the machining time. In this method, the design freeform surface was processed and split into two different
domains based on surface geometry as shown in Figure 23. Low order surface geometry with a total stroke of 0.5 mm was machined
by a STS system and the higher order part with a total stroke of 20 mm was machined by an FTS system.
The second method gives us motivation to search for the solutions for long-standing research issues on long-stroke FTS systems.
Hence, we have developed a novel hybrid FTS/STS diamond turning method with layered tool trajectories. The details of this hybrid
FTS/STS technique shall be discussed in the next section.

11.04.5 Hybrid FTS/STS

This section presents our works on a novel hybrid FTS/STS diamond turning method with layered tool trajectories (100) to
overcome the limited stroke length without modifying an existing FTS system. This is the first work to study on the diamond
machining of microstructured surfaces whose heights are greater than the stroke length of FTS.
Traditional FTS diamond turning utilizes three controlled axes, namely, X, C, and W-axes. X-axis is the radial movement
which controls the feed toward the spindle center and is also perpendicular to spindle axis (Z-axis). C-axis is the spindle
rotational direction about the Z-axis. W-axis is the FTS stroke which controls the feed direction into the workpiece surface and is
parallel to Z-axis.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fast and Fine Tool Servo for Ultraprecision Machining 79

Figure 23 Data processing of freeform surface by geometry splitting. Reproduced from Brecher, C.; Niehaus, F.; Merz, M.; Schmidt, K. Machining
Microstructures on Free-form Surfaces. Prod. Res. Micro NanoTechnol. 2008, 6–13.

Figure 24 Archimedes spiral tool trajectory.

The tool trajectory for FTS turning is commonly defined as Archimedes spiral (see Figure 24) and is represented by polar
coordinate system as
x ¼ R cos q
y ¼ R sin q [1]
W ¼ f ðR; qÞ ¼ f ðx; yÞ
where R is the radial position of the X-axis (relative to center of the machine spindle), q is the rotational angle of the machine
spindle, and W is the stroke of FTS for tool trajectory with respect to a cutting point (x, y).
The proposed layered tool trajectory method is incorporating an additional fourth working axis (Z-axis) as described in
Figure 25. This fourth or Z-axis plays a critical role for extending the stroke length of FTS in this proposed method. During FTS
diamond turning, a stepper motor controller stops C-axis motion (spindle) at a position where the FTS tool reaches its upper limit
of maximum stroke zone (ii). Then, the Z-axis motor controller retracts the workpiece surface to the lower limit of maximum stroke
zone of FTS Wmax. After the workpiece surface (Z-axis) retracted, the FTS re-extends its tool to the lower limit of its stroke zone (ii*)
and continues machining in the feed direction (iii). This cycle shall be repeated until the whole process completes the machining of
the desired surface profile and the FTS tool reaches the top of the surface profile. Therefore, the stroke of FTS for tool trajectory in the
eqn [1] shall be rewritten as
Wi ¼ f ðR; jÞ  Wmax  i; i ¼ 0; 1; 2. [2]

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
80 Fast and Fine Tool Servo for Ultraprecision Machining

Figure 25 Schematic diagram of layered tool trajectory.

Figure 26 Formation of layered tool trajectory.

where Wmax is maximum stroke length of FTS or maximum stroke zone and i is ith Z-axis retraction during layered tool trajectory.
Figure 26 shows an illustration of layered tool trajectory which is projected onto the ZX plane. Figure 26(a) displays a series of
Z-axis retraction in the tool trajectory whenever the FTS tool reaches the upper limit of FTSmax stroke zone. Hence, the original tool
trajectory is being modified into the layered tool trajectory as shown in Figure 26(b).
In this proposed layered tool trajectory method, there are two important factors which could not be ignored for having a good
tool trajectory, which are (1) the transition point Pi and (2) the amount of Z-axis retraction. The point Pi locates at the point where
the controller retracts the Z-axis movement and the FTS controller re-extends its tool. If this Pi is randomly allocated as illustrated in
Figure 27(a), an overcutting problem may occur due to dynamic response of C-axis controller.
Fortunately, this overcutting problem can be easily overcome by allocating every exit/reentry points P* i to the outermost radii of
surface outline, which lies within FTS stroke zone Wmax. This allows the recovery of dynamic response in the C-axis movement. On
the undercutting issue, this only occurs at negative region of the layered tool trajectory as shown in Figure 28(a) and is due to over-
retraction in the Z-axis movement. These undercuts can be avoided with a correct amount of retraction as illustrated in Figure 28(b)
without having a negative region.
This amount of Z-axis retraction Zr i can be found by determining the maximum Z-axis boundary Zbmax within a circumscribed
radius R* with respect to Pi* as shown in Figure 29 and are described as
ZR
¼ arg min½f ðR
; gÞ; for g ˛ ½0 ; 360 
Zbmax ¼ Zi
 ZR
[3]
Zr i ¼ Zi
 Zbmax

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fast and Fine Tool Servo for Ultraprecision Machining 81

Figure 27 Exit and reentry points on upper limit of FTS stroke zone: (a) XY plane projection; (b) 3D projection.

Figure 28 Effect of Z-axis retraction; (a) Over-retraction; (b) correct retraction.

where Zi* is the z-coordinate for point Pi*and ZR* is the minimum value for intersection point of surface and cylindrical region
within R*. Hence, the effective stroke zone varies in each layer of tool trajectory and is described as
X
Wi
¼ f ðR; qÞ  Zr i ; where 0 < Zr i < Zi
[4]

11.04.5.1 Experiments and Discussions

Several experiments were conducted to study the feasibility of the proposed layer tool trajectory method for the fabrication of
a single micro-prism with the cutting conditions as shown in Table 1. The layered tool trajectory has been simulated to avoid
unnecessary overcutting by compensating for tool nose radius and good Z-retractions as shown in Figure 30.
Figure 31 shows that the machined micro-prism appeared to have a profile distortion caused by the tool center-alignment error
of 10 microns in y-direction. The height and surface roughness of the micro-prism are measured using a white light interferometer.
The height of micro-prism was measured as 25.09 microns (see Figure 32) and an error of 0.66 microns (2.56%) compared
to the designed height of 25.75 microns. The error in the height is most likely caused by the dynamic response of Z-axis motion.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
82 Fast and Fine Tool Servo for Ultraprecision Machining

Figure 29 Schematic diagram for calculating Z-retraction.

Table 1 Fabrication parameters

Machining conditions Feedrate: 1 mm rev1

Spindle speed: 50 rpm
Sampling number: 720 rev1
Wmax: 5 mm
Cutting tool SCD with 0.2 mm tool nose radius
Material of workpiece Brass
Diameter of workpiece 3 mm
Micro-prism Cyclic hexagonal base with circumscribed radius of 0.25 mm
Height of micro-prism 25.75 mm

Figure 30 Simulated three-dimensional layered tool trajectory.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fast and Fine Tool Servo for Ultraprecision Machining 83

Figure 31 Machined micro-prism.

Figure 32 Overall height measurement of fabricated micro-prism.

This dynamic response leads to the positioning error for each Z-axis retraction and causes P* i points to overlap one another. Thus,
a dynamic response compensation for this Z-axis retraction in the tool trajectory generation would be considered in the future study.
The next Figure 33 shows only a single face of micro-prism is selected for its surface roughness measurement since it is tedious
to tilt every face of micro-prism to measure surface roughness. The surface roughness can be assumed as constant for every face.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
84 Fast and Fine Tool Servo for Ultraprecision Machining

Figure 33 Surface roughness measurement on a single face of micro-prism.

The measured average surface roughness (Ra) was about 195 nm, which could be relatively high for diamond turning application.
This unsatisfactory surface roughness is likely due to sparser cutting points by a low sampling number and can be improved by using
a larger sampling number. Lastly, the proposed layered tool trajectory method has successfully demonstrated that the extension of
effective FTS stroke length was about five folds of its maximum stroke length without replacing the existing FTS system.
Long-stroke FTS may be one of the possible ways to replace any existing FTS system with limited stroke length in order to fulfill
greater surface height requirements. In this section, a novel method has been proposed to extend the limited stroke length without
modifying an existing FTS system by generating layered tool trajectory in the hybrid FTS/STS system. This proposed layered tool
trajectory method has demonstrated the feasibility study of fabricating a micro prism by FTS diamond turning. The height of
fabricated micro prism has an error of 0.65 mm as compared to designed height requirement. Importantly, the proposed method has
successfully extended an effective stroke length of an existing FTS system to about five folds of its maximum possible stroke length.
Although the novel layered tool trajectory method has exhibited its capability to extend the effective stroke length of an existing
FTS system, there are few areas of improvement for this proposed technique that should be conducted in the future. They include the
following:
1. Only a single micro-prism has been validated for the capability of the hybrid FTS/STS method using layered tool trajectory. More
freeform designs are to be studied to expand the effectiveness of this proposed method (i.e., array, off-axis, and nonsymmetrical
designs).
2. The dynamic response in the Z-axis retraction affects the geometrical accuracy and surface quality of the workpiece. Hence, it
would be great to consider dynamic response compensation in the layered tool trajectory.

11.04.6 More Promising Challenges

Today’s ultraprecision machining systems are coupled with state-of-art technologies in controls, drives, and feedback devices.
Important technological advances in a number of disciplines – first of all process technologies, but also including measuring and
testing, quality assurance, as well as the production environment – are necessary for further advances including the development of
ultraprecision machining techniques, machines, and control systems which can achieve nanometer tolerances and subnanometer
surface finishes as well as new analytical techniques that can observe, measure, and provide three-dimensional images of features at
the nanometer level. Although several machining challenges in FTS diamond machining techniques for freeform optical surfaces
have been highlighted in the previous sections, there are much more challenges which have yet been fully explored to expand the
machining barriers and some of these challenges shall be discussed in this section.
First of all, diamond machining techniques for freeform optical surfaces are limited to nonferrous materials due to catastrophic
wear of diamond tools when machining ferrous (101,102). From Figure 34, a periodic table containing elements has been well
distinguished for diamond turnable and nonturnable (102). In order for diamond machining of ferrous metals, a layer of elec-
trodeless nickel has to be coated onto the prefinished surface or prefinished surface has undergone a nitriding treatment process
(103). This leads to longer lead time and higher machining cost. Elliptical diamond machining techniques also may be employed
for machining ferrous materials without nickel coating to reduced tool wear (104). Hence, the search for solutions in the precision
machining of steel is a promising and extensively long-time investigated research to be fully explored to expand the machining
barriers of optical freeform surfaces.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fast and Fine Tool Servo for Ultraprecision Machining 85

Figure 34 Scanning force microscope images of the cutting edge of a diamond tool after cutting (left) iron and (right) copper. Reproduced from
Brinksmeier, E.; Preuss, W. Micro-machining (Review). Philos. Trans. Roy. Soc. A 2012, 370, 3973–3992. http://dx.doi.org/10.1098/rsta.2011.0056

Secondly, there is also a new challenge which is gaining interest by researchers and manufacturers on alternative methods for
polishing brittle materials, ductile regime machining (DRM), to obtain a high quality surface finish by a ductile or plastic material
removal process (105–109). Although DRM has quite well matured in past decades, DRM is relatively new in the micromilling area.
Early works of DRM (105,106) in the micromilling area have exhibited their capability to machine optical brittle materials leaving
a mirror surface quality. It has been known that DRM can be achieved in any brittle material if the depth of cut/chip thickness is less
than its critical value (107,108). For the case of micromilling, this critical chip thickness can be defined as critical feed per cutting
edge (108). There is a strong possibility that DRM could be also achieved when the ratio of depth-of-cut and tool edge radius a/r is
small enough (at about 0.2) (110). Hence, a new possible focus of interest will study DRM in-depth which can contribute toward
the possibility of machining three-dimensional and freeform surfaces on brittle materials.
The next challenge shall be the improvement of computer-aided machining (CAM) support for ultraprecision accuracy often
drives the needs for customized CAM software solutions (111). However, it may come with a high cost price of these customized
software solutions. Ironically, most commercial CAD software solutions have already demonstrated their capability to model
a greater degree of freeform optical surface designs with ultraprecision accuracies. Hence, a built-in application programmable
interface in these commercial software solutions may be considered to be able to overcome these CAM support issues (112).
Lastly, the fundamental understandings of chip formation in freeform diamond turning have yet been fully comprehended and
a great deal of research has been carried out to achieve superior surface quality (113–117). However, the effective tool rake angle
changes all the time during the diamond turning of freeform surface, even with a zero rake diamond tool. From Figure 35, the

Figure 35 Effective tool rake angles along the cutting direction.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
86 Fast and Fine Tool Servo for Ultraprecision Machining

effective rake angle geff in a diamond turning of freeform surfaces with a zero rake diamond tool still varies from positive to negative
values depending on the slopes of surface.
It is deemed necessary to consider the effects of effective rake angles on surface quality, accuracy, etc. Therefore, future research
for effective rake angles in the mechanism of chip formation would be promising and extensively long-time investigated research to
be fully explored to expand the barriers of ultraprecision machining techniques.

11.04.7 Conclusions

This chapter has summarized a short review on FTS. The growing markets of optical devices in telecommunication, medical
imaging, and surveillance systems yield a high potential for engineering tasks in ultraprecision machining. In the future, ultra-
precision machine developments will continue to be driven by market requirements and progress in the ultraprecision machining
area will ignite many industrial applications. Several future works have also been paved to address the challenges and barriers. At the
same time, they will unleash the hidden potential of FTS in the manufacturing areas.

References

1. Monkman, G. J.; Hesse, S.; Steinmann, R.; Schunk, H. Robot Grippers; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, 2007.
2. Rabenorosoa, K.; Haddab, Y.; Lutz, P. Micro-assembly Technologies and Applications. In Ratchev, S., Koelemeijer, S., Eds.; Springer: Boston, 2008; Vol. 260, pp 235–242.
3. Aravind Raghavendra, M. R.; Senthil Kumar, A.; Nikhil Bhat, J. Design and Analysis of Flexure-hinge Parameter in Microgripper. Int. J. Adv. Manuf. Technol. 2010, 49 (9),
1185–1193.
4. Howell, L. L. Compliant Mechanisms; John Wiley & Sons: New York, NY, 2001.
5. Smith, S. T. Flexures: Elements of Elastic Mechanisms; CRC, 2000.
6. Lobontiu, N. Compliant Mechanisms: Design of Flexure Hinges; CRC, 2002.
7. Awtar, S. Synthesis and Analysis of Parallel Kinematic XY Flexure Mechanisms. Ph.D. Diss., Massachusetts Institute of Technology, 2003.
8. Howell, L. L.; Midha, A. The Development of Force-deflection Relationships for Compliant Mechanisms. In Machine Elements and Machine Dynamics, 23rd ASME Biennial
Mechanisms Conference 1994, DE-Vol. 71; pp 501–508.
9. Schellekens, P.; Rosielle, N.; Vermeulen, H.; Vermeulen, M.; Wetzels, S.; Pril, W. Design for Precision: Current Status and Trends. CIRP Ann. – Manuf. Technol. 1998, 47 (2), 557–586.
10. Douglass, S. A Machining System for Turning Non-axis-symmetric Surfaces. Ph.D. Dissertation, The University of Tennessee, Knoxville, 1983.
11. Wang, X.; Yuan, Zh. Experimental Research on the Linear Motor Micro-feed Device with High-frequency Response, Long Travel and High Accuracy. CIRP Ann. – Manuf.
Technol. 1991, 40 (1), 379–382.
12. Weck, M.; Oezmeral, H.; Kehlkopp, K.; Terwei, T. A New Hybrid Concept for a Long Stroke Fast-tool-servo System. In Proceedings of the 9th Annual Meeting – American
Society for Precision Engineering; 1995; pp 211–214.
13. Weck, M.; Oezmeral, H.; Kehlkopp, K.; Terwei, T. A Long Stroke Fast-tool Servo with Air Bearings. In Progress in Precision Engineering and Nanotechnology; Kunzmann, H.,
Ed.; 1997; pp 393–396.
14. Greene, W.; Shinstock, D. Design of a Linear Voice Coil Actuator for Fast Tool Servo Applications. In Proc. of Annual Meeting – ASPE; 1997.
15. Todd, M. W.; Cuttino, J. F. Development of a Long Range, Traction Drive Fast Tool Servo for Diamond Turning Applications. In Proc. of Annual Meeting – ASPE; 1997.
16. Lu, X. Electromagnetically-driven Ultra-fast Tool Servos for Diamond Turning. Ph.D. Thesis, Massachusetts Institute of Technology, 2005.
17. Chopra, I. Review of State of Art of Smart Structures and Integrated Systems. AIAA J. 2012, 40 (11).
18. Butler, J. L.; Butler, S. C.; Clark, A. E. Unidirectional Magnetostrictive Piezoelectric Hybrid Transducer. J. Acoust. Soc. Am. 1990, 88, 7–11.
19. Carman, G. P.; Mitrovic, M. Nonlinear Constitutive Relations for Magnetostrictive Materials with Applications to 1-D Problems. J. Intell. Mater. Syst. Struct. 1995, 6, 673–683.
20. Dapino, M. J.; Smith, R.; Faidley, L. E.; Flatau, A. B. A Coupled Structural Magnetic Strain and Stress Model for Magnetostrictive Transducers. J. Intell. Mater. Syst. Struct.
2000, 11, 135–152.
21. Quandt, E.; Claeyssen, F. Magnetostrictive Materials and Actuators. In Proc Actuator; June 2000; pp 100–105.
22. Park, G.; Bement, M. T.; Hartman, D. A.; Smith, R. E.; Farrar, C. R. The Use of Active Materials for Machining Processes: A Review. Int. J. Mach. Tool. Manuf. 2007, 47 (15),
2189–2206.
23. Michler, J. R.; Moon, K. S.; Sutherland, J. W.; Kashani, A. R. Development of a Magnetostriction Based Cutting Tool Positioner. Trans. North Am. Manuf. Res. Inst. SME 1993, 421–427.
24. Sturos, T. J.; Sutherland, J. W.; Moon, K. S.; D.Liu; Kashani, A. R. Application of an Actively Controlled Magnetostrictive Actuator to Vibration Abatement in the Turning
Process. In IMECE Proceedings of the ASME Dynamic Systems and Control Division, San Francisco, CA; 1995; pp 539–544.
25. Liu, D.; Sutherland, J. W.; Moon, K. S.; Sturos, T. J.; Kashani, A. R. Surface Texture Improvement in the Turning Process via Application of a Magnetostrictively Actuated Tool
Holder. ASME J. Dyn. Syst., Meas., Control 1998, 120, 193–199.
26. Eda, H.; Ohmura, E.; Sahashi, M.; Kobayashi, T.; Ikawa, N. Ultra-precise Machine Tool Equipped with a Giant Magnetostrictive Actuator: Development of New Materials,
TbxDy1-x (FeyMn1-y) N and Their Application. Ann. CIRP 1992, 41, 421.
27. Yamamoto, Y.; Eda, H.; Shimizu, J. Application of Giant Magnetostrictive Materials to Positioning Actuators. In Advanced Intelligent Mechatronics, 1999. Proceedings. 1999
IEEE/ASME International Conference on IEEE; 1999; pp 215–220.
28. Eda, H.; Shimizu, J.; Zhou, L. Study on Ultra-precision Machining of Ceramics for Optical Components. Proc. SPIE – Int. Soc. Opt. Eng. 1999, 3740, 420–423.
29. Tang, Z.; Xiang, Z.; Lv, F. Modeling and Control Method Study of Magnetostrictive Micropositioner and Its Application. In Paper presented at the Systems, Man and Cybernetics,
IEEE International Conference on; 10–13 Oct. 2004.
30. El-Sinawi, A. H.; Kashani, R. Improving Surface Roughness in Turning Using Optimal Control of Tool’s Radial Position. J. Mater. Process. Technol. 2005, 167, 54–61.
31. Al-Zaharnah, I. T. Suppressing Vibrations of Machining Processes in Both Feed and Radial Directions Using an Optimal Control Strategy: The Case of Interrupted Cutting.
J. Mater. Process. Technol. 2006, 172, 305–310.
32. Thoe, T. B.; Aspinwall, D. K.; Wise, M. L. H. Review on Ultrasonic Machining. Int. J. Mach. Tool. Manuf. 1997, 38 (4), 239–255.
33. Higuchi, T.; Yamaguchi, T.; Tanaka, M. Development of a High Speed Noncircular Machining NC-IATHE for Cutting a Piston-head of a Reciprocating Engine by Use of a New
Servomechanism Actuated by Electromagnetic Attractive Force. J. Jpn. Soc. Precis. Eng. 1989, 62 (3), 453.
34. Gutierrez, H. M.; Stancil, B. A.; Ro, P. I. Design of a Magnetic Servo-levitated Fast Tool Servo System for Precision Turning. In IMECE Proceedings of the ASME Dynamic
Systems and Control Division, San Francisco, CA; 1995; pp 491–496.
35. Stancil, B. A.; Gutierrez, H. M.; Ro, P. I. Design of a Long Range Fast Tool Servo System Using Magnetic Servo Levitation. In Proceedings of the 9thAnnual Meeting – American
Society for Precision Engineering; 1995; pp 301–304.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fast and Fine Tool Servo for Ultraprecision Machining 87

36. Gutierrez, H. M.; Ro, P. I. Parametric Modeling and Control of a Long-range Actuator Using Magnetic Servo-levitation. IEEE Trans. Magn. 1998, 34 (5).
37. Gutierrez, H. M.; Ro, P. I. Sliding-mode Control of a Nonlinear-input System: Application to a Magnetically Levitated Fast Tool Servo. IEEE Trans. Indus. Electron. 1998, 45 (6).
38. Lu, Xiaodong; Trumper, David L. Electromagnetically Driven Fast Tool Servo. In Proceedings from ASPE 2003 Annual Meeting; 2003; pp 103–106.
39. Lu, X.; Trumper, D. L. High Bandwidth Fast Tool Servo Control. In American Control Conference 2004; 2004.
40. Montesanti, R. C. High Bandwidth Rotary Fast Tool Servos and a Hybrid Rotary/linear Electromagnetic Actuator. Ph.D. Thesis, Massachusetts Institute of Technology, 2005.
41. Lu, X.; Trumper, D. L. Electromagnetically Driven Ultrafast Tool Servo. In Proceedings from ASPE 2004 Annual Meeting; 2004; pp 269–272.
42. Tran, H. D.; De Bra, D. B. Design of a Fast Short-stroke Hydraulic Actuator. Ann. CIRP 1994, 43 (1), 469–472.
43. Tsao, T. C.; Tomizuka, M. Robust Adaptive and Repetitive Digital Tracking Control and Application to Hydraulic Servo for Noncircular Machining. ASME J. Dyn. Syst., Meas.,
Control Mar 1994, 116, 24–32.
44. Tsao, T. C. Non-circular Turning for Camshaft Machining – Fast Response Actuators, Sensors, and Motion Control for Cam Profile Tracking. In Boudreaux, J. C., Lettieri, T. R.,
Eds. October 1997, Vol. NISTIR 6079.
45. Tsao, T. C.; Hanson, R. D.; Sun, Z.; Babinski, A. Motion Control of Non-circular Turning Process for Camshaft Machining. In Proceedings of the Japan-USA Symposium on
Flexible Automation; 1998.
46. Sun, Z.; Tsao, T. C. Adaptive Repetitive Control Design with Application to an Electrohydraulic Servo. In Fluid Power Systems and Technology: Collected Papers 1997,
Vol. FPST-Vol. 4/DSC-Vol. 63; pp 39–53.
47. Kim, D. H.; Tsao, T. C. Robust Performance Control of Electrohydraulic Actuators for Camshaft Machining. In Fluid Power Systems and Technology: Collected Papers 1997,
Vol. FPST-Vol. 4/DSC-Vol. 63; pp 33–38.
48. Kim, D. H.; Tsao, T. C. Identification and Control of Electrohydraulic Actuator Modeled as a Linear Periodic System. In Proceedings of the American Control Conference; 1998.
49. Meinel, A. B.; Meinel, M. P.; Stacy, J. E.; Saito, T. T.; Patterson, S. R. Wavefront Correctors by Diamond Turning. Appl. Opt. 1989, 25.
50. Luttrell, D. E. Machining Non-axisymmetric Optics. In ASPE Proceedings, Annual Meetings; 1990.
51. http://www.iiviinfrared.com.
52. Evans, C. J.; Bryan, J. B. “Structured”, “Textured” or “Engineered” Surfaces. CIRP Ann. – Manuf. Technol. 1999, 48 (2), 541–556.
53. Sodtke, C.; Stephan, P. Spray Cooling on Micro Structured Surfaces. Int. J. Heat Mass Transfer 2007, 50 (19–20), 4089–4097.
54. Gao, W.; Araki, T.; Kiyono, S.; Okazaki, Y.; Yamanaka, M. Precision Nanofabrication and Evaluation of a Large Area Sinusoidal Grid Surface for a Surface Encoder. Precis. Eng.
2003, 27 (3), 289–298.
55. Obikawa, T.; Kamio, A.; Takaoka, H.; Osada, A. Micro-texture at the Coated Tool Face for High Performance Cutting. Int. J. Mach. Tool. Manuf. 2011, 51 (12), 966–972.
56. Brinksmeier, E.; Riemer, O.; Twardy, S. Tribological Behavior of Micro Structured Surfaces for Micro Forming Tools. Int. J. Mach. Tool. Manuf. 2010, 50 (4), 425–430.
57. Tsujino, K.; Matsumura, M. Formation of a Low Reflective Surface on Crystalline Silicon Solar Cells by Chemical Treatment Using Ag Electrodes as the Catalyst. Sol. Energy
Mater. Sol. Cells 2006, 90 (10), 1527–1532.
58. Pop, A. P.; Ungur, P.; BejinaruMihoc, G. Applications of Microlens Arrays. Fascicle Manage. Technol. Eng. 2007, VI (XVI), 675–680.
59. Patterson, S. R.; Magrab, E. B. Design and Testing of a Fast Tool Servo for Diamond Turning. Precis. Eng. 1985, 7 (3), 123–128.
60. Ludwick, S. J., Jr. A Rotary Fast Tool Servo for Diamond Turning of Asymmetric Optics. Ph.D. Thesis, Massachusetts Institute of Technology, June 1999.
61. Falter, P. J.; Dow, T. A. A Diamond Turning Apparatus for Fabrication on Non-rotationally Symmetrical Surfaces. In Ultraprecision in Manufacturing Engineering: Proceedings
of the International Congress for Ultraprecision Technology, Aachen, FRG; May 1988; p 187.
62. Fawcett, S. C. Small Amplitude Vibration Compensation for Precision Diamond Turning. Precis. Eng. 1990, 12, 91–96.
63. Dow, T. A.; Miller, M. H.; Falter, P. J. Application of a Fast Tool Servo for Diamond Turning of Nonrotationally Symmetric Surfaces. Precis. Eng. Oct 1991, 13 (4), 243–250.
64. Falter, P. J.; Dow, T. A. Design and Performance of a Small-scale Diamond Turning Machine. Precis. Eng. Oct 1987, 9 (4), 185–190.
65. Falter, P. J. Diamond Turning of Non-rotationally Symmetric Surfaces. Ph.D Thesis, Department of Mechanical Engineering, North Carolina State University, 1990.
66. Cuttino, J. F.; Miller, A. C.; Schinstock, D. E. Performance Optimization of a Fast Tool Servo for Single-point Diamond Turning Machines. IEEE/ASME Trans. Mechatron. 1999,
4 (2), 169–179.
67. Okazaki, Y. A Micro-positioning Tool Post Using a Piezoelectric Actuator for Diamond Turning Machines. Precis. Eng. 1990, 12 (3), 151–156.
68. Hara, Y.; Motonishi, S.; Yoshida, K.; Ikawa, N. A New Micro-cutting Device with High Stiffness and Resolution. Ann. CIRP 1990, 39, 375–378.
69. Horiuchi, O.; Kanebako, H.; Ohira, N. Compensation of the Relative Motion Errors between Tool and Work in Ultra Precision Machining. In International Progress in Precision
Engineering: Proceedings of the 8th International Precision Engineering Seminar; 1995; pp 253–256.
70. Rasmussen, J. D.; Tsao, T. C.; Hanson, R. D.; Kapoor, S. G. A Piezoelectric Tool Servo System for Variable Depth of Cut Machining. In Precision Machining: Technology and,
Machine Development and Improvement, Vol. PED-58; ASME, 1992; pp 119–130.
71. Rasmussen, J. D.; Tsao, T. C.; Hanson, R. D.; Kapoor, S. G. Dynamic Variable Depth of Cut Machining Using Piezoelectric Actuators. Int. J. Mach. Tool. Manuf. 1994, 34 (3),
379–392.
72. Kouno, E.; McKeown, P. A. A Fast Response Piezoelectric Actuator for Servo Correction of Systematic Errors in Precision Machining. CIRP Ann. – Manuf. Technol. 1984,
33 (1), 369–372.
73. Zhu, W. H.; Jun, M. B.; Altintas, Y. A Fast Tool Servo Design for Precision Turning of Shafts on Conventional CNC Lathes. Int. J. Mach. Tool. Manuf. 2001, 41, 953–965.
74. Woronko, A.; Huang, J.; Altintas, Y. Piezoelectric Tool Actuator for Precision Machining on Conventional CNC Turning Centers. Precis. Eng. 2003, 27, 335–345.
75. Taniguchi, N. Current Status in, and Future Trends of, Ultraprecision Machining and Ultrafine Materials Processing. CIRP Ann. – Manuf. Technol. 1983, 32 (2), 573–582.
76. Takasu, S.; Masuda, M.; Nishiguchi, T.; Kobayash, A. Influence of Study Vibration with Small Amplitude upon Surface Roughness in Diamond Machining. CIRP Ann. – Manuf.
Technol. 1985, 34 (1), 463–467.
77. Kim, J. D.; Kim, D. S. Waviness Compensation of Precision Machining by Piezoelectric Micro Cutting Device. Int. J. Mach. Tool. Manuf. 1998, 38, 1305–1322.
78. Kim, H. S.; Kim, E. J. Feed-forward Control of Fast Tool Servo for Real-time Correction of Spindle Error in Diamond Turning of Flat Surfaces. Int. J. Mach. Tool. Manuf. 2003,
43 (12), 1177–1183.
79. Kohno, T.; Okazaki, Y.; Ozawa, N.; Mitui, K.; Omoda, M. In-process Measurement and a Workpiece-referred Form Accuracy Control System (WORFAC): Concept of the Method
and Preliminary Experiment. Precis. Eng. 1989, 11 (1), 9–14.
80. Miller, M. H.; Garrard, K. P.; Dow, T. A.; Taylor, L. W. A Controller Architecture for Integrating a Fast Tool Servo into a Diamond Turning Machine. Precis. Eng. 1994, 16 (1),
42–48.
81. Pahk, H. J.; Lee, D. S.; Park, J. H. Ultra Precision Positioning System for Servo Motor–piezo Actuator Using the Dual Servo Loop and Digital Filter Implementation. Int. J. Mach.
Tool. Manuf. 2001, 41 (1), 51–63.
82. Gao, W.; Tano, M.; Araki, T.; Kiyono, S.; Park, C. H. Measurement and Compensation of Error Motions of a Diamond Turning Machine. Precis. Eng. 2007, 31 (3), 310–316.
83. Xu, W. L.; Han, L. Piezoelectric Actuator Based Active Error Compensation of Precision Machining. Meas. Sci. Technol. 1999, 10 (2), 106.
84. Gan, S. W. Development of Hybrid Fine Tool Servo System for Nano-machining. Doctoral Dissertation, National University of Singapore, 2009.
85. Kim, H. S.; Lee, K. I.; Lee, K. M.; Bang, Y. B. Fabrication of Free-form Surfaces Using a Long-stroke Fast Tool Servo and Corrective Figuring with On-machine Measurement.
Int. J. Mach. Tool. Manuf. 2009, 49 (12–13), 991–997.
86. Rakuff, S.; Cuttino, J. F. Design and Testing of a Long-range, Precision Fast Tool Servo System for Diamond Turning. Precis. Eng. 2009, 33, 18–25.
87. Sosnicki, O.; Pages, A.; Pacheco, C.; Maillard, T. Servo Piezo Tool SPT400MML for the Fast and Precise Machining of Free Forms. Int. J. Adv. Manuf. Technol. 2010, 47,
903–910. http://dx.doi.org/10.1007/s00170-009-2140-6.
88. Amplified Piezo Actuator, http://www.cedrat-technologies.com.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
88 Fast and Fine Tool Servo for Ultraprecision Machining

89. Buescher, N.; Dow, T. A.; Sohn, A.; Norlund, B.; Roblee, J. Live-axis Turning. In ASPE Proceedings, Annual Meeting; 2004.
90. Buescher, N.; Dow, T. A.; Sohn, A. Live-axis Turning. In ASPE Proceedings, Annual Meeting; 2005.
91. Zdanowicz, E. M. Design of a Fast Long Range Actuator-flora II. Master Dissertation, NCSU, 2009.
92. Marten, B. F. Design and Control of a Long Stroke Fast Tool Servo. Ph.D. Dissertation, MIT, 2005.
93. Fast Tool Servo (FTS 1000 and FTS 500), http://www.precitech.com/machine-applications/fast-tool-servo/.
94. Nanotech Fast Tool Servo (NFTS-6000), http://www.nanotechsys.com/accessories/nanotech-250upl-factory-options/.
95. Weck, M. Limits of Workpieces Accuracy Caused by the Geometrical Behaviour of Ultraprecision Diamond Turning Machines. In Ultraprecision in Manufacturing Engineering:
Proceedings of the International Congress for Ultraprecision Technology, Aachen, Germany; May 1988; pp 153–170.
96. Liu, Q.; Zhou, X.; Wang, L. A New Hybrid Macro- and Micro-range Fast Tool Servo. In 2010 International Conference on Mechanic Automation and Control Engineering
(MACE); 2010; http://dx.doi.org/10.1109/MACE.2010.5535544.
97. Liu, Q.; Zhou, X.; Xu, P.; Zou, Q.; Lin, C. A Flexure-based Long Stroke Fast Tool Servo for Diamond Turning. Int. J. Adv. Manuf. Technol. 2012, 59, 859–867.
98. Tohme, Y. E.; Lowe, J. A. Machining of Freeform Optical Surfaces by Slow Slide Servo Method. In Proceedings of the ASPE, Annual Meetings; 2003.
99. Brecher, C.; Niehaus, F.; Merz, M.; Schmidt, K. Machining Microstructures on Free-form Surfaces. Prod. Res. Micro NanoTechnol. 2008, 6–13.
100. Neo, W. K.; Kumar, A. S.; Rahman, M. A Novel Method for Layered Tool Path Generation in the Fast Tool Servo Diamond Turning of Non-circular Microstructural Surfaces.
Proc. Inst. Mech. Eng., Part B 2013, 227 (2), 210–219.
101. Paul, E.; Evans, C. J.; Mangamelli, A.; McGlauflin, M. L. Chemical Aspects of Tool Wear in Single Point Diamond Turning. Precis. Eng. 1996, 18 (1), 4–19.
102. Brinksmeier, E.; Preuss, W. Micro-machining (Review). Philos. Trans. Roy. Soc. A 2012, 370, 3973–3992. http://dx.doi.org/10.1098/rsta.2011.0056.
103. Brinksmeier, E.; Autschbach, L. Ball-end Milling of Free-form Surfaces for Optical Mold Inserts. In Proceedings of ASPE, Annual Meetings; 2004.
104. Moriwaki, T.; Shamoto, E. Ultrasonic Elliptical Vibration Cutting. Ann. CIRP 1995, 44 (1), 31–34.
105. Takeuchi, Y.; Sata, T. Ultraprecision 3D Micromachining of Glass. Ann. CIRP 1996, 45 (1), 401–404.
106. Matsumura., T.; Ono, T. Cutting Process of Glass with Inclined Ball End Mill. J. Mater. Process. Technol. 2008, 200, 356–363.
107. Bifano, T. G.; Dow, T. A.; Scattergood, R. O. Ductile-regime Grinding – A New Technology for Machining Brittle Materials. J. Eng. Indus. 1991, 113.
108. Arif, M.; Rahman, M.; Wong, Y. S. Analytical Model to Determine the Critical Feed per Edge for Ductile–Brittle Transition in Milling Process of Brittle Materials. Int. J. Mach.
Tool. Manuf. 2011, 51 (3), 170–181.
109. Neo, W. K.; Kumar, A. S.; Rahman, M. A Review on the Current Research Trends in Ductile Regime Machining. Int. J. Adv. Manuf. Technol. 2012; http://dx.doi.org/10.1007/
s00170-012-3949-y.
110. Woon, K. S.; Rahman, M. Extrusion-like Chip Formation Mechanism and Its Role in Suppressing Void Nucleation. CIRP Ann. – Manuf. Technol. August 2010, 59 (1),
129–132.
111. Steinkopf, R.; Dick, L.; Kopf, T.; Gebhardt, A.; Risse, S.; Eberhardt, R. Data Handling and Representation of Freeform Surfaces. Proc. SPIE 2011, 8169, 81690X.
112. Lee, J. M.; Kim, S. J.; Kang, H. C.; Lee, D. Y. Fast Tool Servo Cutting Simulation Using CATIA API. In Proceedings of the Fifth International Symposium on Mechanics,
Aerospace and Informatics Engineering; 2011.
113. Komanduri, R. Some Aspects of Machining with Negative Rake Tools Simulating Grinding. Int. J. Mach. Tool. Des. Res. 1971, 11 (2), 223–233.
114. Günay, M.; Korkut, I.; Aslan, E.; Şeker, U. Experimental Investigation of the Effect of Cutting Tool Rake Angle on Main Cutting Force. J. Mater. Process. Technol. 2005, 166 (1),
44–49.
115. Biddut, A. Q.; Rahman, M.; Neo, K. S.; Rahman, K. M. R.; Sawa, M.; Maeda, Y. Performance of Single Crystal Diamond Tools with Different Rake Angles during Micro-grooving
on Electroless Nickel Plated Die Materials. Int. J. Adv. Manuf. Technol. 2007, 33 (9–10), 891–899.
116. Pramanik, A.; Neo, K. S.; Rahman, M.; Li, X. P.; Sawac, M.; Maeda, Y. Ultra-precision Turning of Electroless-nickel: Effect of Phosphorus Contents, Depth-of-cut and Rake
Angle. J. Mater. Process. Technol. 2008, 208 (1–2), 400–408.
117. DeVries, W. R. Analysis of Material Removal Processes, 1st ed.; Springer-Verlag: New York, 1992.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 11.05 Compound and Hybrid Micromachining Processes
M Rahman, ABMA Asad, and YS Wong, National University of Singapore, Singapore
MP Jahan, Western Kentucky University, Bowling Green, KY, USA
T Masaki, Masaki Giken, Osaka, Japan
 2014 Elsevier Ltd. All rights reserved.

11.05.1 Introduction 89
11.05.2 Characteristics of the Compound and Hybrid Micromachining Processes 90
11.05.3 Requirements of Compound and Hybrid Processes 90
11.05.4 Compound Micromachining Processes and Related Applications 92
11.05.4.1 Micro-EDM and Microturning 92
11.05.4.2 Micro-EDM and Microgrinding 93
11.05.4.3 Micro-EDM and Micromilling 94
11.05.4.4 Micro-EDM and Micro-ECM 94
11.05.4.5 X-ray Lithography, Electroplating, and Molding (LIGA) and Micro-EDM 96
11.05.4.6 Sequential Laser and Micro-EDM Drilling 97
11.05.4.7 Micro-EDM and Laser Welding 97
11.05.4.8 Sequential Laser and Mechanical Microdrilling 99
11.05.4.9 Combined Micro-EDM Milling and Laser Ablation Process 100
11.05.5 Hybrid Micromachining Processes and Related Applications 101
11.05.5.1 Combined Microgrinding and Microelectrochemical Machining 101
11.05.5.2 Micro-EDM and Micro-USM Combined Process 102
11.05.5.3 Vibration-Assisted Micro-EDM 103
11.05.5.4 Powder-Mixed Micro-EDM 104
11.05.5.5 Microelectrochemical Discharge Machining (Micro-ECDM) 105
11.05.5.6 Magnetic-Assisted Micro-EDM 106
11.05.5.7 Laser-Assisted Microturning 106
11.05.5.8 Laser-Assisted Micromilling 108
11.05.5.9 Laser-Assisted Microgrinding 109
11.05.5.10 Laser Microdrilling with Jet Electrochemical Machining 110
11.05.6 Summary 111
References 111

11.05.1 Introduction

The trend in miniaturization of products is pervasive in areas such as information technology, biotechnology, environmental
technology, and the medical industries (1). Micromachining is the key supporting technology that has to be developed to meet the
challenges posed by the requirements of product miniaturization. However, in many cases, a single micromachining process
cannot fulfill all of the requirements, due to the limitation of that process. For example, the machining of difficult-to-cut materials,
at both the macro- and microscales, has become a challenge in the manufacturing industries. The conventional micromachining
processes (e.g., microturning, micromilling, and microgrinding) cannot machine these materials easily due to their extreme
hardness, brittleness, and toughness. However, several nonconventional machining processes (e.g., microelectrodischarge
machining (micro-EDM), microelectrochemical machining (micro-ECM), and laser micromachining) are found to be capable of
machining difficult-to-cut materials irrespective of their hardness. However, these nonconventional machining processes also have
several disadvantages, such as lower machining speed, a heat-affected zone (HAZ), higher tool wear, and poor surface finish.
Therefore, the development of micromachining processes combining nonconventional micromachining with conventional
micromachining is of prime importance.
In recent years, compound and hybrid micromachining has become the most promising technology for the production of
miniaturized parts and components. This technology is becoming increasingly more important and popular because of a growing
demand for industrial products, with an increased number not only of functions but also of reduced dimensions, higher
dimensional accuracy, and better surface finish. Compound and hybrid machining is the combination of processes and/or
machines to produce parts in a more efficient and productive way (2). Although the terms compound micromachining and hybrid
micromachining are often used as if they have the same meaning, there are differences in the meanings of the two processes.
Compound machining is defined as the combination of two different machining processes in a single setup applied one after
another; the hybrid machining process is defined as the integrated application or combination of different physically active

Comprehensive Materials Processing, Volume 11 http://dx.doi.org/10.1016/B978-0-08-096532-1.01105-5 89

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
90 Compound and Hybrid Micromachining Processes

principles in a single process. In a CIRP (College International pour la Recherche en Productique, or International Academy for
Production Engineering) keynote presentation, the hybrid manufacturing processes are defined as follows: “Hybrid manufacturing
processes are based on the simultaneous and controlled interaction of process mechanisms and/or energy sources/tools having
a significant effect on the process performance” (2).
The hybrid material removal processes can machine microfeatures on a wide range of materials and machined shapes, so that flat
surfaces, arbitrary curvatures, and long shafts can be machined, which are required for moving parts and guiding structures (3). The
hybrid micromachining techniques are able to fabricate microparts and components, with potential applications in the electronics,
optics, biotechnology, automotive, communications, and avionics industries.

11.05.2 Characteristics of the Compound and Hybrid Micromachining Processes

The development of compound and hybrid micromachining processes by integrating conventional and nonconventional processes
on the same machine has several advantages over a single process. Some of the important objectives of developing compound and
hybrid micromachining processes are as follows (2):
l To machine materials that could not be machined earlier
l Lower process forces
l Less tool wear
l Higher productivity
l Required shapes of the microfeatures

Some of the important characteristics of the compound and hybrid machining processes are as follows (2):
l Sequential or simultaneous and controlled interaction
l Interaction that is more or less in the same machining zone and at the same time
l Interaction of process mechanisms, energy sources, and tools
l Significant effect on the process performance
l Better material removal, lower process forces, less tool wear, improved surface finish, and overall improved performance

11.05.3 Requirements of Compound and Hybrid Processes

In order to achieve effective implementation of compound and hybrid micromachining techniques, four important areas need to be
addressed (1):
l Development of a machine tool capable of both conventional and nonconventional micromachining
l Motion and process control
l Process development to achieve the necessary accuracy and quality
l On-machine measurement and inspection

Figure 1 shows the technologies required for successful development of compound and hybrid micromachining processes. One
of the main difficulties in compound micromachining is the availability of an appropriate machine tool that can be used for the
development of compound and hybrid micromachining processes. Most machine tools capable of nonconventional machining are
not designed to perform conventional machining processes. Furthermore, most of the machine tools do not facilitate the
measurement of fabricated products on machine, which has the potential to be used as feedback and to compensate tool trajectory

Figure 1 Technologies required for compound and hybrid micromachining.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining Processes 91

online. Another constraint is imposed by the precision required for such fabrication, which most conventional computer numerical
control (CNC) machine tools cannot fulfill. Mechanical and thermal deformation, chatter vibration, tooling, and its clamping
usually constrain making small, superprecise parts with large, conventional machines. Ultraprecision machines that provide a high
degree of motion accuracy are extremely expensive and mostly do not include facilities for compound processes. Therefore, the first
and the most important requirement for the development of compound and hybrid machining processes is the development of
a multipurpose machine tool or the integration of facilities for performing two or more manufacturing processes in one single setup
and platform.
Attempts were made to perform compound micromachining by modifying a machine tool good for one process as an adap-
tation for the supportive compound process (e.g., modification of a micro-EDM machine for micromilling) (4–6). Although the
capability of compound and hybrid micromachining was demonstrated, the kinds of machines used in these processes encounter
performance difficulties in the full scale due to the setup being too weak to support all the processes equally. Moreover, the
conversion of a machine tool, which was developed for a dedicated process, to perform another machining process requires
a considerable amount of effort and investment.
Therefore, for successful compound and hybrid micromachining processes, there is a need for a unique dedicated platform to be
called a universal miniature machine tool for performing multiple processes at the lower boundary of the micromachining domain
readily on a single platform. This would ensure that the underlying equipment hardware is capable of benefitting the multiple-
process needs for execution and realization of the art of micromachining. On a miniaturized machine tool, the thermal expan-
sion can be minimized, and advanced sensors and instruments can be employed to compensate for positioning errors. In addition
to significant contributions in accuracy, it would also enable tool-based micromachining as a completely feasible option. Besides, it
would also speed up the integrated fabrication process by saving the reclamping time and relieving operators from the handling
needs during the interprocess transfer of such miniaturized components. Figure 2 shows an example of such a multipurpose
machine tool developed for compound and hybrid micromachining processes (1).
With the development of a precision mechanical structure, a specialized precision motion control system is another major
requirement for compound and hybrid micromachining processes. For conventional micromachining, the CNC system is expected
to provide the function of synchronized servo feed control based on real-time monitoring of process control parameters, which are
sometimes difficult to integrate into the nonconventional machine tool. For example, the accuracy and speed of micro-EDM,
a nonconventional process, depend a great deal on the gap control performance of the motion controller. This might be some-
times very difficult to implement with a commercially available motion controller. Similarly, the motion required for micro-EDM

Figure 2 An example of a multipurpose machine tool developed for compound and hybrid micromachining (reproduced from Rahman, M.; Asad, A. B.
M. A.; Masaki, T.; Saleh, T.; Wong, Y. S.; Kumar, A. S. A Multiprocess Machine Tool for Compound Micromachining. Int. J. Mach. Tools Manuf. 2010, 50
(4), 344–356). The machine is capable of performing conventional machining (e.g., microturning, micromilling, and microgrinding) and nonconventional
machining (e.g., micro-EDM, micro-ECM, and micro-wire electrodischarge machining (micro-WEDM)) in a single setup.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
92 Compound and Hybrid Micromachining Processes

milling can be another example of a special trajectory requirement. An open-architecture motion controller can be adopted for
compound micromachining, which can be programmed to provide necessary trajectory control of the machine tool required for the
nonconventional machining. The open-architecture motion controller has a multiprocessor-based design with a universal asyn-
chronous receiver–transmitter (UART)-based bus network communication to distribute processing tasks between multiple
processors. Thus, programs and scheduling can be developed for real-time process control accordingly to meet the synchronized
servo motion required for some of the micromachining processes. The motion controller can execute a program downloaded from
the host computer independently using high-speed communication, which creates a very user-friendly environment for the operator
on a standard PC.
Research work needs to be undertaken toward the understanding of process physics to provide relevant background data for
modeling, measurement, identification of control parameters, and application of feedback control for successful implementation of
compound and hybrid manufacturing processes. The difficulties of microfabrication using available techniques need to be realized,
and then compound processes need to be proposed to complement the weaknesses of different processes. Finally, an on-machine
measurement system needs to be developed for on-machine inspection of machined features by one process before further
machining operations by other processes.

11.05.4 Compound Micromachining Processes and Related Applications

11.05.4.1 Micro-EDM and Microturning
Microturning has the capability to produce three-dimensional (3D) structures on a microscale. The major drawback of the
microturning process is the limit of machinable sizes and the fact that the cutting forces influence machining accuracy (7). It is very
difficult to achieve straight shaft below 100 mm diameter, and in many cases the tool either breaks or starts to wobble due to
excessive radial cutting force on the microshaft. Therefore, a compound process has been developed where the commercial cutting
tool is modified using the micro-EDM process to reduce the force component responsible for breaking of the shaft (7,8).
This compound process is the combination of micro-EDM and microturning in a single setup. First, a commercially
available polycrystalline diamond (PCD) tool is modified by the micro-EDG process to reduce the nose radius of the cutting
tool, thus minimizing the force component that causes shaft deflection during microturning. Commercially available PCD
inserts, designed for a light finishing cut, have a relatively large tool nose radius (e.g., 100 mm). This tool nose resolves the
cutting force on the shaft into two components, namely Fx and Fy, as can be seen in Figure 3(a) (8). The Fy component of the
cutting force does the actual cutting, while the Fx component causes deflection of the microshaft. A commercially available PCD
insert is modified using the micro-EDG process to achieve a very sharp cutting edge, so as to reduce the Fx component of the
cutting force significantly. This modification of the cutting tool makes it possible to achieve a straight shaft of much smaller
diameter. The compound process combining micro-EDG and microturning is presented schematically in Figure 3(b) (8).
A dual-cutter setup is arranged for microturning, one with a round nose for initial turning up to 100 mm, then a sharp tool for
a final cut up to 20 mm.
After the microturning process, the fabricated microelectrodes are used in machining of small and higher aspect ratio microholes
by micro-EDM on the same machine (Figure 1(b)). Therefore, this compound process is in fact a combination of three steps:
modifying the cutting tool using micro-EDG, fabricating microshafts using microturning, and applying fabricated shafts in micro-
EDM drilling. Figure 4 illustrates the concept of the microturning–micro-EDM compound machining process. An electrode of
required dimension is first fabricated by microturning prior to micro-EDM (7). Using this compound process, clamping error can be
avoided, and deflection of electrode can be minimized; consequently, the accuracy of machining can be improved. The fabricated
microelectrode and machined microholes using the microelectrode are presented in Figure 5(a) and 5(b), respectively (1).

Figure 3 (a) Modification of a conventional cutting tool using the micro-EDG (variant of micro-EDM) process; and (b) a schematic representing
the compound process combining modification of a cutting tool by micro-EDM and turning of a microshaft by a modified tool tip. Reproduced from
Asad, A. B. M. A.; Masaki, T.; Rahman, M.; Lim, H. S.; Wong, Y. S. Tool-based Micro-machining. J. Mater. Process. Technol. 2007, 192–193, 204–211.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining Processes 93

Figure 4 Compound process of fabricating a microelectrode using microturning and applying a fabricated microelectrode in the micro-EDM drilling.
Reproduced from Lim, H. S.; Kumar, A. S.; Rahman, M. Improvement of Form Accuracy in Hybrid Machining of Microstructure. J. Electron. Mater. 2002,
31 (10), 1032–1038.

Figure 5 (a) A 19 mm graphite electrode of 0.5 mm length fabricated by a micro-EDG–microturning compound process; and (b) fabricated microholes
with the microelectrode obtained by the microturning–micro-EDM drilling compound process. Reproduced from Rahman, M.; Asad, A. B. M. A.; Masaki, T.;
Saleh, T.; Wong, Y. S.; Kumar, A. S. A Multiprocess Machine Tool for Compound Micromachining. Int. J. Mach. Tools Manuf. 2010, 50 (4), 344–356.

11.05.4.2 Micro-EDM and Microgrinding

The fabrication of a miniaturized (sub-100 mm) grinding tool for difficult-to-cut materials like PCD and tungsten carbide (WC)
is necessary for machining of microchannels with improved surface finish. The micro-EDM process is found to be capable of
machining any difficult-to-cut materials down to the desired dimension. Therefore, a compound process is developed to solve
the issues by combining the micro-EDM process with the microgrinding process. In this compound machining process, a PCD
tool is fabricated on a machine in a desired shape using the block micro-EDG process. The PCD tool contains randomly
distributed protrusions of diamond particles with dimensions around 1 mm that serve as the cutting edges for micromachining
on glass. When the dimension of the PCD tool is reduced to the required dimension of the grinding tool by the micro-EDG
process, the binder materials (usually nickel or WC) are removed because they are conductive, thus protruding the diamond
particles, which are nonconductive. PCD with a cobalt binder, which can be shaped with micro-EDG, is emerging as a tool
material for microgrinding of hard and brittle materials. The cobalt binder provides an electrically conductive network that can
be removed with EDM (5). The diamond cutting edges are exposed as the discharges erode away the cobalt binder. In addition
to microgrinding, reaming of microholes, grinding of microslots, and machining of V-grooves with a fabricated PCD tool have
been reported (9). Figure 6 shows the different steps of the micro-EDM–microgrinding compound process with a machining
example in BK-7 glass (8). As can be seen from Figure 6(c) and 6(d), the fabricated slot has a very fine and smooth surface,
which is comparable to the surface obtained from ductile mode cutting of glass in macroscale. Figure 7 shows the same
compound process for machining slots and microfeatures in ultralow-expansion (ULE) glass (5). However, the grinding tool
has been fabricated by the wire electrodischarge grinding (WEDG) process instead of the block micro-EDG process used in
Figure 6.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
94 Compound and Hybrid Micromachining Processes

Figure 6 (a) A schematic diagram showing the block micro-EDG process (a variant of micro-EDM); (b) a PCD tool before the micro-EDG process;
(c) fabrication of a microgrinding tool with the micro-EDG process; (d) microchannels on glass machined by the microgrinding process with a fabricated
PCD tool; and (e) surface finish of the microchannel in glass. Reproduced from Asad, A. B. M. A.; Masaki, T.; Rahman, M.; Lim, H. S.; Wong, Y. S. Tool-
based Micro-machining. J. Mater. Process. Technol. 2007, 192–193, 204–211.

Figure 7 (a) A PCD scratch tool produced with the WEDG process; (b) a scratch in ULE glass produced with the tool shown in (a); (c) a cylindrical
50 mm PCD tool used to cut pockets in ULE glass; and (d) a slot ground in ULE glass using the tool shown in (c). Reproduced from Morgan, C. J.;
Vallance, R. R.; Marsh, E. R. Micro-machining and Micro-grinding with Tools Fabricated by Micro Electro-discharge Machining. Int. J. Nanomanuf. 2006,
1 (2), 242–258.

11.05.4.3 Micro-EDM and Micromilling

This compound process is a combination of the micro-WEDG process and micromilling. In this process, first the micromilling tool
is fabricated out of the difficult-to-cut tool materials with the help of the micro-WEDG process. After that, the fabricated cutting tool
is used for a conventional micromilling operation in the same setup. The use of the micro-WEDG for the production of milling tools
has several advantages. The geometry can be changed quite easily, and the potential of scaling down the size of the milling tools is
very high (6). In comparison to other contactless machining technologies, micro-EDM has an acceptable machining time, and the
resulting costs for the machining are tolerable. An advantage of using micro-EDM with the milling process is the prevention of
inaccuracy by rechucking processes (6). Figure 8(a) shows the schematic diagram of the micro-WEDG process of fabricating
a milling cutting tool. Figure 8(b) and 8(c) shows the fabricated cutting tool and machined slot, respectively (6). Figure 9(a) shows
a scanning electron micrograph (SEM) of a 100 mm diameter WC microtool (5). The fabricated microslot and surface roughness of
the microslot after milling are shown in Figure 9(b) and 9(c), respectively (5). WC was chosen as the tool material in this
compound process because of its high hardness and low wear rate. Three-fourths of the cylinder was removed to provide a single
cutting edge, and then a 45 slice was also removed from the nose of the tool to provide clearance for various micromilling
applications. Microtools fabricated by WEDG have been used to remove material by mechanical cutting, rather than with electrical
discharges, to achieve better surface finishes and a higher material removal rate (MRR).

11.05.4.4 Micro-EDM and Micro-ECM

This compound process is the combination of micro-EDM and micro-ECM processes in a single setup. The objective is to improve
the surface finish generated by micro-EDM using micro-ECM as a postprocessing step. The surface machined by micro-EDM is

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining Processes 95

Figure 8 (a) A schematic diagram of the micro-WEDG process used to fabricate a micromilling tool; (b) a micromilling tool fabricated by micro-WEDG;
and (c) a microslot machined by micromilling using a fabricated microtool. Reproduced from Fleischer, J.; Masuzawa, T.; Schmidt, J.; Knoll, M. New
Applications for Micro-EDM. J. Mater. Process. Technol. 2004, 149 (249), 246.

Figure 9 (a) WC microtool by micro-WEDG of 100 mm diameter; (b) a micrograph of a square groove machined in AA3003 aluminum using the
fabricated microtool in (a); and (c) Ra of 121 nm on the bottom of the groove. Reproduced from Morgan, C. J.; Vallance, R. R.; Marsh, E. R. Micro-
machining and Micro-grinding with Tools Fabricated by Micro Electro-discharge Machining. Int. J. Nanomanuf. 2006, 1 (2), 242–258.

relatively rough due to microcraters and microcracks produced by the microdischarges. Hence, the process consisting of micro-
EDM followed by micro-ECM can be a suitable solution to improve the machined surface (10). The deionized water used in the
micro-EDM process can serve as an electrolyte medium for micro-ECM at certain machining conditions (11). The surface after
applying micro-ECM becomes much smoother compared to that of micro-EDM, and the peak-to-valley distances of craters (Rmax)
reduce significantly. Micro-ECM can also be applied for finishing the slot machined by micro-EDM milling. In addition to
a sequential micro-EDM and micro-ECM process, a combined or concurrent micro-EDM and micro-ECM process has also been
reported (11,12) that could be considered as a hybrid process. The difference is that in a hybrid process, the discharging of
dissolution takes place in the same cycle during machining, thus applying micro-EDM and micro-ECM concurrently. Figure 10
shows the steps of the micro-EDM–micro-ECM compound process (11) and the change in gap and current during the EDM and
ECM processes of the compound system (10). The improvements of surface finish with both the compound and hybrid processes
are presented in Figure 11 (12).

(a) (b) (c)

Figure 10 (a) A schematic diagram showing a micro-EDM and micro-ECM compound or hybrid process (reproduced from Nguyen, M. D.;
Rahman, M.; Wong, Y. S. Simultaneous Micro-EDM and Micro-ECM in Low-resistivity Deionized Water. Int. J. Mach. Tools Manuf. 2012, 54–55,
55–65); (b) the variation of gap distance with time for micro-EDM and micro-ECM in the compound process; and (c) the change of current due to
discharge and dissolution in two stages of EDM and ECM. Reproduced from Campana, S.; Miyazawa, S. Micro-EDM and ECM in DI Water. In
Proceedings of Annual Meeting of American Society of Precision Engineering (ASPE); 1999.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
96 Compound and Hybrid Micromachining Processes

Figure 11 (a) Surface generated by micro-EDM; (b) surface generated by micro-ECM followed by micro-EDM (a compound process); (c) surface
generated by the micro-EDM milling process; and (d) surface generated by the micro-EDM and micro-ECM combined milling process. Reproduced from
Zeng, Z.; Wang, Y.; Wang, Z.; Shan, D.; He, X. A Study of Micro-EDM and Micro-ECM Combined Milling for 3D Metallic Micro-structures. Precis. Eng.
2012, 36 (3), 500–509.

11.05.4.5 X-ray Lithography, Electroplating, and Molding (LIGA) and Micro-EDM

Despite its excellent capability in terms of precision, surface quality, and complex 3D formation, micro-EDM has not achieved
widespread use in product manufacturing primarily because of its productivity drawbacks (13). For example, in order to machine
arrays of microholes, a single electrode is used to machine one hole each time. However, if the numbers of holes are even higher,
then the electrodes need to be changed in the middle of the machining process due to the electrode wear. On the other hand,
photolithographic methods offer various paths to the fabrication of such arrays with arbitrary patterns on a substrate. The arrays of
electrodes fabricated by the photolithography process are precisely arranged on the substrate and have high structural uniformity
across the arrays, offering high precision and uniformity in the machined products. Moreover, only one electrode is used for
machining one hole, in contrast to the conventional serial-processing method of micro-EDM (13). Therefore, a compound
micromachining process has been developed that combines micro-EDM and LIGA (from the German “Lithographie, Galvano-
formung, Abformung”). In this process, the LIGA process fabricates arrays of microelectrodes, and then those microelectrodes are
applied by machining high-aspect-ratio microholes or microstructures using micro-EDM. The LIGA process uses X-ray lithography
to form high-aspect-ratio molds for electroplated structures. Micro-EDM produces 3D microstructures in any electrically conductive
materials. The steps of the LIGA process are presented in Figure 12 (13). In this compound process, first an array of negative-type
electrodes was fabricated in nickel using the LIGA process. After that, a positive-type patterned structure is produced by micro-EDM
using the arrays of microelectrodes. Figure 13(a) shows an example of a 20  20 array of high-aspect-ratio electrodes of electro-
plated copper with a 20 mm diameter, 60 mm pitch, and 300 mm structural height. The LIGA-fabricated array of 400 Cu electrodes
with 20 mm diameter was used to machine through-holes in 50 mm thick stainless steel using micro-EDM as shown in Figure 13(b)
(13). The machining time was about 5 min, which is 600 times less than that required for serial machining by a single electrode.
Arrayed electrodes of even complicated cross-section shapes, like hexagonal and gear shapes, can be fabricated using LIGA with
a high aspect ratio.

Figure 12 Schematic representation showing the step-by-step mechanism of the LIGA–micro-EDM compound process. Reproduced from
Takahata, K.; Gianchandani, Y. B. Batch Mode Micro-electro-discharge Machining. IEEE/ASME J. Microelectromech. Syst. 2002, 11, 102–110.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining Processes 97

Figure 13 (a) A 20  20 array of LIGA-fabricated copper electrodes; (b) through-holes batch machined in 50 mm thick stainless steel by the
micro-EDM process using the array electrodes shown in (a). Reproduced from Takahata, K.; Gianchandani, Y. B. Batch Mode Micro-electro-discharge
Machining. IEEE/ASME J. Microelectromech. Syst. 2002, 11, 102–110.

11.05.4.6 Sequential Laser and Micro-EDM Drilling

The current practice of using micro-EDM drilling for fabricating fuel injection nozzles is limited in terms of the hole size that it can
produce effectively and the length of time needed to drill. In addition, the tooling cost is also high (14). Attempts have been made
by a number of research groups and companies to use nanosecond pulsed laser drilling instead of EDM. However, the hole quality
cannot meet the industrial standard, as laser drilling typically produces larger recast layers and HAZs (15). Therefore, a compound
micromachining process combining laser machining and micro-EDM drilling one after another has been developed (16). In this
technique, pilot holes are first drilled by a nanosecond pulsed laser beam. Micro-EDM drilling then finishes the holes. The reason for
using laser machining for drilling pilot holes has two advantages. First, it will enhance the machining speed by reducing the overall
machining time, as laser machining is faster than the micro-EDM process. Second, the recast layer formation is higher in laser
machining, and hence it should not be used as the final process. Moreover, finish machining through micro-EDM could reduce the
HAZ generated by laser machining.
This hybrid process can be applied for faster machining of high-aspect-ratio microholes in difficult-to-cut materials with the
same dimensional accuracy and surface finish obtained by the micro-EDM drilling process. Using this hybrid process, the cycle time
of EDM could be reduced from 40 to 20 s by using a laser-drilled pilot hole. The initial laser hole was approximately 60 mm in
diameter, and the final EDM hole was 140 mm in diameter. It was found that this hybrid process has eliminated the problems of
recast and HAZs that are typically associated with the laser drilling process. The new process has enabled a 70% reduction in total
drilling time compared to standard EDM drilling, as less material is removed by the EDM. The quality of the holes is as good as with
direct EDM drilling, thus eliminating the need for recertification of the drilling process. The microholes fabricated using this
compound process can be used for the fuel injection nozzle, giving the minimum total drilling time and the best quality holes. The
technique has enabled valuable cost savings and increases in production capacity for next-generation fuel injection nozzle
manufacture. Figure 14 shows a schematic presentation of the sequential process and inner surface of the micro-EDM after laser
machining and the compound process (16). The improvement in surface finish after implementing the compound process is seen in
Figure 14(c).

11.05.4.7 Micro-EDM and Laser Welding

Microassembly is a feasible method to make 3D micrometal parts. However, the first challenge is how precisely to position the
microparts assembled. Online assembly may be one possible solution. The micro-EDM process can fabricate microparts and
components with desired accuracy. Also, laser welding is a very effective technique of microassembly. Therefore, for the online
assembly of parts, it is possible to integrate two processes in a single setup. A novel process has been developed to solve the issues by
combining the micro-EDM and Nd-YAG laser welding workstation in a single setup (17–19). The micro-EDM process fabricates the
assembled parts, and the Nd-YAG laser performs microjoining to produce the assembly, so the whole processes from micro-
fabrication to microassembly can be completed on the same system. The Nd-YAG laser possesses the advantageous features of
heating concentration and rapid cooling and is well suited to microwelding (17). This can overcome the problem of small-
dimensional assembly. The system, having precision stages and ultraprecision motion control technology, can control the posi-
tional accuracy precisely between the assembled parts. In this compound process, parts can be machined and fuse-welded in an
online process, and, at the same time, precise positioning can be attained by controlling the ultraprecise motion of stages on the
machine. Furthermore, the fusion-welded process can attain sufficient bonding strength. A laser can also be utilized to separate the
assembled parts, so some redundant forces can be avoided during the separation process. Therefore, an adequately high aspect ratio
can be easily attained. Figure 15 shows the steps of the laser–micro-EDM combined process (18). Figure 16 shows examples of
assembled microfeature parts using this compound process (17,19).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
98 Compound and Hybrid Micromachining Processes

Figure 14 (a) Schematic representation of the concept of a sequential laser and micro-EDM process; (b) the inner surface of the microhole generated
after the pilot hole by laser machining; and (c) the inner surface of the microhole after sequential a laser and micro-EDM process. Reproduced
from Lin, L.; Diver, C.; Atkinson, J.; Giedl-Wagner, R.; Helm, H. J. Sequential Laser and EDM Micro-drilling for Next Generation Fuel Injection Nozzle
Manufacture. Ann. CIRP 2006, 55 (1), 179–182.

Figure 15 The process of pin–plate microassembly. (a) Making the pin with WEDG; (b) the pin held by the spindle for micro-EDM; (c) making
a hole under micro-EDM; (d) reworking the pin end to eliminate wear during hole machining; (e) the first spot at point ‘a’; and (f) fusing and separating
the pin from the back via only one laser beam emission. Reproduced from Kuo, C.-L.; Huang, J.-D.; Liang, H.-Y. Precise Micro-assembly through an
Integration of Micro-EDM and Nd-YAG. Int. J. Adv. Manuf. Technol. 2002, 20, 454–458.

Figure 16 (a) Pin–plate microassembly (microjoining of two microrods that has been fabricated by micro-WEDG) (reproduced from Huang, J.-D.;
Kuo, C.-L. Pin-plate Micro Assembly by Integrating Micro-EDM and Nd-YAG Laser. Int. J. Mach. Tools Manuf. 2002, 42, 1455–1464); and
(b) a microscopic view of a tungsten pin assembled into an SUS 304 plate. Reproduced from Kuo, C.-L.; Huang, J.-D.; Liang, H.-Y. Fabrication of 3D
Metal Microstructures Using a Hybrid Process of Micro-EDM and Laser Assembly. Int. J. Adv. Manuf. Technol. 2003, 21, 796–800.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining Processes 99

11.05.4.8 Sequential Laser and Mechanical Microdrilling

Laser percussion drilling is inherently associated with poor geometry and thermal defects. The challenge for laser drilling is the
control of drilling quality in terms of minimizing the recast layer and HAZ (20). In addition, drilling hollow parts such as airfoil
blades and fuel injector nozzles without damage to the back wall is one of the biggest challenges in laser drilling. Although
mechanical microdrilling produces good-quality holes, premature drill breakage and drilling holes at acute angles are the areas of
major concern. During the drilling of microholes at an acute angle, the microdrill tip is subjected to high lateral force at the
contact point with the workpiece surface. This lateral force makes the tip diverge from the required position and causes the
microdrill to bend and eventually break the tool. In addition, burr formation in mechanical drilling is another unavoidable issue,
which affects workpiece accuracy and quality (21). Therefore, a compound process combining both laser drilling and mechanical
microdrilling has been developed in order to reduce the challenges with both the processes (22). The compound process can
overcome the limitations of tip divergence and low tool stiffness in pure mechanical microdrilling, especially for drilling at acute
angles. It also can solve the issues of poor geometry, HAZs, recast layer formation, and back-wall damage problems associated
with laser microdrilling.
In this compound process, a laser beam first drills a pilot hole, then an end mill is used to machine the diffuser portion of
the hole and provide a flat surface for the drill entrance side. Micromechanical drilling then finishes the holes. During the
machining of the pilot hole, the laser beam was focused to a beam diameter of 240 mm at the workpiece surface. In order to
drill inclined holes at an inclination angle of 30 to the surface, the nozzle and the axis of the laser beam were inclined, while
the workpiece remained in a horizontal position. For subsequent drilling, the centers of the two holes were accurately aligned.
The workpiece was mounted on a fixed jig on the machine, and then four alignment holes were drilled on the workpiece
corners. The workpiece was removed and fixed on the laser machine CNC table. A visible laser guide was then used to locate the
four holes to be the reference points for the CNC table. A matrix of laser holes was drilled in the workpiece. After that,
the workpiece was mounted on the same place on the fixed jig on a drilling machine for the mechanical drilling process. The
recorded eccentricity between the laser hole and the drill was less than 10 mm. Figure 17 clearly shows the reduction of the HAZ
and improvement of surface finish from the single laser drilling process to the compound process (22). The improvement of
tool life in the compound process can be realized from Figure 18 (22).

Figure 17 Comparison of HAZ and burr formation during the machining of an inclined hole using (a) laser drilling; (b) mechanical microdrilling;
and (c) sequential laser and mechanical microdrilling. Reproduced from Okasha, M. M.; Mativenga, P. T.; Driver, N.; Li, L. Sequential Laser and
Mechanical Micro-drilling of Ni Superalloy for Aerospace Application. CIRP Ann. – Manuf. Technol. 2010, 59, 199–202.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
100 Compound and Hybrid Micromachining Processes

Figure 18 Comparison of the tool life between the mechanical microdrilling process and the sequential laser and mechanical microdrilling
process. Reproduced from Okasha, M. M.; Mativenga, P. T.; Driver, N.; Li, L. Sequential Laser and Mechanical Micro-drilling of Ni Superalloy for
Aerospace Application. CIRP Ann. – Manuf. Technol. 2010, 59, 199–202.

11.05.4.9 Combined Micro-EDM Milling and Laser Ablation Process

One of the issues with laser beam micromachining (LBM) with ultrashort laser pulses is that the machining times are slow if good-
surface qualities are required. In contrast, micro-EDM is capable of removing material nearly without process forces, but with
a higher processing speed if structure sizes exceed a certain value. With the combination of both the processes, specific advantages
can be realized while disadvantages can be partially eliminated. Therefore, a compound machine tool was developed that combines
the two processes of ultrashort-pulsed laser ablation and micro-EDM milling (23). No reclamping is necessary in this compound
process. The limitations and challenges of the single processes have been reduced in the proposed combined process (23). The
developed process can be used for the efficient and economic manufacturing of microstructures in materials that are difficult to
machine by conventional processes.
Both processes have different requirements for the kinematics. Micro-EDM needs only very low velocities of the axes. Due to the
higher pulse frequency, the LBM process requires axes with considerably higher dynamics. Furthermore, the control of the axes for the
micro-EDM process is relatively complex because the distance of the electrode to the workpiece has to be readjusted constantly.
Therefore, control of the vertical axe and the generator from the EDM device has been integrated into the hybrid machine. The
developed novel hybrid machine tool is shown in Figure 19 (23). To calculate the optimal sequence of the two processes, an algorithm
was developed that evaluates all possible combinations of manufacturing micro-EDM with the different electrode diameters and pulse
shapes: LBM with different pulse energies extracts the sequence with the lowest manufacturing costs. A given structure is therefore
divided into different parts that can be machined only by a particular process (e.g., if the structure contains details that are smaller than
100 mm, it can be machined only by micro-EDM with a 50 mm electrode or by LBM). A test structure that consists of a square-shaped
part of 300  300  40 mm3 and a channel of 100  200  40 mm3 was machined by LBM only and the combined process of micro-
EDM and LBM, as shown in Figure 20 (23). Although there is no significant change in the surface finish, the manufacturing cost has
been reduced from V22.6 to V9.75 (57%), and machining time has been reduced from 1344 to 578 s (23) in the combined process.

Figure 19 CAD schema (left) and machine compartment (right) for the developed combined micro-EDM and LBM process. Reproduced from
Weber, P.; Haupt, S.; Schulze, V. Hybrid Machining of Microstructures Using a Combination of Electrical Discharge Machining Milling and Laser Ablation.
J. Vac. Sci. Technol. B 2009, 27 (3), 1327–1329.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining Processes 101

Figure 20 Test structure in TSF44 machined (a) only by LBM and (b) by the micro-EDM and LBM combined process. Reproduced from Weber, P.;
Haupt, S.; Schulze, V. Hybrid Machining of Microstructures Using a Combination of Electrical Discharge Machining Milling and Laser Ablation. J. Vac.
Sci. Technol. B 2009, 27 (3), 1327–1329.

11.05.5 Hybrid Micromachining Processes and Related Applications

11.05.5.1 Combined Microgrinding and Microelectrochemical Machining
This is a hybrid process of grinding and electrochemical removal for machining of precise small holes in hard-to-machine materials
(24). In the process, a spherical metal rod with coated diamond abrasives is used as a cathode tool, which rotates at high speed and
removes material electrochemically and mechanically for a premachined pilot hole. The diameter of the pilot hole is smaller than
the required diameter and is usually machined by micro-EDM drilling. This process has been developed for machining difficult-
to-cut materials. However, the process can be applied to other materials also as long as the material is electrically conductive.
The tool core should be electrically conductive, and the abrasive needs to be electrically nonconductive. During the machining
process, the tool acts as the cathode and the workpiece performs as the anode. The abrasive diamond particles in the tool protrude
beyond the conductive bond surface (i.e., the nickel layer). This establishes a small gap between the tool’s nickel layer and the hole’s
side wall. Figure 21 shows a schematic representation of the hybrid process of microelectrochemical machining and mechanical
microgrinding (24).
In this hybrid process, material removal occurs in two phases. The electrolytic action begins when the gap is filled with an
electrolyte and the tool is electrically charged, and it results in electrochemical dissolution. Therefore, phase 1 is entirely electro-
chemical action. In this phase, passivation film occurs on the microhole surface due to the application of passive NaNO3 as an
electrolyte. Phase 2 is a combination of electrochemical action and mechanical grinding. As the abrasive tool goes downward, the
gap decreases until the abrasives on the tool base come into contact with the workpiece. The abrasive grains remove the soft,
nonreactive passivation layer by mechanical grinding action, thus exposing fresh metal for an electrolytic reaction. Simultaneously,
the electrolyte trapped between the protruding abrasive grains and the workpiece forms tiny electrolytic cells, thus electrochemical
dissolution of workpiece materials occurs. Phase 2 ends at the point of maximum tool diameter. For obtaining sharp edges and
high-dimensional accuracy of holes, the tool should be insulated except for the first half of the sphere. This is done in phase 3, and

Figure 21 Schematic representation of the hybrid process combining electrochemical removal and mechanical grinding. Reproduced from Zhu, D.;
Zeng, Y. B.; Xu, Z. Y.; Zhang, X. Y. Precision Machining of Small Holes by the Hybrid Process of Electrochemical Removal and Grinding. CIRP Ann. –
Manuf. Technol. 2011, 60, 247–250.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
102 Compound and Hybrid Micromachining Processes

there is no material removal in phase 3. If the tool is not insulated, the electrochemical dissolution will continue for the machined
hole surface (in phase 3) because of the electrical field between the machined hole surface and the tool, resulting in a taper hole.
This hybrid process is able to machine microholes with sharp edges and without any burrs in difficult-to-cut materials. In
addition, the aspect ratio of the microholes can also be increased due to the combined action of electrochemical dissolution and
mechanical material removal. The proposed processing is also employed to remove the recast layer of an engine component in
Ni-based superalloys, which has an unclosed internal cylindrical surface with a high aspect ratio of 16.25 premachined by wire-cut
electrodischarge machining (24). After being machined by the proposed hybrid process, the cylindrical hole was enlarged, and the
machined surface finish was improved. Therefore, the recast layer produced by WEDM could be totally removed with this hybrid
process. Figure 22 shows the entrance side and cross-section of a microhole machined using the hybrid process (24). The improved
surface finish at the inside surface of the microhole can be realized from the image.

11.05.5.2 Micro-EDM and Micro-USM Combined Process

This hybrid machining process combines micro-EDM and micro-ultrasonic vibration machining (micro-USM) (25). In this process,
the holes and cavities are machined in hard or difficult-to-cut materials using high-frequency mechanical motion abrasive slurry in
association with EDM. The abrasive used in this process should be harder than the material being machined. USM can be used alone
for any hard materials; however, for the combined process, the workpiece must be conductive. The material removal is from the
combined action of the electrical discharging and mechanical polishing of the abrasive slurry. As a result, the MRR of the hybrid
process is higher than that of the single USM or micro-EDM process. The high-frequency pumping action of the vibrating surface of
the electrode accelerates the slurry circulation, giving smaller machining times. The pressure variations in the gap lead to more
efficient discharges, which remove more melted metal. The affected layer is reduced, thermal residual stresses are modified, fewer
microcracks are observed, and fatigue resistance is increased due to the abrasive action of slurries. The MRR and surface finish of the

Figure 22 SEM images of the hole machined by a combined electrochemical and mechanical grinding hybrid process: (a) the entrance side of the
hole; and (b) a cross-sectional view of the hole showing an improved surface finish around the edge. Reproduced from Zhu, D.; Zeng, Y. B.; Xu, Z. Y.;
Zhang, X. Y. Precision Machining of Small Holes by the Hybrid Process of Electrochemical Removal and Grinding. CIRP Ann. – Manuf. Technol. 2011, 60,
247–250.

Figure 23 (a) Schematic diagram showing the setup for the micro-EDM–USM hybrid process; and (b) magnified view showing the working principle
of the hybrid process. Reproduced from Lin, Y. C.; Yan, B. H.; Chang, Y. S. Machining Characteristics of Titanium Alloy (Ti-6Al-4V) Using a Combination
Process of EDM with USM. J. Mater. Process. Technol. 2000, 104 (3), 171–177.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining Processes 103

Figure 24 Microhole fabricated by the micro-EDM–USM hybrid process: (a) top view of the fabricated microhole; (b) a cross-section of the hole shown
in (a); (c) the inner surface of the hole machined by micro-EDM only; and (d) the inner surface of the hole machined by the micro-EDM and USM
hybrid process. Reproduced from Lin, Y. C.; Yan, B. H.; Chang, Y. S. Machining Characteristics of Titanium Alloy (Ti-6Al-4V) Using a Combination
Process of EDM with USM. J. Mater. Process. Technol. 2000, 104 (3), 171–177.

process depend on the size of the abrasive particles used in micro-USM (25). Figure 23 presents the working principle of the micro-
EDM and USM hybrid process (25). Comparison of the microhole inner surfaces produced by micro-EDM and the micro-
EDM–USM hybrid process is presented in Figure 24 (25).

11.05.5.3 Vibration-Assisted Micro-EDM

The application of micro-EDM in deep-hole drilling or in the fabrication of high-aspect-ratio microfeatures is still limited due to
improper flushing out of the debris and unstable machining (26). The process is a combination of micro-EDM and vibration to the
workpiece or electrode at the same time. The process improves the flushing conditions, the removal of debris, and the machining
stability, thus reducing the machining time significantly. Unlike the micro-EDM–USM combined process, the material removal in
this process is by the micro-EDM process only. The workpiece or electrode vibration just assists in enhancing the material removal
from the workpiece. The process is suitable for deep-hole drilling in hard and difficult-to-cut materials. Depending on the exper-
imental design and objective, the vibration can be applied to the tool electrode (27) or workpiece (26,28). The vibration can be
a low-frequency vibration or ultrasonic vibration. Tool vibration is comparatively more difficult to apply in micro-EDM, as the tool
electrode is only several microns in diameter; hence, there is a chance for tool deflection. Therefore, in recent years, research has been

Figure 25 (a) Mechanism of applying vibration to the tool electrode (reproduced from Endo, T.; Tsujimoto, T.; Mitsui, K. Study of Vibration-assisted
Micro-EDM – The Effect of Vibration on Machining Time and Stability of Discharge. Precis. Eng. 2008, 32 (4), 269–277) and (b) to the workpiece
(reproduced from Tong, H.; Li, Y.; Wang, Y. Experimental Research on Vibration Assisted EDM of Micro-structures with Non-circular Cross-
section.J. Mater. Process. Technol. 2008, 208 (1–3), 289–298).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
104 Compound and Hybrid Micromachining Processes

Figure 26 Application of electrode vibration-assisted micro-EDM: (a) fabrication of a square shaft without vibration; (b) fabrication of a microshaft
with vibration; and (c) comparison of the machining times for fabricating a square shaft without and with vibration. Reproduced from Endo, T.;
Tsujimoto, T.; Mitsui, K. Study of Vibration-assisted Micro-EDM – The Effect of Vibration on Machining Time and Stability of Discharge. Precis. Eng.
2008, 32 (4), 269–277.

Figure 27 Application of workpiece vibration-assisted micro-EDM: (a) a microhole (diameter 60 mm, depth 0.5 mm) without vibration; (b) a microhole
(diameter 60 mm, depth 1.0 mm) with vibration; and (c) a comparison of machining time without and with vibration. Reproduced from Jahan, M. P.;
Saleh, T.; Rahman, M.; Wong, Y. S. Development, Modeling, and Experimental Investigation of Low Frequency Workpiece Vibration-assisted Micro-EDM
of Tungsten Carbide. J. Manuf. Sci. Eng. 2010, 132 (5), 054503 (8 pp).

carried out on the feasibility of workpiece vibration-assisted EDM for the fabrication of microparts (28) and high-aspect-ratio
microholes (26). Figure 25 shows the schematic representation of the developed devices for generating tool vibration (26) and
workpiece vibration (28) during machining. The application of vibration-assisted micro-EDM in the fabrication of small and high-
aspect-ratio microstructures and microholes is presented in Figures 26 (28) and 27 (26), respectively. In addition, Figures 26 and 27
indicate a significant amount of reduction in machining time in vibration-assisted micro-EDM compared to that of the micro-EDM
process without vibration.

11.05.5.4 Powder-Mixed Micro-EDM

In recent years, to improve the quality of the micro-EDM machined surface and to reduce surface defects, several investigators have
found the addition of powder particles in the dielectric as an effective process (29). In this hybrid process, the electrically conductive
or semiconductive powder is mixed in the dielectric, which reduces the insulating strength of the dielectric fluid and increases the
spark gap between the tool and workpiece. The enlarged spark gap makes the flushing of debris easier. As a result, the process
becomes stable, improving the MRR and surface finish (30). The sparking is uniformly distributed among the powder particles in
the spark gap, reducing the intensity of a single spark, which results in uniform shallow craters instead of a single broader crater.

Figure 28 Comparison of the material removal mechanism for micro-EDM (left) without powder and (right) with powder. For micro-EDM without
powder: a lower spark gap, higher gas explosive pressure, and higher single crater size; and for powder-mixed micro-EDM: a larger spark gap, lower gas
explosive pressure, and lower single crater size. Reproduced from Tzeng, Y.-F.; Chen, F.-C. Investigation into Some Surface Characteristics of Electrical
Discharge Machined SKD-11 Using Powder-suspension Dielectric Oil. J. Mater. Process. Technol. 2005, 170, 385–391.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining Processes 105

Figure 29 Comparison of surface topography and crater height for micro-EDM (a) without and (b) with the addition of powder in the dielectric.
The improvement of surface finish and reduction of crater height are visible in powder-mixed micro-EDM (right-side images). Reproduced from Jahan,
M. P.; Anwar, M. M.; Wong, Y. S.; Rahman, M. Nanofinishing of Hard Materials Using Micro-EDM. Proc. Inst. Mech. Eng., Part B 2009, 223, 1127–1142.

Thus, the surface finish improves. There may be some abrasive actions of the powder particles during the finishing that reduce the
crater boundary heights, making the surface shinier. Figure 28 shows the differences in working principle between the conventional
micro-EDM and powder-mixed micro-EDM processes (31). It can be seen from Figure 28(a) that in powder-mixed micro-EDM,
instead of a single spark with more energy, the sparking is distributed among the powder particles, thus reducing the strength of
a single spark and making uniform discharge and a surface with uniform craters. Figure 28(b) explains that the addition of powders
leads to an increase in gap size that subsequently results in a reduction in electrical discharge power density and in gas explosive
pressure for a single power pulse (Figure 28(b)). The improvement of surface finish and reduction of crater heights in powder-
mixed micro-EDM are presented in Figure 29 (30).

11.05.5.5 Microelectrochemical Discharge Machining (Micro-ECDM)

The micro-ECDM process involves a complex combination of the electrochemical (EC) reaction and electrodischarge (ED) action.
The combined process provides lower electrode wear and a higher MRR compared to the single micro-EDM process (32). Although

Figure 30 (a) Working principle of the micro-ECDM process (reproduced from Bhattacharyya, B.; Doloi, B. N.; Sorkhel, S. K. Experimental Investi-
gations into Electrochemical Discharge Machining (ECDM) of Non-conductive Ceramic Materials. J. Mater. Process. Technol. 1999, 95, 145–154);
(b) machining of nonconductive Pyrex glass using the micro-ECDM process (Ra: 1.8 mm); and (c) machining of the same Pyrex glass using the micro-
ECDM process with SiC powder (Ra: 1.0 mm). (reproduced from Yang, C. T.; Song, S. L.; Yan, B. H.; Huang, F. Y. Improving Machining
Performance of Wire Electrochemical Discharge Machining by Adding SiC Abrasive to Electrolyte. Int. J. Mach. Tools Manuf. 2006, 46, 2044–2050).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
106 Compound and Hybrid Micromachining Processes

micro-EDM only machines electrically conductive materials, this hybrid process can be applied to conductive metals as well as
nonconductive ceramic (32).
In this hybrid process, the electrochemical action helps in the generation of the positively charged ionic gas bubbles (e.g.,
hydrogen). The electrical discharge action takes place between the tool and the workpiece due to the breakdown of the insulating
layer of the gas bubbles. The DC power supply voltage is applied between the tool (or cathode) and the anode, resulting in material
removal due to melting, vaporization of the workpiece material, and mechanical erosion (33). Figure 30(a) shows the schematic
representation of the material removal mechanism for the ECDM process (34). The electrolyte cell used in the ECDM process is
similar to that used in electrochemical machining (ECM). In ECDM, the anode is made up of inert material, while the cathode
normally is made of copper. Dilute hydrochloric acid (HCl) is generally used as the electrolyte. When a voltage is applied to the cell
in proper polarity (i.e., a positive terminal to the anode and a negative terminal to the cathode), reduction of electrolyte with
liberation of hydrogen gas takes place at the cathode tip. When the applied voltage is increased beyond a threshold value, hydrogen
gas bubbles evolve in large numbers at the tip of the cathode and grow in size. Their nucleation site density increases, and the current
path gets restricted between the cathode and electrolyte interface, causing discharge to occur at this interface instantly. Thus,
discharge in ECDM always occurs when the voltage in an electrolytic cell is increased beyond a threshold value (35).
Application of ECDM gives the possibility to get a machined surface with good quality, when there is high efficiency and a lack
of electrode wear. However, the accuracy can be lower than with the single micro-EDM process (32). An expanded version of
micro-ECDM with conductive powder-mixed electrolyte has been found to produce improved surface finish and integrity
compared to the single micro-ECDM process. Figure 30(b) and (c) shows a comparison of the surface finish between ECDM and
abrasive-mixed ECDM (36).

11.05.5.6 Magnetic-Assisted Micro-EDM

This hybrid process involves a complex combination of micro-EDM and magnetic field assistance in order to improve machining
performance by enhancing removal of debris and increasing the MRR. One of the problems associated with micro-EDM of high-
aspect-ratio and/or blind features is that the flushing of debris from the machined zone becomes difficult. Those debris particles
result in unstable machining by generating arcing and short-circuiting, and they reduce the MRR and surface quality. In order to
solve those issues, a magnetic field has been introduced in the micro-EDM process to improve debris circulation (37,38). Imple-
menting magnetic force perpendicular to the electrode’s rotational force produces a resultant force that is efficient in transporting
debris out of the hole during machining. A debris particle in a magnetic field-assisted micro-EDM is subjected to two kinds of forces:
the magnetic force and the centrifugal force. The resultant force on the debris particle is given by the vector addition of the magnetic
force and centrifugal force, which helps to flush out the debris particles from the machine zone, thus improving the machining
stability and MRR, reducing tool wear, and overall improving micro-EDM performance (37). Magnetic field-assisted micro-EDM can
produce higher aspect ratio holes compared with the conventional micro-EDM process under similar working conditions (38). The
application of magnetic fields helps in gap cleaning in micro-EDM due to increased debris transport out of the gap. The enhanced
debris removal due to the application of magnetic fields leads to an increase in MRR. It has been reported that for a magnetic
material, the MRR was nearly three times higher than that of a hole cut without the magnetic field (37). Figure 31(a) shows the
schematic representation of magnetic field-assisted micro-EDM (37). The increment in the aspect ratio of a microhole in magnetic
field-assisted micro-EDM can be understood from Figure 31(b) and (c) (38).

11.05.5.7 Laser-Assisted Microturning

Laser-assisted mechanical microturning offers the ability to machine difficult-to-cut materials like superalloys and ceramics more
efficiently and economically by providing the local heating of the workpiece prior to material removal by a cutting tool. Laser-
assisted machining has an edge over conventional machining methods due to various advantages, which include a lower cutting

Figure 31 (a) Working principle of magnetic field-assisted micro-EDM (reproduced from Heinz, K.; Kapoor, S. G.; DeVor, R. E.; Surla, V. An Investigation
of Magnetic-field-assisted Material Removal in Micro-EDM for Nonmagnetic Materials. J. Manuf. Sci. Eng. 2011, 133, 021002 (9 pp)); (b) the cross-section
of a microhole machined by conventional micro-EDM; and (c) the cross-section of a microhole machined by magnetic field-assisted micro-EDM using the
same machining conditions as in (b). (reproduced from Yeo, S. H.; Murali, M.; Cheah, H. T. Magnetic Field Assisted Micro Electro-discharge Machining. J.
Micromech. Microeng. 2004, 14, 1526–1529).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining Processes 107

Figure 32 (a) Schematic representation and (b) photograph of the laser-assisted microturning system. Reproduced from Shin, Yung C. Laser Assisted
Machining. In Industrial Laser Solutions for Manufacturing; 01/01/2011 (last accessed on 21 February 2013), www.industrial-lasers.com/articles/print/
volume-26/issue-1/features/laser-assisted-machining.html.

Figure 33 Comparison of (a) cutting force and (b) residual stress generated during the machining of Inconel 625 with conventional microturning
and laser-assisted microturning. Reproduced from Samanta, A.; Teli, M.; Singh, R. K. Surface Integrity in Laser Assisted Mechanical Micro-machining of
(LAMM) of Inconel 625. In Proceedings of the 7th International Conference on Micromanufacturing (ICOMM 2012) [CD ROM proceedings].

Figure 34 Surface profile for (a) conventional and (b) laser-assisted microturning (25 mm uncut chip thickness and 30 mm min1 cutting speed).
Reproduced from Samanta, A.; Teli, M.; Singh, R. K. Surface Integrity in Laser Assisted Mechanical Micro-machining of (LAMM) of Inconel 625. In
Proceedings of the 7th International Conference on Micromanufacturing (ICOMM 2012) [CD ROM proceedings].

force at higher cutting speed, low specific cutting energy, a smooth surface finish, less tool wear, and increased compressive surface
residual stress (39).
Figure 32 shows a schematic representation and photograph of the laser-assisted microturning hybrid process (39). In laser-
assisted microturning processes, the workpiece is heated locally by a laser past its thermal softening point, and then ductile mode
machining is carried out on the thermally softened surface (40). The laser beam can pass through the diamond tool, thus heating the
surface just below the tool tip in the chip formation zone (41). The laser beam can be applied from a separate source at an angle, but

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
108 Compound and Hybrid Micromachining Processes

Figure 35 The hybrid laser-assisted micromilling setup: 1 – the rotary stage for orienting the laser; 2 – the stacked linear stages – X, Y, and Z;
3 – spindle assembly; 4 – a fiber-optic cable; and 5 – collimator and micrometer assembly. Reproduced from Melkote, S.; Kumar, M.; Hashimoto, F.;
Lahoti, G. Laser Assisted Micro-milling of Hard-to-machine Materials. CIRP Ann. – Manuf. Technol. 2009, 58, 45–48.

care should be taken to ensure that the laser beam does not interact with the cutting tool tip. The heating effect produced is at the
microscopic scale, and hence the laser power required to heat the workpiece is less than in macro-laser-assisted machining (LAM)
processes. During LAM processing, the workpiece is deformed below the fracture strength, thereby enabling a viscoplastic flow rather
than a brittle fracture (42). The reduction of cutting forces and residual stresses in laser-assisted microturning compared to
conventional turning can be understood from Figure 33 (43). However, the machined surface obtained by laser-assisted micro-
turning has slightly higher roughness than that of conventional microturning, as can be seen from Figure 34 (43).

11.05.5.8 Laser-Assisted Micromilling

The part feature accuracy rate and MRR in micromilling of difficult-to-machine materials are limited by the machine-tool system
stiffness (especially for small-footprint machines), and the low flexural stiffness and strength of the microtools normally used.
Rapid tool wear is another issue during the machining of hard and difficult-to-cut materials, since it negatively impacts part feature
accuracy and finish (44). Those shortcomings can be overcome by introducing laser assistance during the machining process. The
laser heating will induce localized thermal softening of the materials to be machined, which helps to reduce the cutting forces and
tool wear. Tool deflection due to the extreme hardness of the materials can be reduced, which will enhance the dimensional
accuracy. The hybrid laser-assisted micromilling process is able to machine freeform 3D microscale features in hard materials (45).
Figure 35 shows the setup for the laser-assisted micromilling process (45).
In this hybrid process, the laser nozzle is set at an angle, so that the laser radiation hits the surface to be machined just before it is
machined by the micromilling process. The objective is the thermal softening of the materials by laser irradiation before machining
with the milling tool. A relatively low-power, Ytterbium-doped, continuous-wave, near-infrared (1.06 mm) fiber laser is used to
achieve highly localized thermal softening of the material immediately in front of the cutting tool during the micromilling process.

Figure 36 Comparison of the surface quality of the microgrooves produced by the (a) micromilling process and (b) laser-assisted micromilling
process. The discontinuity along the grooves can be noticed when using the micromilling process alone. Reproduced from Melkote, S.; Kumar, M.;
Hashimoto, F.; Lahoti, G. Laser Assisted Micro-milling of Hard-to-machine Materials. CIRP Ann. – Manuf. Technol. 2009, 58, 45–48.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining Processes 109

Figure 37 Comparison of groove depths for the micromilling process without and with the assistance of laser. Higher depth of grooves results
from the laser-assisted micromilling process. Reproduced from Melkote, S.; Kumar, M.; Hashimoto, F.; Lahoti, G. Laser Assisted Micro-milling of Hard-
to-machine Materials. CIRP Ann. – Manuf. Technol. 2009, 58, 45–48.

No assist gas is used. The laser spot size can be adjusted depending on the dimensions of the machined feature. The laser, spindle,
and four axes are controlled simultaneously via a common computer interface. The dry micromilling is performed by TiAlN-coated
WC four-flute ball end mills of 250 mm diameter. The laser-assisted micromilling process can increase the MRR by increasing the
depth of groove or depth of cut during the micromilling process. In addition, it can provide a comparatively better surface finish
with less discontinuity and burr around the edge of the slots, as can be seen from Figure 36 (45). This is due to the fact that the
workpiece materials become softer after laser irradiation before the final machining by the milling cutter. Moreover, the MRR and
depth of groove also increase in the hybrid process due to the softening action of laser processing (Figure 37).

11.05.5.9 Laser-Assisted Microgrinding

The grinding of difficult-to-cut materials (e.g., structural ceramics) is a major challenge due to their extreme hardness, toughness,
and wear resistance. The material removal mechanism in the grinding of hard ceramics involves brittle fracture, which often results
in the formation of surface microcracks because of the low fracture toughness of ceramics (46). In general, high surface cutting
speeds and low depths of cut are recommended to minimize subsurface damage in ceramics, which leads to an increase in
machining time (46). This limitation can be addressed by adopting a hybrid process, including a two-step strategy to reduce the
grinding forces by first locally weakening the ceramic material through the introduction of thermal cracks and subsequently
removing it mechanically. The thermal cracks are induced in ceramic by laser irradiation. By suitably adjusting the laser power, spot
size, and speed, it is possible to create and confine the cracks to a controlled volume of material and then remove those materials by
a mechanical microgrinding process (47).

Figure 38 Schematic diagram showing the mechanism of the laser-assisted microgrinding process. Reproduced from Kumar, M.; Melkote, S.;
Lahoti, G. Laser-assisted Microgrinding of Ceramics. CIRP Ann. – Manuf. Technol. 2011, 60, 367–370.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
110 Compound and Hybrid Micromachining Processes

Figure 39 Reduction of average and maximum grinding forces in laser-assisted microgrinding compared to the microgrinding process alone.
Reproduced from Kumar, M.; Melkote, S.; Lahoti, G. Laser-assisted Microgrinding of Ceramics. CIRP Ann. – Manuf. Technol. 2011, 60, 367–370.

In this hybrid process of laser-assisted microgrinding, the to-be-machined surface is scanned through the laser radiation in order
to induce localized thermal stresses in the materials. The purpose of the laser irradiation is to induce and confine thermal cracks in the
regions of interest. The thermal cracks are produced in the irradiated region due to tensile stresses generated by rapid heating and
cooling of the ceramic. After that, a microgrinding tool is used to mechanically remove the laser-affected (weakened) region at higher
MRRs than possible in conventional microgrinding. This step is followed by a series of finish grinding steps to improve the surface
finish. This two-step strategy is particularly advantageous since coolants can be used without occlusion of the laser beam. Figure 38
shows a schematic representation of the material removal mechanism in the laser-assisted microgrinding process (47). The reduction
in the grinding forces in laser-assisted microgrinding compared to mechanical microgrinding can be understood from Figure 39 (47).

11.05.5.10 Laser Microdrilling with Jet Electrochemical Machining

A novel hybrid process incorporating laser drilling with jet electrochemical machining (JECM-LD) has been developed in order to
overcome the drawbacks accompanied intrinsically by conventional laser drilling, such as the formation of an HAZ and recast layer
(48). Integrating the merits of both the laser drilling and jet electrochemical machining, the hybrid method is implemented by
directing an electrolyte jet coaxially aligned with a focused laser beam onto the workpiece surface. The laser beam needs to transmit
in the jet electrolyte before being focused on the machining area. Electrolyte is a neutral salt solution, which is attenuable to laser
energy by absorption and scattering. The property of laser attenuation in electrolyte is the key factor of JECM-LD.

Figure 40 Schematic diagram showing the principles of hybrid laser drilling with jet electrochemical machining. Reproduced from Zhang, H. Laser
Drilling Assisted with Jet Electrochemical Machining. In Nd YAG Laser; Dumitras, Dan C., Ed.; In Tech, 2012; pp 299–318; Chapter 15.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining Processes 111

Figure 41 Micrograph of the peripheral surface of a penetrated hole (a) laser-drilled in air (200 mJ, 10 s); and (b) drilled with the JECM-LD
hybrid process (200 mJ, 40 V, 20 s). Reproduced from Zhang, H. Laser Drilling Assisted with Jet Electrochemical Machining. In Nd YAG Laser; Dumitras,
Dan C., Ed.; In Tech, 2012; pp 299–318; Chapter 15.

JECM-LD combines two different sources of energy simultaneously: energy of photons (laser drilling) and energy of ions (ECM).
The main aim of combining a jet electrolyte with a laser beam is to obtain high process quality by reducing the recast layer and
spatter produced in laser drilling. The jet electrolyte is aligned coaxially with a focused laser beam and creates a noncontact tool-
electrode. The focused laser beam and the jet electrolyte are acting on the same surface of workpiece synchronously. In the course of
JECM-LD, mainly laser drilling removes material. The defects are overcome by the effects of the jet electrolyte, which consist of
effective cooling to the workpiece, transporting of debris, and electrochemical reaction with materials in the interpulse of the laser.
Figure 40 illustrates the principles of JECM-LD (49).
The JECM-LD hybrid process provides high machining quality with reduced recast layers and spatters. The laser beam takes the
chief responsibility of removing the scrap material during JECM-LD, while the jet electrochemical machining serves the auxiliary
function of eliminating the recast layers and spatters. As can be seen in Figure 41(a), a large irregular area comprising spattering
deposits and resolidified molten layers encircles the surfaces at the edge of the microhole that was machined by laser drilling. In
contrast, the hybrid process provides microholes with a comparatively lower HAZ and fewer spatters, as can be seen from
Figure 41(b). This may be due to the high-speed electrolyte jet, which effectively cools the material to be processed and discharges
scraps. Figure 41(b) exhibits better surface quality and a smoother hole periphery for the hybrid process compared to a single laser-
machining process (49). However, an obvious annular electrochemical overcut can be observed at the entrance surface for the
JECM-LD process.

11.05.6 Summary

Today’s manufacturing industry is facing challenges from advanced difficult-to-machine materials (tough superalloys, ceramics,
composites, etc.), stringent design requirements (high precision, complex shapes, high surface quality, etc.) for microparts and
components, and machining costs associated with the difficult-to-cut materials. As a single conventional or nonconventional process
is unable to solve all of the issues faced during the machining of these difficult-to-cut materials due to their improved thermal,
chemical, and mechanical properties, the necessity of developing innovative compound and hybrid processes is becoming more
important. Hybrid micromachining has the potential to combine the strengths and complement the weaknesses of different processes.
In this chapter, a comprehensive overview on recently developed compound and hybrid micromachining processes has been
provided. In addition to a description of the process mechanisms, applications of different compound and hybrid machining
processes have been discussed. Most of the compound and hybrid machining processes include one conventional machining and
one nonconventional machining process. The compound machining processes are mainly a combination of sequential processes,
whereas the hybrid machining offers two or more simultaneous actions that are responsible for the material removal. Among the
nonconventional machining processes, micro-EDM and laser machining are used more widely with different conventional
machining processes (e.g., turning, milling, and grinding) to develop compound and/or hybrid processes. These compound and
hybrid micromachining processes have enormous potential for the fabrication of microstructures. They can be used for the
fabrication of components for microelectromechanical systems, micromolds, microfluidic channels, microprobes, and patterns in
glass substrates for lab-on-chip devices or biomedical arrays. Finally, due to the effectiveness of hybrid processes at the macro- and
microscales, further research is ongoing to develop innovative hybrid machining processes at the nanoscale.

References

1. Rahman, M.; Asad, A. B. M. A.; Masaki, T.; Saleh, T.; Wong, Y. S.; Kumar, A. S. A Multiprocess Machine Tool for Compound Micromachining. Int. J. Mach. Tools Manuf. 2010,
50 (4), 344–356.
2. Lauwers, B. Surface Integrity in Hybrid Machining Processes, keynote paper. In 1st CIRP Conference on Surface Integrity, Bremen 30/01–01/02; 2012.
3. Rahman, M.; Asad, A. B. M. A.; Masaki, T.; Wong, Y. S.; Kumar, A. S. Integrated Hybrid Micro/Nano-machining. In ASME 2007 International Manufacturing Science and
Engineering Conference (MSEC2007); October 15–18, 2007. Atlanta, Georgia, USA; Paper No. MSEC2007-31009; pp 197–209.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
112 Compound and Hybrid Micromachining Processes

4. Morgan, C. J.; Vallance, R. R.; Marsh, E. R. Micro Machining Glass with Polycrystalline Diamond Tools 95 Shaped by Micro Electro Discharge Machining. J. Micromech.
Microeng. 2004, 14, 1687–1692.
5. Morgan, C. J.; Vallance, R. R.; Marsh, E. R. Micro-machining and Micro-grinding with Tools Fabricated by Micro Electro-discharge Machining. Int. J. Nanomanuf. 2006, 1 (2),
242–258.
6. Fleischer, J.; Masuzawa, T.; Schmidt, J.; Knoll, M. New Applications for Micro-EDM. J. Mater. Process. Technol. 2004, 149 (249), 246.
7. Lim, H. S.; Kumar, A. S.; Rahman, M. Improvement of Form Accuracy in Hybrid Machining of Microstructure. J. Electron. Mater. 2002, 31 (10), 1032–1038.
8. Asad, A. B. M. A.; Masaki, T.; Rahman, M.; Lim, H. S.; Wong, Y. S. Tool-based Micro-machining. J. Mater. Process. Technol. 2007, 192–193, 204–211.
9. Wada, T.; Masaki, T.; Davis, D. W. Development of Micro Grinding Process using Micro EDM Trued Diamond Tools. In Proceedings of the Annual Meeting of ASPE; 2002.
10. Campana, S.; Miyazawa, S. Micro-EDM and ECM in DI Water. In Proceedings of Annual Meeting of American Society of Precision Engineering (ASPE); 1999.
11. Nguyen, M. D.; Rahman, M.; Wong, Y. S. Simultaneous Micro-EDM and Micro-ECM in Low-resistivity Deionized Water. Int. J. Mach. Tools Manuf. 2012, 54–55, 55–65.
12. Zeng, Z.; Wang, Y.; Wang, Z.; Shan, D.; He, X. A Study of Micro-EDM and Micro-ECM Combined Milling for 3D Metallic Micro-structures. Precis. Eng. 2012, 36 (3), 500–509.
13. Takahata, K.; Gianchandani, Y. B. Batch Mode Micro-electro-discharge Machining. IEEE/ASME J. Microelectromech. Syst. 2002, 11, 102–110.
14. Pham, D.-T.; Dimov, S.-S.; Bigot, S.; Ivanov, A.; Popov, K. Micro-EDM – Recent Developments and Research Issues. J. Mater. Process. Technol. 2004, 149, 50–57.
15. Dausinger, F. Precise Drilling with Short Pulsed Lasers, in High-power Lasers in Manufacturing. Proc. SPIE 2000, 3888, 180–187.
16. Lin, L.; Diver, C.; Atkinson, J.; Giedl-Wagner, R.; Helm, H. J. Sequential Laser and EDM Micro-drilling for Next Generation Fuel Injection Nozzle Manufacture. Ann. CIRP 2006,
55 (1), 179–182.
17. Huang, J.-D.; Kuo, C.-L. Pin-plate Micro Assembly by Integrating Micro-EDM and Nd-YAG Laser. Int. J. Mach. Tools Manuf. 2002, 42, 1455–1464.
18. Kuo, C.-L.; Huang, J.-D.; Liang, H.-Y. Precise Micro-assembly through an Integration of Micro-EDM and Nd-YAG. Int. J. Adv. Manuf. Technol. 2002, 20, 454–458.
19. Kuo, C.-L.; Huang, J.-D.; Liang, H.-Y. Fabrication of 3D Metal Microstructures Using a Hybrid Process of Micro-EDM and Laser Assembly. Int. J. Adv. Manuf. Technol. 2003,
21, 796–800.
20. Li, L.; Low, D. K. Y.; Ghoreshi, M.; Crookall, J. R. Hole Taper Characterisation and Control in Laser Percussion Drilling. CIRP Ann. – Manuf. Technol. 2002, 51 (1), 153–156.
21. Chae, J.; Park, S. S.; Freiheit, T. Investigation of Micro-cutting Operations. Int. J. Mach. Tools Manuf. 2006, 46 (3–4), 313–332.
22. Okasha, M. M.; Mativenga, P. T.; Driver, N.; Li, L. Sequential Laser and Mechanical Micro-drilling of Ni Superalloy for Aerospace Application. CIRP Ann. – Manuf. Technol.
2010, 59, 199–202.
23. Weber, P.; Haupt, S.; Schulze, V. Hybrid Machining of Microstructures Using a Combination of Electrical Discharge Machining Milling and Laser Ablation. J. Vac. Sci. Technol. B
2009, 27 (3), 1327–1329.
24. Zhu, D.; Zeng, Y. B.; Xu, Z. Y.; Zhang, X. Y. Precision Machining of Small Holes by the Hybrid Process of Electrochemical Removal and Grinding. CIRP Ann. – Manuf. Technol.
2011, 60, 247–250.
25. Lin, Y. C.; Yan, B. H.; Chang, Y. S. Machining Characteristics of Titanium Alloy (Ti-6Al-4V) Using a Combination Process of EDM with USM. J. Mater. Process. Technol. 2000,
104 (3), 171–177.
26. Jahan, M. P.; Saleh, T.; Rahman, M.; Wong, Y. S. Development, Modeling, and Experimental Investigation of Low Frequency Workpiece Vibration-assisted Micro-EDM of
Tungsten Carbide. J. Manuf. Sci. Eng. 2010, 132 (5), 054503 (8 pp).
27. Endo, T.; Tsujimoto, T.; Mitsui, K. Study of Vibration-assisted Micro-EDM – The Effect of Vibration on Machining Time and Stability of Discharge. Precis. Eng. 2008, 32 (4),
269–277.
28. Tong, H.; Li, Y.; Wang, Y. Experimental Research on Vibration Assisted EDM of Micro-structures with Non-circular Cross-section. J. Mater. Process. Technol. 2008, 208 (1–3),
289–298.
29. Kansal, H. K.; Singh, S.; Kumar, P. Technology and Research Developments in Powder Mixed Electric Discharge Machining (PMEDM). J. Mater. Process. Technol. 2007, 184,
32–41.
30. Jahan, M. P.; Anwar, M. M.; Wong, Y. S.; Rahman, M. Nanofinishing of Hard Materials Using Micro-EDM. Proc. Inst. Mech. Eng., Part B 2009, 223, 1127–1142.
31. Tzeng, Y.-F.; Chen, F.-C. Investigation into Some Surface Characteristics of Electrical Discharge Machined SKD-11 Using Powder-suspension Dielectric Oil. J. Mater. Process.
Technol. 2005, 170, 385–391.
32. Skrabalak, G.; Skrabalak, M. S.; Ruszaj, A. Building of Rules Base for Fuzzy-logic Control of the ECDM Process. J. Mater. Process. Technol. 2004, 149, 530–535.
33. Sorkhel, S. K.; Bhattacharyya, B.; Mitra, S.; Doloi, B. Development of Electrochemical Discharge Machining Technology for Machining of Advanced Ceramics. In International
Conference on Agile Manufacturing; February 1996; pp 98–103.
34. Bhattacharyya, B.; Doloi, B. N.; Sorkhel, S. K. Experimental Investigations into Electrochemical Discharge Machining (ECDM) of Non-conductive Ceramic Materials. J. Mater.
Process. Technol. 1999, 95, 145–154.
35. Kulkarni, A.; Sharan, R.; Lal, G. K. An Experimental Study of Discharge Mechanism in Electrochemical Discharge Machining. Int. J. Mach. Tools Manuf. 2002, 42, 1121–1127.
36. Yang, C. T.; Song, S. L.; Yan, B. H.; Huang, F. Y. Improving Machining Performance of Wire Electrochemical Discharge Machining by Adding SiC Abrasive to Electrolyte. Int. J.
Mach. Tools Manuf. 2006, 46, 2044–2050.
37. Heinz, K.; Kapoor, S. G.; DeVor, R. E.; Surla, V. An Investigation of Magnetic-field-assisted Material Removal in Micro-EDM for Nonmagnetic Materials. J. Manuf. Sci. Eng.
2011, 133, 021002 (9 pp).
38. Yeo, S. H.; Murali, M.; Cheah, H. T. Magnetic Field Assisted Micro Electro-discharge Machining. J. Micromech. Microeng. 2004, 14, 1526–1529.
39. Shin, Yung C. Laser Assisted Machining. In Industrial Laser Solutions for Manufacturing; 01/01/2011 (last accessed on 21 February 2013), www.industrial-lasers.com/articles/
print/volume-26/issue-1/features/laser-assisted-machining.html.
40. Virkar, Saurabh, R.; Patten, John A. Micro-laser Assisted Machining: Numerical Simulations and Analysis of the Thermal Effects on Silicon Carbide. In Proceedings of ASME
2009 International Manufacturing Science and Engineering Conference (MSEC2009) October 4–7, 2009, Vol. 2; pp 409–416. West Lafayette, Indiana, USA.
41. Patten, J. A.; Jacob, J.; Bhattacharya, B.; Grevstad, A. Comparison between Numerical Simulation and Experiments for Single Point Diamond Turning of Silicon Carbide. In
NAMRC Conference; Society of Manufacturing Engineers, 2007; p 2.
42. Shin, Y. C.; Pfefferkorn, F. E.; Rozzi, J. C. Experimental Evaluation of the Laser Assisted Machining of Silicon Nitride Ceramics. ASME J. Manuf. Sci. 2000, 122, 666.
43. Samanta, A.; Teli, M.; Singh, R. K. Surface Integrity in Laser Assisted Mechanical Micro-machining of (LAMM) of Inconel 625. In Proceedings of the 7th International Conference
on Micromanufacturing (ICOMM 2012) [CD ROM proceedings]: Evanston Il.
44. Bissacco, G.; Hansen, H. N.; De Chiffre, L. Micromilling of Hardened Tool Steel for Mould Making Applications. J. Mater. Process. Technol. 2005, 167, 201–207.
45. Melkote, S.; Kumar, M.; Hashimoto, F.; Lahoti, G. Laser Assisted Micro-milling of Hard-to-machine Materials. CIRP Ann. – Manuf. Technol. 2009, 58, 45–48.
46. Malkin, S.; Ritter, J. E. Grinding Mechanism and Strength Degradation for Ceramics. ASME J. Eng. Ind. 1989, 111, 167–173.
47. Kumar, M.; Melkote, S.; Lahoti, G. Laser-assisted Microgrinding of Ceramics. CIRP Ann. – Manuf. Technol. 2011, 60, 367–370.
48. Zhang, H.; Xu, J. Modeling and Experimental Investigation of Laser Drilling with Jet Electrochemical Machining. Chin. J. Aeronaut. 2010, 23, 454–460.
49. Zhang, H. Laser Drilling Assisted with Jet Electrochemical Machining. In Nd YAG Laser; Dumitras, Dan C., Ed.; InTech, 2012; pp 299–318; Chapter 15.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 11.06 Compound and Hybrid Micromachining
Part II – Hybrid Micro-EDM and Micro-ECM
M Rahman, YS Wong, and MD Nguyen, National University of Singapore, Singapore
 2014 Elsevier Ltd. All rights reserved.

11.06.1 Introduction 113

11.06.2 Overview of Micro-EDM and Micro-ECM 114
11.06.2.1 Micro-EDM 114
11.06.2.2 Micro-ECM 115
11.06.2.3 Advantages and Disadvantages of Micro-EDM versus Micro-ECM 116
11.06.3 Sequential Micro-EDM and Micro-ECM 117
11.06.4 Micro-EDM Using Deionized Water 121
11.06.5 Simultaneous Micro-EDM and Micro-ECM Drilling 122
11.06.5.1 Principle of SEDCM Drilling 125
11.06.5.2 Improved Performances of SEDCM Drilling 126
11.06.5.3 Main Factors of SEDCM Drilling 127
11.06.5.3.1 Short Voltage Pulses 127
11.06.5.3.2 Feed Rate 128
11.06.5.4 Modeling of Radial Gap in SEDCM Drilling 128
11.06.5.4.1 Radial Gap Model 129
11.06.5.4.2 Polarization of Double Layer Capacitor 130
11.06.5.4.3 Determination of Dissolution Rate 131
11.06.5.4.4 Simulation of Radial Gap Distance over Time 131
11.06.5.4.5 Simulation and Experimental Verification 131
11.06.6 Simultaneous Micro-EDM and ECM Milling 133
11.06.6.1 Principle of SEDCM Milling 133
11.06.6.2 Main Factors of SEDCM Milling 134
11.06.6.2.1 Scanning Feed Rate 134
11.06.6.2.2 Short Voltage Pulses 136
11.06.6.2.3 Layer Depth 136
11.06.6.3 Improved Performances of SEDCM Milling 136
11.06.6.3.1 Surface Integrity 136
11.06.6.3.2 Dimensional Accuracy 137
11.06.6.4 Critical Conditions for Transition of Micro-EDM/SEDCM/Micro-ECM Milling 139
11.06.6.4.1 Development of the Model 139
11.06.6.4.2 Criteria for Transitions of Material Removal Mechanisms 140
11.06.6.4.3 Dissolution Time 141
11.06.6.4.4 Dissolution Rate 142
11.06.6.4.5 Feed Rate for Critical Conditions 143
11.06.6.5 Simulation and Experimental Verification 143
11.06.6.6 Analysis of the Three Typical Material Removal Mechanisms 145
11.06.7 Summary 148
References 148

11.06.1 Introduction

In recent years, the demands of microfeatures and microshapes coming from electronics, medical applications, and aviation
industries has increased rapidly (1). Miniaturization is an indispensible and vital direction to obtain thinner, smaller, and lighter
products. The applications include microholes for fiber optics, micronozzles for jet engines, micromold, and die for micro-optic and
microfluidic devices, etc. (2). Hence, the well-established machining processes need to be innovated for these microapplications (3).
Although special manufacturing processes such as photolithography, focus-ion-beam, and electron-beam lithography could be
used to fabricate microstructures, such methods require high expenditure for equipment and maintenance (4,5). Therefore, there
have been many attempts to develop low-cost macromachining processes for micro- and nanomachining applications. Conven-
tional metal cutting processes such as turning, milling, and grinding could generate surfaces with nanofinish. However, there is size
limitation of cutting tools to fabricate complex microfeatures due to the existence of cutting forces during machining (3). Among the

Comprehensive Materials Processing, Volume 11 http://dx.doi.org/10.1016/B978-0-08-096532-1.01329-7 113

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
114 Compound and Hybrid Micromachining

tool-based machining processes, micro-electrical discharge machining (EDM) and micro-electrochemical machining (ECM) are
highly favorable due to the fact that they are noncontact machining processes (6,7). During machining, the electrode and workpiece
are separated by a fine gap. As a result, the cutting force is negligible, whereby a very fine electrode could be used to fabricate
microshapes and intricate features (8,9).
Each of these two machining processes has its own advantages and disadvantages. The primary weakness of micro-EDM is high
surface roughness, whereas relatively lower material removal rate (MRR) and dimensional accuracy are the main drawbacks of
micro-ECM. Nevertheless, surface finish and machining accuracy are both of prime importance for microfeatures and products.
Hence, there is a need to associate these two processes to exploit their strengths and reduce their adverse effects. However, the
difference in machining fluid used is a challenging issue. Micro-EDM operates in nonconductive dielectric fluid, whereas micro-
ECM employs conductive electrolyte during machining. For that reason, micro-EDM and micro-ECM have usually been used as
sequential machining processes. Although this approach is feasible, it has certain practical disadvantages. If they are carried out on
different machine tools, the change of machine set-up after micro-EDM is problematic and impractical for microshapes. In addition,
when being performed on the same machine tool, the electrolyte and dielectric fluid need to be alternated, tending to cause
contamination of the machining fluid easily.
This chapter introduces the recent advances in hybrid EDM/ECM and presents an approach to combine micro-EDM and micro-
ECM as a unique hybrid machining process, referred to as simultaneous micro-EDM and micro-ECM (SEDCM), which is expected
to be capable of fabricating microshapes with enhanced surface integrity and dimensional accuracy.

11.06.2 Overview of Micro-EDM and Micro-ECM

11.06.2.1 Micro-EDM
EDM is an electrothermal machining process in which the electroerosion phenomenon is exploited to remove undesirable material
from the workpiece (10). During machining, a series of discrete electric discharges is precisely controlled to occur in the fine gap
between the electrode and workpiece, which are immersed in dielectric fluid, as shown in Figure 1. Each discharge removes a small
material amount, leading to the formation of a discharge crater on the machined surface.
Figure 2 illustrates the working mechanism of the EDM process. Firstly, the gap voltage is applied across the tool electrode and
workpiece; thus, the electric field is created in the gap between them. The electrode is then driven by a servo controller to reduce the
gap distance leading to the increase of electric field. As the gap meets the critical value, the electric field is stronger than the dielectric

Figure 1 Concept of EDM. Reproduced from Kunieda, M.; Lauwers, B.; Rajurkar, K. P.; Schumacher, B. M. Advancing EDM through Fundamental
Insight into the Process. CIRP Ann. – Manuf. Technol. 2005, 54, 64–87.

Figure 2 EDM process mechanism. Reproduced from Rajurkar, K. P.; Levy, G.; Malshe, A.; Sundaram, M. M.; McGeough, J.; Hu, X., et al. Micro and
Nano Machining by Electro-physical and Chemical Processes. CIRP Ann. – Manuf. Technol. 2006, 55, 643–666.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining 115

Table 1 Capabilities of micro-EDM

Micro-EDM variant Geometric complexity Minimum feature size Maximum aspect ratio Surface quality Ra (mm)

Drilling 2D 5 mm w25 0.05–0.3

Die-sinking 3D w20 mm w15 0.05–0.3
Milling 3D w20 mm w10 0.5–1
WEDM 21/2 D w30 mm w100 0.1–0.2
WEDG Axi-sym. 3 mm 30 0.8

Reproduced from Rajurkar, K. P.; Levy, G.; Malshe, A.; Sundaram, M. M.; McGeough, J.; Hu, X., et al. Micro and Nano Machining by Electro-physical and
Chemical Processes. CIRP Ann. – Manuf. Technol. 2006, 55, 643–666.

strength. There is then a breakdown of dielectric fluid and the spark occurs. The plasma column grows, within which the electrons
move toward the anode and the positive ions move toward the cathode. When the electrons hit the anode and the positive ions
reach the cathode, their kinetic energies are converted into heat. It is reported that extremely high temperature (8000–12 000  C) is
created in the plasma column (11). The material is thus melted and vaporized. Besides, dielectric fluid is also evaporated forming
dielectric gases. At the end of discharge, the plasma column disappears. The heated dielectric gas envelope collapses, ejecting
material from the electrodes in the form of debris. A discharge crater is thus formed on the machined surface. With the flushing of
fresh dielectric fluid, the debris generated is carried away. Another discharge occurs, and the process repeats.
Micro-EDM is the innovation of EDM for micromachining (8). The mechanism of micro-EDM is similar to EDM. However, there
are some differences between micro-EDM and EDM. Firstly, micro-EDM is used for fabricating microfeatures so the electrode used
usually has a smaller size (<500 mm). Secondly, the discharge energy is lowered (<100 mJ) to reduce the crater size (7,13). Therefore,
the RC-type pulse generator is more favorable for micro-EDM because it can give short pulse duration and relatively constant pulse
energy (12). Lastly, precise movement mechanisms are required to improve the dimensional accuracy (7,10). Micro-EDM can be
used to drill simple microholes or fabricate complex micromolds (13). In general, micro-EDM can be classified into five main types,
of which the capabilities are summarized in Table 1.

11.06.2.2 Micro-ECM
Electrochemical machining is a material removal process based on the dissolution of metal during the electrolysis of the electro-
chemical cell (14). The illustration of the electrochemical machining process is given in Figure 3.
When the voltage is applied across the anode and the cathode is immersed in the electrolyte, a current passes through them
because the electrolyte acts as a current carrier (15). The anode is dissolved and the shape of the workpiece is approximately the
negative image of the tool (16).
The mechanism of micro-ECM is also similar to ECM. However, the dissolution zone must be localized in micro-ECM to assure
the dimensional accuracy (7). As a result, it requires some modifications such as using smaller electrode size, applying ultrashort
voltage pulses, and lower current and voltage (9,16). In general, micro-ECM can be categorized into four main types (12):
l Micro-ECM drilling (17)
l Micro-ECM using mask (18,19)
l Micro-ECM milling (20)
l Die-sinking micro-ECM

Figure 3 Schematic illustration of the electrochemical machining process. Reproduced from Kalpakjian, S. Manufacturing Processes for Engineering
Materials, 3rd ed.; Addison Wesley, 1997.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
116 Compound and Hybrid Micromachining

11.06.2.3 Advantages and Disadvantages of Micro-EDM versus Micro-ECM

Table 2 summaries the advantages and disadvantages of micro-EDM against micro-ECM. These characteristics mainly stem from
the material removal mechanism of each process. During machining by micro-EDM, material is removed by vaporization and
melting. As a result, the machined surface is made up with thermally damaged layers consisting of the white layer and the heat-
affected zones, as can be seen in Figure 4 (21–23). Microcracks and residual stresses are also observed in these distinctive layers
(24–27). Consequently, the fatigue strength of the product is highly reduced. Besides, after each discharge, a small amount of
material is removed, forming a crater on the surface. The generated surface is thus covered by a multitude of overlapping dis-
charged craters (28–30). Therefore, the surface machined by EDM usually has high surface roughness due to its asperity. The
topography and roughness of machined surface is mainly constituted by the crater size, which is dependent on the discharge
energy (31). On the other hand, the material is removed not only from the workpiece but also from the electrode, which has been
known as electrode wear (32,33). This influences the machining shape and accuracy, especially in micro-EDM drilling and
milling. However, the MRR of micro-EDM is considerably higher, and its accuracy could be controlled better than micro-ECM
(34,35).
In micro-ECM, the material is removed based on the dissolution of metal from the anode. As a result, the dissolution rate of
electrochemical reaction is relatively low. Furthermore, ultrashort pulse and low voltage current must be used in micro-ECM to
improve accuracy by reducing the interelectrode gap (9,16,36). Hence, the MRR of micro-ECM process is considerably lower than
micro-EDM. Although the throwing power is small, the dissolution could occur in an area larger than the facing zone of the
electrode (7). The material is unanticipatedly removed from workpiece, leading to the distortion of machined shapes. Therefore,
accuracy is an obstacle in micro-ECM.
However, micro-ECM has some valuable advantages. Because the material is removed by electrochemical reaction, the surface
generated by micro-ECM is very smooth (7,12,37). Due to the nature of ionic dissolution, there is no thermally affected layer made
up on the machined surface. As a result, it is stress free and there is no burr or microcrack. In addition, during the process, only gas
evolution occurs at the cathode surface. Consequently, there is no tool wear during machining (38).

Table 2 Comparison of micro-EDM and micro-ECM

Micro-EDM Micro-ECM

Advantages Higher accuracy No tool wear

Higher MRR (No heat-affected zone)
Good surface finish
Disadvantages Tool wear Lower accuracy
Thermally damaged layers Lower MRR
High surface roughness

Figure 4 Section observation of micro-EDMed surface. Reproduced from Ekmekci, B.; Sayar, A.; Öpöz, T. T.; Erden, A. Geometry and Surface Damage
in Micro Electrical Discharge Machining of Micro-Holes. J. Micromech. Microeng. 2009, 19, 105030.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining 117

11.06.3 Sequential Micro-EDM and Micro-ECM

Surfaces generated by micro-EDM incur poor surface integrity due to the overlapping of numerous discharge craters and the
formation of distinct thermally damaged zones. These inherent characteristics stem from the nature of material removal by electric
sparks. Therefore, there have been many attempts to enhance the integrity of EDMed surfaces in recent years. One of the approach
directions is using EDM and ECM as sequential machining processes. The primary aims of these studies are to lower the surface
roughness induced by overlapping discharge craters and to remove the thermally damaged zones created during EDM process.
Pioneering work was carried out to perform the finishing of wire EDM (WEDM) products by using ECM (39). The remaining part
of the wire-cut process was used as a mate-electrode in the electrochemical machining step, as shown in Figure 5. The NaNO3
electrolyte was controlled to flow through the gap created by the WEDM step. Within a few seconds, the maximum surface
roughness Rmax was dramatically reduced from over 20 mm to 2–4 mm only. The samples were made from SKD11, SKD61, SUS304,
and brass. A similar method was also used to smooth the surface made from tungsten carbide (40). A smooth surface was obtained
without heat-affected zones or cracks. A specially designed pulse train was applied to uniformly dissolve tungsten carbide. However,
it requires the proper selection of electrode material to prevent the dissolution from the electrode during the reverse voltage pulse. In
addition, different electrolytes for finishing the EDMed surface by ECM have also been investigated (41). An acidic medium is found
to have better smoothing and polishing effects on the surface topography. For the environmental aspect, sodium nitrate also yields
a good polishing rate, but the current density must be identified.
For microapplication, the product size is small. Therefore, low-conductivity electrolyte is required to localize the dissolution
during the micro-ECM step (16). To enhance the surface finish of micropins used for micronozzle fabrication, deionized water with
0.6 MU cm specific resistivity has been used as a weak electrolyte in a new wire electrochemical grinding process, as shown in
Figure 6 (16,37). The set-up used is similar to that of wire electrodischarge grinding, but the electric discharge is simply replaced by
the electrochemical reaction (42). By applying a voltage of 40 V, which is higher than that of normal electrochemical machining,
and using low-feed speed and large depth of cut, the mirrorlike surface was obtained. Other attempts also used deionized water to
reduce the surface roughness of microholes during the sequential ECM step. Deionized water owning 5  104 U cm resistivity was
used as both the dielectric and the electrolyte for machining microholes (43). After EDM, microholes were machined by ECM for
a fixed period of time. The average surface roughness Ra decreased from 0.6 mm to less than 0.05 mm after 60 s machining time. The
optimum duration for ECM was found to be between 40 and 60 s. For higher ECM time, the machining shapes were severely
distorted due to excessive material removal, as shown in Figure 7.
A similar attempt was also performed for through microholes but with higher resistivity of deionized water, 2 MU cm, to prevent
the distortion of microhole (44). After 6 min machining time, the surface roughness was significantly reduced from 0.225 to
0.066 mm Ra, as can be seen in Figure 8. It is reported that the usage of deionized water with resistivity as low as 0.1 MU cm could
lead to the distortion at the entrance and exit of microholes due to excessive dissolution, notwithstanding that its middle area is still
covered with discharge craters, as shown in Figure 9.

Figure 5 Principle of the mate-electrode method. Reproduced from Masuzawa, T.; Sakai, S. Quick Finishing of WEDM Products by ECM Using a Mate-
electrode. CIRP Ann. – Manuf. Technol. 1987, 36, 123–126.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
118 Compound and Hybrid Micromachining

Figure 6 Wire electrochemical grinding process. Reproduced from Bhattacharyya, B.; Munda, J.; Malapati, M. Advancement in Electrochemical Micro-
machining. Int. J. Mach. Tools Manuf. 2004, 44, 1577–1589.

Figure 7 Progressive effect of ECM polishing. Reproduced from Campana, S.; Miyazawa, S. Micro EDM and ECM in DI Water. In ASPE Annual
Meeting, 1999.

Figure 8 Inner hole surface (a) after micro-EDM, (b) after finishing. Reproduced from Chung, D. K.; Shin, H. S.; Kim, B. H.; Park, M. S.; Chu, C. N.
Surface Finishing of Micro-EDM Holes Using Deionized Water. J. Micromech. Microeng. 2009, 19, 045025.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining 119

Figure 9 Hole with reverse barrel shape when ECM finishing using 0.1 MU cm deionized water. Reproduced from Chung, D. K.; Shin, H. S.; Kim, B. H.;
Park, M. S.; Chu, C. N. Surface Finishing of Micro-EDM Holes Using Deionized Water. J. Micromech. Microeng. 2009, 19, 045025.

Microholes on high nickel alloys machined by micro-EDM were also smoothed by the electropolishing process (45). Electrolyte
solution with 85% H3PO4 was used in the electropolishing step with the set-up diagram shown in Figure 10. In view of the high
conductivity of the electrolyte, only low voltage was applied, from 1 to 5 V. After 5 min machining time at the electrolytic voltage of
2 V, taper and burrs were reduced and the surface roughness dropped from 2.11 to 0.69 mm Rmax, as shown in Figure 11. Recently,
dilute electrolyte has also been used in the sequential micro-ECM step to enhance the surface finish of some 3D microshapes. The
surface roughness of the hemisphere is found to be reduced from 0.08 mm Ra to about 0.03 mm Ra after performing micro-ECM
using 0.1 M H2SO4 electrolyte (35). Similarly, the surface finish of several 3D metallic microstructures has also been improved
from 0.707 to 0.143 mm Ra by using an electrolyte solution consisting of 3% wt NaClO3, as shown in Figure 12 (46).
With a view to further enhancing the integrity of the surface generated by micro-EDM, hybrid micro-ECM/lapping has also been
used during the finishing step. The main objective of these attempts is to associate the dissolution effect of the electrochemical
reaction and the polishing effect of abrasive grains. One example of the earliest research was reported by Takahata (47). Fine
abrasive grains Al2O3 were mixed with colloidal aqueous electrolyte. During machining, beside the metal dissolution by electro-
chemical reaction, the movement of abrasive grains impacted by the rotating electrode increases the efficiency of mechanical
polishing, as illustrated in Figure 13. The mirrorlike surface with 32 nm Rmax was obtained after 120 s machining time, as shown in
Figure 14.
Another attempt using a similar approach was made to obtain a smoother surface of hardened steel after micro-EDM (48). The
abrasive grains Al2O3 were also used but with different grain sizes from 2 to 13 mm. A surface with 0.06 mm Ra was obtained after the
ECM/lapping process. It is reported that the surface roughness after ECM/lapping was lower than that after ECM or polishing alone
(Figure 15).

Figure 10 Schematic diagram of electropolishing after micro-EDM. Reproduced from Hung, J. C.; Yan, B. H.; Liu, H. S.; Chow, H. M. Microhole
Machining Using Micro-EDM Combined with Electropolishing. J. Micromech. Microeng. 2006, 16, 1480.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
120 Compound and Hybrid Micromachining

Figure 11 Comparison with the electrodischarge-machined and electropolished surface. Reproduced from Hung, J. C.; Yan, B. H.; Liu, H. S.; Chow, H.
M. Microhole Machining Using Micro-EDM Combined with Electropolishing. J. Micromech. Microeng. 2006, 16, 1480.

Figure 12 SEM photos of square cavity machined by (a) micro-EDM and (b) combined milling. Reproduced from Zeng, Z.; Wang, Y.; Wang, Z.; Shan,
D.; He, X. A Study of Micro-EDM and Micro-ECM Combined Milling for 3D Metallic Micro-structures. Precis. Eng. 2012.

Figure 13 Diagram of machining part around the electrode with Al2O3 lump. Reproduced from Takahata, K.; Aoki, S.; Sato, T. Fine Surface Finishing
Method for 3-Dimensional Micro Structures. In An Investigation of Micro Structures, Sensors, Actuators, Machines and Systems, 1996.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining 121

Figure 14 Comparison between machined surfaces.

Figure 15 Surface of machined holes. Reproduced from Kurita, T.; Hattori, M. A Study of EDM and ECM/ECM-lapping Complex Machining Technology.
Int. J. Mach. Tools Manuf. 2006, 46, 1804–1810.

11.06.4 Micro-EDM Using Deionized Water

Deionized water is an alternative dielectric fluid to hydrocarbon oil for the EDM process (49). It is an ecofriendly substance that
could provide a better and safer environment when working with EDM since it does not release harmful gases such as CO or CH4.
Especially, the salient characteristic that makes deionized water superior to hydrocarbon oil is that it yields higher MRR and lower
electrode wear.
The earliest attempt to use distilled water in EDM carbon steel was performed by Jeswani (50). Under the same machining
conditions and in high pulse energy (72–288 mJ), EDM using distilled water results in higher MRR and lower electrode wear
compared to kerosene. It is also reported that the surface finish is better, but the machining accuracy is poor. Then, the performance
of different water qualities in EDM was also investigated (51). Tap water (0.25  104 U cm resistivity), distilled water
(0.32  105 U cm resistivity), and a mixture of the two with a 25–75% ratio were used. It was observed that tap water yielded the
best machining rate. It also showed the possibility of zero electrode wear when a copper electrode with negative polarity was used in
EDM with water. It is also reported that the erosion in water has higher thermal stability, and thus higher power could be input in
the discharge (52). Accordingly, the MRR when water is used as a dielectric fluid could be greatly increased.
In addition, the comparison of EDM titanium alloy Ti–6Al–4V in kerosene and distilled water was also performed (53). MRR
was also found to be higher, and tool wear is also observed to be lower when distilled water was used as a dielectric fluid. A similar
observation is also obtained when machining microslits on titanium alloys (54). It is found that when water is used as a dielectric
fluid, a thin layer of TiO is generated on the machined surface. On the contrary, a thick TiC layer is formed when kerosene was used.
The melting point of TiC is 3150  C, which is much higher than that of TiO, 1750  C. For that reason, the TiC needs higher energy
density to be removed; thus, the MRR when kerosene is used is much lower. It is also reported that the debris size in distilled water is
greater, but the impulsive force of discharge is smaller and more stable than in an oil medium (53).
Deionized water has also been attempted to be used for micro-EDM. The B0.1 mm microholes with high aspect ratio were
fabricated on S45C carbon steel (55). It was found that using purified water also resulted in higher MRR and lower tool wear. Later,

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
122 Compound and Hybrid Micromachining

Figure 16 Microcolumn fabricated by micro-EDM milling using deionized water with different resistivity: (a) 0.1 MU cm and (b) 12 MU cm.
Reproduced from Chung, D. K.; Kim, B. H.; Chu, C. N. Micro Electrical Discharge Milling Using Deionized Water as a Dielectric Fluid. J. Micromech.
Microeng. 2007, 17, 867.

a horizontal micro-EDM set-up was carried out to improve the flushing effect of dielectric fluid with the aim of drilling deep
microholes (56). Microholes obtained have an aspect ratio as high as 10. However, in those studies, high-resistivity deionized water
(106–107 U cm) must be used to sufficiently suppress the electrochemical reaction. It was also mentioned that the reliability and the
repetition rate of discharge in deionized water are better than oil. Recently, water has also been used in micro-EDM milling (57).
High-resistivity deionized water (12 MU cm) was also used to prevent distortion of the machining shape due to the excessive
unanticipated dissolution of the material. A similar observation of MRR and electrode wear was also obtained. The MRR was found
to be higher and tool wear was observed to be lower when machining in deionized water.
Although deionized water is capable of yielding higher MRR and lower electrode wear, the stray material dissolution during
machining deteriorates the dimensional accuracy of machined shapes, as can be seen in Figure 16 (56,57). Hence, it has been
considered to be the main drawback of micro-EDM using deionized water. Consequently, many attempts have been made to
mitigate the excessive dissolution caused by electrolysis in deionized water during machining. Antielectrolysis power supplies
were developed to reduce the dissolution of the workpiece material (58–60). In general, these power supplies applied AC voltage
instead of DC voltage. The polarity of the electrode and the workpiece was permuted continuously after each pulse to reverse the
electrochemical reaction. As a result, it was reported that the metal dissolution was reduced. However, electrode wear was also
found to be very high because the polarity of the electrode was positive for half of the machining time. For that reason, deionized
water has only been used for wire EDM because the fresh running wire is continuously supplied during the machining process
(10,61).
Micro-EDM of tungsten carbide was also performed in deionized water (61–63). It was reported that there is severe dissolution
of cobalt binder from tungsten carbide notwithstanding that the resistivity of deionized water used was in the 1.6–1.8 MU cm range.
In order to eliminate this problem, a bipolar pulse combined with a modified-shape electrode has been attempted for micro-EDM
drilling using deionized water, as shown in Figure 17 (61). It was found that electrolytic corrosion is less when a 125 kHz bipolar
pulse with a duty factor of 25% and a negative voltage of 20 V were used. However, the electrolysis was not completely suppressed.
Therefore, an electrode with a circular cross-section was modified to be a rectangular, square, and triangular cross-section, as can be
seen in Figure 18. In combination with bipolar pulses, the triangular cross-section electrode was observed to be the most effective
among the different electrode shapes in suppressing electrolytic corrosion. This is due to the fact that it has the least side area
compared to the others.
On the other hand, because the side area of this electrode type is very small, the MRR is also too low and the electrode wear is
violent in micro-EDM milling, resulting in the deformation of machining shapes. In addition, the triangular-section electrode takes
a long time to be fabricated. For that reason, deionized water spray and a bipolar pulse combination have been applied for micro-
EDM drilling and milling of tungsten carbide (62,64). In these studies, deionized water and compressed air are mixed together in the
form of a mist which was then used as a dielectric medium to prevent corrosion. Some water drops go into the narrow machining
gap while the other drops on the workpiece surface are blown away by the compressed air, as illustrated in Figure 19. As a result, it
breaks the continuous electrical connection between the electrode and the surface of workpiece adjacent to the machined hole.
Consequently, the high-quality microgrooves could be machined on WC–Co, as can be seen in Figure 20. Recently, it has also been
reported that by using high frequency bipolar pulses, microholes without electrolytic corrosion could be also obtained (65).

11.06.5 Simultaneous Micro-EDM and Micro-ECM Drilling

Micro-EDM operates in a nonconductive dielectric fluid, whereas micro-ECM employs a conductive electrolyte. Because of two
rather divergent requirements, micro-EDM and micro-ECM are usually used sequentially. This section presents a method to
combine micro-EDM and micro-ECM in a unique hybrid machining process to achieve improved performance in both surface
finish and machining accuracy by using low-resistivity deionized water, which exhibits both characteristics of a slightly conductive
fluid and a dielectric fluid.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining 123

Figure 17 Operation of bipolar pulse generator: (a) schematic of bipolar pulse generator circuit and (b) bipolar pulse waveform. Reproduced
from Song, K. Y.; Chung, D. K.; Park, M. S.; Chu, C. N. Micro Electrical Discharge Drilling of Tungsten Carbide Using Deionized Water. J. Micromech.
Microeng. 2009, 19, 045006.

Figure 18 Various types of electrodes: (a) cylindrical, (b) rectangular, (c) square, and (d) triangular. Reproduced from Song, K. Y.; Chung, D. K.; Park,
M. S.; Chu, C. N. Micro Electrical Discharge Drilling of Tungsten Carbide Using Deionized Water. J. Micromech. Microeng. 2009, 19, 045006.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
124 Compound and Hybrid Micromachining

Figure 19 Schematic diagram of spray ED milling. Reproduced from Song, K. Y.; Chung, D. K.; Park, M. S.; Chu, C. N. Micro Electrical Discharge
Milling of WC-Co Using a Deionized Water Spray and a Bipolar Pulse. J. Micromech. Microeng. 2010, 20, 045022.

Figure 20 Scanning electron microscopic images of machined slots. Reproduced from Song, K. Y.; Chung, D. K.; Park, M. S.; Chu, C. N. Micro
Electrical Discharge Milling of WC-Co Using a Deionized Water Spray and a Bipolar Pulse. J. Micromech. Microeng. 2010, 20, 045022.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining 125

11.06.5.1 Principle of SEDCM Drilling

Figure 21 shows the principle of SEDCM drilling in low-resistivity deionized water (66). In this process, short voltage pulses are
applied instead of a continuous voltage in the normal micro-EDM using an RC-type pulse generator. When the process begins, the
short voltage pulses are applied across the electrode and the workpiece. Then, the electrode is lowered down to reduce the gap
between the electrode and the workpiece, as is shown in Figure 21(a). As soon as the gap meets a critical value, there is a breakdown
of deionized water and sparks occur, as shown in Figure 21(b). Material is removed from the workpiece by melting and vapor-
ization. Each discharge leaves a crater on the machined surface. Therefore, the surface is covered with a multitude of overlapped
discharge craters; as a result, it is rather rough. After a period of time, the gap width between the electrode and the workpiece
increases due to material removal by the sparks and no further discharges occur. With the supplied voltage pulses, an electro-
chemical reaction occurs owing to the slight conductivity of deionized water, as illustrated in Figure 21(c). Anodic dissolution takes
place at the workpiece surface, producing electrons that are transferred to the cathode by low-resistivity deionized water, which acts
as a current carrier. Material is dissolved from the workpiece, and its surface roughness decreases. Due to the use of short voltage
pulses, the dissolution of material is localized within a certain distance, which is marked with a dotted line in Figure 21(c). Hence,
the machining gap is confined, and the machining accuracy could be maintained. After a period of time, the electrode is lowered
down further through feeding and the process repeats the cycle as illustrated in Figure 21(d). In order for the electrochemical
reaction to have enough time to dissolve the material, the feed rate must be sufficiently low. In short, this principle simultaneously
combines micro-EDM and micro-ECM in a unique hybrid machining process with three approach directions: low-resistivity
deionized water as a bicharacteristic fluid, low feed rate to promote the electrochemical reaction, and short voltage pulses to
localize the dissolution zone.
The usage of a conventional RC-type pulse generator is not suitable for micro-EDM in deionized water due to excessive material
removal, especially in deionized water with low resistivity. It can lead to distortion of the machining shape, resulting in poor
machining accuracy. Hence, short voltage pulses are used in this method to localize material dissolution and thus assure accuracy.
Figure 22 shows the model of the electrochemical cell in terms of circuit elements. When a voltage is applied across two electrodes
immersed in deionized water, an electrochemical reaction occurs because deionized water acts as a weak electrolyte. A double layer
is formed at the interface of the electrodes and the electrolyte. It was reported that this electrode–solution interface behaves as a two-
parallel-plate capacitor; thus, this interfacial region could be modeled as a capacitor (as shown in Figure 22). After the voltage has
been applied, this double layer capacitor is charged. The charging time is the product of the resistance of electrolyte and the
capacitance of the double layer:
s ¼ R$CDL ¼ r$dg $CDL

where R is resistance of the electrolyte. It depends on the path of the current and can be determined by multiplying the gap distance
between the anode and the cathode (dg) and the specific resistance of the electrolyte (r). CDL is the capacitance of the double layer.

Figure 21 Principle of SEDCM drilling. Reproduced from Nguyen, M. D.; Rahman, M.; Wong, Y. S. Simultaneous Micro-EDM and Micro-ECM in Low-
resistivity Deionized Water. Int. J. Mach. Tools Manuf. 2012, 54–55, 55–65.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
126 Compound and Hybrid Micromachining

Figure 22 Model of electrochemical cell in terms of circuit elements. Reproduced from Schuster, R.; Kirchner, V.; Allongue, P.; Ertl, G. Electrochemical
Micromachining. Science 2000, 289, 98–101.

This double layer capacitor is only considerably charged if the pulse duration (ton) exceeds this charging time constant (ton > s).
On the other hand, the electrochemical reaction rate is exponentially dependent on the potential drop across this double layer
capacitor. Hence, the electrochemical reaction could be localized by controlling the polarization of this double layer capacitor. By
giving short pulses, the electrochemical reaction only occurs in a restricted area. As shown in Figure 22, the double layer capacitor at
the bottom of electrode is charged faster than its side. This is due to the different charging time constant caused by various gap
distances. The area facing the bottom of the electrode has lower resistance (Rshort < Rlong), so the double layer is polarized earlier and
the electrochemical reaction rate is higher. By supplying an adequately short voltage pulse (tlong > s > tshort), the double layer is
highly polarized at the bottom of the electrode while it is only weakly polarized at its side surface. As a result, the electrochemical
reaction could be localized within a small gap distance.

11.06.5.2 Improved Performances of SEDCM Drilling

Figure 23 shows the fabricated microholes at 60 V, using a DC regime with two different feed rates: 10 mm s1 and 0.2 mm s1. As
shown in Figure 23(a), in the case of a 10 mm s1 feed rate, the microhole surface is observed to be covered with overlapped
discharge craters. In particular, the recast material can be visibly seen at its sharp edge. On the contrary, the cylindrical surface of
the microhole machined at a 0.2 mm s1 feed rate is found to be smooth and free of craters, as can be seen in Figure 23(b).
However, micropits appear on the surface adjacent to the rim of the microhole, and the microhole diameter is conspicuously
expanded. This is induced by the excessive dissolution of material under continuous voltage supplied in low-resistivity deionized
water. When a 10 mm s1 feed rate is used, the feed is rather fast so there is not much time for the electrochemical reaction to
dissolve the material after the sparks have stopped. In contrast, a 0.2 mm s1 feed rate is relatively low; thus, ionic dissolution can
happen on the side surface of the machined microhole. However, under continuously applied voltage, the electrochemical
reaction drastically occurs and is not restricted. Accordingly, it excessively dissolves the material and increases the machining gap
between the electrode and the workpiece. As a consequence, the machining shape is distorted, and the machining accuracy
deteriorates.
Hence, short voltage pulses were applied instead of the DC regime so as to contain the material dissolution during machining.
Figure 24 shows the microhole fabricated using pulses at the frequency of 500 kHz with 30% duty ratio and 0.2 mm s1 feed rate.
The obtained microhole is found to be free of craters on its side surface. Moreover, there is no exceeding expansion of diameter as
well as the appearance of micropits adjacent to the rim. This shows the capability of SEDCM to obtain both good surface finish and
high machining accuracy. It also reveals that low-resistivity deionized water, low feed rate, and short voltage pulses are three integral
factors of SEDCM.

Figure 23 Microholes machined using the DC regime at 60 V with different feed rates: (a) 10 mm s1, (b) 0.2 mm s1. Reproduced from Nguyen, M.
D.; Rahman, M.; Wong, Y. S. Simultaneous Micro-EDM and Micro-ECM in Low-resistivity Deionized Water. Int. J. Mach. Tools Manuf. 2012, 54–55,
55–65.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining 127

Figure 24 Microhole fabricated by SEDCM using 500 kHz pulses with 30% duty ratio at 0.2 mm s1 feed rate. Reproduced from Nguyen, M. D.;
Rahman, M.; Wong, Y. S. Simultaneous Micro-EDM and Micro-ECM in Low-resistivity Deionized Water. Int. J. Mach. Tools Manuf. 2012, 54–55, 55–65.

11.06.5.3 Main Factors of SEDCM Drilling

11.06.5.3.1 Short Voltage Pulses
Figure 25 shows the microholes fabricated using pulses at three different frequencies of 100, 300, and 500 kHz, with duty ratios
ranging from 15 to 70%. It can be observed that with a smaller duty ratio (or shorter pulse-on time), the diameter of the microhole
tends to decrease. Similarly, at higher frequency used, smaller microholes are also formed. This can be seen with the reduction of
machining gap in Figure 26. This phenomenon could be explained by using the double-layer model. When a voltage pulse is
applied, the double layer starts to be charged. However, due to the limited pulse width, the double layer is only considerably
charged on the area where the distance between the electrode and workpiece is under a certain value. Therefore, the current density
and the material dissolution rate are significant merely in that area. As a consequence, the electrochemical reaction could be
localized, and the machining gap is restricted within that distance. When a shorter pulse-on time is applied, the machining gap
becomes thinner. This explains why the machining gap is reduced when a higher frequency as well as smaller duty ratio is used.

Figure 25 SEM images of microholes machined with different pulse frequencies and duty ratios. Reproduced from Nguyen, M. D.; Rahman, M.; Wong,
Y. S. Simultaneous Micro-EDM and Micro-ECM in Low-resistivity Deionized Water. Int. J. Mach. Tools Manuf. 2012, 54–55, 55–65.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
128 Compound and Hybrid Micromachining

Figure 26 Machining gaps corresponding to different pulse parameters.

It is also observed that when the pulse-on time is reduced to as short as 300 ns, the cylindrical surface of machined microholes is
found to be covered with overlapped discharged craters although the identical feed rate 0.2 mm s1 is used, as seen in Figure 26(f)
and 26(i). Figure 27 is a zoom-in view of the microhole where the recast material can be seen at its edge. Although the 0.2 mm s1
feed rate is low enough to promote the micro-ECM effect on the machined surface, there is no visible sign of material dissolution in
this case. The cylindrical surface of the microhole is entirely covered with discharge craters. This shows that the electrochemical
reaction is localized when voltage pulses are used; however, the 300 ns pulse-on time is too short so the localized dissolution
distance is even smaller than the spark gap consisting of critical distance and discharge depth. Therefore, the electrochemical
reaction is mostly suppressed. It reveals that short voltage pulses could localize the material dissolution and even suppress the
electrochemical reaction when the pulse width is less than a critical value. Hence, the voltage pulses applied in SEDCM must be
adequately long to promote the electrochemical reaction within a small gap distance, not to absolutely suppress it.

11.06.5.3.2 Feed Rate

For examining the effect of feed rate on the efficiency of SEDCM, comparison is carried out between the microholes machined at the
same 500 kHz frequency, 30% duty ratio but with two different feed rate values: 1.2 mm s1 (Figure 28) and 0.2 mm s1 (Figure 24).
It can be seen that when the 1.2 mm s1 feed rate is used, the side surface of machined microholes is also entirely covered with crater
marks. It is different from the microhole machined at 0.2 mm s1 in Figure 24, of which the side surface is smooth and crater free.
The variation in feed rate is accounted for by this discrepancy. When the electrode moves faster, there is not much time for the
electrochemical reaction to take place and dissolve the material from machined surface after the discharge has stopped. As a result,
the effect of micro-ECM is less, and the machined surface is still covered with a multitude of discharge craters. Accordingly, the
machining gap will be thinner and the machining time will be shorter. This affirms that the feed rate is one of the key elements in
SEDCM and that choosing the appropriate feed rate is important to enhance the efficiency of this hybrid machining process.

11.06.5.4 Modeling of Radial Gap in SEDCM Drilling

For enhancing the surface finish, the uneven material layer is further dissolved from the machined surface owing to the effect of the
electrochemical reaction in SEDCM drilling. In order to maintain the dimensional accuracy of microholes, short voltage pulses are
applied to localize the material dissolution zone. Therefore, the additional thickness of the material layer removed is of prime
importance in deciding the final dimension of microholes. This section presents the modeling of the radial gap distance in SEDCM

Figure 27 Microhole machined using pulses at 300 kHz, 15% duty ratio, and with 0.2 mm s1 feed rate. Reproduced from Nguyen, M. D.; Rahman, M.;
Wong, Y. S. Simultaneous Micro-EDM and Micro-ECM in Low-resistivity Deionized Water. Int. J. Mach. Tools Manuf. 2012, 54–55, 55–65.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining 129

Figure 28 Microhole fabricated at 500 kHz, 30% duty ratio and with 1.2 mm s1 feed rate.

drilling by predicting the thickness of the material layer further dissolved by electrochemical reaction. The analytical model
incorporating the double-layer theory, the Butler–Volmer equation, and Faraday’s law of electrolysis is used to simulate the radial
gap distance for different pulse parameters.

11.06.5.4.1 Radial Gap Model

In conventional micro-EDM drilling, material is removed by the discharge through melting and vaporization. Therefore, the
machining gap formed consists of the critical distance and the discharge depth (67). However, in SEDCM drilling, a thin layer of
affected material on the lateral surface generated by the electric sparks is further removed to enhance the surface integrity of the
microhole, as illustrated in Figure 29 (68). As a result, aside from the critical distance and the discharge depth, the radial gap in this
hybrid process also consists of the dissolution depth, which stems from the electrochemical reaction. Hence, the final radial gap in
SEDCM drilling is also dependent on the thickness of the material layer which is further dissolved. This model focuses on the
material dissolution characteristic after the discharge to perform the modeling of radial gap of obtained microholes. For that reason,
the side gap formed after micro-EDM is considered the initial gap for material dissolution.
Owing to the slight conductivity of low-resistivity deionized water, it could be considered to be a weak electrolyte. Therefore, the
side gap between the electrode and the workpiece could be modeled as an electrochemical cell. When a voltage is applied across two
electrodes immersed in deionized water, the ions in the solution move toward the electrode surface and the double layer is formed
at the interface of the electrode and the electrolyte. It was reported that this electrode–solution interface behaves as a two-parallel-
plate capacitor (9,69). Thus, it could be modeled as a capacitor, as shown in Figure 30 (70,71). In this model, Rsol is the resistivity of
deionized water. The CDL0 and CDL00 are the capacitance of the double layer at the electrode and the workpiece surfaces, respectively.
Likewise, the RF0 and RF00 are the Faradic resistance (or transfer resistance) representing the current density of electrochemical

dg

Electrode
Material removed
by discharge

Workpiece Material removed

by dissolution

Figure 29 Illustration of the radial gap in SEDCM drilling. Reproduced from Nguyen, M. D.; Rahman, M.; Wong, Y. S. Modeling of Radial Gap Formed
by Material Dissolution in Simultaneous Micro-EDM and Micro-ECM Drilling Using Deionized Water. Int. J. Mach. Tools Manuf. 2012.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
130 Compound and Hybrid Micromachining

Figure 30 Model of electrode–workpiece side gap in terms of circuit element. Reproduced from Nguyen, M. D.; Rahman, M.; Wong, Y. S. Modeling of
Radial Gap Formed by Material Dissolution in Simultaneous Micro-EDM and Micro-ECM Drilling Using Deionized Water. Int. J. Mach. Tools Manuf. 2012.

reaction at the surfaces of the electrode and the workpiece, respectively. Since the thickness of the double layer is significantly
smaller than the radial gap distance, the solution resistance Rsol could be expressed as follows:
Rsol ¼ rdg [1]

where r is the specific resistivity of deionized water and dg is the electrode–workpiece gap distance.

11.06.5.4.2 Polarization of Double Layer Capacitor

Prior to the model development, certain assumptions are made:
l The capacitance of the double layer is constant during the machining process.
l The transfer resistance and the capacitance of the double layer at the electrode–solution and workpiece–solution interfaces are
the same (RF0 ¼ RF00 ¼ RF and CDL0 ¼ CDL00 ¼ CDL).
l There is no material dissolution during pulse-off time.
l The roughness of the electrode and workpiece surfaces is neglected in simulating the radial gap distance.

Following the model in Figure 30, the current density flowing through the workpiece–electrolyte interface comprises two routes:
the charging current density iC (to charge the double layer capacitance CDL) and the Faradic current density iF (flows through the
transfer resistance RF). The charging current density iC is given by the equation
dh
iC ¼ CDL [2]
dt
in which t is the time variable and h is the double layer polarization.
It has been reported that the Faradic current density is exponentially dependent on the polarization of the double layers. Hence,
originating from the Butler–Volmer equation (69), the Faradic current density is given by the following equation:
iF ¼ io ½expðazf hÞ  expðazf hÞ [3]
where io is the exchange current density at the equilibrium condition, a is the transfer coefficient, z is the number of electrons
exchanged during electrochemical reaction, and
f ¼ F=RT [4]
in which F is the Faraday constant, R is the gas constant, and T is the absolute temperature.
In eqn [3], the former term corresponds to the anodic current density whereas the latter term refers to the cathodic current
density. These anodic and cathodic reactions happen on the same electrode. In the application that exploits the anodic dissolution
of metal, the double layer polarization h is considerably high and thus the cathodic current density is significantly small. Therefore,
it can be neglected, and the Faradic current density flowing through the transfer resistance RF could be simplified to be
iF ¼ io expðazf hÞ [5]
Hence, the current density flowing from node A to B in Figure 30 is given by the equation
dh
IAB ¼ CDL þ io expðazf hÞ [6]
dt
The current density flowing from node B to C is calculated as follows:
U  UAB  UCD U  2h
IBC ¼ ¼ [7]
Rsol rdg

where U is the amplitude of the applied voltage pulses.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining 131

Here IAB ¼ IBC, so the following equation can be derived from eqns [6] and [7]:
 
dh 1 U  2h
¼  io expðazf hÞ [8]
dt CDL rdg

Now, the polarization of double layer h could be obtained. From that, the Faradic current density iF could be determined using
eqn [5].

11.06.5.4.3 Determination of Dissolution Rate

In order to compute the dissolution rate, the current density needs to be determined. However, in this process, the short voltage
pulses are applied instead of a continuous voltage. Therefore, the average current density must be calculated through the total
electric charge per unit area q. During one pulse, the total electric charge per unit area passing through the substance could be
determined by integrating the current density iF over the pulse-on time ton:
Zton
q¼ iF dt [9]
0

The eqn [9] yields the total electric charge per unit area passing per single voltage pulse. Therefore, the average current density per
second could be obtained by dividing the total electric charge by the pulse period tp as expressed in the equation:
q
ia ¼ [10]
tp

in which
tp ¼ 1=frequency [11]

Then, originating from Faraday’s law of electrolysis (69), the average dissolution rate per second is given by the following
equation:
ia M
n¼ [12]
zF
where M is the molar volume of workpiece material.

11.06.5.4.4 Simulation of Radial Gap Distance over Time

From eqns [5] and [8], it can be seen that the current density is a function of the electrode–workpiece radial gap dg. When the gap
increases, the current density is slightly smaller and thus the dissolution rate is changed. Hence, to simulate the change of the radial
gap over time, the iteration method shown in Figure 31 was used to update the new dissolution rate after each time step Dt.

11.06.5.4.5 Simulation and Experimental Verification

The simulation parameters used in this study are shown in (Table 3).
Figure 32 exhibits the simulated data of current density for different duty ratios during one pulse period. In this case, the
frequency is set at 500 kHz so the pulse period is calculated to be 2 ms. It is noteworthy that during the first 0.3 ms, although the
voltage has been applied, the current density remains steady near zero. Then, it has the sharp rise to reach the peak 0.3 A cm2 after
0.6 ms. This is attributed to the double layer charging characteristic. When the voltage is applied, the ions in the solution move
toward the workpiece surface to form the double layer. The Faradic current is exponentially dependent on the polarization of this
double layer, as indicated in eqn [5]. It takes a certain time for the double layer to be fully charged so the Faradic current merely
remains near zero at the beginning and increases sharply after 0.3 ms. When the current density peaks and remains stable at
0.3 A cm2, it is indicated that the double layer is fully charged.
It can also be observed that with the smaller duty ratio (or shorter pulse-on time), the width of the current density profile
becomes thinner. After the pulse-on time duration, the applied voltage drops to zero and thus the current falls off. In
addition, the dissolution rate is dependent on the total charge transferred during one pulse, as indicated in eqn [9].
Therefore, the higher duty ratio would result in a higher dissolution rate, as plotted in Figure 32(b). It is also noted that
when the duty ratio is too small, the dissolution rate is negligible because the pulse-on time is too short for the double layer
to be significantly charged. This is the situation in which there is no visible effect of material dissolution to improve the
surface finish of the microhole.
Figure 33(a) shows the simulated data of the current density for different initial gap distances during one pulse period. The pulse
frequency and duty ratio are fixed at 500 and 0.3 kHz, respectively. It could be seen that the current density could reach the peak for
the 5-mm gap. It means that the double layer is considered to be fully charged. On the contrary, it is observed that for a higher gap
distance (7 and 9 mm), the current density is reduced and it could not even attain the peak value. This is due to the fact that a 0.6 ms
pulse-on-time is not long enough for the double layer to be fully charged. As a result, the dissolution rate is significantly reduced for
a higher gap distance, as indicated in Figure 33(b). This explains how the material dissolution is localized and thus the dimensional
accuracy is improved when short voltage pulses are used.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
132 Compound and Hybrid Micromachining

Figure 31 Iterative algorithm to simulate the change of radial gap over time.

Table 3 Simulation parameters for radial gap distance

Initial gap distance, dinitial (mm) 5

Specific conductivity of solution, r (MU cm) 0.4
Double layer capacitance, CDL (mF cm2) 0.5
Transfer coefficient, a 0.5
Valency number of ions, z 2
Faraday constant, F (C mol–1) 96 485
Gas constant, R (J mol–1 K1) 8.314
Temperature, T (K) 298.15
Exchange current density, io (mA cm2) 70
Pulse amplitude, U (V) 60
Pulse frequency (kHz) 100, 300, 500
Pulse duty ratio 0.15 to 0.9
Molar volume, M (cm3 mol1) 7.11
Time step, Dt (s) 10
Dissolution time (s) 300

Figure 34 compares the experimental and analytical results of the radial gap for different duty ratios and frequencies. It can be
observed that there is a reasonable agreement between the simulated data and the experimental results for all the frequencies: 100,
300, and 500 kHz. It is found to be in accordance with the aforementioned analysis in which the radial gap is increased when higher
pulse duty ratio is used. In addition, the obtained experimental results also show that the radial gap is slightly smaller when higher
frequency is applied.
In order to further demonstrate the effectiveness of short voltage pulses in localizing the material dissolution zone, the radial gap
is investigated when microhole is exposed to long dissolution time. After finishing the drilling of the microhole, the electrode is
forced to dwell at that position for additional certain duration instead of being retracted. Figure 35 outlines the change in the radial
gap over time and compares the changes with the theoretical results. It could be observed that the experimental results have the
same increasing trend with the simulated data. This confirms the effectiveness of short voltage pulses in maintaining high
dimensional accuracy for SEDCM.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining 133

Figure 32 Simulation of current density (a) and dissolution rate (b) for different duty ratios (frequency ¼ 500 kHz, dinitial ¼ 5 mm).

Figure 33 Simulation of current density (a) and dissolution rate (b) for different initial gap distances (frequency ¼ 500 kHz, duty ratio ¼ 0.3).

11.06.6 Simultaneous Micro-EDM and ECM Milling

Micro-EDM milling is a promising method to fabricate microtooling for replication technologies such as microinjection molding or
hot-embossing. This section details SEDCM milling for the fabrication of intricate microshapes with enhanced surface integrity and
dimensional accuracy with a view to strengthening the capability of micro-EDM for microtooling manufacture.

11.06.6.1 Principle of SEDCM Milling

The working mechanism of SEDCM milling is fundamentally depicted in Figure 36 (72). In this method, material is removed layer-
by-layer to maintain the original shape of the electrode. The electrode–workpiece gap is filled with low-resistivity deionized water,
which exhibits both characteristics of a dielectric fluid and a slightly conductive electrolyte (37,66).
When the process starts, power is supplied in terms of short voltage pulses. For each layer, the tool electrode is lowered by
a certain layer depth, as shown in Figure 36(a). As the interelectrode gap meets the critical value, the discharge takes place, as
illustrated in Figure 36(b). Each discharge forms a crater, and thus the machined surface is covered with numerous overlapped
craters. Due to the material erosion, the gap width expands and no further discharge occurs. At that time, due to the slight
conductivity of deionized water, the voltage pulses induce a weak electrochemical reaction. Anodic dissolution occurs at the
workpiece surface whereby its roughness is reduced, as illustrated in Figure 36(c). On the other hand, due to the short duration of
voltage pulses applied, dissolution zone is localized within a fine gap width (marked with dotted line) (9). Hence, the machined
shape is prevented from distortion, resulting in improved dimensional accuracy. Then, through feeding, the electrode moves
horizontally following the preset tool path pattern to remove material of the entire layer, as shown in Figure 36(d). The discharge
occurs and the cycle is repeated.
Figure 37 illustrates the tool paths used where d is the depth of each layer and Vf is the horizontal scanning feed rate. To attain
hybrid SEDCM milling, the machining speed needs to be judiciously adjusted and short voltage pulses must be used. In the layer-
by-layer removal strategy, machining speed is affected by the thickness of each layer and the horizontal scanning feed rate. Hence,
the scanning feed rate, layer depth, and voltage pulses are the main factors.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
134 Compound and Hybrid Micromachining

Figure 34 Comparison of experimental data and simulated results of the radial gap for different pulse frequencies: (a) 100 kHz, (b) 300 kHz, and (c)
500 kHz.

Figure 35 Change of radial gap over time (frequency ¼ 500 kHz, duty ratio ¼ 0.3).

11.06.6.2 Main Factors of SEDCM Milling

11.06.6.2.1 Scanning Feed Rate
Figure 38 shows 5-mm deep microslots fabricated using different feed rates ranging from 50 to 10 mm s1. The depth of each layer is
0.2 mm, and the voltage is applied in terms of 500 kHz pulses with 30% duty ratio. As can be seen in Figure 38(a), in the case of
a 50 mm s1 feed rate, the machined surface is entirely covered with discharge craters. When the feed rate is gradually reduced, it can
be observed that there appears a smooth zone wherein the craters are diminished. The smooth zone becomes larger as the feed rate is
lowered. This is attributed to material dissolution from the machined surface induced by electrochemical reaction. At a low feed
rate, after the discharge has stopped, deionized water infiltrates the fine gap. The small amount of ions in the low-resistivity

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining 135

Figure 36 Principle of SEDCM milling. Reproduced from Nguyen, M. D.; Rahman, M.; Wong, Y. S. Enhanced Surface Integrity and Dimensional
Accuracy by Simultaneous Micro-ED/EC Milling. CIRP Ann. – Manuf. Technol. 2012, 61, 191–194.

Figure 37 Tool paths to fabricate microslots (a) and microcavities (b).

Figure 38 Microslots fabricated using short voltage pulses at different scanning feed rates: (a) 50 mm s1, (b) 30 mm s1, (c) 20 mm s1, and
(d) 10 mm s1.

deionized water could act as a current carrier to enable the electrochemical reaction. Material is dissolved from the recast layer,
forming a smoother surface. In contrast, when the electrode moves at a higher feed rate, there is less time for the electrochemical
reaction to remove the micro-EDMed surface. As a result, the efficiency of the surface finish improvement is reduced and it becomes
negligible when the feed rate is as high as 50 mm s1, as shown in Figure 38(a). Hence, this indicates that low feed rate is a requisite
for facilitating an electrochemical reaction.
At a 10 mm s1 feed rate, the smooth zone is seen to be widespread on most of the machined surface, as exhibited in
Figure 38(d). However, there are a few discharge craters left at the two ends of the fabricated slot. This is likely to be due to the lack
of ions stemming from the difficulty of fresh deionized water in infiltrating the fine gap after the discharge has stopped. During the
electrochemical reaction, the electrolyte acts as a current carrier to transfer the electrons created during anodic dissolution from
anode to cathode. Therefore, a lack of electrolyte would result in the inhibition of the electrochemical reaction.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
136 Compound and Hybrid Micromachining

In order to improve the replenishment of fresh deionized water in the fine gap, the electrode was lifted up a few microns after
each layer. However, the effect is found to be insignificant, and the craters still exist at the two ends of the microslot. This may stem
from the fact that the dissolution zone is localized within a thin gap width owing to the short voltage pulses. So the uplift of the
electrode also increases the frontal gap, resulting in low efficiency of the electrochemical reaction. Hence, to stimulate the infiltration
of fresh deionized water into the fine gap as well as to hold that gap width under a certain value, the electrode is forced to dwell at
the two ends of the microslot for a few seconds before it starts a new layer. With this technique, the obtained microslot is found to be
smooth on the entire machined area, as shown in Figure 39(a).

11.06.6.2.2 Short Voltage Pulses

Figure 39(b) exhibits the microslot fabricated at the same 10 mm s1 feed rate, but using a continuous voltage supply (conventional
RC-type pulse generator). Compared with Figure 39(a), wherein short voltage pulses are used, the machined surface is also seen to
be relatively smooth with no trace of crater. However, the slot is observed to be conspicuously enlarged due to the unrestrained
material dissolution. This shows that the usage of short voltage pulses is a vital factor to localize the dissolution zone and thereby
prevent the machined shape from distortion.

11.06.6.2.3 Layer Depth

Figure 40 shows the microslots machined at the identical 10 mm s1 feed rate but with higher layer depths, 0.5 mm and 1 mm,
respectively. It is observed that the discharge craters start to show up when the layer depth is increased. This is a consequence of the
increase of material volume that needs to be removed in every feed. Accordingly, the discharge has occurred for longer duration to
remove material, resulting in less machining time for the dissolution. As seen in Figure 40(b), the machined slot is mostly covered
with discharge craters when the layer thickness is set to 1 mm. This shows that small layer depth is another essential factor in SEDCM
milling. This requirement is also preferred in a layer-by-layer removal strategy because a thinner layer depth would yield better
flatness of the machined surface (73).

11.06.6.3 Improved Performances of SEDCM Milling

11.06.6.3.1 Surface Integrity
Figure 41 shows the SEM micrographs of machined microslots and microshapes. At a 50 mm s1 feed rate, the machined surface is
entirely covered with discharge craters, and it is considered to be conventional micro-EDM milling. At a 10 mm s1 feed rate, the
surface of the obtained microslot and shape is observed to be smooth, and it represents SEDCM milling.

Figure 39 Microslots fabricated at a 10 mm s1 feed rate using different power regimes: (a) 500 kHz voltage pulses and (b) continuous voltage.

Figure 40 Microslots fabricated at a feed rate of 10 mm s1 with different layer depths: (a) 0.5 mm and (b) 1 mm.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining 137

Figure 41 SEM micrographs of microslots machined at different feed rates: (a, c, e) 50 mm s1 and (b, d, f) 10 mm s1. Reproduced from
Nguyen, M. D.; Rahman, M.; Wong, Y. S. Enhanced Surface Integrity and Dimensional Accuracy by Simultaneous Micro-ED/EC Milling. CIRP Ann. –
Manuf. Technol. 2012, 61, 191–194.

Figure 42 exhibits the topography of the generated surfaces. Overlapping craters having a diameter of 3–4 m can be seen in
Figure 42(a), whereas a relatively smooth surface with no visible crater is obtained in Figure 42(b). For quantitative comparison of
surface finish improvement, the surface profiles of these two surface textures are also plotted in Figure 42. In the case of micro-EDM
milling (Vf ¼ 50 mm s1), the average surface roughness (Ra) is 142 nm. For SEDCM milling (Vf ¼ 10 mm s1), the Ra is found to be
22 nm only. This substantiates that SEDCM milling yields a better surface finish owing to the effect of the electrochemical reaction.

11.06.6.3.2 Dimensional Accuracy

Figure 43 outlines the profilographs of the microslots and microshapes fabricated under the two conditions. For SEDCM milling
(Vf ¼ 10 mm s1), the width of the machined slot is found to be a few microns larger than that of micro-EDM milling
(Vf ¼ 50 mm s1). The slot width in the case of SEDCM milling is around 112.99 mm, while it is about 110.35 mm for micro-EDM
milling. It reveals that the SEDCM milling yields a smoother surface because a thin layer formed by overlapped discharge craters is
further dissolved from the machined surface by material dissolution.
On the other hand, it should be highlighted that the machining shape is still preserved, and the dissolution is seen to be
contained within a micron-thick layer only, whereby high dimensional accuracy can be attained. This advantage is the result of short
voltage pulses applied. As seen in Figure 44, for SEDCM milling, the shorter the pulse duration (or the smaller the duty ratio), the
thinner the machining gap. At the same 30% duty ratio, the machining gap of SEDCM milling is larger than that of conventional
micro-EDM milling by Dg. Therefore, the Dg here denotes the thickness of the material layer further removed by the electrochemical
reaction, and it is about 1–1.5 mm for the 30% duty ratio. When the duty ratio is increased, the Dg becomes higher. Hence, it
demonstrates that the SEDCM milling could enhance the dimensional accuracy by using short voltage pulses to constrain the
thickness of the material layer further dissolved from the machined surface.
Figure 45 shows two further examples of intricate 3D microshapes machined by micro-EDM milling and SEDCM milling,
demonstrating the feasibility and capability of the SEDCM milling process.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
138 Compound and Hybrid Micromachining

Figure 42 SEM micrographs and profiles of surfaces generated at different feed rates: (a) 50 mm s1 and (b) 10 mm s1.

Figure 43 Profiles of microslots (a) and microcavities (b) fabricated at different feed rate.

Figure 44 Machining gaps of different machining conditions.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining 139

Figure 45 SEM images of 3D microcavities fabricated by different machining conditions: (a, c) micro-EDM milling and (b, d) SEDCM milling.

11.06.6.4 Critical Conditions for Transition of Micro-EDM/SEDCM/Micro-ECM Milling

11.06.6.4.1 Development of the Model
Figure 46 illustrates the material removal mechanism in each horizontal feeding. When a horizontal feed is given to the electrode,
the spark starts to occur in the area where the electrode–workpiece distance is less than the critical distance. The material is removed
through melting and vaporization. In these areas, the interelectrode gap increases. After a time period, the electrode–workpiece gap
is higher than the critical distance and the spark no longer occurs. Then, the electrochemical reaction takes place and the material is
removed from the workpiece by ionic dissolution. Because short voltage pulses are used to confine the machining zone, material
dissolution merely occurs within the limited distance from the surface of the electrode. The electrode is then fed to move hori-
zontally and the process repeats.
For obtaining simultaneous SEDCM milling, both electric discharge and electrochemical reaction must occur in every feed. In
order to attain this hybrid condition, the feed rate and the layer depth must be properly adjusted. If the feed rate is too high or the
layer thickness is too thick, the discharge occurs most of the time in every feed, resulting in a negligible effect of the electrochemical
reaction. On the contrary, when the feed is too slow or the layer depth is too thin, the electrochemical reaction dominates and the
material is entirely removed by ionic dissolution. In this case, the obtained surface is smooth, but the machining speed is low.
Figure 47 depicts the approach used to predict the critical feed rate for transitions of different machining modes. Firstly, the
criteria for material removal mechanism transition are analyzed and identified. Secondly, the remaining time for material disso-
lution when the electrode passes through the surface is estimated. Then, the current density and the dissolution rate are calculated
using a model originating from the double layer theory, the Butler–Volmer equation, and Faraday’s law of electrolysis. Lastly, the
feed rate for the transition of the material removal mechanism is obtained from the critical dissolution time yielded by the criteria
identified in the first stage.

Figure 46 Material removal mechanism in SEDCM milling.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
140 Compound and Hybrid Micromachining

Figure 47 Summary of approach to predict the transition of material removal mechanism.

11.06.6.4.2 Criteria for Transitions of Material Removal Mechanisms

Figure 48 outlines the schematic of the interelectrode gap between the electrode and the workpiece together with the material layer
removed by the discharge and the electrochemical reaction.
Let the original electrode–workpiece gap after vertical feeding be do. When the electrode moves horizontally, the discharge
removed a layer of material with a thickness of ds whereas the electrochemical reaction dissolves a layer of which the thickness is dd.
After finishing that layer, the electrode is lowered down for the next layer by a depth of d. Hence, the new interelectrode gap for
the next layer is given by the equation
d1 ¼ do þ ds þ dd  d [13]
Let dinitial be the distance that the sparks stop and the electrochemical reaction starts to occur. So it will be given by the equation
dinitial ¼ do þ ds [14]
The new interelectrode gap of the next layer could be rewritten as the following equation:
d1 ¼ dinitial þ dd  d [15]
Now, the critical conditions for the transition of the material removal mechanism could be hypothesized based on the eqn [15];
thus there are three distinct situations:
Case 1: dd  d
In this case, the layer of material dissolved by the electrochemical reaction is thicker than the layer depth.
Hence, from eqn [15], we can obtain
d1  dinitial [16]
From eqn [16], it can be seen that the new interelectrode gap d1 is higher than dinitial, which is the gap that the discharge stops.
This indicates that there will be no more discharge, and the material is entirely removed by ionic dissolution from the next layer.
Hence, this case could be considered pure micro-ECM milling.
Case 2: dd  dthres (dd z 0)
In this case, the amount of dissolved material is very small and negligible (dd z 0). Therefore,
d ¼ ds þ dd zds [17]

Figure 48 Schematic of material removal in SEDCM milling.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining 141

Figure 49 Illustration of hypothesized conditions for the transitions of different material removal mechanisms.

The material is entirely removed by the discharges. Therefore, this circumstance could be seen as micro-EDM milling alone.
Ideally, this situation only appears when dd ¼ 0. However, the surface generated by micro-EDM is not perfectly flat in view of the
overlapping of craters. Hence, it is implied that when the dissolved thickness is less than the roughness of the surface generated by
the sparks, the effect of material dissolution is considered to be negligible. Therefore, the threshold value dthres could be decided
based on the average roughness of surface generated by electric discharges.
Case 3: dthres  dd  d
In this situation, the SEDCM milling condition is attained because the material is removed by both the electric discharge and
electrochemical reaction during machining.
In brief, it could be observed that the criteria for the transition of material removal mechanism depends on the thickness of the
material layer that the electrochemical reaction could dissolve when the electrode is scanned over the surface. Hence, the critical
condition for material removal mechanism transition could be obtained based on the prediction of dd for different feed rates and
layer depths.
Figure 49 illustrates the hypothesized graph for transitions of the three aforementioned cases.

11.06.6.4.3 Dissolution Time

Prior to the development of the model, it is assumed that the surface roughness of the workpiece and surface generated by the sparks
are neglected for the gap distance. The volume of material removed per horizontal feed is given by the equation
V ¼ DdVf [18]
where D is the diameter of electrode and Vf is the horizontal scanning feed rate.
Hence, the volume removed by the discharge per feed is
V ¼ Dds Vf [19]
During machining, the horizontal feeding is given to the electrode in every second. Therefore, the percentage of time for
discharge in 1 s could be obtained by the following equation:

Dds Vf Dðd  dd ÞVf

zs ¼ ¼ ð%Þ [20]
MRR s MRR s
in which MRRs is the maximum material removal rate of the sparks. As a result, the time percentage for material dissolution in 1 s is

zd ¼ 1  zs [21]
In contrast to electric discharge, which stops when the interelectrode gap exceeds a specific distance (dinitial), the electrochemical
reaction still occurs on a certain area provided that it is still facing the bottom surface of the electrode. However, it was also reported
that the material dissolution gradually progresses from the outside toward the center of the facing area due to the escape of gases
generated during the sparks (66). Hence, the effective dissolution time for one point on the workpiece surface when the electrode
passes through it is presumed to be
D D
td ¼ zd gd ¼ ð1  zs Þgd [22]
Vf Vf
where gd is the dissolution time coefficient.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
142 Compound and Hybrid Micromachining

The critical conditions are decided based on the thickness of material layer dissolved by the electrochemical reaction. This
amount is decided by the dissolution time td and the dissolution rate. Hence, in the next stage, the dissolution rate needs to be
determined.

11.06.6.4.4 Dissolution Rate

Low-resistivity deionized water could be considered a weak electrolyte due to its slight conductivity. When a voltage is applied
across two electrodes immersed in deionized water, a double layer is formed at the interface of the electrode and the electrolyte. This
electrode–solution interface is reported to behave as a two-parallel-plate capacitor, and the dissolution rate is exponentially
dependent on the polarization of this double layer (9,69). Thus, the interelectrode gap could also be modeled as shown in Figure 50
(70,71).

Figure 50 Model of electrode–workpiece gap in terms of circuit element.

Figure 51 Iterative algorithm to determine the critical feed rate.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining 143

Using the identical approach presented in the previous section, the average dissolution rate per second is given by the equation
ia M
n¼ [23]
zF

11.06.6.4.5 Feed Rate for Critical Conditions

Based on the critical thickness of material layer removed by material dissolution and the dissolution rate n, the critical dissolution
time td could be determined. Accordingly, the feed rate for critical conditions could be calculated by the equation derived from
eqns [20] and [22], that is,
Dgd
Vf ¼   [24]
D2 gd ðd  dd Þ
td þ
MRR s
It should be noted that the current density is a function of electrode–workpiece dg. Hence, to determine the dissolution time td
for a certain layer depth dd, the iteration algorithm is used to update the new dissolution rate after each time step Dt, as shown in
Figure 51.

11.06.6.5 Simulation and Experimental Verification

The primary machining conditions used for experimental tests are given in Table 4, whereas Table 5 summarizes the parameters
used for simulation works.
Figure 52 shows the SEM images of microslots machined for different layer depths and feed rate. It is observed that for
1 mm layer depth and 20 mm s1 feed rate, the machined area is fully covered with overlapping discharge craters, which
denotes that pure micro-EDM milling mode is obtained. When the feed rate is reduced or smaller layer depth is used,
a smooth zone appears on the machined area. This reveals that the electrochemical reaction has occurred to dissolve the
affected material layer on the micro-EDMed surface. In other words, there is the transition from micro-EDM milling to

Table 4 Machining conditions for verification of material removal mechanism

Electrode diameter (mm) 120

Deionized water resistivity (MU cm) 0.5
Electrode rotation speed (rpm) 500
Voltage (V) 60
Resistor (U) 220
Capacitor (pF) 265
Frequency (kHz) 500
Duty ratio (%) 35
Layer depth (mm) 0.2, 0.5, 1
Horizontal feed rate (mm s1) 1, 3, 5, 7, 10, 15, 20

Table 5 Simulation parameters for transitions of material removal mechanism

Initial gap distance, dinitial (mm) 5

Electrode diameter, D (mm) 120
Pulse amplitude, U (V) 60
Pulse frequency (kHz) 500
Pulse duty ratio 0.35
Maximum removal rate of sparks, MRRs (mm3 s1) 2364
Specific resistivity of solution, r (MU cm) 0.5
Double layer capacitance, CDL (mF cm2) 0.5
Transfer coefficient, a 0.5
Valency number of ions, z 2
Faraday constant, F (C mol–1) 96485
Gas constant, R (J mol–1 K1) 8.314
Temperature, T (K) 298.15
Exchange current density, io (mA cm2) 70
Molar volume, M (cm3 mol1) 7.11
Time step, Dt (s) 0.5
Dissolution depth threshold, dthres (mm) 0.1
Dissolution time coefficient, gd 0.9

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
144 Compound and Hybrid Micromachining

Figure 52 Microslots machined with different feed rates and layer depths.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining 145

SEDCM milling. It is also noted that the craters still exist at the two ends of the microslots, especially on the left area
where the electrode stops. This is due to the fact that there is less dissolution time for these two corners. Hence, when the
craters merely exist on the two semicircles at the two ends of the microslots, the SEDCM milling mode is considered to be
attained. Accordingly, from Figure 52, this hybrid machining mode is obtained when the feed rate is reduced to 15 mm s1
for a 0.2 mm layer depth. For a layer depth of 0.5 and 1 mm, it is observed to be 10 and 7 mm s1, respectively. It should
be highlighted that with a larger layer depth, the SEDCM milling mode is attained at a lower feed rate. This is in
accordance with the aforementioned simulated results because more material needs to be removed by the sparks when
feeding is given to the electrode. In addition, it is also seen that when the feed rate is reduced to 5, 1, 1 mm s1 for 1, 0.5,
0.2 mm layer depth, respectively, the discharge craters at the two ends of the microslots are found to vanish from machined
area, especially on the left side, which is the last position of the electrode. This is probably a consequence of long
dissolution time yielded by a low feed rate. This implies that the material removal mechanism is likely converted to pure
micro-ECM.
The experimental results of machining modes for different feed rates and layer depths are superimposed on simulated data, as
plotted in Figure 53. Comparison between the experimental and simulated results establishes that the SEDCM milling is only
attained at moderate feed rates. When the feed rate is too high, sparks occur most of the time and micro-EDM milling dominates.
On the contrary, when the feed rate is too low, machining mode is converted to pure micro-ECM milling.

11.06.6.6 Analysis of the Three Typical Material Removal Mechanisms

To further analyze the material removal mechanism of the three typical machining modes, further experimental tests were carried
out. The parameters representing three distinct machining modes were selected based on the simulated data and the experimental
results obtained in the previous section. For pure micro-ECM milling, the feed rate was set to be 1 mm s1 and the layer depth used
was 0.5 mm. For micro-EDM and SEDCM milling, the layer depth was fixed at 1 mm for both cases while the horizontal feed rate was
set to be 5 mm s1 for SEDCM milling and 20 mm s1 for micro-EDM milling.
Figure 54 exhibits the SEM images of obtained microslots with the superimposition of the electrode location. The electrode
moves horizontally from right to left, and it is purposely lifted up at the middle of the slot to examine the occurring material
removal phenomenon. For micro-EDM milling mode, it could be observed that the area of the electrode position is entirely covered
with discharge craters, as shown in Figure 54(a). In particular, the overlapping craters are found to be left over on the zone that the
electrode has scanned through.
In case of the SEDCM milling mode, it is observed that the discharge craters exist on a part of the area where the electrode
locates, as exhibited in Figure 54(b). However, it should be highlighted that the discharge craters have been swept away on the
machined area that the electrode scanned over. This provides the insight into the material removal mechanism of SEDCM
milling. Both electric discharge and electrochemical reaction occur during one feed, so discharge craters could be found on a part
of the electrode facing area. However, the machined surface is found to be smooth when the electrode has gone through, owing to
the effect of the electrochemical reaction. Figure 55 magnifies the machining area at the electrode location. The feeding direction
is right-to-left. Hence, the newly formed craters could be seen on the left area, whereas the material dissolution is found to be
occurring on the right side. It is also noted that the dissolution progresses from the outside toward the central area. This
phenomenon stems from the escape of gases generated during the sparks and the difficulty of deionized water in infiltrating the
fine gap.

Figure 53 Modes of machining obtained from experimental works.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
146 Compound and Hybrid Micromachining

Figure 54 SEM images of microslots machined for three typical machining modes: (a) micro-EDM milling, (b) SEDCM milling, and (c) micro-ECM
milling.

Figure 55 SEM image of area machining by SEDCM milling.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining 147

Figure 54(c) shows the machining zone in the case of the micro-ECM milling mode. It should be highlighted that no visible
crater could be traced on the machined zone as well as at the electrode position. This indicates that no spark occurs, and the material
is entirely removed by the electrochemical reaction for this case.
Figure 56 compares the topography of surfaces machined under three typical machining modes. The profiles of these
surfaces are also plotted in Figure 57. From these two figures, it is observed that the surface roughness is significantly
reduced when material removal mechanism is converted from micro-EDM milling to SEDCM milling (from 0.12 mm Ra to
23.6 nm Ra).

Figure 56 Topology of surfaces generated under the three typical machining modes: (a) micro-EDM milling, (b) SEDCM milling, and (c) micro-ECM
milling.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
148 Compound and Hybrid Micromachining

Figure 57 Profilographs of surfaces generated under three different machining modes: (a) micro-EDM milling, (b) SEDCM milling, and (c) micro-ECM
milling.

11.06.7 Summary

This chapter gives a comprehensive overview of hybrid EDM/ECM and presents an approach to combine micro-EDM and micro-
ECM as a unique hybrid machining process. The chapter starts with a brief overview of micro-EDM and ECM. Then, various
approaches to combine micro-EDM and ECM as sequential processes to improve the surface finish are presented. In addition, recent
advances in micro-EDM using deionized water are also discussed. Finally and especially, SEDCM drilling and milling are introduced
in detail. With the appropriate combination of micro-EDM and ECM in a unique machining process, it will be a potential method to
fabricate intricate microshapes for molds and dies, which entail both good surface finish and high dimensional accuracy.

References

1. Alting, L.; Kimura, F.; Hansen, H. N.; Bissacco, G. Micro Engineering. CIRP Ann. – Manuf. Technol. 2003, 52, 635–657.
2. Altan, T.; Lilly, B.; Yen, Y. C. Manufacturing of Dies and Molds. CIRP Ann. – Manuf. Technol. 2001, 50, 404–422.
3. Uriarte, L.; Herrero, A.; Ivanov, A.; Oosterling, H.; Staemmler, L.; Tang, P. T., et al. Comparison between Microfabrication Technologies for Metal Tooling. Proc. Inst. Mech. Eng.,
Part C 2006, 220, 1665–1676.
4. Reyntjens, S.; Puers, R. A Review of Focused Ion Beam Applications in Microsystem Technology. J. Micromech. Microeng. 2001, 11, 287.
5. Vieu, C.; Carcenac, F.; Pepin, A.; Chen, Y.; Mejias, M.; Lebib, A., et al. Electron Beam Lithography: Resolution Limits and Applications. Appl. Surf. Sci. 2000, 164, 111–117.
6. Lim, H. S.; Wong, Y. S.; Rahman, M.; Edwin Lee, M. K. A Study on the Machining of High-aspect Ratio Micro-structures Using Micro-EDM. J. Mater. Process. Technol. 2003,
140, 318–325.
7. Masuzawa, T. State of the Art of Micromachining. CIRP Ann. – Manuf. Technol. 2000, 49, 473–488.
8. Masaki, T.; Kawata, K.; Masuzawa, T. Micro Electro-discharge Machining and Its Applications. In An Investigation of Micro Structures, Sensors, Actuators, Machines and
Robots: IEEE; 1990.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Compound and Hybrid Micromachining 149

9. Schuster, R.; Kirchner, V.; Allongue, P.; Ertl, G. Electrochemical Micromachining. Science 2000, 289, 98–101.
10. Kunieda, M.; Lauwers, B.; Rajurkar, K. P.; Schumacher, B. M. Advancing EDM through Fundamental Insight into the Process. CIRP Ann. – Manuf. Technol. 2005, 54, 64–87.
11. Ho, K. H.; Newman, S. T. State of the Art Electrical Discharge Machining (EDM). Int. J. Mach. Tools Manuf. 2003, 43, 1287–1300.
12. Rajurkar, K. P.; Levy, G.; Malshe, A.; Sundaram, M. M.; McGeough, J.; Hu, X., et al. Micro and Nano Machining by Electro-physical and Chemical Processes. CIRP Ann. –
Manuf. Technol. 2006, 55, 643–666.
13. Uhlmann, E.; Piltz, S.; Doll, U. Machining of Micro/miniature Dies and Moulds by Electrical Discharge Machining – Recent Development. J. Mater. Process. Technol. 2005, 167,
488–493.
14. McGeough, J. A. Principles of Electrochemical Machining; Chapman and Hall: London, UK, 1974.
15. Kalpakjian, S. Manufacturing Processes for Engineering Materials, 3rd ed.; Addison Wesley: Reading, MA, 1997.
16. Bhattacharyya, B.; Munda, J.; Malapati, M. Advancement in Electrochemical Micro-machining. Int. J. Mach. Tools Manuf. 2004, 44, 1577–1589.
17. Kim, B. H.; Na, C. W.; Lee, Y. S.; Choi, D. K.; Chu, C. N. Micro Electrochemical Machining of 3D Micro Structure Using Dilute Sulfuric Acid. CIRP Ann. – Manuf. Technol. 2005,
54, 191–194.
18. Madore, C.; Piotrowski, O.; Landolt, D. Through-mask Electrochemical Micromachining of Titanium. J. Electrochem. Soc. 1999, 146, 2526.
19. Kern, P.; Veh, J.; Michler, J. New Developments in Through-mask Electrochemical Micromachining of Titanium. J. Micromech. Microeng. 2007, 17, 1168.
20. Kim, B. H.; Ryu, S. H.; Choi, D. K.; Chu, C. N. Micro Electrochemical Milling. J. Micromech. Microeng. 2005, 15, 124.
21. Pandey, P. C.; Jilani, S. T. Plasma Channel Growth and the Resolidified Layer in EDM. Precis. Eng. 1986, 8, 104–110.
22. Lee, L. C.; Lim, L. C.; Wong, Y. S.; Lu, H. H. Towards a Better Understanding of the Surface Features of Electro-discharge Machined Tool Steels. J. Mater. Process. Technol.
1990, 24, 513–523.
23. Ekmekci, B.; Sayar, A.; Öpöz, T. T.; Erden, A. Geometry and Surface Damage in Micro Electrical Discharge Machining of Micro-holes. J. Micromech. Microeng. 2009, 19,
105030.
24. Guu, Y. H.; Hocheng, H.; Chou, C. Y.; Deng, C. S. Effect of Electrical Discharge Machining on Surface Characteristics and Machining Damage of AISI D2 Tool Steel. Mater. Sci.
Eng. A 2003, 358, 37–43.
25. Kruth, J. P.; Stevens, L.; Froyen, L.; Lauwers, B. Study of the White Layer of a Surface Machined by Die-sinking Electro-discharge Machining. CIRP Ann. – Manuf. Technol.
1995, 44, 169–172.
26. Ekmekci, B. White Layer Composition, Heat Treatment, and Crack Formation in Electric Discharge Machining Process. Metall. Mater. Trans. B 2009, 40, 70–81.
27. Bleys, P.; Kruth, J. P.; Lauwers, B.; Schacht, B.; Balasubramanian, V.; Froyen, L., et al. Surface and Sub-surface Quality of Steel after EDM. Adv. Eng. Mater. 2006, 8, 15–25.
28. Lee, H.; Rehbach, W.; Tai, T.; Hsu, F. Surface Integrity in Micro-hole Drilling Using Micro-electro Discharge Machining. Mater. Trans. 2003, 44, 2718–2722.
29. Kurnia, W.; Tan, P. C.; Yeo, S. H.; Tan, Q. P. Surface Roughness Model for Micro Electrical Discharge Machining. Proc. Inst. Mech. Eng., Part B 2009, 223, 279–287.
30. Ekmekci, B.; Elkoca, O.; Erden, A. A Comparative Study on the Surface Integrity of Plastic Mold Steel due to Electric Discharge Machining. Metall. Mater. Trans. B 2005, 36,
117–124.
31. Yu, Z. Y.; Rajurkar, K. P.; Narasimhan, J. Effect of Machining Parameters on Machining Performance of Micro EDM and Surface Integrity. In ASPE Annual Meeting; Portland OR,
USA, 2003. http://aspe.net/technical-publications/proceedings/aspe-annual-meetings/.
32. Tsai, Y. Y.; Masuzawa, T. An Index to Evaluate the Wear Resistance of the Electrode in Micro-EDM. J. Mater. Process. Technol. 2004, 149, 304–309.
33. Mohri, N.; Suzuki, M.; Furuya, M.; Saito, N.; Kobayashi, A. Electrode Wear Process in Electrical Discharge Machinings. CIRP Ann. – Manuf. Technol. 1995, 44, 165–168.
34. Rajurkar, K. P.; Zhu, D.; Mc Geough, J. A.; Kozak, J.; De Silva, A. K. M. New Developments in Electro-chemical Machining. CIRP Ann. – Manuf. Technol. 1999, 48, 567–579.
35. Jeon, D. H.; Kim, B. H.; Chu, C. N. Micro Machining by EDM and ECM. Korean Soc. Precis. Eng. 2006, 23, 52.
36. Kozak, J.; Rajurkar, K. P.; Makkar, Y. Selected Problems of Micro-electrochemical Machining. J. Mater. Process. Technol. 2004, 149, 426–431.
37. Masuzawa, T.; Kuo, C. L.; Fujino, M. A Combined Electrical Machining Process for Micronozzle Fabrication. CIRP Ann. – Manuf. Technol. 1994, 43, 189–192.
38. Crichton, I. M.; McGeough, J. A.; Munro, W.; White, C. Comparative Studies of ECM, EDM and ECAM. Precis. Eng. 1981, 3, 155–160.
39. Masuzawa, T.; Sakai, S. Quick Finishing of WEDM Products by ECM Using a Mate-electrode. CIRP Ann. – Manuf. Technol. 1987, 36, 123–126.
40. Masuzawa, T.; Kimura, M. Electrochemical Surface Finishing of Tungsten Carbide Alloy. CIRP Ann. – Manuf. Technol. 1991, 40, 199–202.
41. Ramasawmy, H.; Blunt, L. 3D Surface Topography Assessment of the Effect of Different Electrolytes during Electrochemical Polishing of EDM Surfaces. Int. J. Mach. Tools
Manuf. 2002, 42, 567–574.
42. Masuzawa, T.; Fujino, M.; Kobayashi, K.; Suzuki, T.; Kinoshita, N. Wire Electro-discharge Grinding for Micro-machining. CIRP Ann. – Manuf. Technol. 1985, 34,
431–434.
43. Campana, S.; Miyazawa, S. Micro EDM and ECM in DI Water. In ASPE Annual Meeting; Monterey, CA, USA, 1999. http://aspe.net/technical-publications/proceedings/aspe-
annual-meetings/.
44. Chung, D. K.; Shin, H. S.; Kim, B. H.; Park, M. S.; Chu, C. N. Surface Finishing of Micro-EDM Holes Using Deionized Water. J. Micromech. Microeng. 2009, 19, 045025.
45. Hung, J. C.; Yan, B. H.; Liu, H. S.; Chow, H. M. Microhole Machining Using Micro-EDM Combined with Electropolishing. J. Micromech. Microeng. 2006, 16, 1480.
46. Zeng, Z.; Wang, Y.; Wang, Z.; Shan, D.; He, X. A Study of Micro-EDM and Micro-ECM Combined Milling for 3D Metallic Micro-structures. Precis. Eng. 2012, 36 (3), 500–509.
47. Takahata, K.; Aoki, S.; Sato, T. Fine Surface Finishing Method for 3-Dimensional Micro Structures. In An Investigation of Micro Structures, Sensors, Actuators, Machines and
Systems; San Diego, CA, 1996. http://ieeexplore.ieee.org/xpl/login.jsp?tp=%26arnumber=493832%26url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%
3Farnumber%3D493832.
48. Kurita, T.; Hattori, M. A Study of EDM and ECM/ECM-lapping Complex Machining Technology. Int. J. Mach. Tools Manuf. 2006, 46, 1804–1810.
49. Mohd Abbas, N.; Solomon, D. G.; Fuad Bahari, M. A Review on Current Research Trends in Electrical Discharge Machining (EDM). Int. J. Mach. Tools Manuf. 2007, 47,
1214–1228.
50. Jeswani, M. L. Electrical Discharge Machining in Distilled Water. Wear 1981, 72, 81–88.
51. Jilani, S. T.; Pandey, P. C. Experimental Investigations into the Performance of Water as Dielectric in EDM. Int. J. Mach. Tool Des. Res. 1984, 24, 31–43.
52. Konig, W.; Siebers, F. J. Influence of the Working Medium on the Removal Process in EDM Sinking. In ASME Winter Annual Meeting. New Orleans, LA, USA;
1993; pp 649–658.
53. Chen, S. L.; Yan, B. H.; Huang, F. Y. Influence of Kerosene and Distilled Water as Dielectrics on the Electric Discharge Machining Characteristics of Ti-6Al-4V. J. Mater. Process.
Technol. 1999, 87, 107–111.
54. Lin, C. T.; Chow, H. M.; Yang, L. D.; Chen, Y. F. Feasibility Study of Micro-slit EDM Machining Using Pure Water. Int. J. Adv. Manuf. Technol. 2007, 34, 104–110.
55. Kagaya, K.; Oishi, Y.; Yada, K. Micro-electrodischarge Machining Using Water as a Working Fluid–I: Micro-hole Drilling. Precis. Eng. 1986, 8, 157–162.
56. Masuzawa, T.; Tsukamoto, J.; Fujino, M. Drilling of Deep Microholes by EDM. CIRP Ann. – Manuf. Technol. 1989, 38, 195–198.
57. Chung, D. K.; Kim, B. H.; Chu, C. N. Micro Electrical Discharge Milling Using Deionized Water as a Dielectric Fluid. J. Micromech. Microeng. 2007, 17, 867.
58. Yan, M. T.; Lai, Y. P. Surface Quality Improvement of Wire-EDM Using a Fine-finish Power Supply. Int. J. Mach. Tools Manuf. 2007, 47, 1686–1694.
59. Yamada, H.; Magura, T.; Sato, K.; Yutomi, T.; Kobayashi, K. High Quality Electrical Discharge Machining Using Anti-electrolysis Power Source. EDM Technol. 1993, 1, 25–30.
60. Ukai, Y; Satou, S. Power Supply System for Applying a Voltage of Both Positive and Negative Polarities in Electric Discharge Machining. US Patent 6727455, 2004.
61. Song, K. Y.; Chung, D. K.; Park, M. S.; Chu, C. N. Micro Electrical Discharge Drilling of Tungsten Carbide Using Deionized Water. J. Micromech. Microeng. 2009, 19, 045006.
62. Song, K. Y.; Chung, D. K.; Park, M. S.; Chu, C. N. Micro Electrical Discharge Milling of WC-Co Using a Deionized Water Spray and a Bipolar Pulse. J. Micromech. Microeng.
2010, 20, 045022.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
150 Compound and Hybrid Micromachining

63. Masaki, T.; Kuriyagawa, T. Study of Precision Micro-Electro-Discharge Machining (3rd Report) – Analysis of Micro-EDM Process with Deionized Water. J. Jpn. Soc. Electr. Mach.
Eng. 2009, 43, 163–171.
64. Song, K. Y.; Chung, D. K.; Park, M. S.; Chu, C. N. Water Spray Electrical Discharge Drilling of WC-Co to Prevent Electrolytic Corrosion. Int. J. Precis. Eng. Manuf. 2012, 13,
1117–1123.
65. Chung, D. K.; Shin, H. S.; Park, M. S.; Chu, C. N. Machining Characteristics of Micro EDM in Water Using High Frequency Bipolar Pulse. Int. J. Precis. Eng. Manuf. 2011, 12,
195–201.
66. Nguyen, M. D.; Rahman, M.; Wong, Y. S. Simultaneous Micro-EDM and Micro-ECM in Low-resistivity Deionized Water. Int. J. Mach. Tools Manuf. 2012, 54–55, 55–65.
67. Kao, C. C.; Tao, J.; Shih, A. J. Near Dry Electrical Discharge Machining. Int. J. Mach. Tools Manuf. 2007, 47, 2273–2281.
68. Nguyen, M. D.; Rahman, M.; Wong, Y. S. Modeling of Radial Gap Formed by Material Dissolution in Simultaneous Micro-EDM and Micro-ECM Drilling Using Deionized Water.
Int. J. Mach. Tools Manuf. 2012.
69. Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications, 2nd ed.; John Wiley: New York, 2001.
70. De Abril, O.; Gündel, A.; Maroun, F.; Allongue, P.; Schuster, R. Single-step Electrochemical Nanolithography of Metal Thin Films by Localized Etching with an AFM Tip.
Nanotechnology 2008, 19, 325301.
71. Kozak, J.; Gulbinowicz, D.; Gulbinowicz, Z. The Mathematical Modeling and Computer Simulation of Pulse Electrochemical Micromachining. Eng. Lett. 2008, 16, 556–561.
72. Nguyen, M. D.; Rahman, M.; Wong, Y. S. Enhanced Surface Integrity and Dimensional Accuracy by Simultaneous Micro-ED/EC Milling. CIRP Ann. – Manuf. Technol. 2012, 61,
191–194.
73. Yu, Z. Y.; Masuzawa, T.; Fujino, M. Micro-EDM for Three-dimensional Cavities – Development of Uniform Wear Method. CIRP Ann. – Manuf. Technol. 1998, 47, 169–172.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 11.07 Micromilling
T Matsumura, Tokyo Denki University, Tokyo, Japan
 2014 Elsevier Ltd. All rights reserved.

11.07.1 Introduction 151

11.07.2 Microend Mill 152
11.07.3 Micromilling Operation 153
11.07.3.1 Cutter Runout 153
11.07.3.2 Dynamic Response 155
11.07.4 Cutting Process 158
11.07.4.1 Chip Formation and Minimum Chip Thickness 158
11.07.4.2 Force in Micromilling 158
11.07.4.3 Micromilling Process 160
11.07.4.3.1 Cutting Model 160
11.07.4.3.2 Intermittent Chip Formation 161
11.07.5 Material 163
11.07.5.1 Crystal Grain Sizes in Materials 163
11.07.5.2 Burr Formation 164
11.07.5.3 Milling of Brittle Materials 165
11.07.5.3.1 Milling Process of Brittle Materials 165
11.07.5.3.2 Effect of Cutter Axis Inclination on Cutting Process 169
11.07.5.3.3 Effect of Cutter Axis Inclination on Surface Finish 169
11.07.5.3.4 Applications 175
11.07.6 Conclusions 177
References 177

11.07.1 Introduction

Miniaturization in manufacturing has been growing with demand of microdevices in several fields that include optics, electronics,
medicine, biotechnology, and communications. Applications of the microdevices are microscale fuel cells, microfluidics systems,
microholes for fiber optics, micronozzles, micromolds, and many more.
Many processes have been applied to micromanufacturing so far. Wet etching, plasma etching, and lithography, electroplating,
and molding (LIGA) are the most popular processes in the last two to three decades as a result of progress in microelectromechanical
systems (MEMS). Recently, laser processing has also been applied to microfabrication. Although microscale parts are manufactured
relatively easily in these processes, there are some issues to be improved in terms of manufacturing performance, economics, and
environmental impacts. In addition, the processes have been limited to basically two or two and one-half dimensional
manufacturing. Therefore, alternative processes have been required for the manufacturing of three-dimensional (3D) parts.
Mechanical processes such as cutting and forming have been economically applied to manufacturing 3D shapes at high
production rates. Furthermore, they are flexible enough to manufacture diversified parts with numerical control systems. Recently
downsizing technologies of mechanical processes have developed with miniaturized tools and micro/nanomotion controls.
Therefore, these mechanical processes have been expected to expand the applications of the micromanufacturing.
In micromilling, some issues which have been ignored in macroscale cutting, have prevailed to control the cutting performance,
including the surface quality. When the milling tool is miniaturized, the material removal volume should be reduced to prevent tool
breakage and vibration. Then, the edge radius of the tool has an influence on the removal mechanism. When the uncut chip
thickness is of the same order or less than the edge radius of the tool, the effective rake angle becomes highly negative. This, in turn,
causes plowing and associated elastic-plastic deformation of the workpiece material, which become much more dominant factors in
the process. Then, a chip may not be formed during each tooth passing. In milling, where the uncut chip thickness varies during
a single engagement of a tooth in the cut, the cutting mechanism may change from plowing-dominated to shearing-dominated and
back to plowing-dominated again within a single excursion of a tooth through the cut.
In conventional removal processes of polycrystalline materials, workpieces have been regarded as isotropic materials. Meanwhile,
in microscale cuttings, the crystal grain sizes become relatively large compared to the removal volumes. The removal processes
depend on the sizes, the shapes, and the orientations of the crystal grains. Therefore, the effect of the crystal grains in materials should
be considered in micromilling. With progress in material developments, the sizes of crystal grains are also reduced to an order of
micrometer or submicrometer. The fine-grained materials could be widely applied to micromanufacturing for high qualities.
This chapter describes micromilling with some applications. The second section introduces the micromilling tools. The
manufacturing of the end mills requires not only sophisticated grinding technologies but also coating technologies. The advanced
technologies of the micromilling tools are shown. In the third section, micromilling is discussed from the point of the machining

Comprehensive Materials Processing, Volume 11 http://dx.doi.org/10.1016/B978-0-08-096532-1.01106-7 151

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
152 Micromilling

operation. Cutter runout is normally accompanied with clamping of the end mill onto the spindle-shank system. The stiffness of the
microtool is also a critical factor to determine the cutting parameters when taking the tool breakage and the machining accuracy into
account. The cutter runout and the stiffness of the microend mill are described to understand the dynamic behavior of the end mill
during rotations. The fourth section shows the cutting process modeling in the micromilling. First, the chip formation is described
to understand the microcutting then the cutting force is modeled for micromilling. In addition, a cutting model in milling is shown
to analyze the cutting thickness, which is associated with the cutting regime in microcutting. Then, the intermittent chip formation
in micromilling is described when introducing a model, which considers deflect of the tool and noncutting under the minimum
chip thickness. The fifth section describes the micromilling in terms of materials. Microcutting and milling of fine-grained steels,
burr removal in micromilling, and micromilling of brittle materials are described to improve the machining quality. In addition, in
micromilling of brittle material, the effect of the cutter axis inclination on the cutting process is discussed with simulation. The last
section concludes the issues of the micromilling for future manufacturing.

11.07.2 Microend Mill

Micromilling has progressed with miniaturizing the end mills. The micromilling requires machining accuracy in the order of
submicrometers and surface roughness in the order of nanometers. Therefore, in the manufacturing of the microend mills, the
accuracy of the edge shape and also the edge roughness should be considered in grinding, coating, and measurement.
Figure 1 shows examples of microsquare and ball end mills, where the diameters of these tools are 0.03, 0.015, and 0.1 mm. The
recent tool manufacturing technology has achieved reduction of end mill diameter, where it is much smaller than human hair.

Figure 1 Microend mill: (a) square end mill, diameter 0.03 mm; (b) square end mill, diameter 0.015 mm; and (c) ball end mill, diameter 0.1 mm.
Reproduced from Hitachi Tool Engineering, Ltd.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micromilling 153

Figure 2 Measurement of microball end mill. From Hitachi Tool Engineering, Ltd.

Because the miniature end mill is easy to be broken, the grinding force in the edge finishing should be controlled to keep small. The
edge sharpness is associated with chip formation in microcutting. Because the material removal volume reduces to the same size or
less than the edge radius, the actual rake angle becomes highly negative. Then, the material cannot be removed under a critical uncut
chip thickness; and plowing occurs with elastic deformation. Because the critical chip thickness is associated with the edge radius,
the grinding of the edge on the end mill has to be considered how sharp it can be finished.
Considering the machining accuracy in micromilling, the manufacturing of the microend mill also requires higher accuracy in
measurement than that of the conventional end mill. The edge shape is normally measured by the optical microscope. When radius
or ball end mill is manufactured, the shape of the end mill is evaluated at the rounded corner or the nose shape, which controls the
machining accuracy. Figure 2 shows the measurement points of the ball end mill in a measuring method, where symbols are an
example of measured radii. The center of the nose radius is determined with reference to the bottom of the tool. Then, the radii are
measured to check the errors within 0.005 mm in orientations from the center to the edge points.
The edge roughness also has an influence on the surface finish in micromilling. Especially in milling of brittle materials,
the edge roughness induces brittle fracture in micromilling. Furthermore, the surface finish depends on the edge roughness.
Coating materials and processing on the microend mill should be controlled to finish fine surfaces, because the uncut chip
thickness is nearly the same order as the coated grains in micromilling. The diamond coating was recently developed for
machining of nonmetallic materials. However, the coating grains are diversified as shown in Figure 3(a). In glass milling,
described in Section 11.07.5.3, brittle fracture cannot be controlled due to the diversified sizes in the grains, as shown in
Figure 3(b). Nanoscale coating grains have recently been developed for micromilling. Figure 4 compares the nanoscale coated
end mill with the conventional coated one. The grain sizes are much smaller than those of the conventional end mill.
Figure 5 shows a crack-free surface finished by the nanoscale coated end mill in glass milling. It is known that the wear
resistance and the delamination of the coated layers depend on the material properties and the grain sizes. The reduction of
coating grain sizes also has an effect on surface finish. The micromilling requires downsizing of not only the end mill size but
also the coating grain sizes.

11.07.3 Micromilling Operation

11.07.3.1 Cutter Runout
Cutter runout is usually accompanied with clamping of the tool, even though the precisely finished end mill is used for the
micromilling operation. The cutter runout in micromilling is relatively large compared to that of the conventional milling of normal
sized parts. The cutter runout is induced by eccentricity and axial inclination of the tool-spindle system. The cutter runout generally
includes shank runout due to alignment error and tool runout due to error in the tool manufacturing. Although detail analysis is
discussed in Refs. (1,2), a simple model is shown to understand the effect of the cutter runout here. At the height of the cutter, the
center of the tool (OT) is located at a distance of ε from the rotation center of the spindle (O), as shown in Figure 6. Then, an edge is
orientated at an angle x when the tool is clamped. Because OT rotates around O in the radius ε at a spindle speed, the rotating radii r1
and r2 of the edges are:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 9
r1 ¼ ε2 þ R2  2εR cosðp=2 þ xÞ ¼ ε2 þ R2 þ 2εR sinðxÞ =
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi [1]
r2 ¼ ε2 þ R2  2εR cosðp=2  xÞ ¼ ε2 þ R2  2εR sinðxÞ ;

where R is the radius of the tool. Therefore, the difference between r1 and r2 results in the cutter runout. When jr1r2j is larger than
the feed per tooth, an edge does not cut the material. Because the feed per tooth should be small to prevent tool breakage and
improve the machining accuracy in micromilling, the cutter runout has a relatively large influence on the chip formation.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
154 Micromilling

Figure 3 Glass milling with diamond coated end mill: (a) diamond coating and (b) surface finish. A microgroove is machined by a 0.4 mm diameter
ball end mill inclined at 45 in feed direction. Cutting conditions: spindle speed, 15 000 rpm; depth of cut, 0.02 mm; feed rate, 0.24 mm min1;
lubrication, water.

Figure 4 Comparison between nanocoated and conventional coated end mills: (a) conventional coated end mill and (b) nanocoated end mill.
Reproduced from Hitachi Tool Engineering, Ltd.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micromilling 155

Figure 5 Surface finish in milling with nanocoated ball end mill: cutting conditions are the same as those in Figure 3.

Figure 6 Cutter runout.

When milling is performed with the cutter runout, the different cutting forces are loaded on the edges with the cutting thick-
nesses. Because the tool wear also depends on the cutting force, different wear progress is observed on the edges.
In micromilling, the cutter runout should be reduced to an order of micrometer or submicrometer for high machining accuracy.
However, it is not so easy to measure the cutter runout directly during rotation. Because the cutter runout is associated with the cutting
force, the tool clamping may be adjusted with measuring the cutting force. Figure 7 shows the measurements of the cutting force
when the tool makes contact with the workpiece, where the cutting force is measured with a piezoelectric dynamometer. Figure 7(b)
shows an example of Z component of the cutting force of glass when the tool contacts the workpiece surface. Different peaks are
measured in the cutting force due to the cutter runout. Therefore, the tool clamping such as the tool orientation and the position may
be adjusted in the clamping device (collet chuck) so that the same maximum cutting force is loaded on each cutting edge.

11.07.3.2 Dynamic Response

In micromilling, the tool displacement induces not only the machining error but also the tool breakage because of small diameter of
the end mill. Therefore, dynamic behavior of the end mill should be evaluated with dynamic response of the tool, the workpiece,
and the machine tool. Generally, in dynamic analysis, the tool displacement x at the time t is given by:
d2 x dx
m þ c þ kx ¼ F [2]
dt 2 dt
where m, c, and k are the equivalent mass, the coefficient of viscous damping, and the stiffness, respectively. F is the cutting force at
the external force, which changes with the cutting thickness and the tool displacement. Although the dynamic response of the end
mill can be estimated numerically in the structure analysis such as in finite element method (FEM), the tool-spindle system, which
includes the end mill, the collet, and the spindle should be evaluated from a practical point of view. Normally the modal

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
156 Micromilling

Oscilloscope
Spindle
End mill

Workpiece plate
Z

Y X

(a) Piezoelectric dynamometer

Z component of cutting force N
2
Difference due to cutter runout
1.5
1
0.5
0
–0.5
–1
0 0.005 0.01 0.015 0.02
Time s
(b)

Figure 7 Adjustment of cutter runout: (a) measurement of cutting force and (b) difference of Z component in cutting force due to cutter runout.

parameters, m, c, and k, for large diameter end mill can be estimated in impulse response tests using the displacement sensor and the
special hummer, on which a force sensor is mounted. If the impulse response test is applied to estimation of the modal parameters
of the miniature end mill, the tool will be broken.
A method is shown to measure the dynamic response of the small diameter end mill, here (3). The dynamic response of the tool-
spindle system mounting the microend mill is measured with the vibration generator, as shown in Figure 8. The vibration tests are
performed to measure the exciting load with changing the frequency of the tool displacement. The end mill is mounted on the
spindle with the cutter axis inclination. The edge is clamped on the excited table made of tungsten carbide, which is small enough to
ignore elastic deformation of the table, as shown in Figure 8(b). The exciting load can be measured at the given displacement with
a piezoelectric dynamometer mounted under the excited table. The displacement of the table is controlled by the sinusoidal wave
generator. A displacement sensor measures the distance between the sensor head and the table with eddy current. The compliance
can be acquired with measuring the power spectrum of the load and that of the displacement on an Fast Fourier transform (FFT)
analyzer. The modal parameters in eqn [2], then, are estimated in the dynamic response.
Figure 9 shows an example of the compliances with the vibration frequencies. The frequency is changed from 0 to 1000 Hz in the
test. The figure shows the dynamic response around the natural frequency. The natural frequency is estimated as 335 Hz based on
the peak of the compliance. Because the vibration test is performed on the tool mounted on the spindle, the measured peak can be
regarded as one of the vibration modes in the tool-spindle system, which includes the tool, the collet, the spindle, and the spindle-
clamping device for mounting on the machining center. Then, the modal parameters for the radial direction of the tool are esti-
mated, as shown in Figure 10. Ff and df are the load and the displacement of the tool in the test with the inclined tool, respectively.
Because the radial stiffness is much larger than the axial stiffness in the tool-spindle system, it is assumed that the tool displacement
is mainly subjected in the radial direction of the tool. The load and the displacement in the vibration direction are transferred to the
components of the tool radial direction as follows:
8
< F ¼ Ff sin f
>
df [3]
>
:d ¼
sin f

where f is the inclination angle of the cutter. d and F are the displacement and the load in the radial direction of the tool,
respectively. Therefore, the compliance G in the tool radial direction can be given by:
Gf
G¼ [4]
sin2 f

Gf is the compliance in the vibration direction, which can be measured in the vibration test shown in Figure 8.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micromilling 157

FFT analyzer Power supply

Displacement
sensor

Amplifier

Charge amplifier Piezoelectric

Function generator dynamometer

Vibration generator
(a)

Spindle clamping device

Displacement sensor

Ball end mill

Piezoelectric dynamometer
(b)

Figure 8 Dynamic response test: (a) vibration test and (b) ball end mill on excited table.

0.003
Gain
0.002
Compliance mm N–1

Real
Imaginary
0.001

–0.001

–0.002

–0.003
250 300 350 400 450
Frequency Hz

Figure 9 Compliance. Inclination angle of the cutter axis, 30 ; the overhang of the end mill from the collet, 15 mm.

Figure 10 Vibration model.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
158 Micromilling

11.07.4 Cutting Process

11.07.4.1 Chip Formation and Minimum Chip Thickness
The principals of micromilling are similar to those of milling normal-sized parts. Chip formation is generally associated with the
uncut chip thickness in micromilling (4). When the uncut chip thickness, h, is less than a critical minimum chip thickness, hm, as
shown in Figure 11(a), elastic deformation occurs and the edge does not remove material. As the uncut chip thickness approaches
the minimum chip thickness, chips are formed in shearing. Some elastic deformation still occurs, as illustrated in Figure 11(b).
Therefore, the removal depth is less than the specified depth. When the uncut chip thickness increases beyond the minimum chip
thickness, the elastic deformation phenomena decreases significantly and the entire depth of cut is removed as a chip, as shown in
Figure 11(c).
In conventional cutting, the rake angle is defined as the inclination angle of the rake face. Chip formation is controlled by the
rake angle in a large uncut chip thickness. However, when the uncut chip thickness reduces to an order of micrometer, the actual
rake angle is controlled by the edge radius. As a result, cutting is performed at a large negative rake angle, which affects the
magnitude of the plowing and shearing forces. The minimum chip thickness is associated with the edge radius and the material.
Figure 12 shows the change in the chip load/cutting force with the uncut chip thickness (5). A sudden change in thrust forces occurs,
shifting from plowing/sliding dominant mode to shearing dominant mode. The minimum chip thickness of steel is estimated to be
0.2 and 0.3 times the edge radius for pearlite and ferrite, respectively (6).

11.07.4.2 Force in Micromilling

The uncut chip thickness changes with the cutter height in helical end mills during a rotation of the tool. Therefore, the cutting forces
are predicted with dividing the flutes into discrete small elements. The cutting force is induced by the plowing force around the edge

Tool

h Edge radius

Workpiece

(a) Elastic recovery

Removed
thickness
Tool
h
Workpiece

(b) Elastic recovery

Removed
thickness
Tool

Workpiece
(c)

Figure 11 Chip formation in microcutting: (a) uncut chip thickness h < minimum chip thickness hm; (b) hyhm ; and (c) h > hm. From Chae, J.; Park, S. S.;
Freiheit, T. Investigation of Micro-Cutting Operations. Int. J. Mach. Tools Manuf. 2006, 46, 313–332.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micromilling 159

Figure 12 Change in chip load/cutting force with cutting thickness. From Liu, X.; DeVor, R. E.; Kapoor, S. G.; Ehmann, K. F. The Mechanics of
Machining at the Microscale: Assessment of the Current State of the Science. Trans. ASME J. Manuf. Sci. Eng. 2004, 126, 666–678.

radius and the shearing force for chip formation. When an edge element locates at an instantaneous angle, q, the cutting force loaded
on the element, dF(q), is estimated as:
dFðqÞ ¼ ke dz þ kc hðqÞdz [5]
where kc and ke are the cutting force coefficient and the edge force coefficient, respectively (7). h(q) is the uncut chip thickness
associated with the instantaneous angle q, as described in the next section. The first term represents the plowing force, which
depends on the edge length in cutting. The second term is associated with the shearing force controlled by the cutting area. Figure 13
shows the geometrical removal area during a rotation of the cutter in milling, where the noncutting due to plowing is not
considered. When the edge engages the workpiece, h(q) is too small to form the chip. Because only plowing force is loaded on the
tool, eqn [5] may be expressed as:
dFðqÞ ¼ ke dz [6]

Workpiece

Current cutter path

Removal area
Feed per tooth

Uncut chip thickness

Feed
Cutting edge

Cutter path of previous tooth

Figure 13 Uncut chip thickness in milling.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
160 Micromilling

Non-cutting

theoretical
Cutting

Edge

Figure 14 Surface profile in micromilling. From Weule, H.; Huntrup, V.; Tritschle, H. Micro-Cutting of Steel to Meet New Requirements in
Miniaturization. Ann. CIRP 2001, 50, 61–64.

After rotating at a certain large angle, h(q) becomes large enough to form the chip and the cutting force is estimated in eqn [5].
However, the first term in eqn [5] is relatively large compared to the second term when the uncut chip thickness is small. In
micromilling, the feed rate is generally low to prevent the tool deflection and the tool breakage. Then, the uncut chip thickness is
small compared to that of the conventional milling of large-sized parts. Therefore, the plowing effect becomes large in estimation of
the cutting force. Considering the noncutting due to plowing in the uncut chip thickness of less than the minimum chip thickness,
a sawtoothlike surface profile may be finished, as shown in Figure 14 (8).

11.07.4.3 Micromilling Process

The cutting process in micromilling changes from plowing to cutting, increasing the uncut chip thickness. The uncut chip thickness
should be controlled by the cutting parameters and the tool geometry. When the tool rotates at an angular velocity of u, the uncut
chip thickness variation h(ut) in milling is analyzed for the cutting time t. Generally, the uncut chip thickness is approximated as:
hðutÞ ¼ fz sinðutÞ [7]
where fz is the feed per tooth. In order to discuss in general case, a mechanistic model is described to analyze the uncut chip thickness
here. Then, the intermittent chip formation in micromilling is discussed with the minimum chip thickness.

11.07.4.3.1 Cutting Model

Figure 15 illustrates the analysis model of milling. The end mill is fed in X direction at a feed rate of f. X–Y–Z is the reference system;
and X0 –Y0 –Z0 rotates with a cutting edge at an angular velocity, u, which is given by the spindle speed. At a time, t, the coordinates of
a cutting Point P on an edge in X0 –Y0 –Z0 are associated with those in X–Y–Z as follows:
8 0
>
> x ¼ ðx  ftÞsinðut  gÞ þ y cosðut  gÞ
<
y0 ¼ ðx  ftÞcosðut  gÞ þ y sinðut  gÞ [8]
>
>
: 0
z ¼z

where g is the delay angle with respect to the edge orientation at the bottom of the tool. Figure 15(b) shows the uncut chip thickness
at Point P on an Edge i with the trajectory of the edges. The uncut chip thickness is given by the segment QP, where Point Q on the
previous Edge j exists on the line from the center of rotation Point O0 to Point P. The coordinates of Point P are (R, 0, h) in the system
X0 –Y0 –Z0 of Edge i. The coordinates of Point Q in X–Y–Z are given as follows:
8      
< xQ ¼ f t þDt  ddij =u þ R sin
>

u t þ Dt  ddij =u  ddij  gQ

yQ ¼ R cos u t þ Dt  ddij =u  ddij  gQ [9]
>
:
zQ ¼ h

where ddij is the lead angle of Edge j with respect to Edge i; and gQ is the delay angle of Q on Edge j. Dt is the time parameter to be
used for determination of Point Q. The coordinates of Point Q ðx0Q ; yQ0 ; z0 Þ in X0 –Y0 –Z0 of Edge i are given by substituting (x , y , z )
Q Q Q Q
into eqn [8]:
8 0  
>
> x ¼ xQ  ft sinðut  gÞ þ yQ cosðut  gÞ
< Q  
0 ¼  x  ft cosðut  gÞ þ y sinðut  gÞ
yQ Q Q [10]
>
>
: z0 ¼ z
Q Q

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micromilling 161

(a)

(b)

Figure 15 Cutting model in milling: (a) end milling and (b) coordinate systems in model.

Because Point Q exists on the line O0 P, the coordinates of Point Q is determined to satisfy the following equations at the time
t þ Dt  ddij/u:
( 0
yQ ¼ 0
[11]
z0Q ¼ h

The uncut chip thickness h(ut) designated in Figure 15(b) is given by:
hðutÞ ¼ R  x0Q [12]

When h(ut) is negative, Point P is outside of the material.

Figure 16 shows an example of the simulated uncut chip thickness in slotting with a two flute square end mill. Because a chip
may not form in uncut chip thicknesses of less than the minimum chip thickness, the actual removal thickness is less than the
simulated thickness.

11.07.4.3.2 Intermittent Chip Formation ( 9)

As described in Section 11.07.4.1, if the feed per tooth is small compared to the cutting edge radius, a chip may not form with each
pass of the tool. As a result, intermittent chip formation occurs. The intermittency in the chip formation process is caused by the lack

6
Index 1
Uncut chip thickness (μm)

5 Index 2

0
0 0.001 0.002 0.003 0.004 0.005
Cutting time (s)

Figure 16 Simulation of uncut chip thickness. Cutting conditions: spindle speed, 32 000 rpm; feed rate, 260 mm min1; depth of cut, 0.02 mm.
Tool: type, square; diameter, 0.4 mm; helix angle, 30 ; flutes, 2.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
162 Micromilling

of material removal when the uncut chip thickness is smaller than the minimum chip thickness. Deflection of the end mill due to
small stiffness is another factor to control the chip formation in micromilling. Chip formation models for conventional-scale
milling have not taken the lack of material removal into account because the feed per tooth is usually much larger than the
minimum chip thickness. A model of chip formation in micromilling is discussed with accounting for the above effects of the
deflection of the end mill and the noncutting process. The model is a static model of a stable cutting process and, therefore, does not
consider the dynamic response of the tool and workpiece. For simplicity, an example of milling uses an end mill with a single flute
with no helix angle.
Figure 17 illustrates the model of milling with tool deflection due to the forces acting on the cutting tooth viewed from the top to
the bottom of the tool. The tool will deflect away from the workpiece due to the cutting forces, which can be decomposed into radial
thrust force FT and tangential cutting force FC. The figure describes the relative displacement of the cutting tooth with respect to the
center of rotation. Although in the actual micromilling, the milling tool deflects in both directions, the displacement in the
tangential direction is neglected. The tool deflection in the milling process can, therefore, be modeled as one-dimensional
displacement of a rigid cutting tooth with respect to a stationary point representing the center of rotation, utilizing a Hookean
spring with the spring constant equated to the milling tool stiffness in radial direction k.
If the machine tool system and the milling tool are completely rigid, the cutting tooth follows the paths denoted with the dashed
arcs in Figure 17 as the tool rotates. The bottom and top dashed arcs represent the path of the tool tip for the previous and current
tooth pass, respectively. At a specific position angle q0, the theoretical uncut chip thickness, h(q0), is given by the procedure in the
previous section, where q0 is associated with ut. In the real process, the tool usually deflects away from the workpiece in the radial

(a)

(b) –

(c)

Figure 17 Intermittent chip formation: (a) model in milling (top view); (b) change in force with uncut chip thickness; and (c) examples of periods
with position angle. From Kim, C. J; Mayor, J. R; Ni, J. A Static Model of Chip Formation in Microscale Milling. Trans. ASME J. Manuf. Sci. Eng. 2004, 126,
710–718.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micromilling 163

direction due to the cutting forces. No chip formation may also occur when the uncut chip thickness is smaller than the minimum
chip thickness. In nth tool pass, the ideal uncut chip thickness, hi(q0, n), is given by adding the noncutting thickness in the previous
tooth passes. Therefore, the actual uncut chip thickness, ha(q0, n), is given by the magnitude of the deflection of the tool in the radial
direction during the current tooth pass. For a given hi(q0, n), ha(q0, n) is achieved when the restoring force in the tool, due to
deflection, k{hi(q0, n)  ha(q0, n)}, is balanced with the radial thrust force FT:
 
FT ¼ k hi ðq0 ; nÞ  ha ðq0 ; nÞ [13]

The thrust force varies depending on material properties of the workpiece, tool geometry, the actual uncut chip thickness, and the
axial depth of the cut. The force is illustrated schematically in Figure 17(b). For a specified hi(q0, n), ha(q0, n) is given in eqn [13]
with obtaining FT in Figure 17(b).
After the milling process settles into a steady state, the magnitude of tool deflection will be constant and, thus, the actual uncut
chip thickness ha(q0, n) will be the same as the projected feed per tooth. When ha(q0, n) is greater than the minimum chip thickness
hm, the tooth pass produces a chip and the process continues in a steady cutting regime:
ha ðq0 ; nÞ ¼ hðq0 Þ hðq0 Þ > hm [14]
If h(q0) is smaller than hm, the formation of a chip at a specific position angle depends on whether ha(q0, n) is greater than hm or
not. Therefore, the ideal uncut chip thickness in the following tooth pass, hi(q0, nþ1), is given by adding the material as thick as the
amount of tool deflection:
)
hi ðq0 ; n þ 1Þ ¼ ½hi ðq0 ; nÞ  ha ðq0 ; nÞ þ hðq0 Þ
[15]
hðq0 Þ < hm ha ðq0 ; nÞ > hm

When no material is removed on the noncutting tooth passes, hi(q0, nþ1) on the following tooth pass is:
)
hi ðq0 ; n þ 1Þ ¼ hi ðq0 ; nÞ þ hðq0 Þ
[16]
hðq0 Þ < hm ha ðq0 ; nÞ < hm

This equation is valid for any noncutting tooth pass for which ha(q0, n) is smaller than hm. In the illustrative example presented in
Figure 17(b), the ideal uncut chip thickness for n ¼ 2, 3, and, 4 can be obtained using eqn [16]. On the cutting tooth pass n ¼ 4,
which is identical with the initial cutting tooth pass n ¼ 0, the material accumulated during previous three noncutting tooth passes
will be removed as a chip, and afterward this series of tooth passes – three noncutting and one cutting – will continuously repeat.
Consequently, the actual uncut chip thickness and the corresponding cutting and thrust forces vary with respect to the number of
tooth passes, with a period of 4 in this particular example. Figure 17(c) shows examples of variations of periodicity as a function of
position angle at feeds per tooth.

11.07.5 Material
11.07.5.1 Crystal Grain Sizes in Materials
The size of crystal grain is generally ranged from 100 nm to 100 mm in engineering materials. When the machining size is reduced to
an order of micrometer, the crystal grains have an influence on the cutting process. Polycrystalline materials have certain distri-
butions in the crystal grain sizes with random crystal orientations. Figure 18(a) shows microstructure of 0.45% carbon steel, which
is normally used in machine shops. The grain size is distributed from 10 to 20 mm. The shapes of the crystal grains are not uniform.
Then, anisotropy may appear in microcutting of the material. If the uncut chip thickness is less than the grain size, the cutting
process depends on each crystal grain to be cut and become unstable with vibration. Recently new materials have been developed to
reduce the crystal grain sizes with progress in material engineering. Figure 18(b) shows microstructure of fine-grained 0.45% carbon
steel. The grains are much smaller than those of the normal grain steel. Then, the workpiece to be cut may be regarded as a uniform
material.
In order to discuss the effect of the grain size, the microplanning process of fine-grained stainless steel is compared with
that of the conventional one (10). Figure 19 shows microstructures of the stainless steels observed by electron backscatter
diffraction (EBSD). Normal stainless steel consists of large crystal grains with different sizes. Meanwhile, crystal grains are
reduced in the fine-grained steel, which are manufactured by repeating plastic deformation and reverse phase transformation.
The average grain sizes are 9.1 mm in the normal-grain steel and 1.52 mm in the fine-grain steel, respectively. The planning
operations are conducted with the single point tools made of single crystal diamond, as shown in Figure 20(a). The tools are
controlled to remove the materials, where the cutting depth gradually increases to the specified cutting depth with the cutting
travel, as shown in Figure 20(b). Figure 21 compares the cutting forces of the normal and the fine-grained steels (because the
slope angle is larger than the clearance angle of the tool, the flank face contacts the workpiece during penetration of the edge.
Therefore, the vertical force before the time 0 is larger than the principal force. The cutting force in the steady process is
performed after time 0). The cutting force of the fine-grained steel is smaller than that of normal grain steel with the vibration
component. Figure 22 shows the shear angles of the normal and the fine-grained steels when changing the cutting depth. The
shear angles of the fine-grained steel are larger than those of the normal one. The reduction of the cutting force is owed to the

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
164 Micromilling

Figure 18 Crystal grains of carbon steels: (a) normal steel (conventional material) and (b) fine-grained steel.

increase of the shear angle. Figure 23 shows the surface finishes and the chip formations. Remarkable difference is observed
on the sides of the chips. The unstable chip formation occurs in cutting of the normal grain steel because the grain existence
and the grain boundaries have a large influence on the chip formation. Furthermore, large waviness is also observed at the
edges of the groove. In cutting of the fine-grained steel, waviness at the edge of the groove reduces with a stable chip
formation.
The same effect of reduction of the grain size is observed in micromilling. Figure 24 compares the microgrooves in milling
of the fine-grained steel with that of the normal grain steel. Because the end mills are fed from the bottom to the top in the
pictures, the cutting edges engage the workpiece at the left of the groove, and exit at the right hand. Remarkable difference of
the surface finishes is observed at the right edges of the grooves. Burr formation along the edge of the groove is reduced with
the crystal grain size.

11.07.5.2 Burr Formation

Burr formation usually occurs in metal cutting. The sizes of burrs are relatively large compared to the machining sizes in
micromilling. Burrs can be usually removed by the postprocess operations with specialized tools. However, burrs formed in
micromilling cannot be removed easily. Figure 25 shows a machined groove on an aluminum alloy plate in a width of 0.04 mm and
a depth of 1 mm. A large burr formation occurs at the edges of the grooves in milling of ductile materials such as aluminum alloy
and deteriorates the surface finish. Such burr cannot be removed by the deburring tools in the manual or the post-milling finishing.
The cutter traces are also observed on the bottom of the grooves due to the displacement of the tool and the clogging of the chips
adhered on the tool. Figure 26 shows an orthogonal pillar array on an aluminum plate, where 0.25 mm deep grooves are machined
with a 0.5 mm diameter square end mill and 0.25 mm  0.25 mm pillars are left in a pitch of 0.5 mm. Figure 26(b) shows the
profile of the pillars scanned along the center of the pillars. Burrs higher than 0.2 mm are formed upward on the pillars, and the
designed shape of the pillars cannot be observed.
The difficulties of burr removal in the micromilling are as follows:
1. Deburring area is relatively small as compared to the removal tool.
2. Because the strength of micropart is low, the part would be deformed or broken in deburring.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micromilling 165

5μm

5μm

Figure 19 Microstructures of stainless steels observed by EBSD: (a) normal steel (conventional material) and (b) fine-grained steel. RD and ND shows
the microstructure in the rolling and that of perpendicular direction, respectively.

Therefore, the effective deburring processes are required without losing the shapes of the microparts. Although many deburring
studies have been done in metal cutting, the additional processes are generally performed to remove burrs in machine shops.
Figure 27 shows the pillars finished in the postprocess, in which the tops of the pillars are finished in polishing and burrs at the
edges of the pillars are removed by the water jet, without slurry (11). No burr is observed on the pillars with keeping the rectangle
pillar shape, as the profile shown in Figure 27(b). Figure 27(c) shows the surface finish on the top of a pillar. The fine surface is
finished with a roughness of 0.17 mm as well as the accurate pillar shape.

11.07.5.3 Milling of Brittle Materials

11.07.5.3.1 Milling Process of Brittle Materials
One of the attractive applications in micromilling is machining of microdevices made of brittle materials because reduction of the
removal volume results in ductile mode cutting. In this part, glass milling is introduced as an example of milling of brittle
materials (12).
Figure 28 shows trajectories of the teeth in a cross-section of a rotating cutter. The shaded area shows the material to be removed
by a tooth. The cutting process starts at Point A in an uncut chip thickness of zero; the uncut chip thickness increases with the cutter
rotation in the up-cut process from Point A to Point B; the uncut chip thickness decreases with the cutter rotation in the down-cut
process from Point B to Point C; and the process finally ends at Point C in an uncut chip thickness of zero. Point A0 and Point C0 are
the engagement and the exit points of the following tooth. The surface of the grooves can be finished in the cutting process from
Point A to Point A0 and from Point C0 to Point C.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
166 Micromilling

60°

6°

30 μm
(a)

Tool

Thrust force

Principal force

Workpiece

Piezoelectric dynamometer
(b)

Figure 20 Microcutting experiment: (a) single crystal diamond tool and (b) tool path.

3
Principal
2.5
Thrust
Cutting force (N)

2
1.5
1
0.5
0
–0.5
–2 –1 0 1 2 3 4 5 6
(a) Time (s)
3
Principal
2.5
Cutting force (N)

Thrust
2
1.5
1
0.5
0
–0.5
–2 –1 0 1 2 3 4 5 6
(b)
Time (s)

Figure 21 Cutting forces of stainless steels: (a) normal steel (average grain size 9.1 mm) and (b) fine-grained steel (average grain size 1.52 mm).
Cutting conditions: depth of cut, 14 mm; cutting speed, 0.5 mm s1.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micromilling 167

40
35 Normal, average grain size 9.1mm
Fine-grained, average grain size 1.52mm

Shear angle (deg)

30
25
20
15
10
5
0
0 5 10 15
Cutting depth (μm)

Figure 22 Shear angles. Cutting conditions are the same as those in Figure 21.

Figure 23 Surface finishes and chip formations: (a) normal steel (average grain size 9.1 mm) and (b) fine-grained steel (average grain size 1.52 mm).
Cutting conditions are the same as those in Figure 21.

Glass can be removed without brittle fracture in an uncut chip thickness of less than a micrometer. Brittle fracture, therefore, does
not occur when the uncut chip thickness is less than a micrometer from Point A to Point A0 and from Point C0 to Point C. Even if
brittle fracture occurs over a critical uncut chip thickness in the process from Point A0 to Point C0 , the fractured area can be removed
during the subsequent cut. The cutting teeth cannot actually move the ideal trajectory given by the cutting conditions and the tool
geometry because the spindle and the end mill have runout and displaced during cutting. However, the change in the uncut chip
thickness follows the above manner with the cutter rotation. As long as the cutting area from Point A to Point A0 and from Point C0
to Point C can be removed in a ductile mode, a crack-free surface can be finished in the grooves. Because the above cutting process
does not depend on the axial depth of cut, micromilling can be applied to machine deep grooves with finishing crack-free surfaces of
the glass workpiece. The same discussion can also be applied to cutting with ball end mills, where the radius of the trajectory

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
168 Micromilling

Figure 24 Surface finishes in micromilling of stainless steels: (a) normal steel (average grain size 9.1 mm) and (b) fine-grained steel (average grain
size 1.52 mm). A microgroove is machined by a 0.8 mm diameter ball end mill inclined at 45 in feed direction. Cutting conditions: spindle speed, 500 rpm;
depth of cut, 0.042 mm; feed rate, 50 mm min1.

Figure 25 Burr formation in micromilling. Cutting conditions: spindle speed, 150 000 rpm; feed rate, 70 mm min1. Tool: type, square; material,
TiSiN coated carbide; diameter, 0.04 mm; flutes, 2; helix angle, 30 ; lubrication, oil.

changes with the cutter height. The crack-free surface can be finished in the groove by the ball end mill. Figure 29 shows machining
of orthogonal grooves with 0.4 mm and 0.5 mm diameter ball end mills made of tungsten carbide. The ball end mills are inclined to
finish fine surfaces at 45 in water, as shown in Figure 29(a). The horizontal and vertical grooves are machined in depths cut of
15 mm and 20 mm, respectively. Crack-free surfaces are finished without brittle fracture at the edges of the grooves, as shown in
Figure 29(b).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micromilling 169

400 μm 200 μ m
(a)

mm

0.20

–0.20
mm
(b) 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8

Figure 26 Burr formation of orthogonal micropillars array: (a) surface finish and (b) surface profile. Cutting conditions: spindle speed, 150 000 rpm;
feed rate, 40 mm min1; lubrication, oil. Tool: type, square; material, TiSiN coated carbide; diameter, 0.04 mm; flutes, 2; helix angle, 30 .

11.07.5.3.2 Effect of Cutter Axis Inclination on Cutting Process

The cutting area is evaluated with calculating the uncut chip thickness. Figure 30 shows simulation of the height in the cutting area
when cutting with a 2 flutes ball end mill during a rotation of the cutter. Because the cutting edges were divided into small segments
in the simulation, the cutting areas are designated by symbols. When the cutter axis is not inclined, each edge cuts the material
alternatively. Therefore, the tool always contacts the workpiece, as shown in Figure 30(a). Figure 30(b) shows the simulation in
cutting with the tool inclined at 45 . The noncutting time, during which neither of the teeth cut the workpiece, appears and the
actual cutting time of each tooth reduces during a rotation of the cutter. Figure 31 shows the cutting force of crown glass measured
by a piezoelectric dynamometer. The force components loaded on the tool are shown in Figure 31(a). The noncutting is observed in
the measured cutting force, as the simulated results shown in Figure 30(b).
The actual cutting time during a rotation of the cutter decreases when increasing the inclination angle of the tool, as shown in
Figure 32(a). Figure 32(b) shows the height of the cutting area removing the material with the inclination angle, where the symbols
show the minimum and the maximum height. The cutting area spreads when increasing the inclination angle. The cutting velocities
increase with the cutter heights corresponding to the rotation radii of the cutting area.
On the thermal effect, the time for the heat generation reduces with the actual cutting time. Because heat conduction requires
time to heat, the temperature rise is expected to be small even if the cutting velocities increase in the cutting area with the tool
inclination. On the other hand, the teeth are cooled well during the noncutting time by supplying water. Consequently, tool wear
can be suppressed at low cutting temperatures when cutting with the inclined tool.
Figure 33(a) shows the range of the cutting velocity in the cutting area. The cutting velocity increases with the inclination angle.
Then, the surface finish can be improved with increasing the cutting speed. Figure 33(b) illustrates the changing rates in the cutter
radius at a lower cutting area and that of a higher cutting area. The changing rate is smaller at a higher cutting area, and the range of
the cutting velocity in the cutting area reduces when increasing the inclination angle. Therefore, a uniform surface can be finished
due to little change of the cutting velocity when the inclination angle is large.

11.07.5.3.3 Effect of Cutter Axis Inclination on Surface Finish (12)

Generally, the edge roughness is not transferred directly onto the surface in metal cutting because of larger elastic deformation than
the edge roughness. However, when cutting brittle materials, the edge roughness has influence on the surface finish because the
material undergoes almost no elastic deformation.
A model is shown to associate the edge roughness with the surface finish in cutting with the ball end mill inclined at an angle, f,
and in the feed direction, as shown in Figure 34(a). Figure 34(b) shows the cutting process with the inclined ball end mill,
where the ball nose radius is r. The tool moves in the X direction. X0 –Y0 –Z0 is the coordinate system inclined with the tool rotation
axis. X0 –Y0 –Z0 rotates with the cutting edge at an angular velocity of u. Point P at a height hp on a tooth is given at the time t:
8
> x ¼ ft þ hp sin f þ Rp sinðut  gÞcos f
< p
yp ¼ Rp cosðut  gÞ [17]
>
:
zp ¼ Ad þ rð1  cos fÞ þ hp cos f  Rp sinðut  gÞsin f

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
170 Micromilling

(a)

mm
(b)
0

–0.10

–0.20

–0.25 mm
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8

(c)

0.5

–0.5
100 200 300 400 500 600

Figure 27 Deburring of orthogonal micropillars array: (a) surface finish; (b) surface profile; and (c) surface finish on top of a pillar.

where f and Ad are the feed rate and the axial depth of cut, respectively. g is the delay angle of Point P with respect to the tooth at the
bottom of the tool, which is associated with the helix angle, and the cutter height. Rp is the radius of rotation at Point P, which is
defined as a function of the cutter height hp:
 
Rp ¼ J hp [18]

Because the milling tool moves with the rotating teeth, the surface profile in the cross-section changes with the position in
the feed direction. The largest removed shape is projected on Y–Z plane, which is different from a surface profile in

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micromilling 171

Workpiece
Trajectory made by the following tooth

Trajectory made by a tooth

A' C'

Feed per tooth

A C

Feed

Cutting edge

Figure 28 Uncut chip thickness in milling.

Spindle head of
machining center Clamping device

Spindle

Pallet

Water pool
Workpiece
Clamping
plate (Glass)

Micro drive stage

in Z axis
(a)

(b)

Figure 29 Glass milling: (a) cutting operation; (b) surface finishes in microgrooves; and (c) depth of grooves. Cutting conditions: spindle speed,
20 000 rpm; feed rate, 0.48 mm min1; depth of cut, 20 mm and 15 mm; lubrication, water. Tool: type, ball; material, TiAlN-coated carbide; diameter,
0.4 mm and 0.5 mm; flutes, 2.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
172 Micromilling

0.15
Edge index 1
Edge index 2

Height (mm)
0.1

0.05

0
(a) 0 0.001 0.002 0.003
Time (s)
Height (mm) 0.15

0.1

0.05
Edge index 1
Edge index 2
0
(b) 0 0.001 0.002 0.003
Time (s)

Figure 30 Cutting area during a rotation of cutter: (a) inclination angle, 0 and (b) inclination angle, 45 . Cutting conditions: spindle speed, 20 000 rpm;
feed rate, 0.48 mm min1; depth of cut, 0.02. Tool: type, ball; diameter, 0.4 mm; flutes, 2; helix angle, 0 .

Z
ϕ = 45°
X

(a)

3
Y component
Z component
2
Force (N)

X component
–1
(b) 0 0.001 0.002 0.003
Time (s)

Figure 31 Cutting force in glass milling with cutter axis inclined at 45 : (a) cutter axis inclination and (b) cutting force. Cutting conditions: spindle
speed, 20 000 rpm; feed rate, 0.48 mm min1; depth of cut, 0.02 mm; lubrication, water. Tool: type, ball; material, TiAlN-coated carbide; diameter,
0.4 mm; flutes, 2.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micromilling 173

0.002

Actual cutting time (s)

0.0015

0.001

0.0005

(a) 0
0 10 20 30 40
Inclination angle (deg)

0.2

0.15
Height (mm)

0.1

0.05

(b) 0
0 10 20 30 40
Inclination angle (deg)

Figure 32 Effect of cutter axis inclination on cutting time and cutting area: (a) actual cutting time and (b) height of cutting area. Cutting conditions are
the same as those in Figure 30.

25
min–1)

20
Cutting velocity (m

15

10

0
0 10 20 30 40
(a)
Inclination angle (deg)

dR

⎡ dR ⎤ ⎡ ⎤
⎢ dh ⎥ > ⎢ dR ⎥ dh
⎣ ⎦ h = h 1 ⎣ dh ⎦ h = h 2
h2

h1 dh
(b) dR

Figure 33 Effect of cutter axis inclination on cutting speed: (a) cutting speed and (b) changing rate of radius in cutting area. Cutting conditions are the
same as those in Figure 30.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
174 Micromilling

(a)

(b)

Figure 34 Surface model in milling with inclined ball end mill: (a) cutting operation and (b) analytical model.

a cross-section. The surface profile in the up-cut milling process is given to maximize the radial position yp in the cutting area
at each axial position zp. The surface profile finished in the down-cut process is given to minimize yp, where yp is a negative
value. At a given zp, yp is determined with changing the height hp in eqn [17], where the radius of rotation Rp changes with hp
in eqn [18].
For example, the edge roughness is defined by the angular parameter qp, which is the wedge angle with respect to the tool axis on
the center of the ball nose curvature:
(       
Rp ¼ J qp ¼ r  A þ A sin 2pqp =lq sin qp
       [19]
hp ¼ G qp ¼ r  r  A þ A sin 2pqp =lq cos qp

where lq is the angular pitch of the notches. Figure 35 shows the edge shape with the notch type roughness, where A, lq, and r
are 1 mm, 1 , and 0.2 mm, respectively. The cutting edge consists of the convex and the concave shapes. Figure 36 shows the
effect of the cutter axis inclination on the surface profile. When the tool is not inclined, the surface finish is profiled corre-
sponding to the notches of edge roughness, as shown in Figure 36(a). Figure 36(b) shows the simulation in cutting with the
inclined tool at 45 . The notched profile disappears because the convex area on the cutting edge removes the material left by
the concave area.
Figure 37 compares the surface finishes of crown glass. When the tool is not inclined, the cutter traces due to edge profile are
observed in the feed direction, as shown in Figure 37(a). The surface is also deteriorated at the bottom of the groove because of low
cutting velocities around the center of the ball end mill. On the other hand, the surface finish is remarkably improved in cutting with
the inclined ball end mill, as shown in Figure 37(b). The cutting velocity is ranged from 17.8 to 23.7 m min1 in cutting with the
tool inclined at 45 according to Figure 33(a). The cutting velocity increases when reducing its range. Therefore, a better surface can
be finished than the without the cutter axis inclination.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micromilling 175

150

Height (μm)
100

50

0
0 50 100 150 200
Radius (μm)

Figure 35 A model of edge roughness.

–5
Depth (μm)

–10

–15

–20
–100 –50 0 50 100
(a) Radial position (μm)

–5
Depth (μm)

–10

–15

–20
–100 –50 0 50 100
(b) Radial position (μm)

Figure 36 Simulation of surface profile: (a) inclination angle, 0 and (b) inclination angle, 45 .

11.07.5.3.4 Applications (13)

Figure 38(a) shows applications of glass milling to microtesting devices. The cutter axis inclination is controlled so as to keep the
inclination angle at 45 with respect to the feed direction. The example is a multilayer microtesting device manufactured in
micromilling and optical bonding, where microholes connect the channels in the upper and lower plates. Micro total analysis
systems (TASs) such as this example have been used for testing small amounts of samples in microchannels and microreservoirs.
Generally, chemical etching has been applied to the manufacturing of glass chips. Conducted with mask printing in lithography and
chemical reaction in hydrofluoric acid, chemical etching requires a long time for the processes with hazardous operations. If the
micromilling were available in machining of the microchannels, NC operations would make the manufacturing process more
flexible without mask printing. The production time would also be reduced in the cutting operation. It would be effective when
a small number of the glass chips are manufactured as the prototypes used in the laboratories. Also, the cutting operation would not
require the special treatments in the waste fluid process.
The milling mechanism for brittle materials can be applied to not only glass but also tungsten carbide. Figure 39 shows
a machining example of tungsten carbide, which is usually used for the insert tool in the turning operation. A microgroove is

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
176 Micromilling

Figure 37 Surface finishes of glass milling: (a) inclination angle, 0 and (b) inclination angle, 45 . Cutting conditions are the same as those in Figure 30.
Tool: type, ball end mill; diameter, 0.4 mm; flutes, 2; helix angle, 30 .

Figure 38 An application of glass milling. Cutting conditions: Spindle speed, 20 000 rpm; depth of cut, 0.02 mm; feed rate, 0.48 mm min1; lubrication,
water. Tool: type, ball end mill; material, TiAlN-coated carbide; diameter, 0.4 mm; flutes, 2; helix angle, 30 ; cutter axis inclination, 45 in feed direction.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micromilling 177

Figure 39 An application of micromilling of tungsten carbide. Cutting conditions: spindle speed, 20 000 rpm; depth of cut, 0.02 mm; feed rate,
0.48 mm min1; lubrication, water. Tool: type, ball end mill; material, cBN; diameter, 0.4 mm; flutes, 2; helix angle, 0 ; cutter axis inclination, 45 in feed
direction.

machined by a 0.4 mm cBN ball end mill. Because the tool wear can be controlled by the cutter axis inclination, as discussed in
Section 11.07.5.3.2, a microgroove 80 mm long can be machined. It is an application of micromolding. The life of the mold is
a critical issue in a large number of productions. Furthermore, high strength materials have recently been developed to reducing the
material weight. Therefore, micromilling of hard materials is expected for the micromolding manufacturing.

11.07.6 Conclusions

Micromilling is discussed from the point of the tool manufacturing, the operation, the process, and the materials in this chapter. The
technical issues in micromilling are discussed with the recent progresses. Downsizing the machining scale requires high machining
accuracy with fine surface finishing. In micromilling, the following factors make the operations difficult:
1. high precision in the shapes of end mills when finishing the edges,
2. cutter runout depends on human factor,
3. low stiffness of small diameter end mills,
4. plowing force in small uncut chip thickness without material removal,
5. large effect of microstructure on cutting process in same orders of material removal volume,
6. difficulties of removing microscale burrs, and
7. requirement of hard material in micromilling.
There are so many issues to be discussed for the manufacturing of microparts with high quality in micromilling. Because the
micromilling operations require consideration in the above issues, the tool manufacturing technology, the machine tool technology
including control and sensors, the material development, and the skilled operations are required for implementation of micro-
milling. In micromilling, the material removal rate is also much lower than the conventional sized machining. Therefore, micro-
milling for large machining areas would take long production time. Improvement of material removal rate is also critical in the
practical micromilling.

References

1. Jun, M. B. G.; Liu, X.; Devor, R. E.; Kapoor, S. G. Investigation of the Dynamics of Microend Milling – Part I: Model Development. Trans. ASME J. Manuf. Sci. Eng. 2006, 128,
893–900.
2. Filiz, S.; Ozdoganlar, O. B. Microendmill Dynamics Including the Actual Fluted Geometry and Setup Errors – Part I: Model Development and Numerical Solution. Trans. ASME J.
Manuf. Sci. Eng. 2008, 130, 031119-1–031119-10.
3. Matsumura, T.; Miyahara, Y.; Ono, T. Dynamic Characteristics in the Cutting Operations with Small Diameter End Mills. J. Adv. Mech. Des. Syst. Manuf. 2008, 2 (4), 609–618.
4. Chae, J.; Park, S. S.; Freiheit, T. Investigation of Micro-Cutting Operations. Int. J. Mach. Tools Manuf. 2006, 46, 313–332.
5. Liu, X.; DeVor, R. E.; Kapoor, S. G.; Ehmann, K. F. The Mechanics of Machining at the Microscale: Assessment of the Current State of the Science. Trans. ASME J. Manuf. Sci.
Eng. 2004, 126, 666–678.
6. Vogler, M. P.; Devor, R. E.; Kapoor, S. G. On the Modeling and Analysis of Machining Performance in Micro Endmilling. Trans. ASME J. Manuf. Sci. Eng. 2004, 126, 685–705.
7. Perez, H.; Vizan, A.; Hernandez, J. C.; Guzman, M. Estimation of Cutting Forces in Micromilling through the Determination of Specific Cutting Pressures. J. Mater. Process.
Technol. 2007, 190, 18–22.
8. Weule, H.; Huntrup, V.; Tritschle, H. Micro-Cutting of Steel to Meet New Requirements in Miniaturization. Ann. CIRP 2001, 50, 61–64.
9. Kim, C. J.; Mayor, J. R.; Ni, J. A Static Model of Chip Formation in Microscale Milling. Trans. ASME J. Manuf. Sci. Eng. 2004, 126, 710–718.
10. Komatsu, T.; Matsumura, T.; Torizuka, S. Effect of Grain Size in Stainless Steel on Cutting Performance in Micro-Scale Cutting. Int. J. Automat. Technol. 2011, 5 (3), 334–341.
11. Matsumura, T.; Konno, T.; Tobe, S.; Komatsu, T. Deburring of Micro-Scale Structures Machined in Milling. In Proceedings of the ASME 2010 International Manufacturing
Science and Engineering Conference, Erie, Pennsylvania, 2010, MSEC2010-34149.
12. Matsumura, T.; Ono, T. Cutting Process of Glass with Inclined Ball End Mills. J. Mater. Process. Technol. 2008, 200, 356–363.
13. Matsumura, T.; Tamura, S. Micro Milling of Brittle Materials. In Proceedings of the 6th International Conference on Micromanufacturing ICOMM/2011, 2011; pp 633–638.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
This page intentionally left blank

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 11.08 Fundamentals and Modeling of Micro-End Milling Operation
M Arif, Swiss Federal Institute of Technology of Lausanne (EPFL), Lausanne, Switzerland
M Rahman, National University of Singapore, Singapore
Published by Elsevier Ltd.

11.08.1 Introduction 179

11.08.2 Size Effect in Micro-Milling 179
11.08.3 Minimum Chip Thickness 181
11.08.3.1 Minimum Energy Approach 182
11.08.3.2 Infinite Shear Strain Approach 184
11.08.4 Burr Formation 185
11.08.4.1 Modeling of Burr Height in Micro-End Milling 185
11.08.4.2 Geometry of Burr Formation 187
11.08.4.3 Mathematical Formulation 187
11.08.5 Mechanistic Modeling of Micro-Milling Forces 189
11.08.5.1 Shearing Dominant Regime 189
11.08.5.2 Plowing Dominant Regime 189
11.08.6 Laser-Assisted Micro-Milling 193
11.08.6.1 Thermal Modeling 193
11.08.7 Chatter Vibrations in Micro-Milling 195
11.08.7.1 Robust Chatter Stability 196
11.08.8 Concluding Remarks 198
References 198

11.08.1 Introduction

Everyone is well familiar with the phrase ‘the world is getting smaller.’ However, it is not just that the world is getting smaller;
practically speaking, everything that is used is getting smaller (1). Computers, gadgets, and medical devices continue to shrink in
size, and monitors and TV displays continue to get thinner.
Miniaturization has become an important measure of technological advancement in the modern-day world. Minimizing the size
of electronics and mechanical components is the key requirement to embed multiple operational features into a small device and/or
functional platform. Miniaturization is typically a bottom-up approach where the building blocks are miniaturized first to realize an
assembled miniaturized device.
Manufacturing of such small products forces a new challenge on the mold makers to manufacture small enough molds for the
production of tiny size parts. To produce miniaturized parts, it is also inevitable to develop manufacturing processes capable of
dealing with the challenge of creating micro features on them. This has triggered extensive research to size down the manufacturing
processes to microscale.
Machining technology is one of the most frequently applied manufacturing processes in the production industries. Milling is one
of the most versatile machining processes capable of producing three-dimensional features and cavities on prismatic parts. Micro-
end milling is the milling process in which the maximum undeformed chip thickness is less than 1 mm. Micro-end milling is
particularly significant to produce smart molds with micro features for plastic part manufacturing. The process of micro-end milling
has seen extensive research on various aspects ranging from meeting high surface finish requirements to the improvement of
productivity. The fundamental concepts in the micro-end milling process from both process mechanics and process capability
viewpoints are discussed here. The modeling of various machining characteristics is presented to support the theoretical
contribution.

11.08.2 Size Effect in Micro-Milling

In macro-scale machining, the cutting edge of the tool is assumed to be perfectly sharp as shown in Figure 1(a). In micromachining,
the undeformed chip thickness is comparable to the cutting edge radius of the tool and grain size of the material. In this way, the
cutting edge cannot be treated as perfectly sharp. Hence, in micromachining, a round cutting edge tends to remove a solid grain of
the workpiece material which consumes more energy (Figure 1(b)). Furthermore, in macro-scale machining, it is assumed that there
is no contact between the cutting tool and workpiece at the tool flank due to the sharp cutting edge theory. Any frictional force
existing at the interface is therefore neglected, being insignificant. Only two deformation zones are, therefore, considered in macro-
scale machining, namely primary and secondary deformation zones as shown in Figure 1(a). In micromachining, however, there
exists a friction zone at the tool flank due to elastic recovery of the newly machined surface of the workpiece at the tail end of the

Comprehensive Materials Processing, Volume 11 http://dx.doi.org/10.1016/B978-0-08-096532-1.01119-5 179

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
180 Fundamentals and Modeling of Micro-End Milling Operation

Figure 1 Schematic of the cutting edge in (a) conventional macroscale and (b) micro-scale cutting. Here, h ¼ undeformed chip thickness and a is
the effective rake angle. Reproduced from Aramcharoen, A.; Mativenga, P. T. Size Effect and Tool Geometry in Micromilling of Tool Steel. Precis.
Eng. 2009, 33, 402–407.

cutting edge. Hence, there exist effectively three deformation zones named primary, secondary, and tertiary deformation zones as
shown in Figure 1(b). The force required to overcome this deformation does not contribute to the chip removal and hence is
commonly named plowing force. Practically speaking, no tool has a perfectly sharp cutting edge and must experience a plowing force
in the real cutting process. But in macro-scale machining, the plowing force is proportionally very small compared to the forces
generated in the primary and secondary deformation zones due to the large values of the undeformed chip thickness. It is believed
that the plowing force remains constant for a given cutting tool. Therefore, if a cutting tool with similar edge radius is used for
removing micro-scale chip (micromachining), the same magnitude of the plowing force becomes proportionally considerable due
to the small overall cutting force and hence can no longer be neglected. Specific cutting energy is typically calculated by the ratio of
undeformed chip area and the resultant machining force acting in the cutting direction. As mentioned earlier, plowing force does
not contribute toward the chip removal and hence, an effectively high specific cutting energy is observed in the micro-scale
machining processes. Plowing force is difficult to measure directly. However, there is an indirect way to observe an increase of
specific cutting force with decrease in undeformed chip thickness (2).
Figure 2 shows the specific cutting force with the ratio of undeformed chip thickness to cutting edge radius for a feed direction.
Specific cutting force was calculated by dividing the feed force by the product of chip load and axial depth of cut. Higher specific
cutting force is observed at the lower ratio of undeformed chip thickness to cutting edge radius (at the lower chip load). Nonlinear
increase of specific cutting force becomes more evident when the feed per tooth is less than the cutting edge radius. The result
supports the size effect phenomenon. This means that when cutting edge radius is higher than the undeformed chip thickness,
plowing will be dominant causing the elastic deformation. The plowing effect plays critical role when the feed per tooth is low.
The ratio of undeformed chip thickness to edge radius also affects the surface roughness of the slot machined as shown in
Figure 3 in micro-milling of hardened steel. It follows that when the undeformed chip thickness is less than edge radius, the surface
roughness decreases with an increase of chip load. This is due to reducing plowing effect for higher chip loads. For second range,
when the undeformed chip thickness is higher than the edge radius, the surface roughness increases with the chip load which is
typical in the machining process. The optimal value of surface roughness occurs when the undeformed chip thickness is equal to the
cutting edge radius of the tool (2). The result of excessive plowing is the increased cutting forces and higher surface roughness and
burr formation (3).

Figure 2 Specific cutting force in feed direction. Reproduced from Aramcharoen, A.; Mativenga, P. T. Size Effect and Tool Geometry in
Micromilling of Tool Steel. Precis. Eng. 2009, 33, 402–407.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fundamentals and Modeling of Micro-End Milling Operation 181

Figure 3 Surface roughness in micro-milling of hardened steel. Reproduced from Aramcharoen, A.; Mativenga, P. T. Size Effect and Tool
Geometry in Micromilling of Tool Steel. Precis. Eng. 2009, 33, 402–407.

It is also believed that besides plowing force and edge-radius size, there is a third factor which also contributes to the size effect
phenomenon in micro-scale machining. It is believed that deformation of the workpiece material in the machining process is due to
the coalescence of microstructural dislocations or flaws. When machining at macroscale, the cutting zone encompasses a higher
number of the microstructural dislocations and shearing of the material is easier, consuming less energy. On the other hand, in
micromachining, the cutting zone encompasses fewer dislocations and hence the shear deformation consumes relatively high
energy resulting in size effect.

11.08.3 Minimum Chip Thickness

It is a well-established fact in machining that a chip will not form if the undeformed chip thickness is less than a critical value called
minimum chip thickness. This critical value of undeformed chip thickness is believed to be in the range of 5–38% depending on the
workpiece material (4,5). It is therefore believed that when the feed per tooth in milling is significantly less than the cutting edge
radius of the tool, a chip may not be formed in each tooth pass. Instead the cutting edge compresses the workpiece material beneath
it resulting in elastic deformation of the workpiece material. The elastic recovery takes place immediately behind the cutting edge as
the cutting edge advances further. As discussed in the preceding section, this elastic recovery creates a frictional contact between the
tool flank and the newly machined surface of the workpiece. An evidence of such friction is the tool flank wear observed in the
machining processes.
In micro-end milling when cutting a slot, the undeformed chip thickness is zero at the beginning of the cut due to up-milling
mechanism. As the cut progresses, the undeformed chip thickness increases but still no chip is formed and material is deformed only
elastically until the minimum chip thickness is reached (Figure 4(a)). When the undeformed chip thickness reaches the minimum chip
thickness, a combination of elastic and plastic deformation (shearing) occurs. A part of the undeformed chip thickness is removed
in the form of the chip coupled with a partial elastic deformation still occurring (Figure 4(b)). Hence, the removed material is less
than the desired value. Finally, as the cutting edge advances further and undeformed chip thickness is larger than the cutting edge
radius, material is removed and the chip is formed distinctly (Figure 4(c)) (2). In the second half of the slot, the milling mechanism

Figure 4 Chip formation relative to the minimum chip thickness in micro-scale machining. (a) h < hm; (b) hyhm ; and (c) h > hm. Reproduced
from Aramcharoen, A.; Mativenga, P. T. Size Effect and Tool Geometry in Micromilling of Tool Steel. Precis. Eng. 2009, 33, 402–407.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
182 Fundamentals and Modeling of Micro-End Milling Operation

switches from up milling to down milling and the undeformed chip thickness starts decreasing, and finally the cutting edge
disengages from the workpiece at the end of the cut. Toward the down-milled side of the slot, the minimum chip thickness effect is
less because the cutting edge is already engaged in the shearing of the workpiece before reaching the minimum chip thickness point.
Therefore, the plowing phenomenon is mitigated due to the typical cutting mechanics of the down-milling cut.
It is also believed that the size effect in micro-scale machining is also contributed by the size effect of material behavior at the
micron level due to specific shear strength increases greatly when the chip thickness is smaller than minimum chip thickness due to
the plowing phenomenon and the accumulation of the actual chip thickness (6). The microhardness and strength of the workpiece
material are modified locally due to the intense cutting pressure, high strains and strain rate, and temperature rise in the cutting zone
contributing to the observation of size effect in micro-scale machining.
Many approaches are used to model the minimum chip thickness in micromachining. Analytically minimum chip thickness has
been modeled based on the principle of minimum required energy and infinite shear strain method (7) and it is presented here.
A schematic of material flow around a tool with a finite edge radius is shown in Figure 5. A stagnant point is assumed on the tool,
below which the material flows downward without any chip formation. This phenomenon is called plowing, which is elastic–plastic
deformation without material removal. Above this stagnation point, the material flows up and forms as chips. The stagnation point,
which happens at a critical or stagnant angle, qm, with an effective negative rake angle (i.e., ae ¼ p/2  qm) determines the value of
the minimum uncut chip thickness (MUCT), namely:
hm ¼ re ð1  cos qm Þ [1]
where re is the edge radius.

11.08.3.1 Minimum Energy Approach

Identification of the MUCT is done by finding the stagnant point, A, in Figure 5 (7).
This can be achieved by modeling the minimum energy at the stagnant point since the stagnant point represents the minimum
power needed for the material removal (7). The total power is formulated based on the cutting forces and velocity and then taking
the derivative equal to zero to find the stagnant point. Two different material deformation mechanisms below and above the MUCT
(plowing and shearing, respectively) are considered in the formulation of the forces. According to (8), the total friction coefficient
(mT) is a combination of an interfacial or adhesion friction coefficient (ma) between the surfaces of the tool and the workpiece, and
the plowing friction coefficient (mp), which is caused by plastic deformation of the workpiece material.
Given the infinitesimal forces on the tool edge during the plowing process, the adhesion friction coefficient is considered
(between the tool and the workpiece). With a simple force model for material plowing below the stagnation point, which assumes
material sliding along the tool edge, the infinitesimal force in the cutting direction is obtained from Figure 6.
dFCp ¼ kare sin qdq þ ma kare cos q dq [2]

where k and ma are the cutting coefficient (in Pa) and adhesion friction coefficient during plowing, respectively, re is the edge radius,
and a is the width of the workpiece.
When the undeformed chip thickness is greater than the MUCT, the material is removed by a shearing. Assuming the
conventional shear cutting force model, the infinitesimal force in the cutting direction can be written as given by (9):

ass cosðbs  ae Þ
dFCs ¼
sin 4c cosð4c þ bs  ae Þ
[3]
are ss sin q cosðbs  ae Þ
dh ¼
sin 4c cosð4c þ bs  ae Þ
where ss is the shear strength, 4c is the shear angle, ae is the effective rake angle, and bs is the friction angle between the workpiece
and the rake face during shearing.

Figure 5 Machining using an edge radius tool. Reproduced from Malekian, M.; Mostofa, M. G.; Park, S. S.; Jun, M. B. G. Modeling of Minimum
UncutChip Thickness in Micro Machining of Aluminium. J. Mater. Process. Technol. 2012, 212, 553–559.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fundamentals and Modeling of Micro-End Milling Operation 183

Figure 6 Plowing forces on cutting edge. Reproduced from Malekian, M.; Mostofa, M. G.; Park, S. S.; Jun, M. B. G. Modeling of Minimum Uncut
Chip Thickness in Micro Machining of Aluminium. J. Mater. Process. Technol. 2012, 212, 553–559.
p
Based on the geometry, ae ¼ q  , the infinitesimal cutting force can be rewritten as:
2
are ss sin q sinðbs  qÞ
dFCs ¼ dq [4]
sin 4c sinð4c þ bs  qÞ

Thus, if the uncut chip thickness is greater than the minimum chip thickness, the total cutting force in the cutting direction is
obtained from:
Zqm Zqex
FC ¼ dFCp þ dFCs [5]
qst qm

where qm is the stagnant angle demonstrating the location of the stagnation point. Using eqns [2] and [4], FC can be written as:
Zqm Zqex
are ss sin q sinðbs  qÞ
FC ¼ kare ðsin q þ ma cos qÞdq þ dq [6]
sin 4c sinð4c þ bs  qÞ
qst qm

The total power needed for the machining operations is:

PT ¼ FC V [7]

According to the principle of minimum energy, the stagnation point happens at a location where the power needed for the
material removal is at its minimum value, i.e.,
dPT
¼0 [8]
dqm

Therefore,
Zqm Zqm
d d are ss sin q sinðbs  qÞ
kare ðsin q þ ma cos qÞdq  dq ¼ 0 [9]
dqm dqm sin 4c sinð4c þ bs  qÞ
qst qex

The shear angle is approximated as the theoretical value derived by (10):

p b  a e q  bs
4c ¼  s ¼ [10]
4 2 2

By substituting eqn [10] into eqn [9], the following equation is obtained:
Zqm Zqm
d d are ss sin q sinðbs  qÞ
kare ðsin q þ ma cos qÞdq þ dq ¼ 0 [11]
dqm dqm sin2 ð1=2Þðbs  qÞ
qst qex

This can be calculated as:

2ss
ðsin qm þ ma cos qm Þð1  cosðbs  qm ÞÞ þ sin qm sinðbs  qm Þ ¼ 0 [12a]
k

or,  
b  qm
sin s ½ð1 þ m cot qm Þtanðbs  qm =2Þ þ ð2ss =kÞ ¼ 0 [12b]
2 |fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}
TE

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
184 Fundamentals and Modeling of Micro-End Milling Operation

The second part (depicted as TE) of eqn [12b] is known as a transcendental equation and the explicit solution for qm does not
exist. Therefore solving the equation for qm, we obtain:
qm zbs [13]
This states that the stagnant angle is equal to the friction angle between the material and the rake face, regardless of the other
parameters involved in the process.
The MUCT model can thus be determined from the stagnant angle that is based on the friction angle with respect to the edge
radius. To investigate further, we employed the infinite shear strain approach to identify the MUCT.

11.08.3.2 Infinite Shear Strain Approach

Shear strain in metal cutting is significantly higher than those found from standard tensile tests and metal forming operations. Shear
strain can be determined from the change in geometry from undeformed chip to sheared chip. The shear strain is defined as the ratio
of deformation over the nominal distance between the deformed and undeformed planes (9) as:
cos ae
ε ¼ cot 4c þ tanð4c  ae Þ ¼ [14]
sin 4c cosð4c  ae Þ
As the undeformed chip thickness decreases relative to the cutting edge radius, the effective rake angle (ae) become highly
negative, resulting in substantial increase in shear strain according to eqn [14] where (Phi as in eqn [14]) 4c is the shear angle.
Further decrease of undeformed chip thickness continues to increase shear strain. At the stagnation point there is no chip formation,
and shear strain approaches to infinite, since the material is stagnant. The shear strain is very large just before the fracture at the
stagnant point. The geometric model of the shear strain does not consider fracture and assumes the shear strain increases to infinite.
In other words, when the denominator of eqn [14] approaches zero, the shear strain will be largest at the stagnant point which is
a reasonable assumption for machining with a rounded edge cutting tool.
By substituting the shear angle using eqn [10], the shear strain can be rewritten as:
cos ae
ε¼ [15]
sinððp=4Þ  ð1=2Þðbs  ae ÞÞcosððp=4Þ  ð1=2Þðbs  ae ÞÞ
By equaling the denominator of the equation to zero and solving the equation, the effective rake angle (ae) at the MUCT (hm)
becomes as:
p
ae ¼ bs  [16]
2
Therefore, the stagnant point occurs when the stagnant angle is at:
p
qm ¼  ae zbs [17]
2
The same conclusion is achieved using the infinite shear stain approach as with the minimum energy approach, where the
stagnant angle is approximately equal to the friction angle. Increasing the friction between the material and tool rake face increases
the stagnant angle, which results in an increasing of the MUCT based on eqn [1].
According to (11), a higher friction between the material and the tool rake face causes a downward flow of material or plowing;
whereas, at low frictions, the material flows up and forms a chip, as shown in Figure 7. This is in agreement with our findings based
on the two approaches, which demonstrate that increasing friction causes more material to be plowed under the tool and more
plowing occurs. However, it contradicts the derivation of (12), where the stagnant angle decreases with increasing the friction angle.
In the formulation of the stagnant angle, several assumptions have been made. The effects of material hardening, cutting
velocity, and change of temperature during the machining operations, which can change the friction coefficient, have not been
directly considered. We have used the theoretical shear angle proposed by Merchant (10) in the formulations, which assumes the
shear zone to be a thin plane. However, the shear zone in large edge radius cutting is not a thin shear zone, but rather a wide shear
zone with the formation of wedges. Furthermore in this study, we assumed that the MUCT and the stagnation point are

Figure 7 Effect of the friction coefficient on material flow: (a) low friction and (b) high friction. Reproduced from Kita, Y. I.; Mamoru, H.
Mechanism of Metal Removal by an Abrasive Tool. Wear 1978, 47, 185–193; Malekian, M.; Mostofa, M. G.; Park, S. S.; Jun, M. B. G. Modeling
of Minimum Uncut Chip Thickness in Micro Machining of Aluminium. J. Mater. Process. Technol. 2012, 212, 553–559.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fundamentals and Modeling of Micro-End Milling Operation 185

equivalent, but this needs further investigation. Obikawa and Ohbuchi (13) investigated large negative rake angle cutting and
found that the stagnation happens not at a point, but a region. Therefore, the MUCT can only be identified approximately from
the experimental cutting tests by identifying the friction angle. Neither the material pile up in front of the tool nor the elastic
recovery beneath the tool clearance face were considered which may also affect the MUCT. These can be included in more
elaborate models. In reality, the material pile up and elastic recovery can affect the MUCT and consequently influence cutting and
plowing forces as well.
In the above-mentioned details, the MUCT was modeled based on the minimum energy principle and the infinite shear strain
method to find the stagnation point with respect to the edge radius. Both proposed approaches resulted in the same conclusion: the
stagnant angle can be approximated as the friction angle; therefore, the MUCT increases when the friction angle increases, resulting
in more plowing.
The proposed models for the MUCT can be effectively employed in the prediction of micro-machining forces and in the selection
of the optimal machining parameters in different machining operations, such as micro-turning and micro-milling operations (7,13).

11.08.4 Burr Formation

Micro-milling finds extensive applications in the manufacturing of dies and molds used in the injection molding of micro-fluidic
devices, prototyping and manufacturing of fuel cells (micro-channels), and production of tubular parts in liquid filtration. Certain
applications in the fields of optics, electronics, medicine, biomedical devices, communications, and avionics require burr-free
components for easy assembly and functioning. However, it is observed that all the micro- as well as macro-machining
processes leave burrs on the machined parts. Though the burrs left on the smaller size parts are smaller, they are known to cause
similar difficulties in assembly of the micro-parts too. Further, the removal of micro-burrs on smaller size (of the order of a few
hundred micrometers) components is far more difficult than those of the corresponding burrs on macro parts (14).
Burrs are usually formed in micro-milling when the cutting edge exits the surface being cut. In micro-milling, there is consid-
erable plastic deformation around the cutting tool edge; this plastic deformation is resisted by the bulk material in front of the edge.
However, when the advancing plastic deformation reaches a free surface there is little resistance to the deformation and hence the
material gets pushed out resulting in a burr (15). Such burrs formed by the exiting cutting tool edge are called exit burrs. This can be
clearly visualized in an orthogonal cutting process (Figure 8(a)). In addition to the exit burrs there are also side burrs formed in
orthogonal cutting. These are also called Poisson burrs formed by side bulging deformation of the material (16).
Another type of burr, known as the top burr, is also formed on top edge of the sidewalls in micro-milling (Figure 8(b)). Here
although the cutting tool edge does not exit a surface, there is considerable burr formation as the chip flows and exits the surface –
this is also called the tear burr. In micro-milling a slot there are differences in the top burrs formed on the two sides of the walls. The
sidewall where the cutting edge enters into the cut (up milling) has smaller burrs than the other wall where the tool edge finishes the
cut (down milling). In the up milling side, the burr is a Poisson burr formed by side bulging action only. On the down milling side,
the top burr is formed by the action of the chip material tearing away as it flows as well as side bulging deformation; hence the down
milling burrs tend to be larger. We can also infer from this that the material tearing action due to flow has more influence than the
side bulging action in the burr formation (17).

11.08.4.1 Modeling of Burr Height in Micro-End Milling

Modeling of burr height is given in Refs (14,16). A flow chart of the theoretical model is shown in Figure 9. The minimum chip
thickness is computed from the chip thickness expressions given by Bao and Tansel (18) for instantaneous chip thickness in micro-
end milling operation based on the true trochoidal trajectory. Note that this chip thickness is precursor to the burr, and conse-
quently after this point there would be no more natural chip formation. The geometry of burr formation in micro-end milling is
shown in Figure 10. This geometry is used for evaluating the negative shear angle and another parameter of the process, w, and
eventually the burr height. The exit angle is determined from the continuity of work, i.e., the work done in chip formation equals the

Figure 8 (a) Exit and side burr formations are easier in the simple case of orthogonal cutting. (b) Exit and top burrs formed in micro-milling.
Reproduced from Saptaji, K.; Subbiah, S.; Dhupia, J. S. Effect of Side Edge Angle and Effective Rake Angle on Top Burrs in Micro-Milling. Precis.
Eng. 2012, 36, 444–450.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
186 Fundamentals and Modeling of Micro-End Milling Operation

Figure 9 Scheme of the proposed model to predict burr height. Reproduced from Lekkala, R.; Bajpai, V.; Singh, R. K.; Joshi, S. S. Characterization
and Modeling of Burr Formation in Micro-End Milling. Precis. Eng. 2011, 35, 625–637.

Figure 10 (a) Exit geometry in micro-end milling. (b) Geometry of burr initiation and formation. Reproduced from Ko, S. L.; Dornfeld,
D. A. Analysis of Fracture in Burr Formation at the Exit Stage of Metal Cutting. J. Mater. Process. Technol. 1996, 189–200.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fundamentals and Modeling of Micro-End Milling Operation 187

work done in burr formation at the transition point. This exit angle is finally used to determine the burr height. In the milling
process, exit angle corresponds to the tool rotation and has significant influence on burr height (19,20).
The burr formation model in micro-end milling is built upon the burr formation model for orthogonal machining using
continuity of work principle by (21,22). Following assumptions have been used in the model development:
l The model is based on the burr formation geometry and the effects of workpiece material properties have been assumed to be
constant throughout the process.
l The tool engagement and the onset of plastic hinging happen simultaneously in the exit region.
l The rotation angle at exit, q, corresponds to the tool rotation angle, where the transition from chip formation to burr formation
occurs. The work done in chip formation and burr formation is equal due to continuity.
l The burr formation in exit zone is modeled as an orthogonal process, consequently, the effect of plunge depth and cutter
geometry is ignored.

11.08.4.2 Geometry of Burr Formation

It is known that the usual positive shear angle transitions to a negative value, when the cutter is about to exit the workpiece as shown
in Figure 10(a) (14). There is a point where the natural chip formation stops and the subsurface plastic deformation begins. The
point at which this phenomenon occurs is termed as the transition point (21,22). The geometry of burr initiation and burr
formation is shown in Figure 10(b). The location of tool corresponding to burr initiation point is denoted by point A. It is char-
acterized by the initial negative shear angle, b0, defined by the angle between the cutting direction and the plane (negative shear
plane) connecting the cutting tool edge, A, to the plastic hinge, B, as shown in Figure 10(b). As the tool progresses further after the
burr initiation from point A to A0 , the negative shear plane also shifts from AB to A0 B. The new shear angle is depicted by b. The initial
negative shear plane determines the location of the plastic hinge point, B. It is located where the initial negative shear plane crosses
the end surface of workpiece and acts as a plastic hinge. It remains unchanged during the development of burr formation (21,22).

11.08.4.3 Mathematical Formulation

Assuming dq and db are very small angles, AA0 can be approximated Rdq and l0 db from Figure 10(a) and triangle AA0 B in
Figure 10(b), respectively (14):
AA 0 ¼ Rdq ¼ l0 db [18]
From triangle APB and A0 PB in Figure 10(b), PB can be expressed as
PB ¼ l0 sin b ¼ w tan b0 [19]
The burr height CD is given by (22),
hf ¼ ðt0 þ w tan b0 Þ sin j [20]
Hinge rotation, j, is equal to b  b0 (see Figure 10(b)),
j ¼ b  b0 [21]
At the exit, b ¼ p/2; hence eqn [20] can be rewritten as,
hf ¼ ðt0 þ w tan b0 Þcos b0 [22]
The chip thickness in micro-end milling operation can be expressed in terms of tool rotation angle (q), number of tooth (N),
cutter radius (R), and feed per tooth (tx) as (23),
N 2 1
tc ¼ tx sin q  t sin q cos q þ tx2 cos2 q [23]
2pR x 2R
where tx ¼ f/n  N; f ¼ feed rate, and n ¼ spindle speed.
Note that the exit burr is formed when q w 0. Chip thickness at exit, t0, can be estimated by substituting the value of q in eqn [23],
N 2 1
t0 ¼ tx q  t q þ tx2 [24]
2pR x 2R
From eqn [19], we get,
w tan b0
l0 ¼ [25]
sin b
Substituting eqn [25] in eqn [18] and integrating eqn [18] with respect to b and q,

Zp=2 Zq  
w tan b0 b
db ¼ Rdq  w tan b log tan 0 ¼ Rq [26]
sin b 2
b0 0

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
188 Fundamentals and Modeling of Micro-End Milling Operation

Rearranging eqn [26]

Rq
w¼ [27]
logðtan b0 =2Þtan b0

Substituting eqn [24] and eqn [27] in eqn [22],

 
N 2 1 Rq
hf ¼ tx q  tx q þ tx2  cos b0 [28]
2pR 2R logðtanb0 =2Þ

Equation [12] shows the final expression of burr height in micro-end milling based on the previous models given by (21–23).
Note that tool rotation angle, q, is the only unknown in eqn [28].
If a small advancement of the tool, corresponding to exit angle q is considered, at the instance of transition from the chip
formation to burr formation as shown in Figure 10, it can be assumed that the work done in cutting (chip formation) should be
equal to the work for burr formation by the continuity of the cutting force (22).
DWc ¼ DWb [29]

where DWc ¼ work done in chip formation and DWb ¼ work done in burr formation.
Total work done in chip formation or cutting could be approximated by the work done by the tangential force (Ft). The cutting
force is acting in tangential direction and the incremental tool rotation angle, dq, is very small. Consequently, incremental tool
distance moved is Rdq (see Figure 10(a)). The total work done in chip formation is,
Zq
DWc ¼ Ft R dq [30]
0

From mechanistic force models the tangential force is proportional to the uncut chip area,
Ft ¼ Kt tc b [31]

where Kt ¼ specific tangential cutting pressure, tc ¼ uncut chip thickness, and b ¼ plunge depth.
Note that Kt is very similar to specific cutting energy and it is known that the specific cutting energy (u) exhibits size effect at the
micro-scale cutting. The specific cutting energy increases with decrease in uncut chip thickness. Typically, specific cutting energy can
be expressed as: u f (tc)m; where m > 0. Similarly, Kt can also be expressed as:
Kt ¼ A0(tc)m for commercially pure aluminum, values of A0 and m are 183.07 and 0.885, respectively.
In the exit region, the total work done in cutting could be approximated by:
Zq
DWc ¼ Kt tc bR dq [32]
0

Substituting the value of tc from eqn [23] in eqn [32],

Zq  
N 2 1
DWc ¼ Kt tx sin q  tx sin q cos q þ tx2 cos2 q bR dq [33]
2pR 2R
0

 
bKt tx 4pR þ ptx q  Ntx sin2 q þ p cos qðtx sin q  4RÞ
DWc ¼ [34]
4p

The energy required in burr formation for orthogonal machining has been estimated by k0 and (22) as,
 
k0 se
dWb ¼ cos2 b0 þ tan b0 w dx [35]
2 4

where k0 is the yield shear strength, ae is the normal yield strength, and dx is the distance the tool has moved since the onset of
plastic hinging.
Extending this to micro-milling for the transition point in the exit region
Zq  
k0 2 se
DWb ¼ cos b0 þ tan b0 w Rdq [36]
2 4
0  
k0 2 se
DWb ¼ cos b0 þ tan b0 w Rq [37]
2 4

Using von Mises criterion for shear strength and substituting w from eqn [27],
 
R 2 q2 se se
DWb ¼  pffiffifficos2 b0 þ tan b0 [38]
logðtan b0 =2Þtan b0 2 3 4

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fundamentals and Modeling of Micro-End Milling Operation 189

Substituting the values of work done from chip formation and burr formation from eqns [34] and [38], respectively, in eqn [29]
an implicit equation with q and b0 can be formulated. Note that the value of initial negative shear angle, b0, has been experimentally
found to be close to 20 irrespective of the workpiece material and processing parameters (21,22).
   
bKt tx 4pR þ ptx q  Ntx sin2 q þ p cos qðtx sin q  4RÞ R2 q2 se se
þ pffiffiffi cos2 b0 þ tan b0 ¼ 0 [39]
4p logðtan b0 =2Þtan b0 2 3 4
The rotation angle, q, can be calculated by finding roots of the implicit equation given in eqn [39]. The burr height can be
predicted from the expression given in eqn [28]:
 
N 2 1 Rq
hf ¼ t x q  tx q þ tx2  cos b0 [40]
2pR 2R logðtan b0 =2Þ
The burr heights predicted by the model are found in good agreement with the experimental results.

11.08.5 Mechanistic Modeling of Micro-Milling Forces

A mechanistic force model is used to predict micro-end milling forces for both the shearing and plowing dominant cutting regimes.
The model assumes that there is a critical chip thickness (minimum chip thickness) that determines whether the cutting is
predominantly shearing or plowing (24). The approach described in (24) is presented here.

11.08.5.1 Shearing Dominant Regime

Beyond this critical chip thickness, the elasto-plastic effects are assumed to be negligible and the cutting regime is considered to be
shearing dominant. In this case, the cutting mechanism is assumed to be similar to the conventional cutting mechanism that
considers the shearing and edge coefficients.
In milling operations, when the uncut chip thickness is greater than the minimum chip thickness value (h > hc), the
tangential (dFts) and radial (dFrs) shearing cutting forces acting on a differential flute element with height dz can be modeled as
follows (24):
dFrs ¼ ½Krc hðqi ðzÞÞ þ Kre  dz
[41]
dFts ¼ ½Ktc hðqi ðzÞÞ þ Kte  dz
where Krc, Ktc, Kre, and Kte are the radial and tangential cutting and edge coefficients, respectively. The cutting coefficients represent
shearing of the workpiece, and the edge components represent friction between the tool and the workpiece. In order to come up
with the micro-cutting force model for the shearing dominant regime, the identification of cutting constants, which can be obtained
from experiments, is imperative.

11.08.5.2 Plowing Dominant Regime

The micro-milling undergoes both plowing and shearing when chip formation does not occur because when the uncut chip
thickness is less than the critical thickness; instead, there is plowing and partial elastic recovery of the material. When the chip
thickness increases due to the rotation, shearing occurs. The plowing forces are modeled as proportional to the volume of inter-
ference between the tool and the workpiece. The elastic recovery of the workpiece is experimentally obtained using scratch testing.
The following equation is formulated in the plowing dominant cutting regime (24):

ðK h þ Kte Þdz when h  hc ðshearingÞ
dFt ¼  tc  
K A þ Kte dz when h < hc plowing
 tp p [42]
ðK h þ Kre Þdz when h  hc ðshearingÞ
dFr ¼  rc  
Krp Ap þ Kre dz when h < hc plowing

where Krp and Ktp are plowing constants; h is the chip thickness, which is a function of elastic recovery, run-out, and dynamics; and
Ap is the interference volume. The friction forces are considered constant for different conditions and modeled with the edge
coefficients, which are obtained from the shearing dominant cutting regime.
A comprehensive chip thickness model developed (23) is used to compute the correct chip thickness, including the effects of
the trochoidal tool path, minimum chip thickness, elastic recovery, and tool vibrations. Figure 11 shows the surface generation
and chip thickness computation in the presence of elastic recovery, which is represented as the shaded region, for an arbitrary axial
slice. Points C and F represent the tool center and cutting edge locations, respectively. The superscript denotes the tooth pass
number, and the subscript represents the rotational angle. Point I is found at the intersection between the previously generated
surface from the previous tooth pass and the line connecting C and F for the current tooth pass. The chip thickness can be
formulated as:
j j j j1
h ¼ maxð0; kCi Fi k  kCi Ii kÞ [43]

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
190 Fundamentals and Modeling of Micro-End Milling Operation

Figure 11 Chip thickness model considering elastic recovery. Reproduced from Jun M. B. G.; Liu X.; de Vor R. E.; Kapoor, S. G. Investigation of
the Dynamics of Micro-End Milling, Part 1: Model Development. Trans. ASME J. Manuf. Sci. Eng. 2006, 128 (4), 893–900.

Figure 12 Plowing area and elastic recovery. Reproduced from Park, S. S.; Malekian, M. Mechanistic Modeling and Accurate Measurement of
Micro-End Milling Forces. CIRP Ann. – Manuf. Technol. 2009, 58, 49–52.

The interference area, Ap, can be obtained by examining the height of the elastic recovery her ¼ peh, where pe is the elastic recovery
rate; and, the effective rake angle ap ¼ cos1(1  h/re) is obtained where re is the edge radius of the tool, as shown in Figure 12. The
plowed area, Ap (shaded area of Figure 12), can be obtained as:
1   1
Ap z re2 ap þ g þ re ðl1  l2 Þ [44]
2 2
where
 
her  re ð1  cos gÞ
l1 z ;
sin g
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi   [45]
 
l2 z re2 þ l21 sin ap þ g þ j ; j ¼ tan1 rl1e

Depending on the critical chip thickness, hc, the computed forces can be summed up among all the engaged axial slices over all
the cutting flutes to obtain the total forces using eqn [42].
The tool–workpiece contact at the flank face is an important contributor of the overall cutting forces in micro-milling. Another
mechanistic model was presented by Kang et al. (25) to take into account flank friction. This model is presented here (24).
Figure 13 shows the difference between macro-cutting and micro-cutting. When the depth of cut is larger than the tool edge
radius, the effect of the tool edge radius can be ignored (24). However, in micro-cutting, this radius has an influence on the cutting
mechanism. In particular, in cases where elastic recovery occurs in the flank face of the workpiece, sliding due to the contact between
tool and workpiece and plowing due to the tool edge are regarded as major cutting mechanisms (26,27).
The tool–workpiece contact length in micro-cutting can be obtained by the relief angle and the springback of the material caused
by the elastic recovery taking place in the flank face of the workpiece. The tool–workpiece contact length Lf in the flank face can be
obtained with the following expression [46] (26).
S
Lf ¼ [46]
sin qf
Here, springback S is k1rtH/E, k1 is a constant, rt is tool edge radius, H and E are Vicker’s hardness and the material elastic
modulus, and qf is relief angle of tool, respectively. When only the shear plane and the contact friction of the flank face are

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fundamentals and Modeling of Micro-End Milling Operation 191

Figure 13 Difference of conventional (a) macro- and (b) micro cutting. Reproduced from Kang, I. S.; Kim, J. S.; Kim J. H.; Kang, M. C.;
Seo, C. Y. W. A Mechanistic Model of Cutting Force in the Micro End Milling Process. J. Mater. Process. Technol. 2007, 187–188, 250–255.

considered in the cutting force of micro-cutting, the force of the normal component acting on the shear plane can be expressed as
expressions [47] and [48].
 pffiffiffi
s= 3 bt0
Fs ¼ [47]
sin f

sbt0
Ns ¼ [48]
sin f
Here, an additional contact frictional force caused by sliding brought about by the tool–workpiece contact following the elastic
recovery in the workpiece flank face can be obtained from the material yield strength and the contact length (28). As for the tool–
workpiece contact frictional force in the flank face, Ffc and Fft, the horizontal and vertical components toward the cutting direction,
can be obtained by expressions [49] and [50].
CY
Ffc ¼ pffiffiffi Lf b [49]
3

Fft ¼ CYLf b [50]

The model of Tlusty and MacNeil obtained the cutting force by experimentally getting specific cutting forces, but this model is
heavily dependent upon the yield strength and flow stress related to the material property. In this study, a cutting force was
obtained through cutting experiments and constant C was determined to compensate for the value determined by the properties of
the material. Then they were multiplied by K and Y in expressions [47]–[50] to make up for the experimental and theoretical
values.
Accordingly, in the micro-cutting that takes into account the tool edge radius, the principle cutting force and thrust cutting force
can be represented as expressions [51] and [52]. They are expressed in the form in which the frictional force of contact on the flank
face is added to the existing two-dimensional cutting.
Fc ¼ Fs cos f þ Ns sin f þ Ffc [51]

Ft ¼ Fs sin f þ Ns cos f þ Fft [52]

The cutting force in micro-end milling is predicted by using principal cutting force Fc and thrust cutting force Ft that considered
the tool edge radius.
Figure 14 shows changes in the chip thickness in end milling. The depth of cut in two-dimensional cutting can be seen as chip
thickness in end milling. Also, chip thickness h can be expressed as the function of tool rotation angle q as in expression [53]
h ¼ ft sin qð ¼ t0 in expression ½2Þ [53]
Accordingly, when expression [53] is substituted for expressions [47] and [48], which represent the forces acting on the shear
plane in cutting expressions [51] and [52], the expressions for the principal cutting force and thrust cutting force can be expressed
again as follows:
sbft cos f Y
Fc ¼ pffiffiffi sin q þ sbft sin q þ pffiffiffi Lf b [54]
3 sin f 3

sbft sbft cos f

Ft ¼ pffiffiffi sin q þ sin q þ YLf b [55]
3 sin f
As shown in Figure 15, the chip thickness varies in end milling and therefore the sizes of the cutting force differ at each position
depending on tool rotation angle q. Therefore, expressions [54] and [55] can be presented as expressions [56] and [57], respectively.
Also dy, the cutting element depending on a Helix angle, can be represented as expression [58].
 
sft r cos f YLf r
dFc ¼ pffiffiffi sin q þ sft r sin q þ pffiffiffi dy [56]
3 sin f 3

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
192 Fundamentals and Modeling of Micro-End Milling Operation

Figure 14 End milling process. Reproduced from Kang, I. S.; Kim, J. S.; Kim J. H.; Kang, M. C.; Seo, C. Y. W. A Mechanistic Model of Cutting
Force in the Micro-End Milling Process. J. Mater. Process. Technol. 2007, 187–188, 250–255.

Figure 15 Contribution of an element of the current edge. Reproduced from Kang, I. S.; Kim, J. S.; Kim J. H.; Kang, M. C.; Seo, C. Y. W. A
Mechanistic Model of Cutting Force in the Micro End Milling Process. J. Mater. Process. Technol. 2007, 187–188, 250–255.

 
sft r cos f sft r
dFt ¼  pffiffiffi sin q sin q þ YLf r dy [57]
sin f 3

r
dy ¼ dq [58]
tan b
In Figure 15, dF can be divided into two components of feed and normal directions. Accordingly, the two cutting force
components in the feed (x) and normal (y) directions can be expressed as follows:
dFx ¼ dFc cos q  dFt sin q [59]

dFy ¼ dFc sin q  dFt cos q [60]

And, when expression [58] is substituted for expressions [56] and [57], dFc and dFt can be presented as follows:
 
sft r cos f sft r YL r
dFc ¼ pffiffiffi sin q þ sin q þ pffiffiffi f dq
3 sin f tan b tan b 3tan b
 
sft r sft r cos f YL r
dFt ¼  pffiffiffi sin q þ sin q þ f dq
3 tan b sin f tan b tan b

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fundamentals and Modeling of Micro-End Milling Operation 193

When the expressions obtained in the above are combined, the final cutting forces in the feed and normal directions can be
obtained as in expressions [61] and [62].
 
sft r cos f sft r YL r sft r sft r cos f YL r
dFx ¼  pffiffiffi sin q cos q  sin q cos q  pffiffiffi f cos q þ pffiffiffi sin2 q  sin2 q  f sin q dq
3 sin f tan b tan b 3 tan b 3 tan b sin f tan b tan b
[61]
 
sft r cos f sft r YLf r sft r cos f YL r
dFy ¼ pffiffiffi sin2 q þ sin2 q þ pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi sin q cos q  sin q cos q  f cos q dq [62]
3 sin f tan b tan b 3 tan b sin f tan b tan b

When expressions [61] and [62] are integrated with the tool start angle qs and tool end angle qe which can be determined
geometrically are set as constants of integration, the final cutting forces in the feed and normal directions can be obtained as follows:
  
Fx ¼ C1 sin2 qe  sin2 qs þ C2 ðsin 2qe  sin 2qs Þ  C4 ðsin qe  sin qs Þ þ C5 ðcos qe  cos qs Þ þ C3 ðqe  qs Þ [63]
  
Fy ¼ C3 sin2 qe  sin2 qs þ 0:5C1 ðsin 2qe  sin 2qs Þ  C5 ðsin qe  sin qs Þ  C4 ðcos qe  cos qs Þ  C1 ðqe  qs Þ [64]

Here,
sft r cos f sft r
C1 ¼  pffiffiffi  ;
2 3sin f tan b 2 tan b
sft r sft r cos f
C2 ¼  pffiffiffi þ ;
4 3 tan b 4 sin f tan b
[65]
sft r sft r cos q
C3 ¼ pffiffiffi  ;
2 3 tan b 2 sin f tan b
YL r pffiffiffi
C4 ¼ pffiffiffi f ; C 5 ¼ 3C 4
3 tan b
The prediction of shear angle 4 by Merchant’s equation can be obtained from expression [65] through the thin shear plane
model (10), and friction angle bf can be obtained from expression [66] in force equilibrium condition.
p a bf
f¼ þ  [66]
4 2 2
 
Ft þ Fc tan a
bf ¼ tan1 [67]
Fc þ Ft tan a
Unlike two-dimensional cutting, the depth of cut changes in end milling. Therefore, the cutting force also changes depending on
the tool rotation angle.

11.08.6 Laser-Assisted Micro-Milling

Micro-milling is increasingly being used to fabricate micro-molds for injection molding of plastic parts. Hardened steels and
ceramics are the preferred materials for micro-molds because of their ability to withstand high thermal and mechanical
cyclic loads (29).
However, the part feature accuracy and material removal rate in micro-milling of such difficult-to-machine materials are limited
by the machine-tool system stiffness (especially for small foot print machines) and low flexural stiffness and strength of the micro-
tools normally used. Also, rapid tool wear is a problem since it negatively impacts part feature accuracy and finish (29).
One approach to overcome these limitations is to use laser heating to induce localized thermal softening. By suitably controlling
the laser power, spot size, and speed, it is possible to produce a sufficiently large reduction in the strength of the work material, and
consequently, the cutting forces and tool/stage deflections (31). This approach is presented here.

11.08.6.1 Thermal Modeling

The temperature distribution in the workpiece surface directly in contact with the tool is analyzed via an experimentally calibrated
moving heat source model to estimate the thermal loads experienced by the tool in laser-assisted micro-milling (LAMM) and to aid
in the selection of suitable cutting tool coatings. A standard moving heat source model is used to calculate the temperature
distribution in the workpiece surface in contact with the tool (termed the material removal surface in this paper). Specifically, the
temperature distribution is determined using an analytical model of a moving point heat source (30,31) over a semi-infinite
medium as given by eqn [68]:
RR 0
Tðx; y; zÞ ¼ a qððx0 ; y0 Þ=2pKsÞeðU=2kÞðsðxx ÞÞ dx0 dy0 ;
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi [68]
s ¼ ðx  x0 Þ2 þ ðy  y0 Þ2 þ ðzÞ2

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
194 Fundamentals and Modeling of Micro-End Milling Operation

Where a is the absorptivity, q is the heat intensity (W m2), K is the thermal conductivity (W m1 K1), k is the thermal
diffusivity (m2 s1), and U is the velocity of the heat source (m s1). The thermal conductivity of the work material is modeled as
a temperature dependent function given by eqn [69] (32).
K ¼ 18:7 þ 0:0138  T [69]
The temperature dependence of thermal conductivity requires an iterative method to solve eqn [69].
The experiments were performed to assess LAMM. The laser beam is modeled as a 280 mm diameter Gaussian heat source and
the laser power is fixed at 18 W. The distance between the trailing edge of the laser spot and the leading edge of the cutting tool is
fixed at w60 mm. The temperature distribution below the workpiece surface calculated from the model assuming unit absorptivity
is shown in Figure 16. The scanned specimen is sectioned, polished, and etched to reveal the microstructure shown in Figure 17. It

Figure 16 Temperature distribution (in 1 C) in the X–Z plane (shown in Figure 18) due to a laser scan along the indicated direction (laser power:
18 W, spot size: 280 mm, scan speed: 100 mm min1, laser-tool distance: 100 mm, absorptivity: 1). Reproduced from Mukund Kumar, M.; Melkote,
S. N.; M’Saoubi, R. Wear Behavior of Coated Tools in Laser Assisted Micro-Milling of Hardened Steel. Wear 2012, 296, 510–518.

Figure 17 Micrograph of laser scanned surface showing the phase transformed region (white color) for the conditions given in Figure 16 (d is the
depth of phase transformed region). Reproduced from Mukund Kumar, M.; Melkote, S. N.; M’Saoubi, R. Wear Behavior of Coated Tools in Laser
Assisted Micro-Milling of Hardened Steel. Wear 2012, 296, 510–518.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fundamentals and Modeling of Micro-End Milling Operation 195

Figure 18 (a) Position of tool with respect to the laser beam (AB indicates the line along which the temperature is predicted). (b) Predicted temper-
ature variation in the workpiece along AB and EF (laser power: 18 W, spot size: 280 mm, scan speed: 100 mm min1, laser-tool distance: 100 mm).
Reproduced from Mukund Kumar, M.; Melkote, S. N.; M’Saoubi, R. Wear Behavior of Coated Tools in Laser Assisted Micro-Milling of Hardened
Steel. Wear 2012, 296, 510–518.

clearly shows the demarcation between the phase transformed or laser hardened region (white color) and the tempered region
(black color) created by the laser scan. The depth of the phase transformed region, d, is measured using an optical microscope. The
temperature at this depth is taken to be the Ac3 temperature of A2 tool steel, which is 793  C. The absorptivity coefficient in the
thermal model is calibrated by dividing the temperature at a given depth d computed from the thermal model by the nominal Ac3
temperature of A2 tool steel.
Using this approach, the absorptivity for the conditions given in Figure 16 is found to be 0.793. Note that it is possible to predict
the depth of the phase transformed region without measuring it each time, provided the temperature dependence of thermal
conductivity and specific heat, and the dependence of surface roughness on the absorptivity of the laser beam are known precisely
(32). However, in practice it is often difficult to obtain these values for numerical computations. Hence, the above model calibration
approach is used in this study (30). Once the calibrated absorptivity is known, the thermal model can be used to determine the
temperature variation in the curved material removal surface at the leading edge of the tool. The relative positions of the laser beam
and the tool used in the LAMM experiments are shown in Figure 3(a). Under these conditions, the temperature distribution is
determined at the front edge of the tool along the lines AB and EF in Figure 18(a). The temperature rise is found to vary between 300
and 450  C on the curved material removal surface, as seen in Figure 18(b). However, this temperature rise is only due to laser heating.
The heat generated due to cutting is not considered in this study. Thus, the temperature rise experienced by the tool will be higher than
shown in Figure 18. It is evident from Figure 18(b) that the temperature rise of w300–450  C in LAMM is significant when compared
to the estimated temperature rise of only 100–200  C in micro-milling of 1018 steel and Al6061-T6 without laser heating (33).
Consequently, the coated tools must be capable of withstanding such high temperatures in the LAMM process for them to be viable.

11.08.7 Chatter Vibrations in Micro-Milling

Due to the fragile nature of the miniature tools, even a minute vibration in micro-milling can lead to part failures. Similar to
macro operations, micro-milling processes also exhibit an unstable phenomenon, called regenerative chatter, due to the

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
196 Fundamentals and Modeling of Micro-End Milling Operation

undulations on a previously cut surface (34). Chatter results in a poor surface finish and reduces the longevity of the tool. Chatter
stability can be expressed by stability lobe diagrams, which plot the boundary that separates stable and unstable machining in the
form of the axial depth of cut limit versus spindle speed for a specific radial width of cut and workpiece/cutting tool combi-
nations (35).
The prediction and avoidance of chatter vibrations, based on the assumptions that the dynamics of the system and cutting
coefficients are time-invariant, have been widely studied. However, micro-milling operations require very high rotational speeds to
maintain productivity; and, micro-cutting behavior is different from that of macro operations, due to large negative rake angles and
size effects. At high rotational speeds, the dynamics of high-speed spindles vary due to centrifugal and gyroscopic effects that affect
chatter (34). When the chip thickness is less than the critical chip thickness, the chip does not form and the workpiece material is
instead plowed. Also, size effects, which increase the specific energy required due the decrease in scale, play an important role in
micro-milling. Since the dynamics and cutting coefficients are the main parameters affecting chatter stability, investigating the effect
of changing these parameters in chatter is imperative. Furthermore, elastic recovery of the workpiece generates a great deal of
friction, which results in an increase of process damping. The process damping results in increasing the critical depth of stable
cutting in milling operation especially at lower spindle speeds (34).

11.08.7.1 Robust Chatter Stability

Micro-milling is described as a two-degrees-of-freedom system, as shown in Figure 19. The equation of motion with considering the
effect of process damping can be described as (34):
Mx€x þ Cx x_ þ Kx x ¼ Fc;x þ Fpd;x ð_xÞ [70]

where Mx, Cx, and Kx are the effective mass, damping coefficient, and stiffness; Fc,x is the cutting force; and Fpd,x is the process
damping force, which is a function of velocity, in X direction (33). The same formulation can be applied to Y direction as well. Since
the depth of cut in micro-milling operation is very small, the effect of the helix angle and axial forces can be neglected.
The resultant cutting forces in chatter stability can be described as (35):
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
F c ðfÞ ¼ Ktc 1 þ Kr2 ahðfÞ ¼ Ks ahðfÞ [71]

where Ktc is the cutting coefficient in the tangential direction; Kr is the ratio of radial to tangential cutting coefficients; Ks is the
resultant cutting force coefficient; a is the depth of cut; and h is the regenerative chip thickness, which is a function of the immersion
angle, f. As it is depicted in Figure 19, the chip thickness, h, can be formulated as (35):
h ¼ c sin f  ðrðt  TÞ  rðtÞÞ [72]
where c is the feed rate, T is the tooth passing period, and r is the displacement in the radial direction (i.e., r ¼ X sinf þ Y cosf). The
static component of the chip thickness is dropped since it does not contribute to chatter.
The resultant process damping force can be obtained as (36):
Kpd
F pd ðfÞ ¼ r_ [73]
Ruspindle

where Kpd is the resultant process damping force coefficient, R is the tool radius, uspindle is the rotational speed, and r_ is the radial
velocity of the tool. Based on the authors’ previous work (36), Kpd is found to be approximately 1.46 by identifying the interference
volume between the tool and the workpiece for micro-milling of Al 7075.

Figure 19 Schematics of micro-milling and surface profile during cutting. Reproduced from Park, S. S.; Rahnama, R. Robust Chatter Stability in
Micro-Milling Operations. CIRP Ann. – Manuf. Technol. 2010, 59, 391–394.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fundamentals and Modeling of Micro-End Milling Operation 197

After transferring the cutting and process damping forces, substituting them in eqn [70] and transfer to the Laplace domain, the
following equations will be obtained
!
 2
  sT
 Kpd s
Mx s þ Cx s þ Kx X ¼  Ks a cosðf  qÞ 1  e  cosðf  bÞ  ðX sinðfÞ þ Y cosðfÞÞ [74]
Ruspindle

where q ¼ tan1(Kr), which is the angle between the resultant cutting force and the tangential cutting force and b is the angle
between radial and tangential components of F pd .
Two-degrees-of-freedom milling processes can be transformed into a pseudo single-degree-of-freedom problem by projection
onto one plane (37,38). Combining the above equations in X and Y directions and separating the terms corresponding to cutting
and process damping forces, the characteristic equation of the system can be derived as:
 
1 þ alim Fc Ks 1  esT þ Fpd Ruspindle
Kpd s
¼0

Zfex
N  
Fc ¼ Fxx cosðf  qÞsinðfÞ  Fyy sinðf  qÞcosðfÞ df [75]
2p
fst

¼ ux Fxx þ uy Fyy

Zfex
N  
Fpd ¼ Fxx cosðf  bÞsinðfÞ  Fyy sinðf  bÞcosðfÞ df
2p [76]
fst

¼ yx Fxx þ yy Fyy

where Fxx and Fyy are the direct transfer functions of the system in the X and Y directions, N is the number of flutes on the miniature tool
(N ¼ 2 for micro-end mills), and ux, uy, vx, and vy are the orientation factors. Resultant dynamics, Fc , and the process damping transfer
function, Fpd , are obtained by integrating over the starting immersion angle (fst) to the exit angle (fex) (38). Equation [75] is applicable
for either micro- or macro-milling cases. The plowing effects in micro-milling contribute to the changing cutting coefficients.
The critical depth of cut (alim) in eqn [75] determines the border of stability and instability. Depths of cut greater than the critical
value cause chatter; whereas, with depths of cut smaller than the critical value, the cutting operation is stable. The tool tip dynamics
are indirectly obtained using the receptance coupling method. To examine the stability of varying parameters, the robust stability
theorem is proposed.
In conventional chatter stability theories, the cutting parameters are considered to be constant. However, some parameters, such
as system dynamics and cutting coefficients, change during micro-milling operations. The robust chatter stability theorem, based on
the edge theorem and the zero exclusion principle, is utilized to find stability within the changing boundaries. The edge theorem is
an extension of Kharitonov’s robust theory that allows us to predict the stability of an uncertain time-delay system, whose
parameters vary within a certain range (39). The edge theorem states that a polynomial, P, which has variable coefficients, is robustly
stable, if and only if, the edges that correspond to each pair of extreme polynomial vertices, pi and pj, are stable (39,40). The edge
theorem guarantees the stability of the edges and within the boundary of the edges.
The polynomial is the characteristic equation (eqn [75]); and, the uncertain parameters are the natural frequency, un, and the
resultant cutting coefficient, Ks, which change within a specific range. The varying parameters have been identified from experi-
mental tests. According to the edge theorem, the polynomials that form the edges can be formulated as (40):

P ¼ fpðs; un ; xÞ : F ˛ ½Fmin ; Fmax ; Ks ˛ ½Ks ; Ks g [77]

Combining eqns [75] and [77], the polynomials can be described as:
   Kpd s  
~ xx þ uy F
~ s 1  esT ux F
pi ðsÞ ¼ 1 þ aK ~ yy þ ~ xx þ yy F
yx F ~ yy [78]
Ruspindle

where i ¼ 1 – 4 for two parameter variations, K s is either Ks,min or Ks,max, and Fxx and Fyy are either Fmin or Fmax in X and Y directions.
The above polynomials are used to form the edges (40):

Ek ¼ ð1  lÞpi ðsÞ þ lpj ðsÞ [79]

for all l ˛ ½1; 0, where 1  i, j  4, and Ek is representing each edge.

In order to find the stability, we employ the zero exclusion method, which is a graphical technique for improving the efficiency of
finding the stability in the edge theorem (39). In each frequency, the characteristic equation forms a family of four different edges
with the extreme boundaries of variations. The edges shape a trapezium, as depicted in Figure 20. The trapezium must be inves-
tigated to determine if it encircles the origin or not: if the edges encircle the origin in a complex plane, the system is unstable;
otherwise, the system is stable (38). In order to automate the stability detection process, we find the angles that result from con-
necting the origin to each of vertices of the trapezium (p1–p4). Connecting the origin to the corners will make ai angles (i ¼ 1 – 4)
between the lines and the positive real axis, as shown in Figure 20. gi (i ¼ 1–4), which is the angle between each two subsequent

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
198 Fundamentals and Modeling of Micro-End Milling Operation

Figure 20 Zero exclusion method (angles in a trapezium). Reproduced from Park, S. S.; Rahnama, R. Robust Chatter Stability in Micro-Milling
Operations. CIRP Ann. – Manuf. Technol. 2010, 59, 391–394.

corners, is defined based on ai. gi ¼ aiþ1  ai for i ¼ 1 – 3 and g4 ¼ a1  a4. The gi angle must be a positive angle. If all the angles gi
are less than 180 , we can deduce that the edges encircle the origin. However, if one of the angles g1 – g4 will be greater than 180 ,
the origin is not encircled by the edges (34).
To find the stability lobes for micro-milling operations, the algorithm sweeps the depths of cut and chatter frequencies at each
spindle speed; and, it checks the stability through the proposed automated zero exclusion method. The first set of unstable
conditions is recorded as the border between the stable and unstable regions, in order to determine the stability lobes (34).

11.08.8 Concluding Remarks

Micro-end milling is a versatile machining process capable of removing material with undeformed chip thickness in the range of
1–999 mm. However, micro-end milling is not simply the scaled-down version of conventional end-milling. There are certain
challenges encountered in micro-end milling that are not associated with conventional end-milling. These challenges originate from
the difference in material response at micro- and macro scales. The material removal at the microscale is characterized by higher
specific cutting energy, burr formation, plowing effect and higher cutting forces. These characteristics must be considered in
modeling of the cutting forces in micro-end milling. Due to the micro-sized tool, the tool failure can be caused by even minute
vibrations. Hence, certain strategies for vibration and chatter suppression must be employed for successful application of micro-end
milling. The material removal rate in micro-end milling of hardened materials can be increased significantly by laser heating.

References

1. Capitalizing on the Growing Demand for Micro-Milling: A Mold Maker’s Guide, 2013. Retrieved from http://toolingtimes.com/MicroMillingGuide.pdf (accessed June 13, 2013).
2. Aramcharoen, A.; Mativenga, P. T. Size Effect and Tool Geometry in Micromilling of Tool Steel. Precis. Eng. 2009, 33, 402–407.
3. Ducobu, F.; Filippi, E.; Rivière-Lorphèvre, E. Chip Formation and Minimum Chip Thickness in Micro-Milling. In Proceedings of the CIRP Conference on Modelling of Machining
Operations, 12; 2009; pp 339–346.
4. Chae, J.; Park, S.; Freiheit, T. Investigation of Micro-Cutting Operations. IJMTM 2006, 45, 313–332.
5. Ohbuchi, Y.; Obikawa, T. Finite Element Modeling of Chip Formation in the Domain of Negative Rake Angle Cutting. ASME 2003, 125, 324–332.
6. Laia, X.; Lia, H.; Lia, C.; Lina, Z.; Ni, J. Modelling and Analysis of Micro Scale Milling Considering Size Effect, Micro Cutter Edge Radius and Minimum Chip Thickness. Int. J.
Mach. Tool Manuf. 2008, 48, 1–14.
7. Malekian, M.; Mostofa, M. G.; Park, S. S.; Jun, M. B. G. Modeling of Minimum Uncut Chip Thickness in Micro Machining of Aluminium. J. Mater. Process. Technol. 2012, 212, 553–559.
8. Sin, H.; Saka, N.; Suh, N. P. Abrasive Wear Mechanics and Grit Size Effect. Wear 1979, 55, 163–190.
9. Altintas, Y. Manufacturing Automation: Metal Cutting Mechanics, Machine Tool Vibrations and CNC Design; Cambridge University Press, 2000.
10. Merchant, M. E. Basic Mechanics of the Metal Cutting Process. J. Appl. Mech. 1944, 11, 168–175.
11. Kita, Y. I.; Mamoru, H. Mechanism of Metal Removal by an Abrasive Tool. Wear 1978, 47, 185–193.
12. Son, S. M.; Lim, H. S.; Ahn, J. H. Effects of the Friction Coefficient on the Minimum Cutting Thickness in Micro Cutting. J. Mach. Tool Manuf. 2005, 45, 529–535.
13. Malekian, M.; Park, S. S.; Jun, M. B. G. Modeling of Dynamic Micro Milling Cutting Forces. Int. J. Mach. Tool Manuf. 2009, 49, 586–598.
14. Lekkala, R.; Bajpai, V.; Singh, R. K.; Joshi, S. S. Characterization and Modeling of Burr Formation in Micro-End Milling. Precis. Eng. 2011, 35, 625–637.
15. Saptaji, K.; Subbiah, S.; Dhupia, J. S. Effect of Side Edge Angle and Effective Rake Angle on Top Burrs in Micro-Milling. Precis. Eng. 2012, 36, 444–450.
16. Gillespie, L. K. Deburring and Edge Finishing Handbook; Society of Manufacturing Engineers: Dearborn, MI, 1999.
17. Lee, K.; Dornfeld, D. A. Micro-Burr Formation and Minimization through Process Control. Precis. Eng. 2005, 29, 246–252.
18. Bao, W. Y.; Tansel, I. N. Modeling Micro-End-Milling Operations. Part I. Analytical Cutting Force Model. Int. J. Mach. Tool Manuf. 2000, 40, 2125–2173.
19. Ma, Y.-S. Tool Path for Face Milling Considering Cutter Tool Entry/Exit Conditions. In Proceedings of the 3rd International Conference on Computer Integrated Manufacturing
1995, Vol. 1; pp 427–434.
20. Hashimura, M.; Hassamontr, J.; Dornfeld, D. A. Effect of In-Plane Exit Angle and Rake Angles on Burr. ASME J. Manuf. Sci. Eng 1999, 121, 13–19.
21. Ko, S. L.; Dornfeld, D. A. A Study on Burr Formation Mechanism. ASME J. Eng. Mater. Technol. 1991, 113, 75–87.
22. Ko, S. L.; Dornfeld, D. A. Analysis of Fracture in Burr Formation at the Exit Stage of Metal Cutting. J. Mater. Process. Technol. 1996, 189–200.
23. Jun, M. B. G.; Liu, X.; deVor, R. E.; Kapoor, S. G. Investigation of the Dynamics of Micro-End Milling, Part 1: Model Development. Trans. ASME J. Manuf. Sci. Eng. 2006,
128 (4), 893–900.
24. Park, S. S.; Malekian, M. Mechanistic Modeling and Accurate Measurement of Micro End Milling Forces. CIRP Ann. – Manuf. Technol. 2009, 58, 49–52.
25. Kang, I. S.; Kim, J. S.; Kim, J. H.; Kang, M. C.; Seo, C. Y. W. A Mechanistic Model of Cutting Force in the Micro End Milling Process. J. Mater. Process. Technol. 2007, 187–
188, 250–255.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Fundamentals and Modeling of Micro-End Milling Operation 199

26. Arcona, C.; Dow, T. A. An Empirical Tool Force Model for Precision Machining. J. Manuf. Sci. Eng. 1998, 120.
27. Lucca, D. A.; Seo, Y. W.; Rhorer, R. L. Energies in the Ultra-Precision Machining of Ductile Materials. In Proceedings of the 1992 NSF Design and Manufacturing System
Conference, Atlanta, Georgia; 1992; pp 123–129.
28. Srinivasa, Y. V.; Shunmugam, M. S. Mechanistic Model for Prediction of Cutting Forces in Micro End-Milling and Experimental Comparison. Int. J. Mach. Tool Manuf. 2013, 67,
18–27.
29. Melkote, S. N.; Kumar, M.; Hashimoto, F.; Lahoti, G. Laser Assisted Micro-Milling of Hard-to-Machine Materials. CIRP Ann. – Manuf. Technol. 2009, 58, 45–48.
30. Carslaw, H. C.; Jaeger, J. C. Conduction of Heating Solids, 2nd ed.; Clarendon Press: Oxford, 1959.
31. Mukund Kumar, M.; Melkote, S. N.; M’Saoubi, R. Wear Behavior of Coated Tools in Laser Assisted Micro-Milling of Hardened Steel. Wear 2012, 296, 510–518.
32. Cverna, F. Thermal Properties of Metals; ASM International: Materials Park, Ohio, 2002.
33. Wissmiller, D. L.; Pfefferkorn, F. E. Micro-End Mill Temperature Measurement and Prediction. J. Manuf. Process. 2009, 11, 45–53.
34. Park, S. S.; Rahnama, R. Robust Chatter Stability in Micro-Milling Operations. CIRP Ann. – Manuf. Technol. 2010, 59, 391–394.
35. Altintas, Y.; Weck, M. Chatter Stability of Metal Cutting and Grinding. Ann. CIRP 2004, 53 (2), 619–642.
36. Rahnama, R.; Sajjadi, M.; Park, S. S. Suppression of Chatter in Micro Milling with Process Damping. J. Mater. Process. Technol. 2009, 209, 5766–5776.
37. Tlusty, J.; Zaton, W.; Ismail, F. Stability Lobes in Milling. Ann. CIRP 1983, 32 (1), 309–313.
38. Opitz, H.; Bernardi, F. Investigation and Calculation of the Chatter Behavior of Lathes and Milling Machines. Ann. CIRP 1970, 18 (1), 335–343.
39. Fu, M. Y.; Olbrot, A. W.; Polis, M. P. Robust Stability for Time-Delay System: The Edge Theorem and Graphical Tests. IEEE Autom. Control 1989, 34 (2), 813–820.
40. Barmish, B. R. New Tools for Robustness of Linear Systems; MacMilllan Publishing Company: New York, 1994.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
This page intentionally left blank

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 11.09 Diamond Turning
XQ Zhang and KS Woon, Singapore Institute of Manufacturing Technology, Singapore
M Rahman, National University of Singapore, Singapore
 2014 Elsevier Ltd. All rights reserved.

11.09.1 Introduction: Background and Definition 201

11.09.2 Products and Applications 202
11.09.3 Machine Tool Development 204
11.09.3.1 Machine Characteristics 204
11.09.3.2 Historical Perspective 204
11.09.3.3 Present State 205
11.09.4 Diamond-Cutting Tool 206
11.09.5 Mechanics of Material Removal 207
11.09.5.1 Tool Edge Radius Effect 207
11.09.5.2 Specific Cutting Force and Cutting Energy 207
11.09.5.3 Minimum Uncut Chip Thickness 208
11.09.5.4 Microstructure and Grain Size Effects 209
11.09.5.5 Ductile-Mode Machining 210
11.09.5.6 Extrusion-Like Chip Formation 210
11.09.6 Surface Finish Characteristics 213
11.09.6.1 Form Error and Roughness 213
11.09.6.2 Residual Stress and Elastic Recovery 214
11.09.6.3 Burr Formation 215
11.09.7 Tool Wear Characteristics 216
11.09.7.1 Types and Mechanism 216
11.09.7.2 Control and Suppression 218
11.09.8 Summary and Outlook 218
References 219

11.09.1 Introduction: Background and Definition

Diamond turning is an ultraprecision machining technology for the generation of complex functional surfaces and extremely fine
microstructures with the use of geometrically defined diamond cutters. The cutters can be natural diamond or synthetic diamond
depending finishing scale of machining and finishing requirements. At present, diamond turning can be readily applied to a wide
range of metallic alloys, polymeric materials, and precious crystals.
The origin of diamond turning can be dated back to the seventeenth century. In 1779, the first documented study on single-
crystal diamond turning was reported by Ramsden (1), who used diamonds to cut hardened steel screws for the assembly of
dividing engines. Then, in 1901, diamond turning was first used at Zeiss to improve surface finishing for aesthetic purpose and later
for the production of aspheric surfaces with optical function (2). Modern diamond-turning process was introduced only in the late
1950s, to achieve near-mirror surface finishing with Rmax less than 0.1 mm on simple cylindrical, spherical, and flat faces. With the
accumulative scientific breakthroughs in machine tool and machining technologies over the last six decades, diamond turning is
now a highly repeatable materials processing technology to achieve high-quality surface finishing and form accuracy on a wide
variety of engineering materials. But, in essence, the superiority of diamond turning originates from the properties of diamond.
Diamond is an allotrope of carbon where atoms are arranged in a specific cubic lattice known as diamond cubic structure.
Diamond cubic structure consists of carbon atoms at the center as well as the four corners of a tetrahedron. Each of these atoms is
strongly bound to another atom through cosharing an electron, leading to the formation of a covalent bond. More details of this
aspect can be found in Ref. (3). The covalent bond is strong, contributing to the extreme hardness of diamond cutters. In conjunction
with other superior physical properties such as high thermal conductivity, low thermal expansion coefficient, and high abrasion and
corrosion resistance (4), diamond is one of the most important tool materials for ultraprecision machining in this modern era.
This chapter aims to provide a comprehensive discussion on diamond turning which covers its significance in modern materials
processing technologies in seven major topics, namely, industrial applications and products, machine tool development, diamond
tool design and fabrication, material removal mechanics, surface finish characteristics, diamond tool wear, and other special dia-
mond-turning techniques. The discussion spans from the review of important fundamental research work in the distant past and the
more recent scientific breakthroughs that tipped off the rapid technological progress in the last three decades. A detailed analysis of
this chapter helps to justify the importance of this materials processing technology through the understanding of its practical

Comprehensive Materials Processing, Volume 11 http://dx.doi.org/10.1016/B978-0-08-096532-1.01110-9 201

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
202 Diamond Turning

capabilities and limitations, working principles and methodology, critical prerequisites of cost-effective operations, and its extended
abilities for complex machining in the future.

11.09.2 Products and Applications

Diamond turning is used primarily to manufacture ultra precision parts for advanced applications, those that call for extremely high
levels of form accuracy and surface finishing. Such applications can be found in a number of industry sectors, including aerospace,
defense, electronics, semiconductor, and biomedical. In the early days, a large fraction of the parts produced with diamond turning
were optical components like reflectors and lenses, mostly machined directly from the stock material. Figure 1 shows a parabolic
reflector with a reflectivity of more than 98%, turned directly with a single-point diamond tool at the Lawrence Livermore National
Laboratory (LLNL) (5). As the technology becomes more mature diamond turning was applied to manufacture mold inserts for the
mass production of high-quality plastic lenses with injection molding.
Indeed, advancement in optical lens manufacturing has driven rapid progression in telecommunication, consumer cameras,
and surveillance technology. Although modern lithography has been successfully deployed in the industry for decades, its
usefulness is limited to semiconductors with two-dimensional and low-aspect-ratio configurations. But three-dimensional form
accuracy and mirror surface finishing are essential for optical components. This makes diamond turning, which has the capa-
bility to achieve form accuracy of up to a fraction of light wavelength, a viable process to make highly precise mold inserts for
high-volume plastic injection molding. Figure 2 shows one of the metallic molds used to produce plastic lenses in mobile
phones.
With its commercial success in electronic applications for the last two decades, diamond turning is becoming more important in
biomedical applications, particularly in the mass production of contact lens. Although large-scale contact lens manufacturing via
spin casting or double-sided molding has been conventionally established, there is a need to customize contact lenses for various
individual requirements while keeping the total cost affordable. With diamond turning, highly accurate contact lenses can be made
without expensive spin casting or double-sided molding, while the quality and dimensions are both reproducible and customized.
More importantly, diamond turning is also the only industrially feasible technique to fabricate mold inserts in toric shapes that are
nonradially symmetric. Figure 3 shows a polymethyl methacrylate (PMMA) contact lens produced directly with diamond turning.
Other than mold inserts manufacturing, diamond turning is also used to produce high-precision roller molds for large-area
printing, a process where additive materials are printed on engineering substrates as microstructures. For example, brightness-
enhancing films of are light crystal displays (LCDs) are manufactured through large-area printing with the key ultraprecision roller
molds produced by diamond turning. Figure 4 shows a drum roll being manufactured on an ultraprecision computer numerical
control (CNC) horizontal drum lathe (6).
As an advanced materials processing technology, diamond turning is gradually replacing conventional grinding and polishing,
as it is capable of producing highly accurate profiles and superior surface finishing faster than traditional methods, and fabricating
optical components with special profiles and features like diffractive and hybrid systems. With the rapid development of the
electronics, optics, and biomedical industries, it is expected that diamond turning will find more and more industrial applications in
the future.

Figure 1 Silver parabolic reflector machined with diamond turning by the Lawrence Livermore National Laboratory. Reproduced from Bryan, J. B.,
et al. Diamond Turning of Parabolic Mirrors, 1974, pp 39–44; (fr 4 to 5).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Diamond Turning 203

Figure 2 A lens mold and the injection-molded camera lenses for mobile phones.

Figure 3 A contact lens fabricated by diamond turning.

Figure 4 A drum roll installed on an ultraprecision CNC horizontal drum lathe. Reproduced from Moore. Nanotech HDL-2000, 2013, Available from:
http://www.nanotechsys.com/machines/nanotech-hdl-2000-horizontal-drum-lathe/.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
204 Diamond Turning

11.09.3 Machine Tool Development

11.09.3.1 Machine Characteristics
The machine tool is an essential element of diamond turning. It commands the highest levels of accuracy and repeatability that
a mechanical system and software can deliver – in short, an ultraprecision machine. The frame and structure of ultraprecision
machine tools, as well as the translational and rotational axes, are stiffened to minimize instability from the surrounding
environment. Most of the machines are mounted on granite blocks or in-ground concrete blocks surrounded with vibration
isolation material. Temperature is kept constant, with maximum variation within 2  F (1  C). Lack of temperature control
leads to uncontrollable drift in positioning, resulting in errors and inaccuracy. This is especially significant in long-time
continuous cutting.
Sharp changes in humidity and atmospheric pressure will also affect the accuracy of machining. To reduce the impact of this
aspect, independent hydrostatic or aerostatic bearing slides are incorporated to the linear axes to enhance the dynamic stiffness and
damping properties. A low-flow and low-pressure hydrostatic oil bearing with closed-loop servo control is preferred as it improves
the stiffness of the machining axes while simultaneously reducing external vibration.
Moreover, the accuracy of machine assembly, like the straightness of linear axes and squareness of mechanical elements, is
critical. The driving mechanism and air-bearing spindle are carefully balanced to minimize chatter between tool and workpiece and
rotational run-out. The positioning feedback system relies greatly on a laser holographic linear scale or an optical-grade glass scale to
provide high-resolution feedback with nanometer accuracy.
In summary, a qualified machine tool for diamond turning should possess the following merits:

l Vibration isolation and constant temperature control

l High stiffness in both axial and radial directions for all axes
l High sliding accuracy and assembly accuracy
l High positioning and repositioning accuracy
l High feedback and programming resolution
l Sufficient spindle power for high-speed cutting
l High servo performance
l High-precision tool and workpiece setup
l Complete process-monitoring system

11.09.3.2 Historical Perspective

The Americans are the pioneers in building ultraprecision machine tools for diamond turning, holding the most advanced
technology. According to the history of the Y-12 National Security Complex facility, the construction of the world’s first
diamond-turning machine was attempted by Union Carbide in 1962 (7). The machine was capable to produce a hemisphere
component with average form accuracy and surface roughness of 0.6 mm and 25 nm, respectively. Over the next five
decades, various machine tools for diamond turning were developed by private companies and institutes. One example is
the building of an ultraprecision diamond-turning machine by Moore, United States, through the modification of a preci-
sion coordinate-measuring system in 1976; in the same year, Ex-Cell-O UK built horizontal (II-G) and vertical (III-B)
diamond-turning machines to fabricate high-precision optical components for the Inertial Confinement Fusion program.
Cranfield Precision, United Kingdom, built the OAGM 2500 diamond-turning machine in collaboration with the British
Science and Engineering Research Council in 1991. Among all the machines built, the most noticeable one was constructed
by the LLNL and its collaborators in 1983, known as the Large Optics Diamond Turning Machine (LODTM), as shown in
Figure 5.
LODTM was designed and built to fabricate large optics with complex surfaces for a laser guiding system of missile
launchers that require extremely high accuracy: a fraction of light wavelength. The metrological loop of LODTM consists of
thermally stable components, laser interferometers, and precise capacitance gauges. Seven laser interferometers were used for
large translational motion measurements, four horizontal and three vertical, while two high-precision capacitance gauges
were used on the edges of the faceplate to measure its axial, radial, and tilt errors. A number of pressurized bearings and
high-precision capstan driving systems were used to actuate motion in the X- and Z-axis. To protect the machine from
temperature fluctuation and mechanical vibration, the main structure was rested on four air isolators, while a dedicated air-
conditioning system was used to maintain the ambient temperature in the enclosure within 0.01  F. On top of that,
circulation of chilled water was also used to regulate the temperature within 0.001  F.
With these advanced features, LODTM was capable of producing an extremely high-precision workpiece that weighed 1360 kg
with a diameter and length of 1650 and 500 mm, respectively (8). The tolerance could be maintained below 28 nm, approximately
a thousand times more accurate than conventional machine tools. LODTM is still in operation to this day, but many customized
components that were designed 30 years ago have been replaced with the standard ones to facilitate its maintenance (8). The front
view of LODTM is shown in Figure 5.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Diamond Turning 205

Figure 5 Large Optics Diamond Turning Machine constructed by the Lawrence Livermore National Laboratory. Reproduced from Lawrence Livermore
National Laboratory. Available from: https://www.llnl.gov/str/April01/Klingmann.html.

11.09.3.3 Present State

Although diamond turning has been widely accepted by the industry for more than three decades, only a few manufacturers in
the world are technologically capable in supplying such machine tools. The two major market shareholders are Ametek Precitech
and Moore Nanotechnology Systems, both located in Keene, New Hampshire. The former was founded in 1992 (9), and the
latter in 1997 (10). Both companies have experienced growth even when the global economy was slowing down, and this was
particularly driven by the strong needs for high-precision optical components and micro-optics systems (7).
At present, these two companies produce two-axis diamond-turning machines for simpler applications like contact lens
making, as well as multiaxis machines for much more complex applications such as functional free-form surface generation.
With the development of more advanced linear motors, hydrostatic slide guides, high-resolution grate scales, and high-
performance CNC motion control systems, the accuracy and performance of diamond turning will be continuously
improved for more complex applications. Figure 6 shows two multiaxis machines built by both Ametek Precitech and Moore,
respectively (11,12).

Figure 6 Examples of multiaxis diamond-turning machine: Nanoform 700 Ultra (left) (reproduced from Precitech. Nanoform 700 Ultra, 2013, Available
from: http://www.precitech.com/products/nanoform700ultra/nanoform_700_ultra.html); and Nanotech 350FG (right) (reproduced from Moore.
Nanotech 350FG, 2013, Available from: http://www.nanotechsys.com/machines/nanotech-350fg-freeform-generator/).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
206 Diamond Turning

11.09.4 Diamond-Cutting Tool

Diamond turning is performed with the use of cutters made of diamonds. Diamonds, either natural or synthetic, are carefully
ground and polished to generate geometrically defined cutting edges at specific crystallographic orientations for material removal.
The cutting performance and surface finishing produced via diamond turning rely greatly on the edge sharpness of diamond cutters.
The edge sharpness, quantitatively defined as the tool edge radius r, is a governing factor of the diamond-turning process due to
the edge radius effect. When the magnitude of cutting or undeformed chip thickness a approaches the size of r in diamond turning,
the cutting-edge radius affects the mechanics and mechanism of chip formation to a great extent. This will be addressed in greater
length in Section 11.09.5.1. Newly prepared cutting edges are usually within the range of several micrometers for most general
applications and tens of nanometers for other special applications (13). But determination of the cutting-edge radius is hard due to
its shape and size. Various techniques such as confocal laser scanning microscopy (14), the nano-indenting method (15), scanning
electron microscopy analysis (16), and atomic force microscopy (AFM) (13) were previously attempted, but none was sufficiently
consistent and accurate for industry adoption. Figure 7 shows AFM 3D and a section view of the indentation profile after using the
nano-indenting method.
As diamond is anisotropic, its mechanical properties vary drastically in different crystallographic planes and orientations in
which the corresponding wear resistance is significantly different (17). Choosing the right plane and orientation as the cutting edge
is critical to ensure sufficient strength for high-performance material removal. It was previously determined that the rake face with
h110i crystal orientation performs better than that of h100i in terms of wear resistance, surface finishing, and cutting forces (18), for
both diamond turning and vibration-assisted turning (19). This has been adopted as the standard for most diamond tools supplied
commercially by Contour Fine Tooling, Apex Diamond, and Osaka Diamond. Figure 8 shows a novel concave diamond tool, of
which the radius size can range from 350 mm to over 200 mm (20).

Figure 7 Atomic force microscopy analysis of the diamond tool edge radius by the nano-indenting method. Reproduced from Lucca, D. A.; Seo, Y. W.;
Komanduri, R. Effect of Tool Edge Geometry on Energy Dissipation in Ultraprecision Machining. CIRP Ann. – Manuf. Technol. 1993, 42 (1), 83–86.

Figure 8 A novel concave diamond tool made by Contour. Reproduced from Contour. Contour Cutting Edge July 2012 News Bulletin, 2012.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Diamond Turning 207

11.09.5 Mechanics of Material Removal

11.09.5.1 Tool Edge Radius Effect
The great difference between the undeformed chip thickness a and the tool edge radius r distinguishes diamond turning from
conventional machining. Established models for conventional machining such as the shear plane model (21) are no longer
applicable to diamond turning, as most of these models assume perfect edge sharpness. Such crucial assumptions might be
acceptable in the modeling of conventional macromachining as a is very much larger than r, by at least three orders of magnitude.
But as a approaches r in micron and nanometric scales, as illustrated schematically in Figure 9, the same assumption does not hold
true. Largely due to negligence of the tool edge radius effect, previous models for conventional machining are incapable to account
for the exponential increase in specific cutting energy at reducing undeformed chip thicknesses, commonly known as the size effect
in metal cutting.
Size effect is claimed to be attributed to a number of factors, including subsurface plastic flow (22), strain inhomogeneity (23),
extensive plastic deformation (24), material plowing and elastic recovery (25), and strain gradient (26). Close examinations of the
downscaling process suggests that the size effect and many other process characteristics are governed by the tool edge radius effect.
But research interest on this effect was not encouraging in the past because the tool edge radius effect is not significant in
conventional macromachining when a is very much larger than r. But it could no longer be ignored in diamond turning because the
magnitudes of a and r are comparable. Indeed, it was approximated by Zorev (27) that the tool edge radius effect is significant when
r is greater than 1/10 of a.
In this regard, Masuko (28) and Albrecht (29) were among the pioneers in investigations of the tool edge radius effect on chip
formation and cutting forces. Subsequently, Bitans and Brown (30) claimed that plastic deformation was governed by the tool edge
radius effect, where thicker shear zones were attributed to larger tool edge radius; while Nakayama and Tamura (22) determined that
the variations in subsurface plastic deformation were due to a simultaneous shear zone extension and shear angle reduction as
governed by the tool edge radius. At critical cutting depths, Basuray et al. (31) reported an irregular increment of normal and
tangential force when large-edge radii were used due to nonuniform displacement of work material. Moriwaki and Okuda (24)
found that the increase in specific cutting resistance was driven by the dominance of plastic deformation rather than sheared cutting.
Additionally, experimental and molecular dynamics (MD) simulation studies as reported by Komanduri et al. (32), Fang et al. (33),
and Li et al. (34) revealed great influence of the tool edge radius in ductile-mode machining of brittle materials.

11.09.5.2 Specific Cutting Force and Cutting Energy

The effects of the tool edge radius on the cutting force was first identified by Masuko (28) and subsequently validated by Albrecht
(29) in a separate experimental study. Through the machining of aluminium alloy and carbon steel, Masuko discovered that the
cutting force consisted of the chip formation component as well as an indentation component that arose by the tool edge radius, as
shown in Figure 10. The extrapolation of the cutting force to zero undeformed chip thickness showed the magnitude of the
indenting action.

Figure 9 The tool edge radius effect is significant when the magnitude of undeformed chip thickness a approaches the size of the tool edge radius r.

Figure 10 (a) Indentation force Rn and resultant force Rc induced by the tool radius and tool face, respectively; and (b) differences in power FP and
feed components FQ with undeformed chip thickness in the cutting force. After Masuko, M. Fundamental Research on Metal Cutting. Bull. Jpn. Soc.
Mech. Eng. 1956, 22, 371–377.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
208 Diamond Turning

Figure 11 Exponential increment in total specific energy with fine reductions in undeformed chip thickness. After Lucca, D. A.; Rhorer, R. L.;
Komanduri, R. Energy Dissipation in the Ultraprecision Machining of Copper. Ann. CIRP 1991, 40 (1), 69–72.

In this regard, Finnie (35) claimed that the indentation effect was not significant when material was removed at several milli-
meters but conceded that such an effect could become increasingly important in the grinding process. Malkin (36) later discovered
that high energy close to the specific melting energy was required for chip formation in grinding. Indeed, Nakayama and Tamura
(22) encountered a nonlinear increase in specific cutting and thrust forces at reducing cutting depths at the micron scale. The authors
attributed such phenomena to the tool edge radius effect through the extension of shear zone and the decrease in shear angle.
Abdelmoneim and Scrutton (37) recognized the importance of the tool edge radius under a unique condition of
a  r (1 þ sin gtool), which was validated by Taminiau and Dautzenberg (38).
At a critical undeformed chip thickness, Basuray et al. (31) claimed that nonuniform material displacement resulted with large r,
which led to irregular increment of machining forces. Subsequently, through MD simulation, Komanduri et al. (39) justified that the
tool edge radius is negligible in conventional cutting but not for nanometric cutting, where specific cutting energy increased rapidly
with the reductions in a and by keeping a constant a/r ratio. From orthogonal fly-cutting experiments, Moriwaki and Okuda (24)
reported that the dominance of plastic deformation rather than sheared cutting was the main cause for the increased specific cutting
resistance; while Lucca et al. (25) pointed out that material plowing and flank face rubbing due to elastic recovery were the main causes,
as shown in Figure 11. These experimental observations agreed well with those of finite element analysis, as reported in Refs. (40,41).
In another orthogonal fly-cutting study of Te–Cu, Lucca and Seo (42) reported the great influences of tool geometries on the
resulting forces and energies. When the tool edge radius was larger than the undeformed chip thickness, Lucca et al. (43) found that
the resultant force shifted gradually from the cutting direction to the thrust direction with the reduction in undeformed chip
thickness. Moreover, Shaw (23,44) claimed that heat generated from concentrated shearing in cutting would be carried away by the
chips while microextrusion in grinding would retain heat in the work, which influenced the energy dispersion. Through finite
element analysis with the strain-gradient effect, Liu and Melkote (45) concluded that strain-gradient hardening was the main factor
for nonlinear increment of specific cutting energy in micromachining.

11.09.5.3 Minimum Uncut Chip Thickness

The smallest magnitude of a where a continuous chip is produced is substantially governed by the tool edge radius. In face turning
of electroplated copper with diamond tools, Ikawa et al. (46) discovered that the minimum a was 1/10 of r. Through MD simu-
lation, Ikawa et al. revealed the generation of a rough surface when copper was removed below the minimum a. Shimada et al. (47)
applied a similar MD simulation technique but encountered the best surface finishing at the minimum a of 1/20 of r.
In the study of the tool–work friction coefficient effect on minimum a, Yuan et al. (48) reported that the minimum a was within
1/4–1/3 of r, ranging from 0.2 to 0.6 mm for Al alloys. It was found that a smaller tool edge radius would enable a smaller order of cutting
depths for diamond tools, as substantiated by an analytical model. Subsequently, Son et al. (49) extended the experimental investi-
gation in this area to pure Al, Te, and Cu, where finer a could be generated with smaller r and a higher friction coefficient to warrant
the best surface finishing. The magnitude of minimum undeformed chip thickness was found to vary for different work materials.
More recently, Weule et al. (50) examined the minimum a in the micro-endmilling of steel and determined it to be 1/3 of r. The
trends of surface roughness and sawtooth-like surface topography were believed to be related to minimum undeformed chip
thickness, cutting velocities, and material properties. Vogler et al. (51) observed the influence of minimum a through the identi-
fication of a peak periodic force component at 1/3 of the actual cutting frequency under a constant small feed rate condition. The
authors explained that material would pile up through multiple tool passes up to a certain thickness greater than the minimum
value, and then the production of a chip would occur.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Diamond Turning 209

Figure 12 Relationships between cutting forces and chip loads. After Liu, X.; Jun, M. B. G.; DeVor, R. E.; Kapoor, S. G. Cutting Mechanisms and
Their Influence on Dynamic Forces, Vibrations and Stability in Micro-endmilling. In ASME International Mechanical Engineering Congress and Exposition,
2004, Anaheim, CA, Paper No. 62416.

In this regard, Kim et al. (52) reported that the measured chip volume was much larger than the nominal chip volume and the
pitch of feed mark was larger than the feed rate where both aspects indicated that chips did not form in every tool pass. These
findings were supported with another mechanistic modeling study by Kim et al. (53) and a finite-element (FE) modeling study by
Lai et al. (54).
Through microstructure-level FE modeling, Vogler et al. (55) identified that ductility could govern the magnitude of minimum
undeformed chip thickness for a given work material. Moreover, Liu et al. (56) found that the cutting and thrust forces increased
rapidly when the magnitude of material removal in steel was greater than the minimum undeformed chip thickness compared to
that above the minimum value. As shown in Figure 12, the transition between the two regimes was signified with a sudden peak in
thrust force where it was attributed to the shift from shearing to plowing.

11.09.5.4 Microstructure and Grain Size Effects

Most of the engineering materials are polycrystalline, anisotropic, and inhomogeneous. The microstructure of these materials is
distributed randomly throughout the bulk matrix with varying grain sizes between 100 nm and 100 mm, depending on the heat
treatment history. When the grain size is larger than the cutting magnitude of undeformed chip thickness, individual grain is
subjected to shear deformation in which chips are formed within the grains. This gives rise to the microstructure and grain size
effects in diamond turning that lead to fluctuations in cutting performance and mechanics (57–60) in terms of surface generation,
chip formation, and forces. Figure 13 shows a schematic illustration for diamond cutting of polycrystalline materials.

Figure 13 Shear deformation of individual grains during diamond turning of polycrystalline materials. After Yuan, Z. J., et al. Effect of Crystallographic
Orientation on Cutting Forces and Surface Quality in Diamond Cutting of Single Crystal. CIRP Ann. – Manuf. Technol. 1994, 43 (1), 39–42.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
210 Diamond Turning

Figure 14 Crack propagation system in single-point diamond turning of hard and brittle materials.

In the study of the relationship between cutting direction and crystallographic orientation in the diamond turning of single-
crystal b-brass, Ueda et al. (57) found a continuous type of chip with lamellar slip structure in most cases and a discontinuous type
in a particular orientation, and the shear angle can vary from 15 to 60 due to the change of crystal orientation. While for single-
crystal aluminium, To et al. (59) reported that the cutting force and surface quality vary significantly in the three cutting directions
attempted (h100i, h110i, and h111i), and continuous chip formation can be achieved in all cases. In the diamond turning of single-
crystal copper, Yuan et al. (60) reported that a large cutting force was observed in h110i, a stark contrast in comparison to the small
resultant force in h001i.

11.09.5.5 Ductile-Mode Machining

Hard and brittle materials are widely used in the electronics industry, especially semiconductor materials like silicon. To enhance the
functionality of silicon-based components, it is more effective and repeatable to realize nanometric surface finishing through
diamond turning (14) as abrasive processes like chemical-mechanical polishing are not deterministic or productive. But it is not
easy to machine hard and brittle materials because these materials tend to undergo cleavage-based fracture, resulting in severe losses
in terms of material scrap and tooling cost. Figure 14 elucidates the crack propagation system involved, where Cl and Cm are the
length of lateral and median cracks, respectively.
Nevertheless, ductile-mode machining is a unique diamond-turning technique where hard and brittle materials are machined in
a ‘ductile manner’ that yields surface finishing that is free from brittle fracture. The phenomenon was first observed in the
indentation test (61), and gradually adapted to diamond turning after the physics involved were better understood over time.
Primarily, when the undeformed chip thickness is below the critical threshold, brittle materials undergo shear deformation similar
to that in metal cutting owing to microscopic plasticity. The critical undeformed chip thickness is governed by intrinsic material
properties, tool geometry, and cutting conditions. Giovanola and Finnie (62) were among the first to discover that glass can be
machined in the ductile mode when the cutting magnitude is sufficiently fine. Similarly, Puttick et al. (63) claimed that brittle
materials can be machined plastically at a critical undeformed chip thickness below which a crack-free machined surface is
produced. Subsequently, Venkatachalam et al. (64) proposed a fracture toughness based model to predict the ductile-brittle
transitional undeformed chip thickness in the machining of brittle materials. They have also found that a negative rake angle
could generate enough hydrostatic compressive stress for ductile-regime machining. More recently, Arif et al. (65) predicted
the critical undeformed chip thickness from the corresponding transition in the representative modes of energy at the ductile-brittle
transition point in nanomachining of brittle materials.
Figure 15 shows schematic and physical views of a groove on silicon generated with a single crystalline diamond tool; both
ductile and brittle machining modes are observed. The critical undeformed chip thickness measured for that particular tool and
workpiece set ranges from 200 to 270 nm, and its value may vary when different diamond tools are applied due to the inconsistent
tool geometries.

11.09.5.6 Extrusion-Like Chip Formation

The effect of the tool edge radius on the mechanics of diamond turning has been addressed in Sections 11.09.4 and 11.09.5.1. This
section concentrates on chip formation transformation driven by such an effect through the simultaneous manipulation of
undeformed chip thickness a and tool edge radius r. At large a with a fixed r, the resultant force R is operating toward the cutting
direction with a saturation of low q. Gradual decrease of a/r leads to a gentle transition in the direction of R toward the thrust
direction with increasing q. Further decreasing a/r will lead to a saturation of high q. The process is seen to transform from a cutting-
dominated mechanism (low q) to a thrust-dominated mechanism (high q), similar to the transition from concentrated shearing to
material plowing due to the amplification of thrust action with decreasing a/r. This transformation is shown in Figure 16.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Diamond Turning 211

Figure 15 Transformation from brittle-mode to ductile-mode finishing through the generation of a groove on silicon with single-point diamond turning.

Figure 16 Transition in the direction of the resultant force vector R from cutting to thrust with the reduction of undeformed chip thickness a and
a fixed tool edge radius r. Reproduced from Lucca, D. A.; Seo, Y. W. Effect of Tool Edge Geometry on Energy Dissipation in Ultraprecision Machining.
Ann. CIRP 1993, 42 (1), 83–86.

Three modes of chip formation (18) with unique contact and deformation behavior could take place at different regimes of a/r,
as illustrated in Figure 17. First, at a/r  1.0, chip flow is extended on the tool rake face, which leads to the formation of the
secondary deformation zone (SDZ), the primary deformation zone (PDZ), and an effective positive rake angle þgeff. Second, the
decrease of a/r to a critical magnitude below 1.0 causes drastic changes in chip formation. The continuous chip produced could not
extend on the tool rake face but was deflected by the tool edge radius to form a small effective negative rake angle geff. Hence, the
formation of the SDZ is suppressed, while the PDZ is solely established, resulting in a complete localization in deformation ahead
of the tool edge radius. Third, a further decrease of a/r to another threshold induces the formation of a highly localized PDZ and
a large effective negative rake angle geff. Discontinuous chips are then produced intermittently.
Figure 17 is compared with Figure 16 to distinguish among different domains of chip formation mechanisms. With the
formation of the SDZ and þgeff, mode (a) has the characteristics of a cutting-dominated mechanism (low q) that resembles
a concentrated shearing process. In contrast, mode (c) bears resemblance to a thrust-dominated mechanism (high q) with a large
geff and a highly localized PDZ, like a material plowing process.
The microextrusion model proposed by the late Professor Milton C. Shaw for finish grinding appears to be a good approxi-
mation (20). As shown in Figure 18, a blocky abrasive grain is idealized as a sphere. At a specific a/r below unity, a continuous chip

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
212 Diamond Turning

Figure 17 Three modes of chip formation at decreasing a/r from (a) to (c). (a) Formation of an effective positive rake angle þgeff, PDZ and SDZ;
(b) formation of a small effective negative rake angle geff and a localized PDZ; and (c) formation of a large effective negative rake angle geff
and a highly localized PDZ. After Woon, K. S.; Rahman, M. Extrusion-like Chip Formation Mechanism and Its Role in Suppressing Void Nucleation.
CIRP Ann. – Manuf. Technol. 2010, 59 (1), 129–132.

Figure 18 The microextrusion model for finish grinding that could be the mechanism of mode (b) chip formation in micromachining. Reproduced from
Shaw, M. C. A New Theory of Grinding. Mech. Chem. Eng. Trans. 1972, 8 (1), 73–78.

Figure 19 Distinguishable deformation behavior between (a) concentrated shearing at a/r > 0.25 and (b) extrusion-like mechanism at a/r  0.25, as
reflected from the nodal displacement plots. After Woon, K. S.; Rahman, M. Extrusion-like Chip Formation Mechanism and Its Role in Suppressing
Void Nucleation. CIRP Ann. – Manuf. Technol. 2010, 59 (1), 129–132.

is ‘extruded’ at a geff, with the thrust component Ft greater than the cutting component Fc. A large volume of material is plastically
deformed for a small volume of material to be formed as a chip. This contributes to the size effect phenomenon.
The typical nodal displacement plot of mode (a) concentrated shearing is shown in Figure 19(a) for 1.00  a/r  0.25. Materials
along the cutting path are subjected to horizontal tool engagement and then displaced toward the cutting direction, with a gradual
change in direction. Further decreasing a/r to 0.25 triggers an abrupt change in the displacement characteristics, as shown in
Figure 19(b). Small amounts of materials begin to escape from the bulk matrix as chips, against the cutting direction, followed by
an upward directional transition along the rounded-edge curvature. The remaining fraction of deformed materials is compressed
back into the bulk material to form the machined surface. These are perceived as the signature characteristics of mode (b) extrusion-
like chip formation at a/r  0.25.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Diamond Turning 213

Figure 20 (a) Extrusion-like mechanism and (b) concentrated shearing after chip growth with the formations of an effective negative rake angle geff
and an effective positive rake angle þgeff, respectively. After Woon, K. S.; Rahman, M. Extrusion-like Chip Formation Mechanism and Its Role in
Suppressing Void Nucleation. CIRP Ann. – Manuf. Technol. 2010, 59 (1), 129–132.

Figure 21 Experimental findings on surface roughness Ra (mm) at varying a/r (V ¼ 100 m min1; gtool ¼ þ10 ; and r ¼ 10 mm). After Woon, K. S.;
Rahman, M. Extrusion-like Chip Formation Mechanism and Its Role in Suppressing Void Nucleation. CIRP Ann. – Manuf. Technol. 2010, 59 (1), 129–132.

The extrusion behavior at a/r  0.25 is maintained, and more materials are extruded to form the chip (66). As illustrated in
Figure 20(a), a small geff greater than the þgtool is eventually formed after chip growth is mature, as the chip is deflected by the
tool edge radius. Tool–chip contact is then established exclusively on the edge curvature. As the chip continues to grow, the upward-
extrusion action is transmitted aggressively across the PDZ, which increases the intensity of material deformation. The PDZ would
then become highly strained, as reflected from the elongated nodal displacement vectors. These characteristics contrast greatly with
that of concentrated shearing at a/r > 0.25, as shown in Figure 20(b), which involves an effective positive rake angle þgeff as the
flow of chip extends onto the rake face.
Due to these phenomena, it was experimentally determined that surface quality improves with decreasing a/r, as shown in
Figure 21. A considerable amount of microcracks were encountered at 1.8  a/r  0.6, while further decrease of a/r to 0.05
deteriorates the surface finish significantly as discontinuous chips were produced. But at a/r ¼ 0.2, exceptional surface finish that
approaches the quality of surface grinding (67) was achieved through the extrusion-like chip formation mechanism.

11.09.6 Surface Finish Characteristics

11.09.6.1 Form Error and Roughness
The functionality and performance of many optical components are governed by their form accuracy and surface roughness. Take
optical lens for measurement, imaging, and lighting as an example. When the resultant accuracy and finishing exceed the critical
design tolerance, light sources will be deflected from their intended path when the components are in operation, leading to
deterioration in metrological accuracy, imaging quality, and focusing capability. With the advantages of diamond turning,
extremely high form accuracy and low surface roughness, within a nanometric scale, have been achieved on many engineering
materials, including copper, aluminium, and electroless nickel.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
214 Diamond Turning

Form accuracy in diamond turning is influenced by many factors, including machine accuracy, programming accuracy, setup
accuracy, cutting-edge quality, tool wear and degradation, operating temperature, humidity, and stability conditions. Moreover,
the control of form accuracy on a large workpiece is significantly more challenging than on smaller ones. Fortunately, the form
accuracy required by most industrial applications can be delivered by most commercially available machine tools. Some special
components used in aerospace or laser systems may need extremely high form accuracy on a large surface, which call for special
machine tools and processing know-how. LODTM, for example, can achieve a form accuracy of 28 nm over the entire 1.6 m
diameter work zone (8).
The capability of diamond turning in delivering mirror-surface finishing on many engineering materials without the need for
postprocessing has been one of its major strengths. Although comparable surface quality on flat surfaces could be achieved with
loose abrasive processes, they are not applicable to complex surfaces, including aspheric, high-aspect-ratio, and free-form ones,
largely due to the non-deterministic nature of these processes.
In diamond turning, the surface roughness is largely governed by the nose radius and feed rate, and still very much affected by
other machining parameters, such as depth of cut, rake angle, tool wear, coolant supply, machine dynamics, and material properties.
According to Ref. (68), the theoretical surface roughness due to feed marks is expressed as follows:

fr2
Rth ¼ [1]
8rn
where fr represents the feed rate per revolution, and rn represents the diamond tool nose radius. From this equation, it can be
predicted that the surface roughness will be reduced by increasing nose radius. But the experimental surface roughness values are
usually several times larger than theoretical values. Sugano et al. (69) found that, as the nose radius becomes smaller, the surface
roughness of aluminium alloy will approach the theoretical values. Although researchers have tried to provide more comprehensive
theoretical models to predict more accurate surface roughness by considering machining dynamics, tool defects, elastic recovery,
and so forth, these models are not efficient and practical for industrial use because the enormous calibration and calculation effort
is not cost effective.
Surface roughness produced in diamond turning is not sensitive to cutting speed. As long as tool geometries, feed rates,
and relevant cutting conditions are optimized, mirror surface finishing could be produced within a wide range of cutting
speeds. Similar to conventional cutting, progressive roughing and finishing procedures are conducted to obtain the best
combination of surface finishing and subsurface damage. The depth of cut in the final finish cut can approach 0.1 mm, and the
feed rate can be less than 1 mm per revolution. According to the authors’ knowledge, the best reported surface roughness till
this day is 1 nm Rz that was previously achieved on oxygen-free high-thermal-conductivity copper with the ultraprecision
lathe in LLNL (70).

11.09.6.2 Residual Stress and Elastic Recovery

Stress is introduced to the machined surface through excessive subsurface deformation and temperature gradient, governed by the
cutting conditions, tool geometry as well as tool wear and degradation (70,71). Residual stress is detrimental to the parts produced
in terms of fatigue strength, corrosion resistance, and form accuracy. For example, the form accuracy of large reflectors used in laser
systems is governed by the residual stress of the machined surface (72). Residual stress on the surface is usually tensile, associated
with subsurface compressive residual stress. But the opposite could occur for different metallic materials. Therefore, some post
processing techniques like stress relief annealing are applied to remove the undesirable residual stress. But due to the low material
removal rates and sharp cutting edges in diamond turning, the magnitude of residual stress generated is a lot smaller than that in
conventional machining.
In the diamond turning of aluminium alloy, Sugano et al. (69) reported that tensile stress is realized during roughing, while the
stress is largely compressive under finishing conditions. The magnitude of such compressive stress increases with the wear on
diamond tools via burnishing. The influences of nose radius and feed rate are also significant. Other experimental (72) and
simulation studies (73) were mostly conducted to study the influence of tool geometries and cutting parameters. Figure 22 shows
simulated effects of the rake and clearance angles on tensile surface residual stress.
In diamond turning, the machined surface is deformed plastically and elastically when the workpiece is subjected to material
removal via fine-scale cutting. When the chip is formed continuously through plastic deformation, the machined surface undergoes
elastic deformation inevitably due to the tool edge radius effect. After the tool pass is made, the machined surface will recover
elastically to the original position. Elastic recovery is undesirable, as it will lead to the formation of burr (74) and deterioration in
the surface finishing (75). As diamond tools are much sharper than conventional cutting tools, the magnitude of elastic recovery is
also comparatively small, but it is still critical due to the demanding requirements on surface quality. Lucca et al. (42) found that the
elastic recovery of workpiece material along the tool flank face plays a significant role when the undeformed chip thickness
approaches the tool edge radius. Kim and Kim (76) developed a comprehensive orthogonal cutting-force model considering the
elastic recovery along the clearance face that is caused by the round diamond tool edge. Kong et al. (77) conducted a comprehensive
experimental and simulation study on the effects of material swelling and recovery on surface generation and roughness prediction
in diamond turning of ductile materials. Figure 23 shows a schematic illustration of the effects of material swelling and recovery
across the diamond tool nose.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Diamond Turning 215

Figure 22 Simulated effects of the rake and clearance angles on tensile residual stress after diamond turning. Reproduced from Zong, W. J., et al. FEM
Optimization of Tool Geometry Based on the Machined Near Surface’s Residual Stresses Generated in Diamond Turning. J. Mater. Process. Technol.
2006, 180 (1–3), 271–278.

Figure 23 Illustration of the effects of material swelling and recovery across the diamond tool nose. After Kong, M. C., et al. A Study of Materials
Swelling and Recovery in Single-point Diamond Turning of Ductile Materials. J. Mater. Process. Technol. 2006, 180 (1–3), 210–215.

11.09.6.3 Burr Formation

Burrs are formed in most materials processing operations that involve materials removal, chiefly governed by the cutting conditions
and different tool–workpiece combinations. The size of burr can be reduced by reducing the effective undeformed chip thickness
and optimizing the cutting conditions and operating procedures (78). In diamond turning, as the effective undeformed chip
thickness can be reduced to the submicron and nanometer scales, burr formation can be greatly reduced. Therefore, burrs are
negligible in diamond turning, because the fine and continuous tool feeding could remove the minor burrs generated during the
process. However, when diamond cutting is applied in micromachining to fabricate microcomponents, burr formation starts
playing an important role, and large-sized burrs are detrimental to its application (79).
According to Fang and Liu (80) burr size can be reduced by using sharper diamond tools and reducing the undeformed chip
thickness. They claimed that burrs are formed only when plastic deformation occurs. Due to the size effects induced by the diamond
tools, there is a minimum undeformed chip thickness for each tool-work and operating parameters combination, below which no
material removal nor burr formation will take place. This is shown in Figure 24. (81)

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
216 Diamond Turning

Figure 24 Burr height versus undeformed chip thickness. Reproduced from Fang, F. Z.; Liu, Y. C. On Minimum Exit-burr in Micro Cutting.
J. Micromech. Microeng. 2004, 14 (7), 984.

11.09.7 Tool Wear Characteristics

11.09.7.1 Types and Mechanism
The wear condition of diamond cutters affects various machining performances of diamond turning. After the diamond tool is
worn, it is more possible to obtain worse surface quality higher cutting force, greater chatter vibration, inaccurate form profile, and
so on. Figure 25 shows the variation of cutting force against cutting distance while machining 6061 aluminum alloy using diamond
tools. Both of the thrust and cutting force components increase with the cutting distance alongside with the development of wear on
the tools.
In general, tool wear develops slowly and steadily in the processing of non-ferrous metals, like copper, aluminium, and so on,
that are generally soft. High quality surface finishing is readily achievable even after continuos cutting for several hundreds of
kilometers. Figure 26 shows an example of the variation in surface roughness with respect to the cutting distance using diamond
tools with different nose radii. But, in many practical cases diamond tools fail unexpectedly during the process with the devel-
opment of micron-scale chipping on the cutting edges due to small vibration or hard impurities in the workpiece. Therefore,
diamond tools should be used only on ultraprecision machine tools, which can minimize both external and internal vibration.
The development of wear on diamond tools could be driven by an individual mechanism or a combination of multiple
mechanisms. Evans et al. (82) classified the wear mechanism as: (1) adhesion and formation of a built-up edge; (2) abrasion,
microchipping, fracture, and fatigue; (3) tribothermal wear; and (4) tribochemical wear. The intensity of most wear mechanisms can
be minimized or eliminated by optimizing the operating conditions, but tribochemical wear that takes places aggressively in
diamond turning of some particular materials (steel for instance), is difficult to avoid.
Thus, the understanding of ‘diamond-turnable’ materials becomes enormously important in diamond turning. Paul et al. (83)
did an extensive study on diamond-turnable materials and, based on their investigation, summarized 13 diamond-turnable
elements (indium, tin, lead, zinc, plutonium, magnesium, aluminium, germanium, silver, gold, copper, beryllium, and silicon),
and 15 non-diamond-turnable elements (uranium, manganese, nickel, cobalt, iron, titanium, chromium, vanadium, rhodium,
ruthenium, niobium, molybdenum, tantalum, rhenium, and tungsten). Through the investigation of the chemical properties of
these elements, they found that the chemical wear of diamond tools is originated from the formation of carbon–metal complexes
with unpaired d-shell electrons from the workpiece. This was concluded with the discovery of the all 13 diamond-turnable elements
do not have unpaired d-shell electrons, while the other 15 non-diamond-turnable elements carry one to five unpaired d electrons
Metals with more unpaired d electrons will expedite wear development on diamond tools. It is necessary to note that iron, which is
highly non-diamond-turnable, has four unpaired d electrons. This theory was experimentally proven and became much more
established compared to other explanations, such as those regarding melting point, crystal structure, and hardness.

Figure 25 Cutting forces measured while cutting 6061 aluminum alloy. Reproduced from Lane, B. M., et al. Diamond Tool Wear when Machining
Al6061 and 1215 Steel. Wear 2010, 268 (11–12), 1434–1441.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Diamond Turning 217

Figure 26 Surface roughness variation with the increase of cutting distance. After Sugano, T., et al. Diamond Turning of an Aluminum Alloy for Mirror.
CIRP Ann. – Manuf. Technol. 1987, 36 (1), 17–20.

In the processing of diamond-tunable materials, abrasive wear which is considered very slow is the dominant tool wear
mechanism, while microchipping is caused by material impurities and vibration. In machining non-diamond-turnable materials
like steel, the tribochemical wear mechanism plays the leading role. Figure 27 shows a comparison of a diamond tool edge after
machining 6061 aluminum alloy and steel 1215. It can be seen that the diamond tool edge is quickly worn after very short cutting of
steel. Evans et al. (82) summarized three types of possible tribochemical wear: oxidation, diffusion wear, and catalyzed
graphitization.
Oxidation does not usually happen in diamond turning as the tool tip temperature is lower than the oxidation temperature of
diamond (900–1000 K) (82). But some researchers believed that the presence of oxygen will promote the chemical reaction
between metal and diamond an oxygen–metal–carbon transition complex (83). But, this is still arguable after Thornton and Wilks

Figure 27 Comparison of the tool edge profile for diamond worn on Al6061 (left) and St1215 (right). Reproduced from Lane, B. M., et al. Diamond
Tool Wear when Machining Al6061 and 1215 Steel. Wear 2010, 268 (11–12), 1434–1441.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
218 Diamond Turning

(84) found that diamond tool wear is less significant when steel is machined under normal atmospheric conditions comparing to
that in vacuum.
On the other hand, diffusion wear takes place when carbon atoms of the diamond matrix enter the workpiece, during which atoms
move into the vacancies within the solid metal lattice of the workpiece (85). The rate of diffusion depends on the type of workpiece
materials, the difference between the carbon concentration, the saturation limit, and the temperature. There is an energy barrier ED that
must be crossed during the atom diffusion, and ED for the diffusion of graphitic carbon into a-iron is found to be 84.1 kJ mol1.
In addition, graphitization is the most commonly reported chemical wear mechanism of diamond tools in machining steel.
From the aspect of chemistry, graphitization is the reverse process of diamond synthesis. During the graphitization process, carbon
atoms in the diamond lattice revert to the more stable graphite form. The uncatalyzed process is extremely slow, because the sp3
hybrid orbitals between the carbon atoms are tightly interlocked. Similar to the diamond synthesis process, the graphitization
process can be promoted with specific metal catalysts, by reducing the energy barrier for the onset of the transformation.

11.09.7.2 Control and Suppression

To minimize wear, diamond tools must be used on machine tools with highly stable spindle rotation and accurate linear axis
movement. With that, the minor vibration at the tool–workpiece interface that is difficult to control can be minimized to avoid
microchipping of the tool edge. The machining conditions should be carefully chosen to avoid large undeformed chip volume,
failure of doing so may lead to serious tool degradation or even catastrophic tool failures.
Nevertheless the inability of diamond turning to some important engineering materials like steel strongly limits its applications.
This has always been an important motivation to develop new and more effective methods in bridging the technological gap.
Effective methods developed over the years include (1) chemical composition modification of work surface, (2) cryogenic cooling
the tool-work interface, and (3) ultrasonic vibration-assisted machining.
The introduction of electroless nickel plating on materials that are not possible to be processed with diamond turning is the most
commercially successful method, despite nickel is not diamond-turnable due to the two unpaired d electrons that induce severe tool
wear (83) Still, electroless nickel can be effectively turned with diamond tools, and the corresponding diamond tool life and surface
quality are comparable to those when using copper or aluminium alloys (86). According to Paul et al. (83), the unpaired d-shell
electrons of nickel will get paired with the p-shell electrons of phosphorus, resulted in fewer free d-shell electrons to react with the
carbon atoms in diamond tools. Thus, chemical wear is substantially reduced, and will get further reduced by increasing phosphorus
content. With this understanding, Brinksmeier et al. (87) introduced a nitriding process to modify the chemical compositions of
steel surfaces to achieve a similar effect with electroless nickel. However, Wang et al. (19) found a negative effect on diamond tools
after nitriding treatment. Although the chemical wear is reduced, microchipping on the diamond tool edge is more easily induced
due to the significant increase in workpiece surface hardness after nitriding.
In addition, it is possible to reduce the chemical wear by cooling the diamond-cutting process as the rate of all chemical reactions
increases exponentially with temperature. Evans et al. (82) applied cryogenic temperature in diamond turning of stainless steel, and
the tool wear condition was found to be drastically improved. Brinksmeier et al. (87) did a similar study on cryogenic diamond
turning of steel and found that tool wear is reduced with the decrease of tool–workpiece temperature. Although cryogenic
machining performs well in suppressing chemical tool wear, it is still not feasible for application in ultraprecision machining,
because the induced thermal drifting is detrimental to achieving high-accuracy form profile of machined parts.
Among the various methods that suppress tool wear development, ultrasonic vibration-assisted machining has been found to be
an effective and practical solution for industrial application due to the economical hardware involved and absolute free of negative
effects on form accuracy, in return for production efficiency and extended tool life (88). For the underlying reason of reduced
diamond tool wear, researchers attributed it to the experimentally measured reduced cutting force, which they believe will lead to
smaller heat generation and reduced temperature in the vibration-assisted machining process (89). This explanation seems to be
reasonable because an increased temperature usually leads to a higher chemical reaction rate and hence a higher wear rate of
diamond tools. But this may require further investigation following Mitrofanov et al. (90) found that the tool tip temperature in
ultrasonic vibration-assisted cutting of Inconel 718 is 15% higher than conventional cutting.

11.09.8 Summary and Outlook

This chapter attempts to provide a review on diamond turning, an important materials processing technology for ultraprecision
applications. Substantial progress has been jointly made in recent years, including machine tool technology, cutting-tool tech-
nology, process control and optimization, unconventional techniques innovation, and fundamental research in cutting mechanics.
Despite the growing industrial adoption of diamond turning, the following five areas demand for further research and development:
l Affordable machine tools for small- and medium-sized parts with comparable levels of accuracy
l Customized diamond cutters with a special edge profile to generate complex surfaces and textures
l Quality improvement in form accuracy, surface finishing, burr generation, and subsurface deformation
l Productivity enhancement through simplified machine setup, tool wear suppression, and process automation
l Process innovation to generate complex surfaces and features on difficult-to-machine materials.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Diamond Turning 219

References

1. Ramsden, J. Description of an Engine for Dividing Straight Lines on Mathematical Instruments: J. Nourse, 1779. http://books.google.co.in/books/about/Description_of_an_
engine_for_dividing_st.html?id=ovm-nQEACAAJ&redir_esc=y.
2. Venkatesh, V. C.; Izman, S. Precision Engineering; McGraw-Hill Professional Publishing: New Delhi, 2008.
3. Rajagopal, K. Engineering Physics; PHI Learning Pvt. Ltd: New Delhi, 2007. http://books.google.co.in/books?id=vdXHcub8fRAC&pg=PR1&source=gbs_selected_
pages&cad=3#v=onepage&q&f=false.
4. Vleeschdrager, E. Hardness 10, Diamond: History, Diamond Cutting, Trade; Continental Publishing, 1998 (fr 1 to 4).
5. Bryan, J. B., et al. Diamond Turning of Parabolic Mirrors, 1974, pp 39–44; (fr 4 to 5). http://spie.org/x648.html? product_id=953772.
6. Moore. Nanotech HDL-2000, 2013, Available from: http://www.nanotechsys.com/machines/nanotech-hdl-2000-horizontal-drum-lathe/
7. Davies, M. A., et al. Application of Precision Diamond Machining to the Manufacture of Micro-photonics Components. Proc. SPIE 2003, 5183 (fr 2 to 1).
8. Lawrence Livermore National Laboratory. Available from: https://www.llnl.gov/str/April01/Klingmann.html.
9. Precitech. Available from: http://www.precitech.com/about/company_history.html.
10. Moore. Available from: http://www.nanotechsys.com/company-profile/.
11. Precitech. Nanoform 700 Ultra, 2013, Available from: http://www.precitech.com/products/nanoform700ultra/nanoform_700_ultra.html.
12. Moore. Nanotech 350FG, 2013, Available from: http://www.nanotechsys.com/machines/nanotech-350fg-freeform-generator/.
13. Lucca, D. A.; Seo, Y. W.; Komanduri, R. Effect of Tool Edge Geometry on Energy Dissipation in Ultraprecision Machining. CIRP Ann. – Manuf. Technol. 1993, 42 (1), 83–86.
14. Fang, F. Z.; Chen, L. J. Ultra-precision Cutting for ZKN7 Glass. CIRP Ann. – Manuf. Technol. 2000, 49 (1), 17–20.
15. Li, X. P., et al. Nano-precision Measurement of Diamond Tool Edge Radius for Wafer Fabrication. J. Mater. Process. Technol. 2003, 140 (1–3), 358–362.
16. Asai, S., et al. Measuring the Very Small Cutting-edge Radius for a Diamond Tool Using a New Kind of SEM Having Two Detectors. CIRP Ann. – Manuf. Technol. 1990, 39 (1),
85–88.
17. Zong, W. J., et al. Analysis for the Wear Resistance Anisotropy of Diamond Cutting Tools in Theory and Experiment. J. Mater. Process. Technol. 2010, 210 (6–7),
858–867.
18. Pramanik, A., et al. Ultraprecision Turning of Electroless Nickel: Effects of Crystal Orientation and Origin of Diamond Tools. Int. J. Adv. Manuf. Technol. 2009, 43 (7–8), 681–689.
19. Wang, Y., et al. Investigation of Tool Wear Suppression in Ultraprecision Diamond Machining of Die Steel. Precis. Eng. 2011, 35 (4), 677–685.
20. Contour. Contour Cutting Edge July 2012 News Bulletin, 2012.
21. Ernst, H.; Merchant, M. E. Chip Formation, Friction and High Quality Machined Surfaces. Surf. Treat. Met. 1941, 29, 299–378.
22. Nakayama, K.; Tamura, K. Size Effect in Metal-cutting Force. J. Eng. Ind. February 1968, 90, 119–126. https://www.google.co.in/search?q=Nakayama%2C+K.%3B+Tamura
%2C+K.+Size+Effect+in+Metal-cutting+Force.+J.+Eng.+Ind.+February+1968%2C+119%E2%80%93126&oq=Nakayama%2C+K.%3B+Tamura%2C+K.
+Size+Effect+in+Metal-cutting+Force.+J.+Eng.+Ind.+February+1968%2C+119%E2%80%93126&aqs=chrome..69i57.809j0j8&sourceid=chrome&espv=210&es_
sm=93&ie=UTF-8.
23. Shaw, M. C. A New Mechanism of Plastic Flow. Int. J. Mech. Sci. 1980, 22, 673–686.
24. Moriwaki, T.; Okuda, K. Machinability of Copper in Ultra-precision Micro Diamond Cutting. Ann. CIRP 1989, 38 (1), 115–118.
25. Lucca, D. A.; Rhorer, R. L.; Komanduri, R. Energy Dissipation in the Ultraprecision Machining of Copper. Ann. CIRP 1991, 40 (1), 69–72.
26. Liu, K.; Melkote, S. N. Material Strengthening Mechanisms and Their Contribution to Size Effect in Micro-cutting. J. Manuf. Sci. Eng. 2006, 128, 730–738.
27. Zorev, N. N. Metal Cutting Mechanics; Pergamon Press: Oxford, 1966.
28. Masuko, M. Fundamental Research on Metal Cutting. Bull. Jpn. Soc. Mech. Eng. 1956, 22, 371–377.
29. Albrecht, P. New Developments in the Theory of the Metal-cutting Process. J. Eng. Ind. November 1960, 82, 348–358. In https://www.google.co.in/search?q=Nakayama%
2C+K.%3B+Tamura%2C+K.+Size+Effect+in+Metal-cutting+Force.+J.+Eng.+Ind.+February+1968%2C+119%E2%80%93126&oq=Nakayama%2C+K.%3B+Tamura%2C+K.
+Size+Effect+in+Metal-cutting+Force.+J.+Eng.+Ind.+February+1968%2C+119%E2%80%93126&aqs=chrome..69i57.809j0j8&sourceid=chrome&espv=210&es_
sm=93&ie=UTF-8#es_sm=93&espv=210&q=New+Developments+in+the+Theory+of+the+Metal-cutting+Process.+J.+Eng.+Ind.+November+1960%2C+348%E2%80%93358.
30. Bitans, K.; Brown, R. H. An Investigation of the Deformation in Orthogonal Cutting. Int. J. Mach. Tool Des. Res. 1965, 5, 155–165.
31. Basuray, P. K.; Misra, B. K.; Lal, G. K. Transition from Ploughing to Cutting during Machining with Blunt Tools. Wear 1977, 43, 341–349.
32. Komanduri, R.; Lucca, D. A.; Tani, Y. Technological Advances in Fine Abrasive Processes. Ann. CIRP 1997, 46 (2), 545–596.
33. Fang, F. Z.; Wu, H.; Liu, Y. C. Modelling and Experimental Investigation on Nanometric Cutting of Monocrystalline Silicon. Int. J. Mach. Tools Manuf. 2005, 45, 1681–1686.
34. Li, X. P.; Cai, M. B.; Liu, K.; Rahman, M. Characteristics of Ductile Mode Chip Formation in Nanoscale Cutting of Brittle Materials. Int. J. Abrasive Technol. 2007, 1 (1), 37–58.
35. Finnie, I. A Comparison of Stress Strain Behavior in Cutting with That in other Materials Tests. Int. Res. Prod. Eng. 1963, 76–82.
36. Malkin, S.; Joseph, N. Minimum Energy in Abrasive Processes. Wear 1975, 32, 15–23.
37. Abdelmoneim, M. E.; Scrutton, R. F. Tool Edge Roundness and Stable Build-up Formation in Finish Machining. J. Eng. Ind. November 1974, 1258–1266.
38. Taminiau, D. A.; Dautzenberg, J. H. Bluntness of the Tool and Process Forces in High-precision Cutting. Ann. CIRP 1991, 40 (1), 65–68.
39. Komanduri, R.; Chandrasekaran, N.; Raff, L. M. Effect of Tool Geometry in Nanometric Cutting: A Molecular Dynamics Simulation Approach. Wear 1998, 219, 84–97.
40. Moriwaki, T.; Sugimura, N.; Luan, S. Combined Stress, Material Flow and Heat Analysis of Orthogonal Micromachining of Copper. Ann. CIRP 1993, 42 (1), 75–78.
41. Kim, K. W.; Lee, W. Y.; Sin, H.-C. A Finite Element Analysis for the Characteristics of Temperature and Stress in Micro-machining Considering the Size Effect. Int. J. Mach.
Tools Manuf. 1999, 39, 1507–1524.
42. Lucca, D. A.; Seo, Y. W. Effect of Tool Edge Geometry on Energy Dissipation in Ultraprecision Machining. Ann. CIRP 1993, 42 (1), 83–86.
43. Lucca, D. A.; Seo, Y. W.; Rhorer, R. L. Energy Dissipation and Tool-Workpiece Contact in Ultra-precision Machining. STLE Tribol. Trans. 1994, 37, 651–655.
44. Shaw, M. C. A New Theory of Grinding. Mech. Chem. Eng. Trans. 1972, 8 (1), 73–78.
45. Liu, K.; Melkote, S. N. Finite Element Analysis of the Influence of Tool Edge Radius on Size Effect in Orthogonal Micro-cutting Process. Int. J. Mech. Sci. 2007, 49, 650–660.
46. Ikawa, N.; Shimada, S.; Tanaka, H.; Ohmari, G. Atomistic Analysis of Nanometric Chip Removal as Affected by Tool-Work Interaction in Diamond Turning. Ann. CIRP 1991b, 40,
551–554.
47. Shimada, S.; Ikawa, N.; Tanaka, H.; Ohmari, G.; Uchikoshi, J.; Yoshinage, H. Feasibility Study on Ultimate Accuracy in Microcutting Using Molecular Dynamics Simulation. Ann.
CIRP 1993, 42 (1), 91–94.
48. Yuan, Z. J.; Zhou, M.; Dong, S. Effect of Diamond Tool Sharpness on Minimum Cutting Thickness and Cutting Surface Integrity in Ultraprecision Machining. J. Mater. Process.
Technol. 1996, 62, 327–330.
49. Son, S. M.; Lim, H. S.; Ahn, J. H. Effects of Friction Coefficient on the Minimum Cutting Thickness in Micro Cutting. Int. J. Mach. Tools Manuf. 2005, 45, 529–535.
50. Weule, H.; Hüntrup, V.; Tritschler, H. Micro-cutting of Steel to Meet New Requirements in Miniaturization. Ann. CIRP 2001, 50 (1), 61–64.
51. Vogler, M. P.; DeVor, R. E.; Kapoor, S. G. Microstructure-level Force Prediction Model for Micro-milling of Multi-phase Materials; ASME International Mechanical Engineering
Congress and Exposition: New York, 2001, 3–10.
52. Kim, C.-J.; Bono, M.; Ni, J. Experimental Analysis of Chip Formation in Micro-milling. Trans. NAMRI/SME 2002, 30, 1–8.
53. Kim, C.-J.; Mayor, J. R.; Ni, J. A Static Model of Chip Formation in Microscale Milling. J. Manuf. Sci. Eng. 2004, 126, 710–718.
54. Lai, X.; Li, H.; Li, C.; Lin, Z.; Ni, J. Modelling and Analysis of Micro Scale Milling Considering Size Effect, Micro Cutter Edge Radius and Minimum Chip Thickness. Int. J. Mach.
Tools Manuf. 2008, 48, 1–14.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
220 Diamond Turning

55. Vogler, M. P.; DeVor, R. E.; Kapoor, S. G. On the Modelling and Analysis of Machining Performance in Micro-endmilling, Part I: Surface Generation. J. Manuf. Sci. Eng. 2004,
126 (4), 684–693.
56. Liu, X.; Jun, M. B. G.; DeVor, R. E.; Kapoor, S. G. Cutting Mechanisms and Their Influence on Dynamic Forces, Vibrations and Stability in Micro-endmilling. In ASME
International Mechanical Engineering Congress and Exposition; 2004. Anaheim, CA, Paper No. 62416.
57. Ueda, K.; Iwata, K.; Nakayama, K. Chip Formation Mechanism in Single Crystal Cutting of b-Brass. CIRP Ann. – Manuf. Technol. 1980, 29 (1), 41–46.
58. Moriwaki, T., et al. A Study on Orthogonal Micromachining of Single Crystal Copper. Trans. NAMRI/SME 1991, 19, 177–183.
59. To, S.; Lee, W. B.; Chan, C. Y. Ultraprecision Diamond Turning of Aluminium Single Crystals. J. Mater. Process. Technol. 1997, 63 (1–3), 157–162.
60. Yuan, Z. J., et al. Effect of Crystallographic Orientation on Cutting Forces and Surface Quality in Diamond Cutting of Single Crystal. CIRP Ann. – Manuf. Technol. 1994, 43 (1),
39–42.
61. Lawn, B. R.; Jensen, T.; Arora, A. Brittleness as an Indentation Size Effect. J. Mater. Sci. 1976, 11 (3), 573–575.
62. Giovanola, J. H.; Finnie, I. On the Machining of Glass. J. Mater. Sci. 1980, 15 (10), 2508–2514.
63. Puttick, K. E., et al. Single-point Diamond Machining of Glasses. Proc. R. Soc. Lond. 1989, 426, 19–30.
64. Venkatachalam, S.; Li, X.; Liang, S. Y. Predictive Modeling of Transition Undeformed Chip Thickness in Ductile-regime Micro-machining of Single Crystal Brittle Materials.
J. Mater. Process. Technol. 2009, 209 (7), 3306–3319.
65. Arif, M., et al. A Predictive Model of the Critical Undeformed Chip Thickness for Ductile–Brittle Transition in Nano-machining of Brittle Materials. Int. J. Mach. Tools Manuf.
2013, 64, 114–122.
66. Woon, K. S.; Rahman, M. Extrusion-like Chip Formation Mechanism and Its Role in Suppressing Void Nucleation. CIRP Ann. – Manuf. Technol. 2010, 59 (1), 129–132.
67. Metcut. Machining Data Handbook, 3rd ed.; Metcut Research Associates Inc.: Cincinnati, 1980.
68. Vyas, A.; Shaw, M. C. Mechanics of Saw-tooth Chip Formation in Metal Cutting. J. Manuf. Sci. Eng. 1999, 121 (2), 163–172.
69. Sugano, T., et al. Diamond Turning of an Aluminum Alloy for Mirror. CIRP Ann. – Manuf. Technol. 1987, 36 (1), 17–20.
70. Matsumoto, Y.; Hashimoto, F.; Lahoti, G. Surface Integrity Generated by Precision Hard Turning. CIRP Ann. – Manuf. Technol. 1999, 48 (1), 59–62.
71. Brinksmeier, E., et al. Residual Stresses – Measurement and Causes in Machining Processes. CIRP Ann. – Manuf. Technol. 1982, 31 (2), 491–510.
72. Zhang, H.; Zhang, X. Factors Affecting Surface Quality in Diamond Turning of Oxygen-free High-conductance Copper. Appl. Opt. 1994, 33 (10), 2039–2042.
73. Zong, W. J., et al. FEM Optimization of Tool Geometry Based on the Machined Near Surface’s Residual Stresses Generated in Diamond Turning. J. Mater. Process. Technol.
2006, 180 (1–3), 271–278.
74. Lee, K.; Dornfeld, D. An Experimental Study on Burr Formation in Micro Milling Aluminium and Copper. Trans. NAMRI/SME 2002, 30, 1–8.
75. Liu, X.; DeVor, R. E.; Kapoor, S. G. Model-based Analysis of the Surface Generation in Microendmilling – Part I: Model Development. J. Manuf. Sci. Eng., Trans. ASME 2007,
129 (3), 453–460.
76. Kim, J.-D.; Kim, D. S. Theoretical Analysis of Micro-cutting Characteristics in Ultra-precision Machining. J. Mater. Process. Technol. 1995, 49 (3–4), 387–398.
77. Kong, M. C., et al. A Study of Materials Swelling and Recovery in Single-point Diamond Turning of Ductile Materials. J. Mater. Process. Technol. 2006, 180 (1–3), 210–215.
78. Nakayama, K.; Arai, M. Burr Formation in Metal Cutting. CIRP Ann. – Manuf. Technol. 1987, 36 (1), 33–36.
79. Dornfeld, D.; Min, S.; Takeuchi, Y. Recent Advances in Mechanical Micromachining. CIRP Ann. – Manuf. Technol. 2006, 55 (2), 745–768.
80. Fang, F. Z.; Liu, Y. C. On Minimum Exit-burr in Micro Cutting. J. Micromech. Microeng. 2004, 14 (7), 984.
81. Lane, B. M., et al. Diamond Tool Wear when Machining Al6061 and 1215 Steel. Wear 2010, 268 (11–12), 1434–1441.
82. Evans, C.; Bryan, J. B. Cryogenic Diamond Turning of Stainless Steel. CIRP Ann. – Manuf. Technol. 1991, 40 (1), 571–575.
83. Paul, E., et al. Chemical Aspects of Tool Wear in Single Point Diamond Turning. Precis. Eng. 1996, 18 (1), 4–19.
84. Thornton, A. G.; Wilks, J. Tool Wear and Solid State Reactions during Machining. Wear 1979, 53 (1), 165–187.
85. Shewmon, P. G. Diffusion in Solids; McGraw-Hill: New York, 1963.
86. Casstevens, J. M.; Daugherty, C. E. Diamond Turning Optical Surfaces on Electroless Nickel, 1978, pp 109–113.
87. Brinksmeier, E.; Gl, R.; Osmer, J. Ultra-precision Diamond Cutting of Steel Molds. CIRP Ann. – Manuf. Technol. 2006, 55 (1), 551–554.
88. Moriwaki, T.; Shamoto, E. Ultraprecision Diamond Turning of Stainless Steel by Applying Ultrasonic Vibration. CIRP Ann. – Manuf. Technol. 1991, 40 (1), 559–562.
89. Zhou, M., et al. Vibration-assisted Precision Machining of Steel with PCD Tools. Mater. Manuf. Process. 2003, 18 (5), 825–834.
90. Mitrofanov, A. V.; Babitsky, V. I.; Silberschmidt, V. V. Thermomechanical Finite Element Simulations of Ultrasonically Assisted Turning. Comput. Mater. Sci. 2005, 32 (3–4),
463–471.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 11.10 High-Speed Machining
Z Wang, Makino Inc., Mason, OH, USA
M Rahman, National University of Singapore, Singapore
 2014 Elsevier Ltd. All rights reserved.

11.10.1 Introduction 221

11.10.2 Typical Cutting Tool Materials Used in High-Speed Machining 224
11.10.2.1 Coated and Uncoated Carbide Tools 224
11.10.2.2 Advanced Tool Materials 225
11.10.3 Development of Machine Tools for High-Speed Machining 225
11.10.3.1 High-Speed Spindles 226
11.10.3.2 Feed Drive and Its Control System 227
11.10.3.3 Tool Holder and Clamping System 227
11.10.3.4 Advanced Cooling Technologies for High-Speed Application 228
11.10.4 Modeling and Simulation of High-Speed Machining 229
11.10.4.1 Cutting Force Models for Machining Processes 229
11.10.4.1.1 Analytical Models 229
11.10.4.1.2 Numerical Models 230
11.10.4.2 Cutting Temperature Models for Machining Processes 230
11.10.4.3 Modeling of Cutting Forces in High-Speed Machining 231
11.10.4.3.1 Overview of Conventional Orthogonal Cutting Force Models 231
11.10.4.3.2 Modeling of Flow Stress Properties of Work Materials 233
11.10.4.3.3 Application of the JohnsonCook Model in the Prediction of Cutting Forces 235
11.10.4.4 Modeling of Cutting Temperature in High-Speed Machining 236
11.10.4.5 Experimental and Simulation Study of the Cutting Force and Temperature 238
11.10.4.5.1 Experimental and Simulation Study of the Cutting Force 238
11.10.4.5.2 Prediction of the Cutting Temperature 239
11.10.5 Economics of High-Speed Machining 242
11.10.5.1 Optimization Model of High-Speed Milling 242
11.10.5.1.1 Objective Functions 242
11.10.5.1.2 Constraints 244
11.10.5.2 Implementation of the Optimization Algorithm 245
11.10.6 Typical Experimental Studies on High-Speed Machining 246
11.10.6.1 High-Speed Machining of Aluminum Aerospace Alloys 246
11.10.6.2 High-Speed Machining of Titanium Alloys  Case Study 1 247
11.10.6.3 High-Speed Machining of Titanium Alloys  Case Study 2 249
11.10.7 Conclusions 252
References 252

11.10.1 Introduction

The concept of high-speed machining (HSM) was first introduced by C. Salomon back in the 1930s. Through a series of tests done
by Salomon, it was found that the cutting temperature reaches its maximum value at a certain cutting speed, and with a further
increase of the cutting speed, machining temperatures start to drop, as shown in Figure 1 (1).
Salomon’s fundamental research showed that there is a certain range of cutting speeds where machining cannot be done due to
excessively high temperatures. For this reason, HSM can also be referred to as cutting speeds beyond that range. In compliance with
modern knowledge, some researchers define high-speed machining as machining whereby conventional cutting speeds are exceeded
by a factor of 510, as shown in Figure 2 (1).
With the wide use of computerized numerical control (CNC) machines together with high-performance computer-aided design/
computer-aided manufacturing (CAD/CAM) systems, HSM has been demonstrated to have superior advantages over other rapid
manufacturing techniques. In addition to increased productivity, HSM is capable of generating high-quality surfaces, burr-free
edges, and virtually stress-free components after machining, and it can be used to machine thin-wall workpieces because the
cutting forces involved in HSM conditions are lower. Another significant advantage of high-speed machining is the minimization of
thermal effects on the machined parts. Most of the cutting heat is taken away by the cutting chips, reducing thermal warping and
increasing the life of the cutting tool. In many cases, the need for a cooling fluid is eliminated. Furthermore, the elimination of
cutting fluids reduces subsequent pollution to the environment and aids in the recovery and recycling of such expensive materials as

Comprehensive Materials Processing, Volume 11 http://dx.doi.org/10.1016/B978-0-08-096532-1.01113-4 221

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
222 High-Speed Machining

Figure 1 Machining temperature in milling at high cutting speeds. Reproduced from Schulz, H. Why High Speed Cutting (HSC). In Proceedings of the
International Conference on High Speed machining; Nanjing: P.R. China, April 2004; pp 120.

fibre-reinforced
plastics
aluminum
alloys
ge
an

bronze,
lr

brass
e
ng
na
tio

ra

cast iron
e
en

ng
tio
nv

-ra
i
ns
co

steel
tra

HS

titanium
alloys
nickel based
alloys
10 100 1000 10000
cutting speed VC (m min–1)

Figure 2 Cutting speed area depends on material. Reproduced from Schulz, H. Why High Speed Cutting (HSC). In Proceedings of the International
Conference on High Speed machining; Nanjing: P.R. China, April 2004; pp 120.

aluminumlithium alloys. Since HSM has so many advantages, it is widely used in the aerospace industry, the automotive industry,
and the precision engineering industry for machine tools, equipment, and tooling used in the manufacture of domestic appliances,
optics, etc.
In typical HSM applications, a very large amount of strain takes place in the region along the shear plane in a very short interval
of time. Depending on the ductility of work materials, two types of chip formations have been observed: continuous and shear-
localized. HSM is more prone to produce continuous chips for aluminum alloys because of their high thermal diffusivity and
good ductility. Often the continuous chips have considerable strength, and control of chip shape is one of the problems confronting
machinists and tool designers. Therefore, cutting parameters should be carefully chosen to break the long chips because it is much
easier to handle short chips and it may result in less tool wear. With the increase in the cutting speed, the cutting temperature
increases, which then softens the work material. Consequently, the cutting forces are reduced, which enables increased material
removal rates and higher feed rates. Previous research has found that in HSM of aluminum alloys, about 75% of the generated heat
exits with the chip, 20% exits through the tool, and 5% exits through the workpiece.
Although high-speed milling of aluminum has been applied successfully in industries for more than a decade, high-speed
applications on difficult-to-cut materials, such as titanium alloys, are still relatively new. Titanium alloys have recently been
widely used in the aerospace, biomedical, and petroleum industries because of their good strength-to-weight ratio and superior

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 High-Speed Machining 223

Table 1 Mechanical properties of three widely used materials in metal cutting

Material Thermal conductivity (W m1 K) Specific heat capacity (J kg1 C) Tensile strength (MPa)

Aluminum 7075-O 173.0 960 96.5

AISI 4340 steel 44.5 475 786
Ti6Al4V STA 6.7 526.3 1100
WC-Co (610%) 6080 200400 1440
Reproduced from MatWeb Material Property Data, http://www.matweb.com/, 2011.

corrosion resistance. However, it is very difficult to machine them due to their poor machinability. Boeing’s military aircraft group
has begun to apply its expertise with aluminum toward faster milling of titanium in St. Louis, Missouri. Jerry Halley was an
engineering researcher there who has helped the company to realize more effective machining of both materials. Halley concluded
that compared to aluminum, titanium imposes certain constraints. Speed is constrained as heat builds up more quickly. But within
those constraints, there is still considerable room for faster cutting.
Diffusive tool wear is one of the dominant wear patterns for WC-Co tools at high cutting speed; it is typically characterized by
diffusion of the tool material into the workpiece/chip. The diffusion wear is temperature-dependent, and high temperatures at the
toolchip interface can cause tool wear to occur. Therefore, the maximum allowable cutting temperature places an upper bound on
the allowable cutting speed to achieve reasonable tool life. Consequently, the maximum available spindle speed is limited, even
though there are large stable zones observed at the right of the stability lobe diagrams in the spindle speed zone. High temperatures
are particularly problematic when the work material has low thermal conductivity. The mechanical properties of three types of work
materials and WC-Co tool material are listed in Table 1. The thermal conductivity of Ti6Al4V is approximately 25 times and
10 times less than 7075 aluminum and WC-Co, respectively. Such a low thermal conductivity of Ti6Al4V causes the heat to remain
at the toolchip interface at high cutting speeds, rather than being conducted away into the workpiece and cutting chips for
aluminum alloys, which accelerates diffusive wear. Thus it is difficult to machine Ti6Al4V at high cutting speed due to the associated
excessive wear rates.
As mentioned above, the cutting temperature at high cutting speed is the dominant effect that determines whether high-speed
application is feasible. It is necessary to investigate how the temperature affects the strength of the cutting tool and the work
materials. Figure 3 shows the mechanical strength of frequently used tools and work materials at different temperatures. Clearly
with the increase of temperature, the yield strength for all these materials is gradually reduced. Once the temperature reaches 800  C
for WC-Co, its mechanical strength is sharply reduced; this temperature is called the reaction temperature. As mentioned before,
there is higher heat concentrated on the cutting edge during the machining of titanium alloys. Thus, the higher the work material
temperature, the easier it is to remove the material. However, at high temperature the mechanical strength of the tool materials is
weakened. There will be an ideal temperature for high-performance machining of titanium alloys. Therefore, it is necessary to
control the cutting temperature to soften the work material and maintain the mechanical strength of the tool materials in order to
achieve high-speed machining of titanium alloys.

Figure 3 Yield strength of tools and work materials at different temperatures. Reproduced from Tlusty, G. Manufacturing Processes and Equipment;
Prentice Hall: USA, 2000.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
224 High-Speed Machining

King and Vaughn (4) stated that as the cutting speed increases above the conventional speed range, new dynamic effects are
encountered in the cutting process, and Taylor’s empirical equations are no longer adequate since they are not velocity-dependent. It
is necessary to investigate the mechanism of HSM of titanium alloys and establish comprehensive models to describe this process.
The objective of this chapter is to give an overview of the more significant contributions to HSM of aerospace alloys, which can
provide a fundamental understanding of the mechanisms that prevail in the field of HSM of aerospace alloys. In general, HSM may
include several processes, including high-speed turning, high-speed milling, high-speed drilling, etc. This chapter focuses mainly on
the most widely used milling processes.

11.10.2 Typical Cutting Tool Materials Used in High-Speed Machining

A wide range of tool materials is available for HSM of aluminum alloys. Cermets (WC-Co), cubic boron nitride (CBN), and
polycrystalline diamonds (PCD) are most widely used as both solid tools and indexable inserts. To reduce the friction coefficient
between the toolchip interface, tools used in HSM are designed with high positive cutting rakes and polished smooth rake faces.
For abrasive castings that contain hard particles, it is favorable to use PCD tools. On the other hand, it is better to use uncoated solid
carbides for thin-walled parts due to their high rigidity and sharp edges, which reduce cutter forces and vibration. In all cases, it is
recommended that the needed diameter and length of tools should be as short as possible. Moreover, all tools should be regularly
inspected for cracks and wear. CBN and PCD tools will be more widely used to achieve longer tool life in the future. Recently, newer
nanostructured tool materials have provided even higher strength and toughness, therefore they have a greater potential to be used
as tool materials for HSM applications.
During the machining of titanium alloys with conventional carbide tools, tool wear progresses rapidly because of their low
thermal conductivity and high chemical reactivity, resulting in higher cutting temperature and strong adhesion between the tool and
the work material (5). Titanium alloys are generally difficult to machine at cutting speeds of over 30 m min1 with high-speed steel
(HSS) tools, and over 60 m min1 with cemented tungsten carbide (WC) tools, resulting in very low productivity. The performance
of conventional tools is poor when machining Ti6Al4V. In 1955, Siekmann (5) pointed out that machining of titanium and its
alloys would always be a problem, regardless of the techniques employed to transform the metal into chips. The poor machinability
of titanium and its alloys have led many large companies (e.g., Rolls-Royce and General Electric) to invest much in developing
techniques to minimize machining cost (6).
The inability to improve cutting tool performance for the machining of titanium alloys by developing new coating materials has
been very frustrating. Likewise, the exploration of new combinations of speeds, feeds, and depths has led to very little improvement
in productivity. Some developments of interest include specially designed turning tools and milling cutters, along with the use of
a special end-mill pocketing technique. With the evolution of a number of new cutting tool materials, advanced tool materials such
as CBN and PCD are being considered to achieve HSM of titanium alloys.

11.10.2.1 Coated and Uncoated Carbide Tools

Ezugwu et al. (7) attempted to machine titanium alloys in an inert argonenriched environment, so that the chemical reaction at
the toolchip and toolworkpiece interfaces may be minimized when machining titanium alloys at higher cutting speeds. They
carried out machining trials with uncoated carbide (ISO K10 grade) tools in an argon-enriched environment under cutting
conditions typical of finished turning operations. For comparison, experiments with a conventional coolant supply were also done
under the same cutting conditions. Experimental results showed that machining in an argon-enriched environment led to a shorter
tool life relative to a conventional coolant supply. Nose wear was the dominant tool failure in all the cutting conditions inves-
tigated. Ezugwu et al. attributed this to argon being a poor conductor of heat, that is, heat generated during machining tends to
concentrate on the cutting region and accelerate tool wear. In addition, argon has poor lubrication characteristics, which leads to
an increase in friction at the cutting interfaces during machining and an increase in cutting forces required for efficient shearing of
the workpiece (7).
Wang et al. (8) used WC inserts for orthogonal continuous and interrupted cutting at conventional and high cutting speeds. In
their study, the tool material H1 (Sumitomo’s tool material reference number) has fine-carbide grain and maintains high wear
resistance. The mechanical properties of H1 are given in Table 2 (9). The properties of CBN and ceramic materials from Ref. (10)
were also cited to highlight the fracture toughness of cemented carbide H1, as shown in Table 2. Table 2 also shows that cemented
carbide H1 has a higher transverse rupture strength (TRS) value and fracture toughness, although its hardness value is lower. More
importantly, cemented carbide tools are much cheaper than CBN and ceramic tools. Therefore, the higher TRS values and greater
thermal conductivity of H1 make it more suitable for interrupted cutting. In their study, cutting performances under dry cutting
conditions, with minimum quantity lubrication (MQL), and with flood coolant have been investigated. For continuous cutting, dry
cutting has been found to be only effective at lower cutting speed, and MQL is a more economical alternative cooling lubricant than
flood coolant. At a higher feed rate of 0.15 mm per rev. and a higher cutting speed in interrupted cutting, cutting forces under MQL
are weaker than those under flood coolant, and MQL and flood coolant have similar effects on the thrust forces. They indicated that
MQL was an effective alternative approach to flood coolant during high-speed turning of Ti6Al4V. Based on the investigation of the
cutting forces with the different coolant lubricants, the mean friction coefficient in the sliding region at the toolchip interface has
been obtained and used in a finite element method (FEM) to simulate the deformation process of Ti6Al4V during turning. From the

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 High-Speed Machining 225

Table 2 Mechanical and thermal properties of BCBN, CBN, carbide, and ceramics

Properties BCBN CBN Cemented carbide 6% Co Ceramic Si3 N4 Hot-press

CBN content (vol.%) >99.9 8590  

CBN grain size (micron) <0.5 13  
Other constituents -(comp. hBN) Binder (Co etc.)  
Processing method Direct conv. CBN þ Binder  
Hardness (GPa) Room temperature 5055 3540 15.69 21.58
1000  C 20 12  
TRS*(GPa) Room temperature 1.35 1.40 2.20 0.793
1000  C 1.60 0.55  
Thermal conductivity (W m1 K) 360400 100130 95 29
Thermal stability (K in air) 1620 1270  
Reproduced from Sumitomo Electric Industries, Performance Cutting Tools; Sumitomo Electric Industries: Japan, 2000.

FEM simulation and Oxley’s predictive machining theory, cutting forces have been estimated under different cooling supply
strategies and verified experimentally.
Lei and Liu (11) developed a new generation of a driven rotary lathe tool for high-speed machining of a titanium alloy Ti6Al4V.
In their study, high-speed cylindrical turning experiments were conducted using the driven rotary tool (DRT) and a stationary
cutting tool with round tungsten carbide inserts. From the experimental results, they found that DRT can significantly increase tool
life, and the increase in tool life with DRT is more than 60 times compared to that with a stationary cutting tool under certain
conditions. The effects of the rotational speed of the insert were also investigated experimentally in their study. Cutting forces were
found to decline slightly with an increase in the rotational speed, and tool wear was observed to increase with the rotational speed in
a certain speed range.

11.10.2.2 Advanced Tool Materials

Muller et al. (12) used a fiber-optic two-color pyrometer with high spatial and temporal resolution to measure temperatures at an
external turning process at conventional and high cutting speeds. SiC-whisker reinforced oxide ceramic inserts were used for high-speed
turning of Ti6Al4V. Based on the experiments, they observed that the measured temperature in turning titanium alloy Ti6Al4V showed
very good reproducibility at conventional cutting speeds. But at much higher speed (such as  5 m s1), the measurement conditions
became more and more difficult due to tool wear or failure, which caused a widespread increase of the measured temperature.
Zoya and Krishnamurthy (13) used CBN tools for high-speed turning of titanium alloys and evaluated the machining perfor-
mance. They concluded that the machining of titanium alloys is a thermally dominant process, and a critical temperature of 700  C
can be a criterion for tool life. They also found that a good surface finish can be achieved with a cutting speed of 185 m min1, and
a cutting speed range of 185220 m min1 can be recommended for the machining of titanium alloys with CBN tools. During the
machining of titanium alloys, deformation at the cutting nose of CBN tools was observed, and it was claimed that wear of CBN tools
can also be due to diffusion wear.
Ezugwu et al. (14) evaluated the cutting performance of different CBN tool grades in finished turning Ti6Al4V (IMI 318) alloy
under high-speed cutting conditions of up to 250 m min1, with various coolant supplies. In their study, tool wear, failure modes,
cutting and feed forces, and surface roughness of machined surfaces were monitored and used to access the performance of the
cutting tools. For comparison, uncoated carbide tools were also used to machine Ti6Al4V at a speed of 150 m min1. They observed
that the cutting performance of CBN tools is poor relative to uncoated carbide tools in terms of tool life at the cutting conditions
investigated. They also found that the possible reason for this is rapid notching and excessive chipping of the cutting edge associated
with a relatively high diffusion wear rate that tends to weaken the bond strength of the tool substrate. It was also found that an
increase in the CBN content of the cutting tool tended to accelerate notch wear rate, consequently diminishing tool life under the
investigated cutting conditions.
Bhaumik et al. (15) used wBN-cBN composite tools to machine Ti6Al4V and investigated the wear mechanism of this type of
tool. Based on X-ray dot mapping of compositional analysis, they indicated that titanium existed on the crater area, and the
adherent layer might contain titanium in some compound form, such as titanium diboride or titanium nitride. When the
adherent materials were taken away, accelerated attrition was observed on both the rake and flank faces. Some tests were also
carried out with K20 grade cemented carbide tools to compare the machined surface quality and the swarf obtained with those
using composite tools. The tool life of cemented carbide tools was found to be limited by rapid cratering on the rake face and
deformation at the tool nose.

11.10.3 Development of Machine Tools for High-Speed Machining

There is a very close interaction between the HSM manufacturing technological process and the associated machine tools, their
components, and cutting tools. Today, a wide spectrum of HSM machine tools is commercially available. In general, the trend for

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
226 High-Speed Machining

higher speeds is continuing, and standard machine tools have become faster in the wake of HSM machines. Spindle speeds of up to
12 000 min1 and feed rates up to 25 m min1 have meanwhile become state of the art, that is, it is already possible, in particular in
machining steel or cast iron, to reach the HSM range with the present standard CNC machine tools.
Higher accuracy at high speed of HSM machine tools requires machine components with high rigidity, thermal stability, and
damping capacity. To increase the strength and damping characteristics of machine foundations and bases, nodular cast iron is often
used. Also, to minimize inertia forces, the moving components (such as carriages) are made of strong and light materials such as
titanium aluminum and fiber-reinforced composites. To maintain the safety of the machine, a number of sensors (vibration and
temperature) have been used to monitor the machine’s infrastructure. Normally, HSM machine tools differ from conventional CNC
machines by the following essential points:
l High-frequency motor spindles
l Fast CNC controls
l High dynamic feed drives
l Light-weight construction
l Safety devices

The details of the recent development of the above key components in HSM application will be described in the following.

11.10.3.1 High-Speed Spindles

The machine tool spindle plays an important role in machining operations because it provides the cutting speed of the tool and is
part of the force chain between the machine tool structure and the tool or the workpiece. High-speed spindles are the heart of HSM
operations with typical ratings of 30 hp/30 000 rpm to 50 hp/20 000 rpm.
The typical elements of a spindle are the tool interface, the drawbar, the spindle shaft, the bearings, the driving system, the
cooling system, and the housing as shown in Figure 4. There are several types of driving systems, basically with a motor coupled,
directly or indirectly, to the spindle or with an integrated motor. High-speed spindles are usually driven by AC asynchronous motors
that are integrated with the spindle shaft, as shown in Figure 4. A solid-state frequency converter provides variable frequency and
voltage to control the speed of the motor. Bearings are mechanical elements designed to reduce friction between an axle and its
support. For this reason, the efficiency of the bearings is especially important for the spindle performance. For HSM machine tools,
the bearings used have to meet the demands of the spindle in terms of rotational speed, load capacity, and life. That is why the
rolling bearings are the most common solution. At high rotational speed, the vibration and heat are the main enemies of high-speed
spindles and the bearings. The bearings used by machine tool manufacturers have traditionally been metallic, but in recent years,
hybrid-ceramic bearings have been gaining popularity due to their improved performance and suitability for high-speed machining.
The inner and outer races are hardened steel, but the rolling elements are made of rigid and durable ceramics. Other bearing
technologies exist, including magnetic, hydrostatic, and air. More importantly, the bearing components are manufactured to achieve
close tolerances and high balance with independent cooling of the stator and bearings. In order to operate correctly, bearings must
be preloaded by applying a constant thrust load. The preload eliminates axial and radial motion, increases the bearing stiffness,

Power limits
Tooling interface: HSK
- rotor peripheral speed <250 m s–1
- rotor resonance frequency
- heat generation

Speed limited by bearings' peripheral speed Balancing limits

- ball bearings ( oil-jet lubrication ): n·dm <3×106 mm·min–1 high balance grades not
- magnetic bearings: peripheral speed <250 m s–1 achievable at very high
- hydrostatic bearings (water): n·dm <2.5×106 mm·min–1 speed
- aerostatic bearings: peripheral speed <600 m s–1

Figure 4 High-speed spindle. Reproduced from Schulz, H. Why High Speed Cutting (HSC). In Proceedings of the International Conference on High Speed
machining; Nanjing: P.R. China, April 2004; pp 1–20.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 High-Speed Machining 227

reduces runout, prevents ball skidding under high accelerations, and reduces contact angle variation at high speeds (16). However,
too much preload leads to excessive heat generation in running conditions and shortens the bearing life. For high-speed appli-
cations, a hydraulic or a pneumatic system can be used to optimize the preload according to the rotating speed. For spindles with
higher power and higher velocity such as 50 hp/50 000 rpm, more ceramic materials will be used to fabricate the bearing. Mean-
while, magnetic bearings may gain more prominence since their lower friction technology allows better adaptive control of the
spindle parameters during cutting operations.
To ensure the safety of machining operation, sensors are often used to monitor and control the performance metrics of the
spindle system, such as velocity, power, temperature, and vibration.

11.10.3.2 Feed Drive and Its Control System

Feed drives are used to position the machine tool components, such as the cutting tool and workpiece, to the desired location; hence
their positioning accuracy and speed determine the quality and productivity of machine tools. HSM machines require multiple axes
with easy motion, fast response, and accurate positioning. The servo ball-screw and linear drives are most commonly used as
machine tool feed drives. Recent advances in electrical motors, power electronics, computers, actuators, and sensors technology
have been pushing the speed limits of spindle and feed drives upward continuously. There are already spindles that reach over
10 000 rpm, which require feed drives traveling over 50 m min1 with 10 g acceleration rates. Micromachining applications use air
spindles up to 50 000 rpm, which require fast and precise feed drives. However, the ball screw drive with a rotary motor has certain
disadvantages in high-speed applications (17):
l Rotary motors have limitations on maximum rotation speed.
l Rotary motor drive systems are comprised of more than 20 parts that add inertia and cause lower efficiency.
l Motor couplings produce wind-up distortion, backlash, and hysteresis.
l Encoder couplings deflect during acceleration and deceleration.
l Backlash in a ball screw drive train limits the amount of gain that can be used to control axis position. (‘Gain’ is a measure of
responsiveness of control.)
Due to its limitation, the traditional ball screw drive is not suitable for high-speed applications. On the other hand, linear motors
are the latest technology for slide drives. A linear motor is basically a rotary motor that has been rolled out flat. With no coupling, no
transmission, and no ball screw, a linear motor provides excellent acceleration and deceleration rates. Use of linear motors as the
main drive technology in high-speed machines is already established and preferred over traditional rotary motor and ball screws for
moving the slides.
Another key enabler for HSM is the CNC controller and its ability to move the machine accurately enough at high speed,
especially for the parts with complicated geometries and an enormous amount of part programs. Sometimes the size of the
programs even reaches tens of megabytes. Storing and processing such a large amount of part programs presents several immediate
challenges to a system’s hardware, including data storage, memory requirements, processing, and communication speeds.
Controllers with ‘look-ahead’ features that control the present speed and acceleration or deceleration of the tool based on where the
tool is going to go are, by all accounts, just as important as driving the spindle at high rates. The standard look-ahead on advanced
CNC controllers is about 6080 blocks of G-Code ahead. One commercial controller package is designed specifically for HSM and
looks 180250 blocks ahead. For the same cutter path geometries, the new controller package is 1530% faster than the traditional
controller it replaces. Moreover, advanced interpolation algorithms, such as NURBS, will improve contouring accuracy while
reducing the size of conventional programs from 1/10th to 1/100th.

11.10.3.3 Tool Holder and Clamping System

In milling, the tool holder is the interface between the cutting tool and the spindle and it ensures the transmission of the rotational
movement of the spindle to the cutting tool. The selection of the tool holder is defined by the performance and possibilities of
a machine tool. It requires a simple system to clamp and unclamp the tool, allows automatic tool changing within the machine tool,
ensures a precise and rigid adjustment with the spindle, and it must guarantee a proper eccentricity and balancing. A defective tool
holder has negative effects on the dimensional precision and surface quality of the workpiece, and it can reduce the service life of the
cutting tool and the spindle (16). Nowadays, hollow tool holders are made in such a way that the cutting fluid can flow through the
spindle to the tool tip.
A tool holder for milling has three main parts: the taper, the flange, and the collet pocket. The taper is the conical part that is
introduced into the spindle. The flange is the part that serves to hold the tool holder when it is in the tool changer storage magazine
of the machine tool. The collet pocket is the part of the tool holder that actually holds the cutting tool.
Several technological solutions have been developed due to the need to standardize how the cutting tool is held in machine

tools. These standards are BT, CAT (also known as V-Flange), ISO (also known as SK), and HSK; a recent alternative is the Big-Plus
dual contact.
The conical forms such as BT, CAT, and ISO are usually used in conventional milling machines. These conventional tool holders
are inappropriate due to their asymmetric designs, which severely distort and vibrate during high-speed operation. In contrast, the
HSK tool holders are more suitable for high-speed milling because they have high symmetry, high balance, and high tool holding

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
228 High-Speed Machining

power. HSK tool holders (E and F) were specifically designed to resist the z-axis displacement because they provide a simultaneous
fit against both the spindle face and the spindle bore. For HSK tool holders in high-speed applications, two major aspects need to be
considered: (1) how to hold the tool to the holder and (2) how to hold the tool/tool holder assembly to the spindle bore securely.
Both are very important because the cutting forces and torque are eventually transmitted between the machine and the workpiece
through these two interfaces. There are three basic designs available to hold the tool to the tool holder: weight-shifting, hydraulic,
and heat shrink-fit. Different from the other two holding mechanisms, heat shrink-fit utilizes the special thermal expansion
properties of metal. Tools can be inserted into the tool holder and then clamped in the holder in place by cooling down the tool
holder. Compared to other chucks such as the collet types, the shrink-fit holder system has stronger holding ability, and can
successfully hold tools during high-load operations. This holder also provides high-speed and high-accuracy tool change without
deflection. Therefore, shrink fit tool holders will gain more favor because of their inherent simplicity, reliability, and economics.

11.10.3.4 Advanced Cooling Technologies for High-Speed Application

HSM operations can generate chip quantities ranging from 800 to 5600 cm3 min1 (48340 in.3 min1). The energy consumed in
removing a unit volume of material is called the specific cutting energy. The specific energy for the often-used material is listed in the
following table.
For example, for an aluminum alloy with a specific cutting energy of 0.2 hp-min in.3 as listed in Table 3, at a material removal
rate of 300 in.3 min1, the generated cutting heat is 60.0 hp. Such a high cutting heat places extreme demands on the coolant and
chip removal systems. In industries, high pressure and high flow rate coolants (4001000 psi) effectively lubricate the cutting
process, evacuate chips, and maintain the thermal stability of the cutting process. The coolant is supplied to the spindle nose and the
tool either externally or internally through the spindle. For high-speed applications, through-the-spindle cooling is widely preferred
to directly deliver the coolant to the toolchip contact interface. Where high pressure coolant sprays are used, mist collectors may be
necessary to meet the mist exposure standard and to ensure the health and safety of the operators.
In metal cutting, the cutting heat removal mainly relies on boiling of a flowing stream of coolant such as water-based coolant,
liquid nitrogen, or carbon dioxide. This cooling process is also called forced convective boiling. For liquid coolants, a change of state
from liquid to vapor requires the input of energy, which is called the latent heat of vaporization. The specific latent heat of vaporization
is the amount of heat required to convert the unit mass of a liquid into vapor without a change in temperature. The specific latent heat
of often used cooling mediums is listed in Table 4. From Table 4, it can be seen that water has the highest specific latent heat among all
liquids, which makes it the most efficient cooling medium. Recently, cryogenic machining has received more attention due to the fact
that it does not cause any environment pollution. However, liquid nitrogen’s efficiency as a coolant is limited by the fact that it boils
immediately on contact with any warmer object (even when it comes into contact with ambient air). Due to the medium boiling, there
is reduced surface area contact between the medium and the materials being cooled (Leidenfrost effect), especially for liquid N2. Water-
based coolant does not have this problem, and it is very effective when the cutting zone temperature is over its boiling point of 100  C.

Table 3 Approximate specific-energy requirements in cutting operations

Specific energy
3
Material W-s mm hp-min in.3

Aluminum alloys 0.41.1 0.150.4

Cast irons 1.65.5 0.62.0
High-temperature alloys 3.38.5 1.23.1
Nickel alloys 4.96.8 1.82.5
Refractory alloys 3.89.6 1.13.5
Stainless steels 3.05.2 1.11.9
Steels 2.79.3 1.03.4
Titanium alloys 3.04.1 1.11.5
At drive motor, corrected for 80% efficiency; multiply the energy by 1.25 for dull tools.
Reproduced from Kalpakjian, Z.; Schmid, S. Manufacturing Processes for Engineering Materials;
Prentice Hall, 2009.

Table 4 Specific latent heats of vaporization of mediums in a liquid state

Substance Specific latent heat of vaporization, kJ kg1 Boiling point,  C ( F)

Water 2258 100 (212)

Ammonia 1369 33 (27)
Ethanol 838 78 (172)
Liquid CO2 574 57 (70)
Ethanoic acid 395 118 (244)
Liquid nitrogen 199 196 (320)

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 High-Speed Machining 229

From an energy consumption point of view, for all compressed gas coolants, such as liquid N2 or CO2, at least the same amount
of energy as the cutting heat is needed to generate compressed gas with enough cooling abilities to take away the cutting heat.
However, for a water-based coolant, due to water’s large latent heat and specific heat capacity, the main energy is consumed on
pumping the coolant and delivering it to the cutting zone. Typically, the energy consumption is less than 1/10th of the energy used
to generate the compressed gas. In addition, less water-based coolant is needed to take away the same amount of cutting heat due to
its superior cooling capability. Therefore, water-based coolant is more suitable for applications with high cutting heat concentrated
at the toolchip interface.

11.10.4 Modeling and Simulation of High-Speed Machining

11.10.4.1 Cutting Force Models for Machining Processes
The implementation of high-speed milling is affected by various factors, such as the cutting force, the cutting temperature, the
cutting power, etc. Among them, the cutting force is an important parameter as it relates to the deflection of the cutter, tool breakage,
as well as basic data for an estimation of chatter vibration and machining error. The generation of mathematical models of high-
speed machining is essential for a deeper understanding of this advanced process. Therefore, an analytical model is needed to
predict the cutting forces for high-speed application of aerospace alloys.
According to the comprehensive survey conducted by the CIRP’s working group (19), over 55 major groups are involved in
modeling efforts. Over 43% of the research groups were active in experimental/empirical modeling followed by 32% involved in
analytical modeling and 18% involved in numerical modeling. Analytical and numerical models are more accurate and more
applicable, and many researchers have developed numerous models of these two types for the machining operations. For simplified
models, an analytic solution with a set of equations or a formula can be obtained to describe how the model will behave under any
circumstances. This is called analytic modeling. However, for more complex models, the math becomes very complicated and it is
difficult to derive an analytic solution. In this case, numerical methods will be used to describe the behavior of the model over time.
Typically, the numerical method starts with the initial values of the variables, and then uses the equations to figure out the changes
in these variables over a very brief time period.

11.10.4.1.1 Analytical Models

According to the methods used, analytical models can be divided into two main categories: mechanistic models and shear plane/
zone models.
Mechanistic models are based on the relationship between cutting forces and the undeformed area of the cut, cutting tool
geometry, cutting conditions, and workpiece geometry. The cutting force is usually assumed to be proportional to the undeformed
area of the cut. For example, the tangential cutting force of the milling process can be predicted using the following equation:
Ft ¼ Kt bh
where Kt is the specific cutting pressure, b is the width of the cut, and h is the thickness of the cut. Basically, based on the experi-
mental results, Kt can be obtained; then it can be used to predict the cutting force very quickly for a set of fixed combinations of
cutting tools and the workpiece. This mechanistic approach works without knowing cutting force mechanics parameters such as
shear angle, shear stress, and friction angle, so it is widely used. However, these models are commonly computer-based and depend
heavily on empirical cutting data for their modeling capacity (19). So strictly speaking, this type of mechanistic model is not purely
an analytical model, because it depends heavily on empirical cutting data.
The shear plane model was developed based on Merchant’s shear plane theory (20,21). This model is based on the assumption
of continuous chip formation in a narrow zone which is idealized as a plane with uniform distributed shear stress. This method has
been used quite successfully in the prediction of forces in several practical machining operations (22,23).
For the shear plane model stated above, a major problem is regarding the uncertain magnitude of toolchip friction and the
shear stress at the shear plane (19). In addition, this model is based on the assumption that workpiece material deforms at constant
flow stress. Considering the dependence of flow stress of metal on strain, strain-rate, and temperature, Oxley (24) developed a more
effective model, which considered the variation of flow stress properties in terms of the strain, strain-rate, and temperature. This
model assumes a thin shear zone, chip equilibrium, and uniform shear stress in the secondary deformation zone at the toolchip
interface. Oxley’s predictive machining theory is widely used for predicting many machining characteristic factors, such as the shear
angle, cutting forces, flow stress, etc. However, one of the major assumptions in Oxley’s predictive machining theory is that the tool
is perfectly sharp. In practice, it is impossible for the cutting tools to be perfectly sharp. In reality, the cutting edge radius can improve
the strength and wear characteristics of the tool. For most end mills, the cutting edge radius is about 5 mm, so Oxley’s machining
theory might need some revision before it is applied to the often used end mills.
Recently, some other researchers also considered the temperature effects on the cutting forces, and then developed analytical
models for the machining processes. Moufki et al. (25) established a model for orthogonal cutting that combined a thermo-
mechanical analysis of the material flow within the primary shear zone and a modeling of friction at the toolchip interface. In their
model, they considered that the primary shear zone was a thin straight band. In addition, they assumed that toolchip contact was
described by use of a mean friction coefficient, which was a decreasing function of the temperature. Jacobus et al. (26) proposed
a thermomechanical model to predict the full in-plane biaxial residual stress profiles at and beneath the newly generated surface

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
230 High-Speed Machining

from the turning process. They claimed that thermal effects lead to the development of tensile stresses in the surface/near-surface
layer and support the presence of thermal and mechanical effects on residual stress that are consistent with the experimental data.
Huang and Liang (27) established a model to investigate the effect of the tool thermal property on cutting forces. By thermal
modeling of both the primary and secondary heat sources, they modified Oxley’s predictive machining theory to analyze the metal
cutting behaviors. Temperature distributions along the primary and secondary shear zones are modeled with the moving heat
source method. To generalize the modeling approach, they applied the modified JohnsonCook (JC) equation in the modified
Oxley approach to represent the workpiece material properties as a function of strain, strain-rate, and temperature. Prediction results
from their study showed that the JC equation worked well as the material constitutive equation.

11.10.4.1.2 Numerical Models

In numerical modeling, FEM techniques were found to be the most dominant tool (19). In this approach, the solution region is first
divided into many smaller elements, so that various tool geometries, cutting conditions, and more sophisticated material and
friction models can be incorporated (28). Then, element equations are formulated. Based on the interconnected relation of
elements, element equations can be assembled into global equations. Finally, after solving the global equations, a numerical
solution to the problem domain is obtained. Two basic approaches are often used to solve the global equations, namely the Newton
Raphson method and the direct iteration. Thus, the main advantage of FEM is its ability to predict chip flow, cutting forces, and
especially a distribution of tool temperatures and stresses for various cutting conditions by simply changing the input data. Ozel and
Altan (29,30) have completed many definitive works in this field. They developed a predictive model for high-speed milling based
on FEM simulations. Using their model, the resultant cutting forces, flow stresses, and temperatures in turning and flat end milling
were predicted primarily. More importantly, with fewer experiments, this method is able to estimate the variations of flow stress and
friction conditions of high-speed machining. In their model, the tooth-path was assumed to be circular; however, this approxi-
mation will cause some error for high-speed milling.
Using the commercial code FORGE2, Ng et al. (31) presented an FE model to simulate orthogonal machining of hardened die
steel with advanced ceramic tools. Unfortunately, their model underestimated the magnitude of the cutting force due to limited data
on the sensitivity of the workpiece material to strain hardening and the strain-rate sensitivity at elevated temperature, and an
oversimplification of the frictional conditions at the toolchip interface. Based on Oxley’s theory, Carrino et al. (32) used a coupled
thermomechanical finite element model to simulate orthogonal cutting of carbon steel C40. In their model, the tangential force
applied along the tool/chip interface was assumed to be a fraction of the shear stress of the material. Good agreement between
experimental and numerical results was found based on cutting forces measurements.
Altintas (28) and his research group at UBC developed an arbitrary LagrangianEulerian (ALE) formulation, which has been
applied for the prediction of cutting variables in machining. In the developed ALE code, the effects of edge radius on the cutting edge
on the cutting forces were considered.
Although rapid progress and better results of FEM have been achieved recently, there are also some problems for FEM which
need to be considered. The most significant problem is to obtain the material properties under the practical cutting conditions.
Nowadays, the data of material properties used in the simulation of FEM are obtained in tensile or compression tests. Obviously,
the real cutting conditions are different from those of tensile or compression tests. In addition, the numerical model requires
significant amounts of computation time. The computational burden is almost unbearable for three-dimensional modeling. There
is still a long way to go before the FEM can be used to simulate practical machining operations with an acceptable degree of accuracy
and reliability and an acceptable amount of effort for daily use (19).

11.10.4.2 Cutting Temperature Models for Machining Processes

Cutting temperature is another important factor in HSM operations, and it plays a major role in tool wear, tool life, and surface
quality. Therefore, it is necessary to establish theoretical models to accurately predict cutting temperature and then optimize
machining parameters. Many research efforts have been made to study the temperature and its distribution in the cutting region.
Lazoglu and Altintas (33) proposed a numerical model based on the finite difference method to predict the steady-state tool and
chip temperature fields as well as transient temperature variation in continuous and interrupted orthogonal machining processes.
Their approach can be utilized to select the optimal cutting speed, feed rate, as well as tool geometries such as tool rake and
clearance angles so as to avoid excessive thermal loading of the tool. Ng et al. (31) used an FE model using FORGE2 to simulate
cutting force and temperature distributions, and they performed experimental tests at 20 000 rpm on an HSM center to verify the
estimated temperature. In their study, cutting temperatures were measured using thermocouple and infrared techniques. Based on
the comparative study, it was concluded that cutting temperature is increased with the cutting speed and no reduction in cutting
temperature was observed when cutting speed increased. This result shows opposite phenomena to Salomon’s theory. Recently in
Ref. (34), the authors proposed a new methodology to the temperature prediction of milling. The temperature of the workpiece,
during the milling operation, has been measured using an infrared camera. During the experiments, cutting speed and feed rate have
been changed. After data analysis, an FEM model of the cutting process during milling was developed. The rheological model is
calibrated using different milling tests. The results of the model have been compared with the experimental data, obtaining a good
agreement. The approach can be useful to the insert tool designer in order to improve the cutting tool performance.
Warnecke et al. (35) proposed a new material model to describe the thermo-viscoplastic flow behavior of workpiece material in
metal cutting. The material parameters that are achieved by using the flow stress data available at low strain-rates are enhanced by

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 High-Speed Machining 231

matching the results of the experimental investigations and finite element simulations of the orthogonal cutting process. The results
of the experimental investigations showed that the model had a wide validity range of strains, strain-rates, and temperature. In
addition, Fang and Jawahir (36) developed a new method to integrate the rigid-plastic slip-line theory with Oxley’s predictive
machining theory, which can fully take the effects of strains, strain-rates, and temperature into consideration through an analytical
rather than a numerical finite element approach. In the proposed model, material thermal properties and velocity-modified
temperature are considered to calculate the flow stress of work materials. The estimated results agreed well with experimental data.
Ueda et al. (37) measured the temperature on the flank face of a cutting tool in high-speed milling and in turning using a two-color
pyrometer with a chalcogenide optical fiber. They investigated the influence of cutting conditions on the temperature and the
influence of cutting speed on tool wear. The temperature of the end milling cutter rises rapidly with an increase in cutting speed, and
it is approximately 770  C at 628 m min1. The heating and cooling characteristics of the cutting tool are examined and the
temperature change during an intermittent cutting operation is approximately 150  C at a cutting speed of 471 m min1 and a feed
of 0.047 mm per tooth.

11.10.4.3 Modeling of Cutting Forces in High-Speed Machining

11.10.4.3.1 Overview of Conventional Orthogonal Cutting Force Models
The widely used shear plane model was first proposed by Ernst and Merchant, based on the assumption of continuous chip
formation in a narrow zone, which is idealized as a plane with uniformly distributed shear stress. The forces in chip formation in the
machining of Ti6Al4V are shown in Figure 5. The area between the boundaries CD and EF is the chip formation zone or the shear
zone. Based on the assumption that the boundaries CD and EF are parallel and equidistant from AB, Oxley (24) developed
a parallel-sided shear zone theory to predict the cutting forces.
He assumed that the shear flow stress kAB along the shear plane is constant; the shear force can be written as
kAB t1 w
FS ¼ kAB wl ¼ [1]
sin f
where w is the width of the cut, l is the length of the shear plane, t1 is the undeformed chip thickness, and f is the shear angle. The
physical meaning of all the parameters is given in Ref. (38).
According to the geometric relations as shown in Figure 5, the following equations can be obtained:
VC sin f
Vchip ¼
cosðf  aÞ
VC cos a
VS ¼
cosðf  aÞ
t2 ¼ t1 cosðf  aÞ=sin f
q¼fþla [2]
FS
FR ¼
cos q
kAB t1 w cosðl  aÞ
FC ¼ FR cosðl  aÞ ¼
sin f cos q
kAB t1 w sinðl  aÞ
FT ¼ FR sinðl  aÞ ¼
sin f cos q
From eqn [2], the cutting forces can be predicted if the shear stress kAB, shear angle f, and the friction angle l are known.
Oxley (24) proposed a theory to estimate kAB, f, and l. In this chapter, the cutting force predictive model is based on this

Figure 5 Cutting force diagram based on the shear plane model. Reproduced from Wang, Z. G. High-Speed Milling of Titanium Alloys: Modeling and
Optimization. Ph.D. Thesis, National University of Singapore, 2005.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
232 High-Speed Machining

method, and a brief overview of it is given as follows. In the shear plane zone, kAB is calculated according to the following
equation:
s1 εn1
kAB ¼ pffiffiffiAB [3]
3

where s1 and n1 are constants, and n1 is also called the strain-hardening index. In reality, s1 and n1 vary with strain-rate and
temperature. For a combination of specific strain-rate and temperature, s1 and n1 give a good fit in the following empirical stress
strain relation:
s ¼ s1 εn1 [4]

where s and ε are the effective flow stress and strain. In order to find the values of s1 and n1, it is necessary to know the value of the
velocity-modified temperature Tmod, which is defined as a function of strain-rate ε_ and temperature TAB as
Tmod ¼ TAB ½1  vlgð_ε=_ε0 Þ [5]

The constants v and ε_ 0 are taken as 0.09 and 1 s1, respectively.

From eqns [3][5], in order to determine kAB, three parameters are needed, namely the temperature at AB TAB together with the
strain εAB and the strain-rate ε_ AB at AB. In Ref. (24), the strain and strain-rate are calculated as
pffiffiffi
εAB ¼ gAB = 3 [6]
pffiffiffi
ε_ AB ¼ g_ AB = 3 [7]

where gAB and g_ AB are maximum strain and strain-rate at AB, and they are assumed to be given by
1 cos a
gAB ¼ [8]
2 sin f cosðf  aÞ

VS
g_ AB ¼ Cn [9]
l

where Cn is the strain-rate constant.

The temperature at the shear plane TAB is calculated from the shear forces, shear velocities, and tool and workpiece material
thermal properties (thermal conductivity and specific heat), shown as the following equation:
TAB ¼ TW þ hDTSZ [10]

where TW is the initial workpiece temperature and DTSZ is calculated from the equation
1  b FS cos a
DTSZ ¼ [11]
rSt1 w cosðf  aÞ

where S is the specific heat, r is the work material’s density, and b is obtained from the following equations:
b ¼ 0:5  0:35lgðRT tan fÞ for 0:04  RT tan f  10:0
[12]
b ¼ 0:3  0:15lgðRT tan fÞ for RT tan f > 10:0

where RT is a nondimensional thermal number, which is given by

RT ¼ rSVC t1 =K [13]

where K is the thermal conductivity of the work material.

Because the cutting forces, thermal conductivity, and specific heat of workpiece material are temperature dependent, an initial
temperature is assumed to start the calculation. For the given assumed temperature, the cutting forces and thermal properties of
workpiece material are determined, which can be used to calculate the temperature at AB. Then, the temperature at AB is updated
with the calculated one. With the replacement of the calculated temperature as the starting temperature, a new calculation starts. This
process is repeated until the difference between the starting temperature and the calculated one is less than a given value, say, 0.1 K.
For the calculation of chip temperatures given by the following equation, a similar iterative procedure is also necessary.
Tint ¼ TW þ DTSZ þ jDTM [14]

where DTM is calculated from the equation

     
DTM RT t2 1=2 R T t2
lg ¼ 0:06  0:195d þ 0:5lg [15]
DTC lc lc

where the average temperature rise in the chip DTC is given by

Ff sin f
DTC ¼ [16]
rSt1 w cosðf  aÞ

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 High-Speed Machining 233

At first, the mean chip temperature is estimated for the initial calculation. Then, it is replaced with the new calculated value. This
process continues until the difference between the estimated and calculated values is less than a certain value. Then, eqn [5] is used
to calculate the value of Tmod at the toolchip interface with the temperature Tint and the average strain-rate ε_ int at the toolchip
interface. The value of ε_ int is calculated as
pffiffiffi Vchip
ε_ int ¼ g_ int = 3 ¼ pffiffiffi [17]
3dt2
Finally, the shear flow stress in the chip at the toolchip interface kchip is given by
s1
kchip ¼ pffiffiffi [18]
3
where s1 is determined by the value of Tmod at the interface. In addition, the resolved shear stress at the toolchip interface is found
from the following equation:
Ff
sint ¼ [19]
lc w
where lc is the toolchip contact length given as
 
t1 sin q Cn n1
lc ¼ 1þ [20]
cos l sin f 3½1 þ 2ðp=4  fÞ  Cn n1 
which is derived by considering moments about B of the normal stresses on AB to find the position of FR. Assuming that normal
stress on the toolchip interface is distributed uniformly, the resultant force FR intersects the toolchip interface with a distance lc/2
from B, and the angle q is given by the following equation:

tan q ¼ 1 þ 2ðp=4  fÞ  Cn n1 [21]

By applying the appropriate stress equation along AB, it can be shown that f should be in the range of (0, p/4).
Based on the above description, for given values of tool rake angle a, the cutting speed VC, the thickness t1, and the width of
cut w of the undeformed chip, together with the thermal and flow stress properties of the workpiece material and the initial
temperature of the work Tw (say, 20  C in all calculations), the following procedure is used. For a given D and Cn, the
equilibrium (when sint is equal to the value of kchip) values of f are found, as shown in Figure 6. If there is more than one shear
angle f that satisfied the equilibrium condition, the higher or highest value of f is chosen as the solution. Then, the required
value of Cn is determined from the stress boundary condition at B of Figure 5. According to the stress equilibrium equation in
Ref. (24) and the theory in which the toolchip interface is assumed to be a direction of maximum shear stress, the normal
stress sN0 at B is given by
sN 0 p
¼ 1 þ  2a  2Cn n1 [22]
kAB 2
Based on the assumption of uniform normal stress along the toolchip interface, the normal stress sN at B is also
given by
FN
sN ¼ [23]
wlc
Then, Cn can be found by fulfilling the condition sN0 ¼ sN.
The above procedure is iterated for a given range of d and Cn until all the equilibrium conditions are fulfilled. For Oxley’s
machining method, three parameters are very important for its accuracy: the shear angle, the strain-rate constant, the ratio of
the toolchip interface plastic zone thickness to the chip thickness. Based on the previous research done by the authors, for
practical computation, if the values of Cn and d are given in the appropriate range, it takes a reasonable amount of time to use
this method to find the accurate shear angle f. However, it is difficult to predict the strain-hardening index n1 for titanium
machining.
In this chapter, an improved method based on Oxley’s theory is introduced. In the proposed model, the flow stress and strain-
hardening index n1 are estimated based on the JC strength model. Using this method can ensure a more accurate estimation of these
two parameters.

11.10.4.3.2 Modeling of Flow Stress Properties of Work Materials

The yield stress of metal under uniaxial conditions is defined as the flow stress or the effective stress, which depends on the strain
and strain-rate, material properties, and cutting temperature. The metals start deforming plastically when the applied stress reaches
the values of the yield stress or flow stress (39).
The flow stress is mostly influenced by temperature, strain, strain-rate, and material properties. Accurate and reliable flow stress
models are very important for describing the workpiece materials’ deforming behavior during practical machining processes. Many
researchers developed several techniques to determine the flow stress of metals.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
234 High-Speed Machining

Figure 6 Diagram of the machining calculation with Oxley’s theory.

The widely used constitutive model of flow stress is the JC strength model, which was proposed by Johnson and Cook (40). The
JC model represents the flow stress s of a material as the product of strain, strain-rate, and temperature, shown as the following
equation:
" !#   
n ε_ T  Tr m
s ¼ ½A þ BðεÞ  1 þ Cln 1 [24]
ε_ 0 Tm  Tr

The parameter A is the initial yield strength of the material at room temperature and a strain-rate of 1 s1, ε is the equivalent
plastic strain, and ε_ is the strain-rate normalized by a reference strain-rate ε_ 0 . The temperature term is valid within the range from
room temperature (Tr) to melting temperature of the workpiece material (Tm). The parameters A, B, C, m, and n are fitted to the
experimental results obtained from the corresponding compression and tension tests. This model sacrifices the potential
coupling of effects, such as temperature effects on strain-rate hardening, but it can be calibrated more easily. Therefore,
researchers chose the JC strength model as the constitutive equation for the deformation behavior of metals at a higher strain-rate

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 High-Speed Machining 235

Table 5 Parameters of JC constitutive model for Ti6Al4V

Reference A B n C m Test

Lee and Lin (41) 782.7 498.4 0.28 0.028 1.0 SHPB
Meyer and Kleponis (42) 862.5 331.2 0.34 0.0120 0.8 SHPB
This chapter 1100.0 331.2 0.34 0.0120 1.0 SHPB
Reproduced from Lee, W. S.; Lin, C. F. High-Temperature Deformation Behaviour of Ti6Al4V Alloy Evaluated by High
Strain-Rate Compression Tests. J. Mater. Process. Technol. 1998, 75 (13), 127136; Majorell, A.; Srivatsa, S.; Picu, R. C.
Mechanical Behavior of Ti-6Al-4V at High and Moderate Temperatures  Part I: Experimental Results. Mater. Sci. Eng. A
2002, 326 (2), 297305.

and temperature. Lee and Lin (41) investigated the deformation behavior of Ti6Al4V using the split Hopkinson bar (SHPB). They
fitted SHPB test results into the JC model at the strain-rate of 2  103 s1. Meyer Jr. and Kleponis (42) also studied high strain-rate
behavior of Ti6Al4V and low-cost titanium. In this study, the JC strength model is chosen to describe the deformation behavior of
Ti6Al4V.
Based on the published data listed in Table 5 (4143), a primitive study was carried out to estimate the parameters for the JC
model and then to check the global convergence and estimation error in comparison to experimental measured cutting forces. For
reference, the parameters found by Lee and Lin (41) are also listed in Table 5.

11.10.4.3.3 Application of the JohnsonCook Model in the Prediction of Cutting Forces pffiffiffi
pffiffiffiflow stress in the shear plane zone kAB can be calculated according to eqn [3] ðkAB ¼ s1 εAB = 3Þ. This is
In Oxley’s model (24), n

replaced by kAB ¼ s= 3, where s is the effective flow stress along AB, which can be calculated using the constitutive eqn [24].
The change rate of flow stress (dk/ds2) normal to AB can be assumed to be only related to the actual strain-rate. Therefore, dk/ds2
can be derived using the following equation Ref. (24):
dk dk dg dk dg dt
¼ ¼ [25]
ds2 dg ds2 dg dt ds2
where t is time. From eqn [24], the following equation can be obtained at AB:
" !#    
ds ε_ T  Tr m nBðεAB Þn1
¼ nBðεAB Þn1 1 þ Cln 1 ¼ sAB [26]
dε ε_ 0 Tm  Tr ½A þ BðεAB Þn 

Then, the first term on the right-hand side of eqn [25] can be obtained as
pffiffiffi
dk ds= 3 nBðεAB Þn1
¼ pffiffiffi ¼ pffiffiffi kAB [27]
dg 3dε 3½A þ BðεAB Þn 
The second term on the right-hand side of eqn [25] is the strain-rate, which is given in eqn [9]. The last term is the reciprocal of
the cutting speed normal to AB, which can be represented as
dt=ds2 ¼ 1=ðVC sinðfÞÞ [28]
By substituting for eqn [25] with eqns [9],[27], and [28], the following relation is obtained, and a similar derivation procedure is
given in Ref. (27):
dk kAB nBðεAB Þn1 Cn VC cos a 1 2kAB Cn nBðεAB Þn
¼ pffiffiffi ¼ [29]
ds2 3½A þ BðεAB Þn  cosðf  aÞl VC sin f l½A þ BðεAB Þn 

According to the stress equilibrium equation along AB from Oxley (24), the following relation exists:
dk
dp ¼ ds1 [30]
ds2
By applying the equation along AB, substituting for dk/ds2 from eqn [30], the next equation is given as
2kAB Cn nBðεAB Þn
pA  pB ¼ [31]
½A þ BðεAB Þn 
where pA and pB are the hydrostatic stresses at points A and B, respectively. Finally, the unknown parameter Cn is given by
ðpA  pB Þ½A þ BðεAB Þn 
Cn ¼ [32]
2kAB nBðεAB Þn
In Oxley’s theory, the angle q made by the resultant force FR with AB is expressed as
p  Dk l
tan q ¼ 1 þ 2 f  [33]
4 2kAB Ds2

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
236 High-Speed Machining

With the empirical stressstrain relation s ¼ s1εn1, eqn [33] is represented as eqn [21], while in this study, by substituting for
eqn [33] with eqn [29], the following equation is obtained:
p 
tan q ¼ 1 þ 2  f  Cn n 1 [34]
4
BεnAB
where n1 ¼ n , which is the same strain-hardening index as that in Oxley’s theory.
A þ BεnAB
Moreover, the shear flow stress in the chip at the toolchip interface kchip can also be estimated based on the strain, strain-rate,
and temperature at the sticking region of the toolchip contact interface, and its length is assumed to be half of the undeformed
chip thickness.
Thus, eqn [21] can be replaced with eqn [34] to estimate the angle between the resultant force and the shear plane.
Based on the above description, for given values of the tool rake angle a, the cutting speed VC, the thickness t1, and the width of
cut w of the undeformed chip, together with the thermal and flow stress properties of the workpiece material and the initial
temperature of the work Tw (say, 20  C in all calculations), an analytical approach as shown in Figure 6 can be employed to predict
the cutting forces.

11.10.4.4 Modeling of Cutting Temperature in High-Speed Machining

The temperature at the contact between the chip and the tool and, in general, near to the tool tip plays an important role in the
machinability of various work materials. Thus, a considerable amount of research has been carried out to analyze the temperature
fields in the chip and in the tool. Among them, Tlusty’s finite difference computation approach (3) was borrowed in this study with
a necessary augmentation for the high-speed milling operation.
As mentioned above, the shear temperature along the shear plane AB TAB can be estimated with eqn [10]. It is assumed that the
heat conducted from the chip into the surrounding air is completely neglected. Consider the chip area ABCD as shown in Figure 7.
The boundary conditions are such that along AD the chip is insulated based on the assumption that no heat is conducted into the
surrounding air from the chip. Along BC over the contact length lc, there is power/heat input p due to chiptool friction. The friction
power is generated in the contact between the chip and the tool, and the distribution of this power follows the distribution of the
normal pressure and the heat power p, which are distributed as shown in Figure 7. The pressure on the tool has been studied
experimentally by Yellowley and found to be maximum and constant from the cutting edge B over half the undeformed chip
thickness t1, and thereafter it decreases linearly toward the end of the contact (3). The heat spreads in the chip by conduction in the
direction of Y and by conduction and mass transfer (chip motion) in the direction of X.
As shown in Figure 7, the basic idea of Tlusty’s finite difference computation approach is described as follows. The whole field of
the chip is divided in incremental slices in both directions; there are a total of 20  KKK elements. Assuming that the temperature is
constant over every element and changes discretely from element to element, the method of finite differences can be applied to the
field. The mass transfer and heat transfer along the X direction are considered. Correspondingly, only the heat conduction along the

Figure 7 Finite difference formulation for temperature in the chip. Reproduced from Tlusty, G. Manufacturing Processes and Equipment; Prentice Hall:
USA, 2000.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 High-Speed Machining 237

(a) (b)
lc
zi
Chip
Δzi
Tool–chip
contact wi
Tool

Tool

Chip flow illustration Thermal field in the tool

Figure 8 Diagram of the finite-difference formulation for the tool: (a) Chip flow illustration; (b) Thermal field in the tool.

Y direction is considered in the computation. Over the contact length, from 1 to KK, it is obtained at its lower end heat input pk and
the heat spreads in this slice in the direction Y by conduction. Heat is also conducted away through the tool.
The initial state of the thermal field in the chip is such that the slide of k ¼ 1 has the following uniform shear plane temperature
TAB, which can be estimated with eqn [10]:
j ¼ 1 to 20; Tj;1 ¼ TAB

At every instant k, the lowest element, j ¼ 1, receives heat input pk and loses heat by conduction to the next higher element, j ¼ 2.
The surplus of heat is used to increase the temperature over the time constant Dt from T1,k to T1,kþ1:
T1;k  T2;k T1;kþ1  T1;k
pk  Dxbk ¼ bDxDyrc [35]
Dy Dt
where Dt is calculated as
Dx lc
Dt ¼ ¼ [36]
Vchip KKVchip

The heat conducted per time step Dt from j ¼ 1 to j ¼ 2 is the second term on the left side of eqn [35], which is proportional to the
temperature gradient (T1,k  T2,k)/Dy, to the area Dxb, and to thermal conductivity k. The right side is the heat capacity of the
element, which is the product of its volume bDxDy and of the density r and specific heat c multiplied by the temperature increase
(T1,kþ1  T1,k) over time Dt.
For the elements j ¼ 219, instead of the heat input pk the heat flow from the next lower element is used.
 
Tj1;k  Tj;k Tj;k  Tjþ1;k Tj;kþ1  Tj;k
 Dxbk ¼ bDxDyrc [37]
Dy Dy Dt
For j ¼ 20, the second term in the brackets on the left side of eqn [37] is removed because heat is neither conducted out of this
element nor convected out of the chip. Equations [35][37] can be used to express the new temperatures at time (kþ1):
"   #
pk T1;k  T2;k k
T1;kþ1 ¼  Dt þ T1;k ; where j ¼ 1 [38]
bDxDyrc ðDyÞ2 rc

  k
T1;kþ1 ¼ Tj1;kþ1 þ Tjþ1;k  2Tj;k Dt þ Tj;k ; where j ¼ 219 [39]
ðDyÞ2 rc

  k
T20;kþ1 ¼ T19;kþ1  T20;k Dt þ T20;k ; where j ¼ 20 [40]
ðDyÞ2 rc
The time increment Dt in eqn [36] must not be chosen too large because then the computation in discrete steps would not
converge.
Obviously, some heat is convected and radiated out of the tool to the surrounding air. Assume that the tool far away from the
cutting zone is at room temperature, and the isotherms in the tool are more or less lines of equal distance from the cutting zone.
Therefore, the tool can be represented as a wedge with the chip/tool contact at the flat top. It will be divided into slices with thickness
Dzi, width wi, and depth (into the paper) b as shown in Figure 8. The thermal field is assumed single dimensional in z, while every
slide has a constant temperature throughout. The sides are assumed isolated and the bottom is at room temperature Tr. From the
point of view of the tool, the temperature along the contact length lc is also constant. It will be taken as equal to the average
temperature along lc. Thus a steady-state single-dimensional heat transfer can be used to calculate the heat transfer into the tool.
Based on the assumption of steady state for the slide with the temperature Ti, the heat first enters into the slice and then
leaves it. The power flowing through the tool is Pt and it is the same for each slice. The first step goes from the top of the

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
238 High-Speed Machining

tool (the interface between the chip and the tool) to the middle of the top slice where temperature T1 is located. The mean
width is
Dz
w1 ¼ lc þ [41]
2
All the following slices step from the middle of one to the middle of the next one with the distance of Dz, and the mean width is
wi ¼ lc þ 2ði  1ÞDz; i ¼ 2; 40 [42]
For the last slice, the step goes from T40 to Tr over a distance of Dz/2:
w41 ¼ lc þ 79:5Dz [43]
Then the thermal resistance for each slice can be calculated based on the same power flowing through the tool.
Tcav  T1 Dz=2
Pt ¼ w1 bk; R1 ¼ [44]
Dz=2 w1

Ti  Tiþ1 Dz
Pt ¼ wi bk; Ri ¼ ; i ¼ 2; 40 [45]
Dz wi

T41  Tr Dz=2
Pt ¼ w41 bk; R41 ¼ [46]
Dz=2 w41
Based on eqns [44][46], the temperature of each slice can be expressed with Tcav as follows:
Pt
T1 ¼ Tcav  R1 [47]
bk

Pt Pt
T2 ¼ T1  R2 ¼ Tcav  ðR1 þ R2 Þ [48]
bk bk

Pt Pt X
iþ1
Tiþ1 ¼ Ti  Riþ1 ¼ Tcav  Ri [49]
bk bk 1

Finally, room temperature can be represented as

Pt r bk
Tr ¼ Tcav  ; and Pt ¼ ðTcav  Tr Þ [50]
bk r
P
where r ¼ 41 1 Ri
Based on the approach mentioned above and the input data from cutting conditions and material properties, the temperature
field in the chip can be estimated with eqns [38][40]. Then the average temperature Tcav at the toolchip contact interface can be
calculated. With Tcav as input, the power flowing through the tool Pt can be estimated. Thereafter, we must update the power escaping
through the tool for the next run and check for the relative difference in the average temperature at the toolchip interface between
two consecutive runs. This procedure will be iterated until the difference is less than a small value, that is, 0.1 ms used in this study.

11.10.4.5 Experimental and Simulation Study of the Cutting Force and Temperature
11.10.4.5.1 Experimental and Simulation Study of the Cutting Force
Then, the horizontal (x direction) and normal (y direction) components of the cutting forces acting on the work material are derived
from the equilibrium diagram shown in Figure 9:
Fx ¼ Ft sin 4  Fr cos 4 [51]

Fy ¼ Ft cos 4 þ Fr sin 4 [52]

Normally, for tangential (Ft(4)) and radial (Fr(4)) cutting forces acting on the cutter in slot milling, the radial component Fr can
be assumed to be xFt, where x is a constant. Assuming that tan 40 ¼ x, then
Ft
Fx ¼ Ft sin 4  Ft tan 40 cos 4 ¼ sinð4  40 Þ [53]
cos 40

Ft
Fy ¼ Ft cos 4 þ Ft tan 40 sin 4 ¼ cosð4  40 Þ [54]
cos 40
Because of the effect of 40, Fx and Fy cannot reach their maximum values when Fr and Ft nearly reach their maximum ones at 4 z
p/2. The varying values of Fx and Fy are simulated in the next section.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 High-Speed Machining 239

Y Y
Ft
Fr –
X –
Fr X

Ft

Figure 9 Mechanics of the milling process.

In the practical experiments, the cutting forces in the x, y, and z directions are recorded. We need to derive the cutting force
components Fr and Ft from measured forces Fx, Fy, and Fz. From Figure 9, the derivation of Fr and Ft is as follows:
Fr ¼ Fx cos 4 þ Fy sin 4 [55]

Ft ¼ Fx sin 4 þ Fy cos 4 [56]

For the mechanical cutting force model, the cutting forces in the tangential and radial directions also can be obtained by
multiplying the specific cutting pressures with the uncut chip area.
Ft ð4Þ ¼ Kt Ac ð4Þ [57]

Fr ð4Þ ¼ Kr Ac ð4Þ [58]

After getting the cutting forces from the simulation, the specific cutting pressures in the tangential and radial direction can be
estimated from the above two equations. In the simulation of the cutting temperature, the estimated specific cutting pressures will
be used to determine the cutting force and machining power.
Figures 1012 show the predicted cutting forces using the developed model and the corresponding experimental cutting forces.
The estimated cutting force values of Fx and Fy are similar to those of the experimental results. This indicates that the cutting forces
can be predicted with good accuracy for these two directions. But for all three cases, the estimated cutting forces of Fz are smaller
than the experimental values. The difference may result from the plowing forces in milling, which are not taken into account in the
cutting force model proposed in this study. However, the difference is not significant, so it means that the cutting forces can be
predicted with reasonable accuracy for all three directions.

11.10.4.5.2 Prediction of the Cutting Temperature

In addition, based on the procedure mentioned in Section 11.10.4.4, the temperature distribution in the chip and along the tool
chip contact interface was simulated to investigate the effects of feed rate and cutting speed on the cutting temperature. To save
simulation time, only the moment when the cutter engagement angle is 90 is considered in the simulation for a slot milling
process. Figure 13 shows the simulation results under v ¼ 50 m min1, ap ¼ 6 mm, and five different feed rates. Obviously,

50
Fx - Experimental
Fy - Experimental
40 Fz - Experimental
Fx - Estimated
Fy - Estimated
Cutting forces (N)

30

20

10

0
0 45 90 135 180
-10 Angular position (deg)

-20

Figure 10 Experimental and estimated cutting forces at a ¼ 0.075 mm, f ¼ 0.075 mm per rev., and v ¼ 350 m min1. Reproduced from Wang, Z.G.
High-speed milling of titanium alloys: modeling and optimization, PhD thesis, National University of Singapore, 2005.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
240 High-Speed Machining

100
Fx - Experimental
Fy - Experimental
80
Fz - Experimental
Fx - Estimated

Cutting forces (N)

60 Fy - Estimated
Fz - Estimated
40

20

0
0 45 90 135 180
-20
Angular position (deg)
-40

Figure 11 Experimental and estimated cutting forces at a ¼ 0.10 mm, f ¼ 0.1 mm per rev., and v ¼ 350 m min1. Reproduced from Wang, Z.G. High-
speed milling of titanium alloys: modeling and optimization, PhD thesis, National University of Singapore, 2005.

80
Fx - Experimental
70 Fy - Experimental
60 Fz - Experimental
Fx - Estimated
Cutting forces (N)

50
Fy - Estimated
40 Fz - Estimated
30
20
10
0
0 45 90 135 180
-10
-20 Angular position (deg)
-30

Figure 12 Experimental and estimated cutting forces at a ¼ 0.10 mm, f ¼ 0.1 mm per rev., and v ¼ 400 m min1. Reproduced from Wang, Z.G. High-
speed milling of titanium alloys: modeling and optimization, PhD thesis, National University of Singapore, 2005.

900

800
Temperature (C)

700

600
fz = 0.06 mm per rev-z
fz = 0.08 mm per rev-z
500 fz = 0.10 mm per rev-z
fz = 0.12 mm per rev-z
400 fz = 0.14 mm per rev-z

300
0.00 0.03 0.06 0.09 0.12 0.15 0.18 0.21
Tool–chip contact length (mm)

Figure 13 Simulated temperature distribution at the toolchip interface at different feed rates. Reproduced from Wang, Z.G. High-speed milling of
titanium alloys: modeling and optimization, PhD thesis, National University of Singapore, 2005.

increasing the feed per tooth or the chip load produces thicker chips, which also increase the length of the toolchip contact
interface and dissipate more heat than the smaller chip loads. This explains why the temperature along the shear slightly decreases
with the increase of the feed rate. However, the friction force/power at the toolchip interface increases with the chip load, which

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 High-Speed Machining 241

Figure 14 Simulated temperature distribution in the chip sliding away at the toolchip interface under v ¼ 50 m min1, ap ¼ 6 mm, and different
feed rates. Reproduced from Wang, Z.G. High-speed milling of titanium alloys: modeling and optimization, PhD thesis, National University of Singapore,
2005.

Figure 15 Simulated temperature distribution in the chip sliding away at the toolchip interface under f ¼ 0.10 mm per rev-z, ap ¼ 6 mm, and different
cutting speeds. Reproduced from Wang, Z.G. High-speed milling of titanium alloys: modeling and optimization, PhD thesis, National University of
Singapore, 2005.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
242 High-Speed Machining

consequently causes the temperature to increase at the toolchip interface as shown in Figure 14, and the position where
maximum temperature occurs at the toolchip interface is increased with the chip load. Figure 15 shows the simulation results
under fz ¼ 0.1 mm per rev-z, ap ¼ 6 mm, and different cutting speeds. With the increase of cutting speed, the friction power at the
toolchip interface increases, which then increases the cutting temperature at the toolchip interface. When the cutting speed
reaches 70 m min1, the maximum cutting temperature is even over 800  C, which is a critical temperature for WC-Co material to
maintain its mechanical strength. Thus, cutting speed is the more influential parameter that increases the cutting temperature.

11.10.5 Economics of High-Speed Machining

The selection of optimal cutting parameters is of great concern in the manufacturing industry, where the economy of
a machining operation plays a key role in the competitive market. Several efforts have been made to optimize the machining
parameters. However, the work done on the optimization of cutting conditions in machining is mostly focused on turning
operations, while multipass milling has received relatively little attention with regard to the optimization of cutting
parameters (44).
Several methods have been used for the optimization of cutting parameters. However, most of the efforts have mainly
focused on single-objective optimization (production rate or machining cost). Little attention has been given to the multi-
objective optimization of the machining process, although the production cost and production time are two main concerns for
machining processes. In this chapter, two objectives of production cost and production time are considered in the optimization
of high-speed milling; thus the optimization is posed as a multiobjective optimization problem. It is clear that a minimum cost
solution is usually not the one with minimum production time, and vice versa. Therefore, in such a scenario, no single solution
is superior to the other one for these two objectives of machining, and such a solution is called a nondominated solution. The
set of all nondominated solutions is called the Pareto set or Pareto front (45). Because there are no efficient multiobjective
optimization algorithms available, the multiobjective optimization problems have traditionally been solved by converting
them into single-objective optimization problems. The often used conversion methods include the weighted sum approach, the
ε-perturbation method, the Tchybeshev method, the minmax method, the goal programming method, and others (46). When
this type of classical method is used, only one solution is obtained, but there is no guarantee that this is a Pareto-optimal
solution. In addition, in order to find many Pareto-optimal solutions, the algorithm needs to be run for a number of times.
Furthermore, the classical methods have been found to be sensitive to the convexity and continuities of the Pareto-optimal
region (46).
Since the genetic algorithm (GA) can maintain a population of solutions, it is logical to use GA to find multiple Pareto-
optimal solutions in a single simulation. However, the successful application of GA depends on the population size or the
diversity of individual solutions in the search space. Otherwise, it may not find the global optimum, and may even result in
premature convergence to a local optimum. Although maintaining diversity is the predominant concern, it also results in
deterioration in the performance of GA. Many attempts have been made to find a tradeoff between the population diversity
and the performance of GA (exploration and exploitation). An alternative approach is to combine GA with other optimization
techniques, such as simulated annealing (SA). SA is a general-purpose stochastic optimization method that has been proven
to be quite effective in finding the global optima for many different NP-hard (nonlinear programming) combinatorial
problems. In this chapter, a hybrid of GA and SA, referred to as genetic simulated annealing (GSA), is used to avoid the
premature convergence of GA by exploiting the local selection strategy of SA. With the use of LINUX clusters and the
workstation network, it is now feasible to move machining optimization problems into a high-performance parallel-
computing environment.

11.10.5.1 Optimization Model of High-Speed Milling

11.10.5.1.1 Objective Functions
The minimum production time and the minimum production cost have been chosen as the objective functions. The production time
for a component is the total time required to produce a component and is composed of the following: (1) setup time Ts, which is the
time it takes to load and unload each component and to return the tool to its starting point; (2) process adjustment and quick return
time Ta; (3) machining time Tm; and (4) tool change time per component Ttc. Therefore, the total production time per component is
Tpr ¼ Ts þ Ta þ Tm þ Ttc. For a multipass operation in which Np passes are required to remove the total depth of cut,

XNp  
Tm
Tpr ¼ Ts þ Tai þ Tmi þ Ttc i [59]
i¼1
T

Recently, with the new developments in cutting tools, the tool lives are long enough to machine the components without
significantly wearing out the tools during the process of machining. Thus, in this study, the replacement of the worn tool can be
neglected. The tool-changing time is determined by the indexing time required by the automatic tool changer (ATC) of the machine
tool to swap a used cutter from the spindle with a tool in the tool magazine. This time is independent of the cutting parameters.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 High-Speed Machining 243

In addition, the setup time, the loading/unloading time, and the process adjustment time are not influenced by the cutting
parameters. Thus, the final optimization model becomes

X
Np
0
Tpr ¼ Tmi [60]
i¼1

The machining time (Tm) for the slot milling operations can be calculated in the following form:
L L L pDL
Tmi ¼ ¼ ¼ ¼ [61]
fri fZi zNi fZi zð1000vi Þ=ðpDÞ 1000fZi zvi
By substituting eqn [61] into eqn [60], the final simplified model for the minimum production time per component can be
expressed as
!
XNp
pDL
0
Tpr ¼ [62]
i¼1
fZi z1000vi

The optimization model for the minimum production cost can be similarly derived. The total production cost per component
comprises the fixed cost and the variable cost. The fixed cost includes the cost of raw material, the setup cost, and the adjustment
cost, which are all independent of machining time. The variable cost is a function of the machining time, which includes the
machining cost and the cost of changing the cutting tool. The setup cost, the adjustment cost, and the machining cost include the
salary of the operator and the corresponding overhead associated with his employment, as well as the depreciation of the machine
tool and the overhead associated with its use. For the tool-changing cost, the cost of the cutter is added to the cost of the operator
and the machine tool depreciation. The tool-changing cost depends on the frequency of tool replacement. Therefore, the average
production cost for each component can be written as
Tm
Cpr ¼ Cmat þ ðCl þ Co ÞTs þ ðCl þ Co ÞTa þ ðCl þ Co ÞTm þ ½ðCl þ Co ÞTtc þ Ct  [63]
T
The first three items are the nonproductive cost, which is constant for a particular operation. Since they are not influenced by the
cutting parameters, the final optimization model for the minimum product cost can be described as
Tm Tm
C 0pr ¼ ðCl þ Co ÞTm þ ðCl þ Co ÞTtc þ Ct [64]
T T
Based on the above analysis, the tool-changing time is independent of the ratio of the machining time to the tool life for
a multitool optimization approach. Thus, the model can be further simplified as
Tm
C 0pr ¼ ðCl þ Co ÞTm þ Ct [65]
T
The empirical tool-life equation vT a f b ag ¼ C is used to estimate the practical tool life of the cutting tool. A multielement
regression method based on the experimental results is used to determine the values of the constant C and the exponents
a, b, g.
In all four parameters, the number of passes Np, corresponding to the depth of cut, the cutting speed, and the feed rate are
involved in the optimization of the milling process. However, the number of passes depends on the depth of cut for every pass and
the total depth of cut. For the given total depth of cut and the number of passes, the depth of cut to be removed per pass is
determined. Then the cutting strategy can be determined by the number of cutting passes and the corresponding depth of cut,
cutting speed, and feed rate for every pass. Finally, the optimization problem is to find the optimal cutting strategy so as to achieve
the selected objective function.
The determination of the number of cutting passes can be considered as a one-dimensional walk problem, in which the whole
distance (total depth of cut) can be divided into possible sections (depth of cuts). The length of each section should always be
smaller than the maximum depth of cut and greater than the minimum depth of cut allowed for the combination of cutting tools
and the workpiece.
For example, the value of the depth of cut changes from 0.075 to 0.150 mm in the example, with the minimum increment of the
depth of cut of 0.025 mm. So the possible values of the depth of cut (length of possible sections) are 0.075, 0.100, 0.125, and
0.150 mm. Therefore, the one-dimensional walk procedure can be described in Figure 16, assuming that the total depth of cut is
0.30 mm. The total depth of cut means the walking distance from point N(0, 0) to any point in the bottom line AB along the
possible ways, which are represented as a solid line. The problem is to find all alternative possible passes, which are composed of a
certain number of possible sections, to the points N(1, 0), N(2, 0), N(3, 0), and N(4, 0). The procedure starts from the point N(0, 0)
(free surface of the workpiece). The stock can be machined from the outer end of the workpiece with the four possible depths of cut:
0.075, 0.100, 0.125, and 0.150 mm. Then the procedure is continued until one of these four destination points is reached. At the
same time, all possible passes are stored for future use. In all, to remove the total depth of cut (0.30 mm), five alternative cutting
strategies are available, as shown in Table 6.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
244 High-Speed Machining

N(0,0)
+
0.125 mm
0.150 mm
0.075 mm
+ 0.100 mm
+
+
+ +
+
+ +
+ +

+ + + +
A N(1,0) N(2,0) N(3,0) N(4,0) B

Figure 16 Schematic representation of sectioning strategy. Reproduced from Wang, Z.G. High-speed milling of titanium alloys: modeling and opti-
mization, PhD thesis, National University of Singapore, 2005.

Table 6 Alternative cutting strategies

Number of sections to be cut for each strategy

Cutting strategy no. Pass 1 (mm) Pass 2 (mm) Pass 3 (mm) Pass 4 (mm)

1 0.075 0.075 0.075 0.075

2 0.075 0.075 0.150 
3 0.075 0.100 0.125 
4 0.100 0.100 0.100 
5 0.150 0.150  
Reproduced from Wang, Z. G.; Wong, Y. S.; Rahman, M. Optimisation of Multi-Pass Milling Using Genetic
Algorithm and Genetic Simulated Annealing. Int. J. Adv. Manuf. Technol. 2005, 24 (910), 727732.

After all the alternative cutting strategies are determined by the procedure analyzed above, the optimization algorithm is used to
find the optimal cutting condition for each pass. After finding the optimal cutting conditions for all possible cutting strategies
shown in Table 6, the best strategy is selected as the optimal one according to the corresponding objective function value. Finally,
the number of passes and the cutting parameters (depth of cut, feed rate, and cutting speed) associated with the optimal strategy can
be used in machining.

11.10.5.1.2 Constraints
For a meaningful optimization of the machining process, a certain number of constraints should be fulfilled. The following
constraints as listed in Table 7 are considered in this chapter.

Table 7 Constraints and their expressions in terms of common variables

Constraint Expression in variables

fmax
1 Feed rate constraint fz min  fz  fz max ¼
znmin
pDnmax
2 Cutting speed constraint vmin  v  v max ¼
1000

3 Horsepower constraint FC v  60000Pm h

Ft da2
4 Arbor strength constraint  ss
4J
Fr L3s
5 Arbor deflection  dmax
3EI
r ð1  arcsinðfz =2r ÞÞ
6 Surface finish Ra ¼  Ra max
4

Reproduced from Wang, Z. G.; Wong, Y. S.; Rahman, M. Optimisation of Multi-Pass Milling Using Genetic
Algorithm and Genetic Simulated Annealing. Int. J. Adv. Manuf. Technol. 2005, 24 (910), 727732.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 High-Speed Machining 245

where

da mm Diameter of the arbor

E MPa Modulus of elasticity of arbor material
FC N Force component in the cutting direction
Fr , F t N Force components in radial and tangential directions
fzmax, fzmin mm per rev. Maximum and minimum feed rate per tooth per revolution
4
I, J mm Moment of inertia of stub arbor
Ls mm Length of the holder
D, R mm Nominal diameter and radius of the cutter
Pm kW Nominal motor power
1
vmax, vmin m min Maximum and minimum available cutting speed
v, VC m min1 Cutting speed
ss Pa Permissible torsional stress of the arbor material
h  Overall efficiency

11.10.5.2 Implementation of the Optimization Algorithm

The hierarchical parallel GA model is used to realize the parallelization of GSA. The details of GSA are described in (44). The general
architecture of the PGSA is shown in Figure 17. A master task starts up at the master node to control the operations of the algorithm
by broadcasting problem parameters, such as constraints and depths of cut. According to the depth of cut, slave processors are
classified into several subgroups. There is no migration between different subgroups.
In each subgroup, the combination model of coarse-grained PGA and master-slave PGA is used to parallelize PGSA. A ladder
neighborhood relation is adopted in each subgroup, and the eight-processor structure of this relation is shown in Figure 18. The
host program, which runs on the host processor, decides on the stop criterion for its subgroup. After certain iterations, the host
processor sends the stop flag to eight slave processors, which are marked from 1 to 8 in Figure 18. According to the stop flag, each
slave processor decides whether to continue to run GSA or not. The stop flag of PGSA is set on the host processor if the following
criterion is fulfilled:

Communication to each other

Host of Host of … Host of

subgroup 1 subgroup 2 subgroup N

Master Node

Figure 17 Schematic diagram of PGSA’s architecture. Reproduced from Wang, Z.G. High-speed milling of titanium alloys: modeling and optimization,
PhD thesis, National University of Singapore, 2005.

8 5

4 1
Host

3 2

7 6

Figure 18 Schematic diagram of the implementation of each subgroup. Reproduced from Wang, Z.G. High-speed milling of titanium alloys: modeling
and optimization, PhD thesis, National University of Singapore, 2005.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
246 High-Speed Machining

k kDk k
fbest  fbest  ε fbest [66]

k
where fbest is best fitness value of a solution on the host process at generation k, and at every Dk generation the stop criterion is
checked; fbest is the global minimum and ε is a constant value of 0.001.
On each slave node within a subgroup, a sequential GSA runs within its own subpopulation. If the migration conditions are
satisfied, each slave node, say the source node, finds out the destination nodes according to the migration topology. Then the
migrant solutions are selected on the source node and sent to the destination nodes. After the migrant solutions are received by the
destination node, the solutions to be deleted are determined and replaced by the received migrant solutions. The same program is
executed on each slave processor, but on different data (their own population) until the global optimum is achieved.
After all the subgroups converge to their optimal cutting parameters, these parameters are sent to the master node. Finally,
the master node decides on the optimal cutting strategy after receiving the optimal solutions from the subgroups. A full
description of the implementation of PGSA can be found in Ref. (47).

11.10.6 Typical Experimental Studies on High-Speed Machining

11.10.6.1 High-Speed Machining of Aluminum Aerospace Alloys
The machinability of aluminum alloys is relatively high. Milling aluminum materials with conventional cutting parameters is easy.
Cutting forces are low and tool wear is relatively small. Tool wear rates of the cutting tools do not normally play a significant role in
the machining of aluminum alloys.
Mikromat evaluated the performance of HSM of three types of aluminum alloys: AlMgSi, AlZn5Mg, and AlZnMgCu. It was
found that the cutting time can be shortened by using higher cutting speeds and feeds. The shortened cutting time for the test sample
workpiece was about 70% of the cutting time of the comparative machining center with a cutter of 416 mm as shown in Figure 19.
More savings could be achieved using a spindle with more spindle power, thus making higher feed rate values possible.
Another typical example is the high-speed milling of aluminum 6061-T651. The blank work material size is 1.500  15.700  4800
(38 mm  400 mm  1220 mm). The feed rates are up to 780 ipm (20 K mmpm) at a rotational speed of 30 000 rpm, and
accordingly the highest MRR reaches up to 250 cu ipm (3900 cc min1). A total of 26 min were used to machine a 110 pound billet
to the finished part with 5 pounds as shown in Figure 20.

Figure 19 High-speed milling of an aluminum part/Mikromat/. Reproduced from Kauppinen, V. High-Speed Milling - A New Manufacturing Technology.
In 4th International DAAAM Conference, Tallin, Estonia, 2004; pp 131-134.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 High-Speed Machining 247

Figure 20 High-speed milling of an aluminum 6061-T651 part.

11.10.6.2 High-Speed Machining of Titanium Alloys  Case Study 1

An example is optimized with the proposed methodology. The final dimension of the workpiece in this example is shown in
Figure 21. Specifications of the required parameters and values of the constants are as follows: Type of machining is pocket milling
the bottom surface of the pocket and the total depth of cut a ¼ 0.6 mm, Pm ¼ 18.5, h ¼ 0.65, da ¼ 18 mm, Ls ¼ 100 mm, ss ¼ 120
MPa, and E ¼ 211 GPa. Cutting speed ranges: 200450 m min1, feed rate ranges: 0.0500.300 mm per rev., tool material is
binderless CBN, z ¼ 1, D ¼ 20 mm, workpiece material is titanium alloys Ti6Al4V, tensile strength is 300 MPa, L ¼ 166 mm,
B ¼ 37 mm, Ts ¼ 10 min, Ta ¼ 0.1 min per part, Co ¼ US$1.45, and Cl ¼ US$0.45, Ct ¼ US$100. Based on the experimental results,
the values of Kt are calculated based on the experimental result and the cutting force model.
In this study, 37 processors have been used. After the termination criterion is satisfied, the Pareto-optimal solutions are shown in
Figure 22 at four different axial depths of cut. All the nondominated points are fitted with polynomial curve fitting methodology,
and the fitted lines are indicated as the Pareto-optimal front as shown in Figure 22. Although there are still some dominated
solutions in the population, Figure 23 shows that more than 80% of the solutions are converted close to the Pareto-optimal front.
Normally, a single solution or several particular solutions are chosen from the obtained Pareto-optimal solutions using tech-
niques such as postoptimal techniques, compromise programming, marginal rate of return, weighted average, etc. (46). If the
optimal solutions at a certain user-specified weighted average of the objective functions are located in a convex Pareto-optimal

Figure 21 The final dimension of the workpiece. Reproduced from Wang, Z.G. High-speed milling of titanium alloys: modeling and optimization, PhD
thesis, National University of Singapore, 2005.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
248 High-Speed Machining

Figure 22 Population after termination of simulation at different depths of cut with PGSA: (a) a ¼ 0.075 mm; (b) a ¼ 0.100 mm; (c) a ¼ 0.125 mm; (d)
a ¼ 0.150 mm. Reproduced from Wang, Z.G. High-speed milling of titanium alloys: modeling and optimization, PhD thesis, National University of
Singapore, 2005.

Nondominated solutions Dominated solutions

800
700
600
500
400
300
200
100
0
0.075 0.100 0.125 0.150

Figure 23 Number of nondominated and dominated solutions under different depths of cut. Reproduced from Wang, Z.G. High-speed milling of
titanium alloys: modeling and optimization, PhD thesis, National University of Singapore, 2005.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 High-Speed Machining 249

region, a weighted average strategy can be used to choose the optimal solution. In practical machining, the cutting strategy with
a high production rate is preferred. It can be seen from Figure 22 that in the smaller machining time region, the corresponding
Pareto-optimal region is convex. In addition, the weighted average method can be implemented easily. Therefore, in this chapter,
this method is used to find the optimal solution from the Pareto-optimal front. Unlike the classical weighted average scheme for
multiobjective optimization, the procedure described here is used to choose the optimal solution from the Pareto-optimal solutions
found. For the classical weighted average scheme, based on the values of the weighted average of the objectives, only one solution
would be found to be optimal. Furthermore, with a different weight vector, a different optimal solution can be found. For the given
example, the values of these two objectives are in a different scale. Thus, data normalization is needed to rescale the values in the
same range. Here, MinMax normalization is used to transform the data into the range [0, 1]. After rescaling the values of these two
objectives, the weighted average strategy is then employed to find the optimal solutions from the Pareto-optimal front. Suppose for
a given weight vector (w1, w2), the optimal solution is found by searching the minimum value of w1 T pri þ w2 Cpri for all the
solutions in the Pareto-optimal front, where T pri and Cpri are normalized values of the machining time and production cost,
respectively. The optimal solutions chosen with a weight vector (w1, w2) ¼ (0.80, 0.20) for all cases are shown in Table 8.
After getting the optimal cutting parameters at a different depth of cut, we need to evaluate the machining time and production
cost under a different depth of cut. Table 8 indicates that with an increase in the depth of cut, the machining time per pass increases
slightly. But at a higher depth of cut, fewer cutting passes are needed, so it takes slightly less time to remove the same amount of
depth of cut with a higher depth of cut. However, the production cost increases significantly or even superlinearly with the increase
in the depth of cut. Therefore, when considering the tradeoff between the machining time and the production cost, it is better to
select the cutting conditions at the lower depth of cut of 0.075 mm than those for the other three cases. So the most optimal cutting
strategy is at a depth of cut of 0.075 mm, a cutting speed of 417.623 m min1, and a feed rate per tooth of 0.0852 mm per rev., and
eight cutting passes are needed to remove the stock.
According to the recommended data listed in (48), a better cutting performance of BCBN tools is found at a higher cutting speed
(400 m min1), a low feed rate per tooth (0.05 mm per rev.), and a low depth of cut (0.05 mm). Under these cutting conditions, the
machining time and production cost are 0.469 min and US$3.058, respectively. Obviously, the cutting performance in terms of
machining time and production cost under these conditions is worse than that under the calculated optimal strategies with PGSA.

11.10.6.3 High-Speed Machining of Titanium Alloys  Case Study 2

Now a more complicated example is considered; its final dimension is shown in Figure 24. The BCBN tool is used for finishing
machining the bottom surface of the pocket and the slot in the inner island, and the total depth of cut is 0.30 mm.
The width of the pocket and slot is greater than the diameter of the cutter, so two passes are needed to remove one layer of the
bottom surface. A contour-parallel tool pass pattern has been chosen as the pocketing strategy. For the pocketing, the first pass is to
reduce the inward for the outer contour, and the second one is to reduce the outward for the inner island. In the first cutting pass, the
radial depth of cut is equal to the diameter of the cutter, and the same procedure as that used in the last section can be used to find
the optimal cutting strategy. For the second pass, the difference between the width of the pocket and the diameter of the cutter is the
actual radial depth of cut. A different radial depth of cut results in a different maximum instantaneous undeformed chip thickness.
The same situation exists for the two cutting passes of the slotting milling in the inner island. In the following, we focus on the
optimization of the second pass, where the maximum value of h(4) is less than that of the feed rate per tooth.
When the radial depth of cut is less than the diameter of the cutter, it is assumed that there is no insert throw and tool
eccentricity. The ideal cutting zone geometry is shown in Figure 25, where ar is the radial depth of cut.
The cutting forces and machining power constraints of the optimization model are more relative to the maximum uncut chip
thickness. Therefore, the maximum uncut chip thickness needs to be predicted based on the cutting zone geometry as shown in
Figure 25. According to the equation of instantaneous chip thickness calculation in (49), the maximum chip thickness BC in
Figure 25 can be derived as follows:
sinðp=2 þ 4Þ  sinðp=2  4  dÞ
BC ¼ R  CD ¼ R [67]
sinðp=2 þ 4Þ
   
p R  ar f p
where 4 ¼ 40  d; 40 ¼  arcsin , Let a ¼ sin þ 4 . Then
2 R R 2

Table 8 Optimal solutions with weighted average strategy

Depth of cut (mm) Cutting speed (m min1) Feed rate per tooth (mm per rev.) Machining time (min) Production cost (US$)

0.075 417.62 0.0852 0.264 2.88

0.100 436.12 0.0734 0.293 6.48
0.125 437.12 0.0595 0.361 17.17
0.150 438.30 0.0502 0.427 48.12
Reproduced from Wang, Z. G. High-Speed Milling of Titanium Alloys: Modeling and Optimization. PhD Thesis, National University of Singapore,
2005.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
250 High-Speed Machining

Figure 24 Geometrical dimension of test part 2.

1 2 3 þ 2p2 3 3 þ 8p2 4 15 þ 100p2 þ 18p4 5  

d¼a a þ a  a þ a þ O a6
2p 12p2 24p3 240p4
In order to get the value of BC, the only unknown parameter is 4. In this study, since f/R is very small, it means that a is very
small, so d can be approximated as a.
If we let a0 ¼ f/R, then the following equation can result:
4 ¼ 40  d ¼ 40  a ¼ 40  a0 cos 4 [68]
After solving the above equation, 4 can be represented as follows:
 
4 ¼ 40  a0 cos 40  sin 40 cos 40 a20 þ O a30 [69]

After obtaining the value of 4, the value of BC can be calculated further. Then, the constraints of the cutting forces and machining
power can be justified. With the same optimization algorithm described before, multiple Pareto-optimal solutions are obtained in
a single simulation as shown in Figure 26. These solutions show that Pareto-optimal regions have a convex shape, so a weighted
average strategy can also be used to choose the optimal solution from the Pareto-optimal regions. With a weight vector (w1, w2) ¼
(0.80, 0.20), the optimal cutting parameters at four axial depths of cut are obtained, as shown in Table 9.
Table 9 shows that the production cost increases significantly or even superlinearly with the increase in the depth of cut.
Although a higher depth of cut can lessen the production time to a certain extent, it results in a significant increase in the production

Workpiece True tool

trajectory
ar B
Tool tip
C
Chip
0 thickness
O A O D E

Figure 25 Cutting zone geometry for the second pass of pocketing. Reproduced from Wang, Z.G. High-speed milling of titanium alloys: modeling and
optimization, PhD thesis, National University of Singapore, 2005.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 High-Speed Machining 251

Figure 26 Population after termination of simulation at different depths of cut for test part 2: (a) a ¼ 0.075 mm; (b) a ¼ 0.100 mm; (c) a ¼ 0.125 mm;
(d) a ¼ 0.150 mm. Reproduced from Wang, Z.G. High-speed milling of titanium alloys: modeling and optimization, PhD thesis, National University of
Singapore, 2005.

Table 9 Optimal solutions to test part 2 with weighted average strategy

Depth of cut (mm) Cutting speed (m min1) Feed rate per tooth (mm per rev.) Production time (min) Production cost (US$)

0.075 450.00 0.100 0.2078 3.2100

0.100 450.00 0.100 0.2078 10.6193
0.125 449.62 0.089 0.2337 32.9934
0.150 450.00 0.078 0.2682 68.3204
Reproduced from Wang, Z. G. High-Speed Milling of Titanium Alloys: Modeling and Optimization. PhD Thesis, National University of Singapore,
2005.

cost. Therefore, cutting conditions at a lower depth of cut are preferred. Finally, cutting parameters at a depth of cut of 0.075 mm are
much more optimal than those for the other three cases.
When considering the production time and production cost of two passes, the optimal cutting strategy is obtained when the depth
of cut is equal to 0.075 mm. For the first pass, the optimal cutting speed and feed rate per tooth are 417.62 mm min1 and 0.085 mm
per rev., respectively, which is same as that obtained in the first application example. For the second pass, the optimal cutting
parameters are at a depth of cut of 0.075 mm, a cutting speed of 450.0 m min1, and a feed rate of 0.100 mm per rev. In all, four cutting
passes are needed to remove the stock. A test component was also tried, and the experimental results were proven to be satisfactory.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
252 High-Speed Machining

11.10.7 Conclusions

Enabled by advances in tooling and control capabilities, high-speed machining is finding wider acceptance in aerospace applica-
tions. Although especially good for aluminum, the technique is finding its place in composites and hard-metal machining as well.
The poor thermal conductivity of titanium alloys results in a concentration of high temperatures at the toolworkpiece and tool
chip interfaces. Consequently, the cutting heat accelerates tool wear and increases manufacturing cost. Traditional cutting tool
materials can only be used for machining titanium alloys at moderate cutting speed. Advanced tool materials, such as PCD and
CBN, are capable of machining titanium alloys at high cutting speed. For all these tool materials, their tool life is limited by
extremely high cutting temperature and high stresses generated at the cutting edge during machining of titanium alloys. All these
tool materials could not maintain their hardness at higher temperature, but the BCBN tools are an alternative tool material that can
withstand the severe cutting temperature and cutting pressure and maintain their hardness at high temperature (around 1000  C).
In addition, the BCBN tools exhibit excellent mechanical properties and superior thermal stability. BCBN would appear to be the
most functionally satisfactory cutting tool material now available for machining titanium alloys.
The generation of mathematical models of high-speed milling is essential for a deeper understanding of this advanced process.
Therefore, an analytical model needs to be established to predict the cutting forces, tool wear, and cutting temperature for high-
speed milling of Ti6Al4V. Based on Oxley’s machining theory, the authors proposed a cutting force model to estimate the
cutting forces as well as the cutting temperature, and this model has been found to predict the cutting forces with good accuracy.
Since it is relatively expensive to use most HSM machine tools, it is necessary to optimize the HSM operation. One advanced
multiobjective optimization algorithm was reviewed in this chapter and then tested with two case studies. It was found that this
algorithm can converge to multiple Pareto solutions. Finally, the process planners can select optimal cutting conditions based on
their targeted objectives.

References

1. Schulz, H. Why High Speed Cutting (HSC). In Proceedings of the International Conference on High Speed machining, Nanjing: P.R. China, April 2004; pp 120.
2. MatWeb Material Property Data, http://www.matweb.com/, 2011.
3. Tlusty, G. Manufacturing Processes and Equipment; Prentice Hall: USA, 2000.
4. King, R. I.; Vaughn, R. L. A Synoptic Review of High-Speed Machining from Salomon to the Present. In High Speed Machining: Presented at the Winter Annual Meeting of
ASME, New Orleans, Louisiana; Komanduri, R., Subramanian, K., von Turkovich, B. F., Eds.; ASME: New York, 1984; pp 113.
5. Siekmann, H. J. How to Machine Titanium, Tool Engineer, 1955, pp 7882.
6. Ezugwu, E. O.; Wang, Z. M. Titanium Alloys and Their Machinability  A Review. J. Mater. Process. Technol. 1997, 68 (3), 262274.
7. Ezugwu, E. O.; Da Silva, R. B.; Bonney, J.; Machado, A. R. The Effect of Argon-Enriched Environment in High-Speed Machining of Titanium Alloy. Tribol. Trans. 2005, 48 (1),
1823.
8. Wang, Z. G.; Rahman, M.; Wong, Y. S.; Neo, K. S., et al. Study on Orthogonal Turning of Titanium Alloys with Different Coolant Supply Strategies. Int. J. Adv. Manuf. Technol.
2009, 42 (78), 621632.
9. Sumitomo Electric Industries. Performance Cutting Tools; Sumitomo Electric Industries: Japan, 2000.
10. Sandvik Hard Materials. Cemented Carbide Rod Blanks for Metal Cutting; Sandvikens Tryckeri, 2001.
11. Lei, S. T.; Liu, W. J. High-Speed Machining of Titanium Alloys Using the Driven Rotary Tool. Int. J. Mach. Tools Manuf. 2002, 42 (6), 653661.
12. Muller, B.; Renz, U.; Hoppe, S.; Klocke, F. Radiation Thermometry at a High-Speed Turning Process. Trans. ASME J. Manuf. Sci. Eng. 2004, 126 (3), 488495.
13. Zoya, Z. A.; Krishnamurthy, R. The Performance of CBN Tools in the Machining of Titanium Alloys. J. Mater. Process. Technol. 2000, 100 (13), 8086.
14. Ezugwu, E. O.; Da Silva, R. B.; Bonney, J.; Machado, Á. R. Evaluation of the Performance of CBN Tools when Turning Ti6Al4V Alloy with High Pressure Coolant Supplies.
Int. J. Mach. Tools Manuf. 2005, 45 (9), 10091014.
15. Bhaumik, S. K.; Divakar, C.; Singh, A. K. Machining Ti-6Al-4V Alloy with a wBN-CBN Composite Tool. Mater. Des. 1995, 16 (4), 221226.
16. Quintana, G.; de Ciurana, J.; Campa, F. J. Machine Tool Spindles. In Machine Tools for High Performance Machining; López de Lacalle, L. N., Lamikiz, A., Eds.; Springer:
London, 2009 (chapter 3).
17. Sharma, I. R. Latest Trends in Machining, 2011. www.drishtikona.com/books/latest-trends-in-machining/ch-all.pdf.
18. Kalpakjian, Z.; Schmid, S. Manufacturing Processes for Engineering Materials; Prentice Hall: Upper Saddle River, NJ; 2009.
19. van Luttervelt, C. A.; Childs, T. H. C.; Jawahir, I. S.; Klocke, F.; Venuvinod, P. K. Present Situation and Future Trends in Modelling of Machining Operations. Progress Report of
the CIRP Working Group ‘Modelling of Machining Operations’. CIRP Ann. 1998, 47 (2), 587626.
20. Merchant, M. E. Basic Mechanics of the Cutting Process. ASME J. Appl. Mech. 1944, 67, 168175.
21. Merchant, M. E. Mechanics of the Metal Cutting Process. J. Appl. Phys. 1945, 16 (6), 318324.
22. Armarego, E. J. A.; Wang, J. P.; Deshpande, N. P. Computer-Aided Predictive Cutting Model for Forces in Face Milling Allowing for Tooth Run-Out. CIRP Ann. 1995, 44 (1),
4348.
23. Altintas, Y.; Budak, E. Analytical Prediction of Stability Lobes in Milling. CIRP Ann. 1995, 44 (1), 357362.
24. Oxley, P. L. B. The Mechanics of Machining: An Analytical Approach to Assessing Machinability; E. Horwood: Chichester (England), 1989; pp 23135.
25. Moufki, A.; Molinari, A.; Dudzinski, D. Modeling of Orthogonal Cutting with a Temperature Dependent Friction Law. J. Mech. Phys. Solids 1998, 46 (10),
21032138.
26. Jacobus, K.; DeVor, R. E.; Kapoor, S. G.; Peascoe, R. A. Predictive Model for the Full Biaxial Surface and Subsurface Residual Stress Profiles from Turning. Trans. ASME
J. Manuf. Sci. Eng. 2001, 123 (4), 537546.
27. Huang, Y.; Liang, S. Y. Cutting Forces Modeling Considering the Effect of Tool Thermal Property  Application to CBN Hard Turning. Int. J. Mach. Tools Manuf. 2003, 43 (3),
307315.
28. Altintas, Y. Modeling Approaches and Software for Predicting the Performance of Milling Operations at MAL-UBC. In Presented in CIRP 2000 Machining Workshop; University of
New South Wales: Sydney, 2000.
29. Ozel, T. Investigation of High Speed Flat End Milling Process: Prediction of Chip Formation, Cutting Forces, Tool Stresses and Temperatures. Ph.D. Thesis, The Ohio State
University, 1998.
30. Ozel, T.; Altan, T. Process Simulation Using Finite Element Method  Prediction of Cutting Forces, Tool Stresses and Temperatures in High-Speed Flat End Milling. Int. J. Mach.
Tools Manuf. 2000, 40 (5), 713738.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 High-Speed Machining 253

31. Ng, E. G.; Aspinwall, D. K.; Brazil, D.; Monaghan, J. Modeling of Temperature and Forces when Orthogonally Machining Hardened Steel. Int. J. Mach. Tools Manuf. 1999,
39 (6), 885903.
32. Carrino, L.; Giuliano, G.; Napolitano, G. Finite Element Simulation of Orthogonal Metal Cutting. In Computational Methods in Contact Mechanics VI; Brebbia, C. A., Ed.; WIT
press: Southampton, 2003; pp 105114.
33. Lazoglu, I.; Altintas, Y. Prediction of Tool and Chip Temperature in Continuous and Interrupted Machining. Int. J. Mach. Tools Manuf. 2002, 42, 10111022.
34. Pittalà, G. M.; Monno, M. A New Approach to the Prediction of Temperature of the Workpiece of Face Milling Operations of Ti-6Al-4V. Appl. Therm. Eng. 2011, 31 (23),
173180.
35. Warnecke, G.; Oh, J. D. A New Thermo-Viscoplastic Material Model for Finite-Element-Analysis of the Chip Formation Process. CIRP Ann. 2002, 51, 7982.
36. Fang, N.; Jawahir, I. S. An Analytical Predictive Model and Experimental Validation for Machining with Grooved Tools Incorporating the Effects of Strains, Strain-Rates, and
Temperatures. CIRP Ann. 2002, 51, 8386.
37. Ueda, T.; Hosokawa, A.; Oda, K.; Yamada, K. Temperature on Flank Face of Cutting Tool in High Speed Milling. CIRP Ann. 2001, 50, 3740.
38. Wang, Z. G. High-Speed Milling of Titanium Alloys: Modeling and Optimization. Ph.D. Thesis, National University of Singapore, 2005.
39. Altan, T.; Oh, S. I.; Gegel, H. L. Metal Forming: Fundamentals and Applications; American Society for Metals: Metals Park, OH, 1983; pp 4546.
40. Johnson, G. R.; Cook, W. H. A Constitutive Model and Data for Metals Subjected to Large Strains, High Strain Rates and High Temperatures. In Proceedings of the 7th
International Symposium on Ballistics, The Hague: Netherlands, 1983; pp 541547.
41. Lee, W. S.; Lin, C. F. High-Temperature Deformation Behaviour of Ti6Al4V Alloy Evaluated by High Strain-Rate Compression Tests. J. Mater. Process. Technol. 1998, 75 (13),
127136.
42. Majorell, A.; Srivatsa, S.; Picu, R. C. Mechanical Behavior of Ti-6Al-4V at High and Moderate Temperatures  Part I: Experimental Results. Mater. Sci. Eng. A 2002, 326 (2),
297305.
43. Lee, W. S.; Lin, C. F. Plastic Deformation and Fracture Behaviour of Ti-6Al-4V Alloy Loaded with High Strain Rate under Various Temperatures. Mater. Sci. Eng. A 1998,
241 (12), 4859.
44. Wang, Z. G.; Wong, Y. S.; Rahman, M. Optimisation of Multi-Pass Milling Using Genetic Algorithm and Genetic Simulated Annealing. Int. J. Adv. Manuf. Technol. 2005, 24 (9
10), 727732.
45. Steuer, R. E. Multiple Criteria Optimization: Theory, Computation, and Application; Wiley: New York, 1986; pp 138164.
46. Deb, K. Multi-Objective Evolutionary Algorithms: Introducing Bias Among Pareto Optimal Solutions. KanGAL Report 99002; Indian Institute of Technology: Kanpur, India, 1999.
47. Wang, Z. G.; Wong, Y. S.; Rahman, M.; Sun, J. Multi-Objective Optimization of High-Speed Milling with Parallel Genetic Simulated Annealing. Int. J. Adv. Manuf. Technol. 2006,
31, 209218.
48. Zareena, A. R. High-Speed Machining of Titanium Alloys. Master Thesis, National University of Singapore, 2002.
49. Wang, Z. G.; Rahman, M.; Wong, Y. S.; Li, X. P. A Hybrid Cutting Force Model for Machining of Titanium Alloys. CIRP Ann. 2005, 54, 7174.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
This page intentionally left blank

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 11.11 Machining with Least Quantity Lubrication
T Obikawa, The University of Tokyo, Meguro-ku, Tokyo, Japan
 2014 Elsevier Ltd. All rights reserved.

11.11.1 Introduction 255

11.11.1.1 Environmentally Conscious Machining Technologies 255
11.11.1.2 Environmental and Cost Adaptabilities of MQL Machining 256
11.11.2 Tribology in MQL Machining 258
11.11.2.1 Cutting Oil 258
11.11.2.2 Capillary Lubrication Models 259
11.11.2.3 Gas-Phase Lubrication 261
11.11.2.4 Liquid-Phase Lubrication 263
11.11.2.5 Stop Distance of an Oil Droplet 264
11.11.3 MQL Machining Technologies 266
11.11.3.1 Oil Mist Generation and Application 266
11.11.3.2 Milling 269
11.11.3.3 Turning 273
11.11.3.4 Drilling 276
11.11.3.5 Micro-MQL Machining and Other Future-Oriented MQLs 278
11.11.4 Concluding Remarks 279
Acknowledgment 279
References 279

11.11.1 Introduction
11.11.1.1 Environmentally Conscious Machining Technologies
Machining with the minimum quantity of lubricant (MQL) is one of the so-called environmentally conscious machining tech-
nologies (ECMTs) and also is one of sustainable machining technologies (1). In MQL machining, only a small amount of oil mist is
sprayed with compressed air to a cutter or a single point cutting tool, as is shown in Figure 1 (2). Oil consumption is typically
around 10 ml h1, which is much less than the normal flow rate of flood coolant. This machining technology can also minimize the
environmental impact through the following striking features: (1) it can reduce waste, especially waste of coolant; (2) it can reduce
electric power and relevant CO2 emission; and (3) for the cutting oil, biodegradable oil with nontoxic additives can be used instead
of mineral oil. Owing to these advantages, MQL machining has so far been widely applied to milling, turning, drilling, hobbing, and
so on for various work materials, and it is thus broadly agreed that MQL machining is generally preferable to other ECMTs.
Shokrani et al. (3) classified ECMTs into three categories: (1) dry machining, (2) MQL machining, and (3) machining with gas-
based coolant including machining with chilled air (4) and cryogenic machining with liquid nitrogen (5). In dry machining, since
the rotation of the work or the cutting tool can exert an extremely weak forced air cooling effect upon the cutting zone, the possible
improvement in machinability should lie solely in heightening the heat resistance of the tool through developing the coating
technology, resulting in the scope of application restricted to certain cutting conditions for some specific work materials. With

Figure 1 Oil mist sprayed to the tip of a single point cutting tool from both sides of the flank and rake faces. Fogging is intentionally caused to
visualize the sprayed mist of cutting oil. Courtesy of Fuji BC Engineering Co., Ltd. Reproduced from http://www.fuji-bc.com/english/ebtool/index.html.

Comprehensive Materials Processing, Volume 11 http://dx.doi.org/10.1016/B978-0-08-096532-1.01114-6 255

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
256 Machining with Least Quantity Lubrication

regard to gas-based coolant, chilled air alone cannot sufficiently cool down the cutting zone and only liquid nitrogen has the ability
to sufficiently cool down it, but there are unfavorable aspects with liquid nitrogen. Liquid nitrogen must always be handled
extremely carefully, and it may cause some troubles with machine tools as well as human bodies due to its extremely low
temperature. Also, if liquid nitrogen is carelessly spilled over the floor, it may result in the deficiency of oxygen and possible fatal
accidents. Note also that liquid nitrogen is not reusable and quite expensive. It can thus be generally concurred that machining with
liquid nitrogen can be suitable only when high value will be added to the end product or when the tool wear can be dramatically
reduced thereby (5). Regardless of these shortcomings, this method is still regarded as a promising alternative to cutting oil from the
viewpoint of sustainable machining technologies.

11.11.1.2 Environmental and Cost Adaptabilities of MQL Machining

Oil consumption in MQL machining operations can hugely be cut down and waste of coolant produced in machining lines can be
reduced as well (1). It thus becomes economically possible to replace mineral oil with expensive biodegradable oils such as
vegetable oils and synthetic esters, which are preferable not only for preserving the global environment but also for eliminating
health hazards in humans as well as improving the work environment. Note that as much oil mist sprayed to the cutting region as
possible must later be recovered with a mist collector, because fine oil droplets can go deep inside human lungs to cause serious
health problems. Hence, the mist collector is always indispensable in MQL machining. Some readers may be concerned about the
extra cost incurred by installing mist collectors, but those readers are encouraged to be reminded that the mist collector is also highly
recommended to introduce in machining with high pressure coolant or flood coolant all the same. Especially in machining with
high pressure coolant, a myriad of fine particles of mineral oil-based coolant will be generated when the coolant impinge upon the
cutting tool, work, spindle, table, and so on, likely resulting in serious health hazards. The author thus believes that the expense for
installing the mist collector is not peculiar to MQL machining alone.
With regard to the waste of coolant, Mazda Motor Corporation, a leading Japanese motor company, disclosed that the share of
waste of coolant before concentration reached as much as 32% of all the waste to be incinerated, as shown in Figure 2 (6). It should
be noted that the waste of water-based coolant requires time-consuming treatments in comparison with that of oil-based coolant
not containing water.
As long as chips can be disposed without flood coolant, MQL machining does not require any coolant pump. This salient feature
further leads to the reduction in electric power necessary for pumps or other coolant-related equipment, which are indispensable in
wet machining. Before Toyota Motor Corporation and Mazda Motor Corporation implemented the MQL machining system in their
machining lines, the extra power used for coolant-related equipment against the total electric power reached as high as 42% in
Toyota, Figure 3 (7), and 34% in Mazda, Figure 4 (6), but the power actually used for driving spindles was only 16% of the total
power, Figure 3, though cleaners of machined parts spent as much as 18%, Figure 4. These data lucidly illustrate that coolant pumps
consume the largest portion of power in the machining lines. This also means that the reduction in electric power by eliminating
coolant pumps is substantial, and hence it is indicated that CO2 inevitably emitted during generating electricity will also be
considerably reduced by replacing the flood coolant method with the MQL machining system. Some forecasts suggest that the
electric power used in machining lines could be cut down roughly by half if cleaning of machined parts also becomes unnecessary
by adopting MQL machining.
Machining costs for wet and MQL machining were estimated by HORKOS Corporation, a leading Japanese machine tool
builder, in Figure 5 (8). The cost reduction by changing wet machining to MQL machining is much larger than expected from the
discussion as to power reduction above, despite biodegradable cutting oils for MQL machining are expensive. This arguably reflects

Others
Machine
(2%)
oil (4%)
Cutting oil before
concentration (32%)
Sludge (19%)

Plastics &
paints (15%)

Paper & wood (28%)

Figure 2 Waste brought to waste incineration equipment. Courtesy of Mazda Motor Corporation. Reproduced from Press Release from the Energy
Conservation Center, Japan: The Minister Prize of Economy, Trade and Industry in Awarding of Successful Case of Energy Conservation in Factory
and Building in 2002 (in Japanese).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Machining with Least Quantity Lubrication 257

Lights
Air-conditioners
Coolant pumps & related
Assembly equipment (42 (53)%)

Others

Mist fans (6 (7)%)

Hydraulic units (6 (7)%)

Air (8 (10)%)

Spindles (16 (20)%)

Figure 3 Electric power consumption in machining lines. Figures in parentheses denote the electric power consumed in machining centers. Cour-
tesy of Toyota Motor Corporation. Reproduced from Ikawa, S.; Iwatsubo, M.; Miyazaki, K. To This Point It Has Come. Dry Machining. Challenge to
Dry Machining for Environmental Correspondence. Mech. Eng. 1999, 47 (5), 18–22 (in Japanese).

Others (21%) Coolant pumps (29%)

Other coolant
equipment (5%)
Machining &
transfer (27%)
Cleaners (18%)

Figure 4 Electric power consumption in machining lines. Courtesy of Mazda Motor Corporation. Reproduced from Press Release from the Energy
Conservation Center, Japan: The Minister Prize of Economy, Trade and Industry in Awarding of Successful Case of Energy Conservation in Factory
and Building in 2002 (in Japanese).

Figure 5 Cost to MQL machining and to wet machining with flood coolant. Reproduced from http://www.horkos.co.jp/english/products/imql01.php.
Courtesy of HORKOS Corp.

the fact that the labor cost to maintain or replace the coolant, as well as to clean the coolant tank, is not marginal at all. The
depreciation of the machine tool in market value can also be significant as the machine tool needs no coolant pump or other
coolant-related equipment. In addition to those various costs, the cost for cleaning the machined products can also be reduced by
MQL machining. Moreover, since the machined products and removed chips in MQL machining are dry and relatively clean, chips
can be sent to recycling without additional processing. Owing to these remarkable advantages, it can generally be considered that
MQL machining is preferable from the environmental and economic points of view. Note that if it is the case that coolant is
necessary for chip disposal, then the reduction in machining cost would be hugely offset by the extra coolant system, and hence, the
strategy on coolant management has to be rectified in such a case.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
258 Machining with Least Quantity Lubrication

11.11.2 Tribology in MQL Machining

In this section, tribological issues related to MQL machining such as cutting oil for MQL machining, the mechanism of lubrication
on the tool–chip and tool–work interfaces, and the flight of oil droplets to the tool face are discussed.

11.11.2.1 Cutting Oil

Various types of vegetable oils, synthetic esters, and water-miscible or immiscible mineral oils can be used as the base oil for MQL
machining. With regard to the biodegradability, vegetable oil is excellent, synthetic ester is good, and mineral oil is poor. Surfactants
and extreme pressure additives are usually added to the base oil so that the cutting oil can retain excellent lubricity even under
tribologically severe conditions in machining. Eisentraeger et al. (9) examined the effects of additives and usage of cutting oil in
machining operations upon the biodegradability in the case of a synthetic ester-based cutting oil that was originally developed as
flood coolant. Chemical compositions of the cutting oil formulated with additives used in their study about flood coolant were, in
percent by mass, 72.8% carbon, 12.3% hydrogen, and 14.9% oxygen, and those for the base oil without additives were 72.9%
carbon, 12.4% hydrogen, and 14.6% oxygen. They spotted that the biodegradation increased up to 74% after 28 days and that the
biodegradability was independent of additives and usage of cutting oil, as shown in Figure 6 (9). They commented that antioxidants
added to the ester would exert a positive effect on the biodegradability of cutting oil that have been exposed to a harsh environment
at high temperatures under high pressures during machining. The biodegradation of the tested cutting oil, i.e., 74% after 28 days, is
almost the same as that of a commercially available cutting oil for MQL machining (kinematic viscosity at 40  C (KV40) is
19 mm2 s1) that exhibits the biodegradability of 72% (OECD 301B) (10). Note that formulated mineral oil for hydraulic fluid
(unused) exhibits much less biodegradation, that is, only 24% (9).
When end milling or drilling an aluminum alloy, the MQL method requires a cutting oil with excellent lubricity because this
alloy tends to severely adhere to end mills or drills under a wide range of cutting conditions. In the view above, Nagata et al. (11)
optimized the composition of MQL cutting oil for end milling of aluminum alloys. They tested the lubricity of candidate oils as well
as the tendency of an aluminum alloy to adhere to a tool by thread rolling with MQL. Note that the test of thread rolling is often
used to evaluate the lubricity of cutting oil. The minimum quantity of oil, qmin, to prevent the aluminum alloy from adhering to the
tool during the test was obtained for ester-based cutting oils, Figure 7 (11), and for mineral oil-based cutting oils, Figure 8 (11). qmin
for a vegetable oil and an alkyl chain surfactant with 12 carbon atoms, called C12 surfactant, are also plotted in Figures 7 and 8.
It can be seen from these figures that the C12 surfactant exhibits a much lower lubricity than the vegetable oil. It can also be
underlined that the lubricity will be improved by adding esters having 12 and 18 carbon atoms to the C12 surfactant and that these
added esters will in turn reduce the minimum quantity of oil, qmin. Note that although the length of the alkyl chain or the carbon
number of an ester does not significantly affect the lubricity of ester-based cutting oil, the lubricity can also be improved by adding
mineral oils to the C12 surfactant. Note that the qmin will be diminished in this case as well. This implies that the molecular weight
of a mineral oil will definitely affect the lubricity of the mineral oil-based cutting oil. In point of fact, it was transpired that a mineral
oil having a small molecular weight of 300 showed the best lubricity among the tested oils. This further indicates that the
permeability as well as the lubricity is essentially important to enhance the performance of cutting oil under a tribologically severe
condition including machining, where metallic contact is inevitable.

Figure 6 Biodegradation of cutting fluids after 28 days; ZDR, machine tool with cutting operations; standard deviation of three parallels; BOD/ThOD,
biological oxygen demand in relation to the theoretical oxygen demand; TIC/ThIC, carbon dioxide production in relation to the theoretical carbon
dioxide production. Reproduced from Eisentraeger, A.; Schmidt, M.; Murrenhoff, H.; Dott, W.; Hahn, S. Biodegradability Testing of Synthetic Ester
Lubricants – Effects of Additives and Usage. Chemosphere 2002, 48, 89–96.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Machining with Least Quantity Lubrication 259

Figure 7 Minimum quantity of ester-based cutting oil to prevent the aluminum alloy from adhering to a tool during thread rolling with MQL.
Reproduced from Nagata, M.; Morita, H.; Yamada, T.; Iwasa, M. Approach to MQL Cutting by Using High Lubrication Cutting Fluid. In Proceedings
of the Third International Conference on Leading Edge Manufacturing in 21st Century, Nagoya; 2005; 127–132.

Figure 8 The minimum quantity of mineral oil-based cutting oil to prevent the aluminum alloy from adhering to a tool during thread rolling with
MQL. Reproduced from Nagata, M.; Morita, H.; Yamada, T.; Iwasa, M. Approach to MQL Cutting by Using High Lubrication Cutting Fluid. In
Proceedings of the Third International Conference on Leading Edge Manufacturing in 21st Century, Nagoya; 2005; 127–132.

Wakabayashi et al. (12) investigated the performance of cutting oil mixed with different alcohols in MQL machining of an
aluminum alloy, JIS AC8A. The base oil was a synthetic polyol ester (KV40 ¼ 49.1 mm2 s1) for MQL drilling and a synthetic ester-
based commercial MQL cutting oil (KV40 ¼ 19 mm2 s1) for MQL end milling. The concentration of alcohol was 20 mass% for
MQL drilling and 10 mass% for MQL milling. In a test case, oleyl alcohol (OAL) (KV40 ¼ 19.4 mm2 s1) and lauryl alcohol (LAL)
(KV40 ¼ 12.1 mm2 s1) were added to the synthetic polyol ester, and the thrust force was reduced as a result by about 20% in MQL
drilling with a diamond-like carbon (DLC) coated carbide drill. This is arguably because the added alcohol had lowered the viscosity
of the mixed cutting oil in which the base oil alone had a high viscosity. In another case, LAL, decyl alcohol (KV40 ¼ 11 mm2 s1),
and benzyl alcohol (KV40 ¼ 8 mm2 s1) were added to the commercial MQL cutting oil in MQL down milling under the condition
that the cutting speed was 110 m min1, the axial depth of the cut was 10 mm, the radial depth of the cut was 5 mm, the feed rate
was 0.3 mm tooth1, and the amount of oil was 20 ml h1. It was observed that the aluminum alloy firmly adhered to an high-
speed steel (HSS) end mill at the early stage of machining, as shown in Figure 9, because the added alcohol had diminished the
viscosity of the mixed cutting oil from its optimized value. In contrast, the addition of OAL with almost the same value of KV40 as
the commercial MQL cutting oil did not allow the aluminum alloy to severely adhere to the end mill until the end of the cutting
experiment up to the cutting length of 1600 mm. Note that the molecular weight of OAL with 18 carbon atoms is 268. Figures 7–9
manifest that a cutting oil with a molecular weight around 300 with a high lubricity and permeability should be most suitable for
MQL machining.

11.11.2.2 Capillary Lubrication Models

In MQL machining, there are two different mechanisms of lubrication: one is the gas-phase lubrication, and the other is the fluid
lubrication with liquid oil. The former is theorized for the tool–chip contact area by Williams and his colleagues (13–15) and can
also be applied to the contact area between the worn tool flank face and machined surface. According to their model, the contact
area is lubricated by absorbing gas that comes from the chip leaving point through a number of capillaries on the interface between
the tool face and chip. A similar model can be assumed for the fluid lubrication that the contact area is lubricated with cutting oil
coming through the capillaries. The locations of capillaries always change, and this will arguably enable absorbed gas or trapped oil
to well pervade across the pertinent surface with some domains that locally adhere to another surface. As a result, the friction

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
260 Machining with Least Quantity Lubrication

Figure 9 Cutting length without severe adhesion of the aluminum alloy to an end mill. Reproduced from Wakabayashi, T.; Atsuta, T.; Tsukuda, A.;
Sembongi, N.; Shibata, J.; Suda, S. Cutting Performance of Oxygen-Including Compounds in MQL Machining of Aluminum. Key Eng. Mater. 2012,
523–524, 967–972.

between the tool–chip and tool–work interfaces will be reduced, leading also to total reduction in tool–chip contact length, cutting
forces, and tool wear.
A schematic diagram of the lubrication mechanism based on the capillary model is shown in Figure 10, in which one capillary
on the tool–chip contact area and another capillary on the interface between the worn flank and machined surfaces are assumed to
be aligned on a longitudinal cross section parallel to both the chip flow direction, x, and cutting direction, x, in orthogonal
machining. Note that the original diagram in Ref. (16) was depicted only for explaining the gas-phase lubrication on the worn flank
face in MQL machining. Note also that neither the tool face nor the machined surface in Figure 10 is assumed to be perfectly flat,
and the fluid lubrication model is illustrated for a capillary on the rake face side but the gas-phase lubrication model is depicted for
a capillary on the flank face side, because this mixed model of fluid and gas-phase lubrication should highly likely grasp the realities
in MQL machining. The maximum roughness of the machined surface may be characterized by the height and width of a capillary, h
and w (13), which is more or less one micrometer, and several times the capillary height, respectively. The mean height of capillaries
is defined as h ¼ h for a rectangular cross section of capillary and as h ¼ ph=4 for a thin elliptic cross section. In the following
analysis, the mean height of capillaries is assumed to be constant. Then, the aspect ratio of a capillary, defined as the ratio of the
length against the height, will be of the order of 10–100.

Figure 10 Capillary lubrication models. Models for gas- and liquid-phase lubrication mechanism are depicted for the flank and rake faces of the
cutting tool, respectively.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Machining with Least Quantity Lubrication 261

Figure 11 Most effective gas-phase lubrication.

Among a huge number of oil droplets sprayed toward the tool tip, only a small portion of droplets will be attached to the
rake, flank, chip, or machined surfaces in the cutting zone. If the attached position of a small oil droplet is close enough to the
tool tip, the droplet will rapidly evaporate due to the high temperature around the tool tip, and the vaporization heat, together
with the compressed air, will in turn reduce the tool temperature. The vaporized oil can permeate into a capillary pushed by the
pressure of vapor to lubricate the interface between the chip and tool rake face or between the machined surface and worn tool
flank face. In attention, the vaporized oil and air permeated into a capillary will rapidly be absorbed into the new surface
generated at the cutting edge, resulting in the reduction in the surface energy of the fresh surface. The pressure inside capillaries is
thus always kept considerably low. This is commonly interpreted, “The bottom end of a capillary is always kept in a vacuum.” If
an oil droplet is relatively large, it may take a little longer to completely evaporate. Such a large oil droplet may be pushed by
compressed air up to a capillary, and if the droplet begins to evaporate at the open mouth of the capillary, as shown in Figure 11,
the gas-phase lubrication with the vaporized oil will appear under higher pressure of vapor than in the case shown in Figure 10,
because oil vapor will be supplied directly and continuously from the oil droplet at the mouth of capillary in this case. If
evaporation of oil is negligible, as can be expected in the case of flood coolant, fluid lubrication with liquid oil will appear. There
may be some readers who think the smaller a droplet is, the better. It is, however, exceedingly difficult for a small droplet to
reach the mouth of a capillary, let alone attach to the tool face, since its inertia is too small to come out of the main airflow, as
shall be described in Section 11.11.2.5.

11.11.2.3 Gas-Phase Lubrication

Williams et al. derived formulas for the permeation distance or permeation depth of gas into a capillary, lg (m) (14,17), as
3
!2
5:5  104 h pg 2 M
3

lg ¼ [1]
Vchip mg qT rlq

for the viscous (Poiseuille) flow and as

!1
1:18  107 2
6
M
lg ¼ h pg [2]
Vchip qT 3 rlq 4

for the Knudsen flow, where mg (Pa s) is the coefficient of viscosity or dynamic viscosity of gas, pg (Pa) is its vapor pressure, rlq
(kg m3) is its density of the liquid phase, M is its molecular weight, Vchip (m s1) is the chip speed, and qT (K) is the absolute
temperature. Note that those constant values in the equations above are different from those in equations by Childs et al. (17)
because the definitions of units are different. For those capillaries on the interface between the worn flank face and machined
surface, chip speed, Vchip, in the equations above are to be replaced with cutting speed, Vc (m s1).
Before actually calculating lg, the dynamic viscosity of vaporized cutting oil should be estimated in advance. Figure 12
illustrates dynamic viscosities of eight different types of hydrocarbons against molecular weight at 100  C under atmospheric
pressure. It can be seen in the diagram that if the molecular weight increases, the viscosity decreases even down to 0.01 mPa s.
Figure 12 shows the dynamic viscosities of acetone, ethyl ether, oxygen, air, carbon monoxide, carbon dioxide, and steam. The
dynamic viscosities of hydrocarbons are smaller than those of air and carbon oxides and are almost comparable to those of
steam, acetone, and ethyl ether. Figure 13 shows the temperature dependence of the dynamic viscosities with regard to eight
hydrocarbons as well as steam. Based on the figure, the dynamic viscosity of vaporized cutting oil mg_oil can linearly be
approximated as follows:
mg_oil ¼ cg_oil qT þ m0g_oil [3]

where cg_oil is a temperature coefficient and m0g_oil is a constant. As for hydrocarbons, these coefficients are assumed to be the
same as those of hexane, i.e. cg_oil ¼ 2.2  108 Pa s K1 and m0g_oil ¼ 0 Pa s. When water-based coolant is used in MQL machining

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
262 Machining with Least Quantity Lubrication

Figure 12 The dynamic viscosity against molecular weight for hydrocarbon gases and other comparative materials at 100  C.

Figure 13 Temperature dependence of the dynamic viscosities of hydrocarbon gases and steam.

to enhance the cooling effect rather than lubrication, the temperature-dependent dynamic viscosity of steam msteam will be
formulated as
msteam ¼ csteam qT þ m0steam [4]
8 1 6
where csteam and m0steam are 3.8  10 Pa s K and 1.37  10 Pa s, respectively. Note that this equation can be used for gas-phase
lubrication of a capillary as well.
Now, the author would like to show an example. For an MQL cutting condition where the cutting speed is Vc ¼ 1.67 m s1
(100 m min1), the molecular weight and density of the cutting oil are M ¼ 300 and 960 kg m3, respectively, let us assume that the
shear angle, rake angle, mean capillary height, h, and vapor temperature, qT are 20 , 0 , 1 mm, and 600 K, respectively. Then, the chip
speed, Vchip, and the dynamic viscosity of vaporized cutting oil, mg_oil, at 600 K can be calculated to be 0.607 m s1 (36.4 m min1)
and 0.0132 mPa s, respectively. Figure 14 shows the permeation depth of vaporized cutting oil into a capillary under the cutting

Figure 14 The permeation depth of vaporized cutting oil into a capillary against vapor pressure at qT ¼ 600 K.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Machining with Least Quantity Lubrication 263

Figure 15 The permeation depth of vaporized cutting oil into a capillary against vapor temperature at pg ¼ 6 kPa.

conditions above under a pressure of vaporized cutting oil in the range of 2 kPa < pg < 20 kPa. Since the pressure of oil vapor is
usually low, the pressure of vaporized cutting oil can vary in a wide range.
Figure 15 shows the permeation depth of vaporized cutting oil against vapor temperature for pg ¼ 6 kPa. When the cutting speed
is revved up, the pressure of vaporized cutting oil in a capillary is to increase since the cutting temperature will rise to accelerate the
evaporation of cutting oil. As the partial pressure of vaporized oil increases, the permeation depth may appear to increase as well.
Yet, as is expected from eqns [1] and [2], an increase in temperature and chip speed will result in a decrease in the permeation depth.
As a consequence, these opposite effects will be set off, with the permeation depth more or less unchanged. When the permeation
depth of vaporized oil into a capillary is assumed, based on Figure 14, to be in the range of 100–200 mm on the rake face, the depth
of permeation on the flank wear land can be calculated about 37–73 mm by substituting Vchip with Vc in eqn [2]. These values of the
permeation depth seem to be large enough for lubricating tool faces.

11.11.2.4 Liquid-Phase Lubrication

When an oil droplet with a density of rlq and a coefficient of viscosity of mlq moves to the open mouth of a capillary from
its original attached position on the tool face, the difference in pressure in and out of the capillary may push the droplet
into a capillary against the chip motion at the speed of Vchip. If the evaporation of oil can be neglected in the capillary, the
flow speed of oil, v, will be distributed in the form of the Couette flow at a longitudinal cross section of the capillary as
follows:
 
1 dp  2  Vchip
v¼ h  hh þ h  v0 [5]
2mlq dx h
where h ð0 < h < hÞ is the vertical distance from the rake face, dp/dx is the gradient of pressure p in the oil droplet along the chip
flow direction, x, and v0 is the speed of oil toward the tool tip at the chip leaving point, which is assumed to be constant. The mean
flow speed of coolant across a capillary, v, is given as
 
1 dp 2 1
v¼ h þ Vchip  v0 [6]
12mlq dx 2

Now, the inside of a capillary can be assumed to be virtually retained in a vacuum of absolute pressure, p ¼ 0. Also, it
is reasonable to assume that the coolant permeating into a capillary is driven only by the atmospheric pressure, patom. For
simplicity, let us further assume that v0 ¼ 0 and that the capillary attraction or the surface tension of coolant is negligible,
since the chip speed is much faster than the permeation speed of coolant due to the capillary action. Let ll denote the
mean permeation depth of coolant into a capillary, then ll can be determined by assuming v ¼ 0 and dp/dx ¼ patom/ll,
namely,
1 patom 2
ll ¼ h [7]
6mlq Vchip

This equation states that the permeation depth of liquid, ll, will increase with the square of capillary height, h, in a similar
manner to the permeation depth of gas, lg, in the Knudsen flow. For a capillary on the interface between the flank wear land and
machined surface, ll will be described as
1 patom 2
ll ¼ h [8]
6mlq Vc

The relationships between the mean permeation depth, ll, and the speed of Vchip or Vc for h ¼ 1 mm are plotted in Figure 16, and
the relationships between ll and h for Vchip ¼ 0.607 m s1 (36.4 m min1) and Vc ¼ 1.67 m s1 (100 m min1) are plotted in

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
264 Machining with Least Quantity Lubrication

Figure 16 The permeation depth of cutting fluid into a capillary of h ¼ 1 mm against the chip velocity or the cutting speed.

Figure 17 The permeation depth of cutting fluid into a capillary against the capillary height h for Vchip ¼ 36.4 m min1 and Vc ¼ 100 m min1.

Figure 17. Note that the coefficient of viscosity of coolant is assumed to be 1.1  103 Pa s for water-based coolant including
emulsion, soluble, and solution types (18) and 18.0  103 Pa s for MQL cutting oil. It is obvious from these diagrams that the
permeation depth will be considerably shallow for MQL cutting oil. Hence, liquid oil in MQL machining cannot well lubricate the
tool–chip contact area or the interface between the flank wear land and machined surface, even if there were a sufficient number of
capillaries generated over these interfaces.
There is a huge difference in the permeation depth into a capillary between the gaseous and liquid lubricants, as seen in Figures
15 and 16. This is the reason why MQL machining is considered much more preferable in lubrication of the tool–chip and tool–
work interfaces than machining with flood coolant. Note that the combination of oil droplets and compressed air in MQL
machining is not as effective in cooling the tool and work as flood coolant usually is, but this is not necessarily considered to be
a negative factor in finish turning with a small depth of cut or in finish milling with a small axial depth of cut. In MQL machining,
the thermal impact due to interrupted cutting is to be alleviated, as shall be explained later, and thus the cutting edge will not have
many thermal cracks. As a result, despite the oil droplets and compressed air do not exhibit a sufficient cooling effect, the tool life
can be rather extended in MQL machining.
Figure 17 indicates that water-based coolant with a high permeability would be capable of permeating deeper into a capillary if
the surface finish is harsh. Also, if the droplet is large and thus takes some time to evaporate as is described in Figure 11, then the
liquid-phase lubrication will arise, since the droplet could permeate into a capillary to a certain extent before its vaporization starts
at the mouth of a capillary.

11.11.2.5 Stop Distance of an Oil Droplet

The oil mist is the key in controlling the machinability and quality of the machined work and also in managing the machining cost.
A droplet of cutting oil sprayed with compressed air must possess an inertial mass large enough to escape from the airflow to reach
the tool face near the tool tip. In other words, fine oil droplets with a small inertial mass cannot fly long to contribute to the
lubrication of cutting. Accordingly, it must be important to understand the flight distance of an oil droplet to determine the nozzle
position for spraying oil mist with compressed air and to perform MQL machining under desirable conditions for both lubrication
and cooling. Note that the relationship between the size of an oil droplet and its stop distance in the Stokes flow was theoretically

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Machining with Least Quantity Lubrication 265

deduced (19); a spherical oil droplet subject to the Stokes’ drag of air can fly in the still air over the stop distance, SS, when gravity can
be neglected, described as
2
R rlq dp rlq dp
SS ¼ ¼ Um [9]
18 ra 18ma
with the Reynolds number, R,
ra Um dp
R¼ [10]
ma
where Um is the initial speed of the droplet, dp is the diameter of the droplet, and ra and ma are the density and the coefficient of
viscosity of the air, respectively. When the stop distance is short, the constraint for an oil droplet by airflow becomes tight, and it
thus becomes difficult for the droplet to freely escape from the airflow. Note that the Reynolds number for an oil droplet in MQL
machining is generally fairly large. This means that we need to know an expression of the stop distance especially for a fast flow with
a high Reynolds number. In the Stokes flow, the drag coefficient for a flow past a spherical droplet CD is given by
24
CD ¼ [11]
R
For a high Reynolds number, Veysey II et al. (20) presented the drag coefficient as
24
CD ¼ CR [12]
R
where
 
192 þ 96R þ 16R2 þ R3 þ 8eR=2 R2  24
CR ¼  [13]
3 4ðR þ 2Þ þ eR=2 ðR2  8Þ

By analogy with the formulation of eqn [9], an approximate equation of the stop distance, SD, for the drag coefficient of eqn [12]
can be yielded (16) as
2
4 rlq dp 1 rlq dp
SD ¼ ¼ Um [14]
3CD ra CR 18ma
This means that the stop distance, SD, can be approximated to be proportional to the square of the diameter of the oil droplet.
Since coefficient CR given by eqn [13] can vary only in the range of 1  CR  8=3, it can be said that the diameter of an oil droplet
will most significantly affect the stop distance. Figure 18 shows the stop distance, SD, plotted against the initial speed of an oil
droplet. The density of cutting oil, rlq, the coefficient of viscosity of the air, ma, and the density of the air, ra, are therein assumed to
be 950 kg m3, 0.0182 mPa s, and 1.20 kg m3, respectively. The diameter of an oil droplet, dp, is set to be 1, 10, or 100 mm. It can be
underpinned that the stop distance will be much shorter than 1 mm when dp ¼ 1 mm, and will be in the range between 0.26 and
10.9 mm for the initial speed of 1–100 m s1 when dp ¼ 10 mm. When dp increases up to 100 mm, the stop distance also increases
and ranges from 14.4 mm to 1087 mm for the initial speeds of 1–100 m s1. Since the external nozzle is usually installed by 30–
50 mm off the cutting point, it is preferable in MQL machining that the diameter of an oil droplet is in the size of at least 10 mm.
If the motion of an oil droplet is considered in moving air, then the initial speed, Um, in eqns [9] and [14] should be substituted
with the relative speed between an oil droplet and the surrounding air stream, DU, and then SD is to be regarded as the relative stop
distance that means the distance a droplet needs to fly till it reaches the same speed as the airflow, i.e.,
2
1 rlq dp
SD ¼ DU [15]
CR 18ma

Figure 18 The stop distance of an oil droplet against the initial speed.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
266 Machining with Least Quantity Lubrication

In other words, SD is a relative mean free path of an oil droplet in the moving air. A large vorticity of air near the tool tip may
increase DU, hugely affecting the stop distance. Also, a large droplet with a large inertia is likely to increase DU. This will promote the
adhesion of oil to the tool face and, as a result, improve the lubrication condition on the interface between worn flank and finished
surfaces. Note that a fine droplet with a pretty small value of SD can hardly reach the tool tip or even the vicinity by escaping from the
airflow. This means that the fine size of droplets would result in an insufficient lubrication condition over the tool–chip and tool–
work contact areas at any rate, leading to short tool life and rough surface finish.

11.11.3 MQL Machining Technologies

In this section, oil mist generation methods and a machine tool spindle specialized for MQL machining shall be first described.
Then performances of milling, turning, and drilling with MQL will be discussed. Other topics include micro-MQL machining for
further reduction in cutting oil and future-oriented MQLs.
Although there is a considerable amount of literature on MQL machining, it is regrettable that much of it does not disclose any
results of machining with flood coolant. This means that there are hardly existing data available to compare MQL machining with
wet machining with regard to the cutting performance. In general, MQL machining can extend the tool life of an end mill. It also can
improve the shape accuracy of drilled holes and extend the tool life of a drill under certain cutting conditions. Still, it remains
difficult to improve the performance of a single point cutting tool in MQL turning as much as in MQL milling, in part because of the
continuous cutting process in turning, whereas the tool life in MQL turning can be longer than in wet turning under certain
conditions all the same.

11.11.3.1 Oil Mist Generation and Application

One method to generate oil mist is spraying dispensed cutting oil with compressed air. Wet mist consisting of oil droplets with
diameters ranging from several micrometers to hundreds of micrometers can be generated with a prescribed amount of cutting oil
intermittently dispensed with a plunger pump at a certain time interval. The oil droplets are directly applied to the cutting region
from the wet oil mist generator through nozzles as shown in Figure 19 (21), and their mean diameter becomes relatively large. Here,
the author will show an example: given the diameter of an oil droplet is 10 mm and the amount of dispensed oil is 10 ml h1, then
the number of oil droplets sprayed per second will be 663 146 in theory. Suppose that the turned area is not less than 2000 mm2 s1
under the cutting condition that the cutting speed is 2 m s1, the depth of cut is 1 mm and the feed rate is 0.2 mm rev1, then the
number of oil droplets per machined area will be c. 300 mm2, which may seem to be large enough. Still, unless the spray is
accurately applied to the cutting point, droplets would not work as only a small portion of the sprayed droplets can reach the cutting
region. The author thus places an emphasis on the importance of the spraying direction that has a direct influence on the
machinability, as is discussed in Ref. (22).

Figure 19 Schematic diagram of milling test setup: 1. Spindle, 2. Nozzle, 3. End mill, 4. Workpiece, 5. Machine bed, 6. Nozzle, and 7. Dynamom-
eter. Reproduced from Rahman, M.; Kumar, A. S.; Salam, M. U. Experimental Evaluation on the Effect of Minimal Quantities of Lubricant in Milling.
Int. J. Mach. Tool Manuf. 2002, 42, 539–547.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Machining with Least Quantity Lubrication 267

If oil mist is to be delivered from the mist generator all the way through a long pipe as well as the spindle and tool holder,
the mist will be dried and thus consist of only such fine oil droplets with the diameter of a few micrometers that the mist would
not be attached or condensed on the internal surfaces en route (23). For generating such dry oil mist, cutting oil should be
sprayed with compressed air into an air chamber, which functions also as an oil reservoir, as shown in Figure 20. The mounting
position of the spray should be a little above the oil surface of the oil pool. Note that the level of the surface will gradually be
going down according as the oil is consumed with time. A large oil droplet with a large inertial mass and a long stop distance
will fly more or less straight in the sprayed direction but will eventually fall into the oil pool under their own weight. If the
chamber is slender, such a large droplet can likely reach the opposite wall of the chamber and then will fall down to the oil
pool along the wall. Consequently, only fine oil droplets will be sucked upwards together with compressed air and be delivered
from the air chamber to the nozzles of the cutting tool. In summary, it can be stated that the amount of cutting oil sent out
from the chamber is not exactly equal to the amount of oil dispensed with the plunger pump in the case of dry mist. Thus, the
optimal condition for sending cutting oil from the chamber should be determined on the trial-and-error basis regardless of the
amount of dispensed oil.
The size of oil droplets in dry mist must have increased by the time they are applied to the cutting point during long
transportation through a piping distribution system, as is often the case with oil mist lubrication of machine elements (24).
A simple way to grow oil droplets in size is to make them come through a narrow nozzle. Figure 21 illustrates a type of orifice
fitting as an example of such a nozzle to convert a dry oil mist to a wet one. The diameter of the narrow nozzle should be less
than about 2–3 mm, examples of which include holes on a drill or an end mill. When fine oil droplets are supplied into a narrow
nozzle, their flow speed is usually accelerated to reach as much as 100 m s1. At such an extremely high speed, droplets will
repeatedly collide with each other to coalesce and eventually grow in size. A Reynolds number for the airflow through a narrow
nozzle, Rh, is given by
rUa D0
Rh ¼ [16]
ma
where r and Ua are the density and the mean flow speed of air through the nozzle, D0 is the diameter of the nozzle, and ma is the
dynamic viscosity of air. As a typical example, Rh can be approximately calculated to be 39 600 with r ¼ 3.6 kg m3 for absolute
pressure 0.3 MPa, Ua ¼ 100 m s1, D0 ¼ 2 mm, and ma ¼ 0.0182 mPa s. It is generally known that the flow in a circular pipe will be
laminar for Rh < 2300, transitional for 2300 < Rh < 4000, and turbulent for Rh > 4000. Therefore, the airflow in the narrow nozzle

Figure 20 Schematic diagram of a dry oil mist generator.

Figure 21 Growth of oil droplets in size through a narrow nozzle.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
268 Machining with Least Quantity Lubrication

in the example above must be well in a turbulent state, where oil droplets should collide with each other to coalesce. The total mass
flow rate of air through all the nozzles, J, can be assumed to be constant,
p
J ¼ rAUa ¼ NrD0 2 Ua ¼ const [17]
4
where A is the total cross-sectional area of nozzles and N is the number of nozzles or oil holes of the tool. Rearrangement of eqn [16]
with eqn [17] yields another expression of Rh as follows;
4J
Rh ¼ [18]
pND0 ma
Since the dynamic viscosity of air can be assumed to be constant for pressure up to 10 MPa, the Reynolds number, Rh, can be
thought to be in effect inversely proportional to the product of the number of nozzles and the nozzle diameter. Note that the
turbulent flow is a necessary condition for converting dry mist to wet one, but not a sufficient condition. In point of fact, if the total
cross-sectional area of the nozzle is too large, it is usually difficult to fully convert a dry oil mist to a wet one even though the airflow
through the nozzles is well in a turbulent state.
It is obvious that the chance for oil droplets to collide will be enhanced if there are more oil droplets contained in unit volume.
The volume of oil per unit volume of the compressed air, q, is defined as
Q0
q¼r [19]
J
where Q0 is the amount of oil supplied or consumed per unit time. Thus, the product of q and Rh can arguably be used as an
objective criterion for the conversion of dry mist to wet one. From eqns [17] to [19],
Q0 4J Q0 4J 4J 16Q0 J 1
qRh ¼ r ¼ ¼ [20]
J pND0 ma J pND0 2 Ua pND0 ma p2 Ua ma N 2 D0 3
Note that this equation indicates that Q0 and J must be large enough when the size of nozzle is also large. Since the flow speed,
Ua, is not as sensitive to pressure as density r is, as is suggested by Bernoulli’s principle for compressible flow, the following
equation,
N 2 D0 3 ¼ C c [21]
can be used as one of possible criteria about the conversion between dry and wet mists for given Q0 and J. Note that Cc is a constant
but of course depends on Q0 and J. Now suppose Cc ¼ 27 mm3, then the diameter of nozzle can be determined as D0 < 3 mm for
N ¼ 1 or D0 < 1.89 mm for N ¼ 2. According to the manual for EcoBooster Type-EB3 (Fuji BC Engineering Co., Ltd.), which is
a distinguished mist generator, tools with the total cross-sectional area of air nozzles smaller than the critical value of Ac ¼ 5 mm2
are recommended for MQL machining to attain sufficient wet mist lubrication, i.e., the acceptable diameter of nozzles will be
D0 < 2.52 mm for N ¼ 1 or D0 < 1.78 mm for N ¼ 2. Since the flow speed, Ua, is not as sensitive to pressure as the density is, it can
virtually be interpreted that the criterion for the total cross-sectional area solely depends on the density of air. The restraint condition
that Ac ¼ 5 mm2 shall be referred to as ‘Ac criterion’ and eqn [21] shall be referred as ‘Cc criterion’ hereinafter.
Table 1 shows that the difference in the critical values of D0 for N ¼ 2 between Cc and Ac criteria under the condition that D0 is
defined to be 3 mm for N ¼ 1 for both criteria. The difference is about 10% that can be considered rather small from the practical
point of view. Note that the Ac criterion is independent of the Reynolds number, Q0 or J, thus this cannot be used to a wide range of
conditions for MQL machining.
It should be worth noting here that large oil droplets may not always be generated by the method explained above for con-
verting a dry oil mist to a wet one even if the airflow is turbulent enough, because the conversion depends also on the density of oil
droplets. Note also that some part of fine droplets in dry mist will not be agglomerated while flowing through a nozzle and will
thus be discharged as the stray mist, which does not contribute to the lubrication of cutting zone at all. For using dry mist, many
restraint conditions must be taken into consideration about tools and the machining system as discussed above, and thus it had
been hoped that a much simpler system without dry mist would be invented. In the mid-1990s, in response to this demand,
HORKOS Corp. developed a new type of spindles with a wet mist generator installed on the side of the tool holder for horizontal
machining centers specialized for MQL machining. Figure 22 shows the schematic diagram of such a spindle with a wet mist
generator, where the distance between the cutting position and the mist generator or the mixing point of oil and air is short, and
thus it would not be a major problem even if the oil mist condensed or attached to the internal surface of the tool during
transportation. Wet mist consisting of large oil droplets can thus be successfully generated as a result. Figure 23 shows

Table 1 D0 for N ¼ 2 for Cc and Ac

criteria when D0 ¼ 3 mm for N ¼ 1

Criterion D0 for N ¼ 2

Cc 1.89 mm
Ac 2.12 mm

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Machining with Least Quantity Lubrication 269

Figure 22 Spindle with a built-in wet mist generator. Courtesy of HORKOS Corp.

Figure 23 Conventional spindle-through oil mist supplying system. Courtesy of HORKOS Corp.

a conventional spindle-through oil mist supplying system, where dry mist generated outside the machine is supplied all the way
through the spindle to the tool.
Spindles with built-in wet mist generators have following advantages: (1) large oil droplets, which are more effective in
improving cutting performance than fine oil droplets, are always applied to the cutting zone, and the consumption of cutting oil can
thus be reduced in total; (2) the size of a nozzle on the tool can be independent of the condition for converting dry oil mist to wet
one; and (3) the quantity of oil applied to the cutting zone is easy to control because it can directly be specified by the amount of
dispensed oil. As a result, this type of spindle can significantly enhance the efficiency of lubrication.

11.11.3.2 Milling
In MQL cutting, compressed air will mildly cool down the cutter to reduce the thermal impact, which is the main cause of the
damage to a cutter in milling processes, upon cutting edges both at the entry to and the exit from the workpiece. The evaporation of
oil droplets on the tool faces will be of some help cooling down the tool as well, though its thermal impact on the cutting edge is
relatively minor. Oil mist by the gas-phase lubrication mechanism will lubricate the tool–chip and tool–work contact areas more
efficiently than water-based flood coolant. If oil mist is not supplied through a cutter, the cutter face is to be exposed to the oil mist
sprayed from a distant nozzle, except when the cutter face is right engaged in cutting. Consequently, oil droplets will repeatedly and
directly be attached to the tool faces and, moreover, a part of them will likely be trapped even between the tool rake face and
a forming chip. This may result in the longer extension of tool life of a cutter than flood coolant can attain. As a matter of fact, this
observation is consistent with the well-known know-how about the selection of cutting oil that water-immiscible cutting oil should
be used for milling processes to extend the tool life of a cutter, especially in machining of difficult-to-machine materials if water-
based coolant is not necessary to wash off chips being produced.
Figure 24 schematically summarizes thermal damages due to temperature and temperature change, TDq and TDDq, against the
cutting speed in MQL milling and wet milling with flood coolant. Note that when the mechanical damage is assumed to be
negligible for simplicity, it can be further assumed that the tool life should simply be determined by the sum of two types of thermal
damages, TD ¼ TDq þ TDDq, which is depicted in solid lines for wet and MQL millings in Figure 24. The critical cutting speed, Vtr,

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
270 Machining with Least Quantity Lubrication

Figure 24 The effect of lubrication/cooling conditions in end milling upon the thermal damages to a tool due to temperature and temperature
change, TDq and TDDq.

under which the thermal damage in MQL milling is smaller than that in flood coolant, will shift to the right in finish end milling
with a small axial depth of cut at a small feed rate per tooth, since the thermal damages will have been considerably alleviated in this
case. On the contrary, Vtr will shift to the left in roughing with a large axial depth of cut at a large feed rate. Note that in a case where
the difference in thermal damage due to temperature change, TDDq, between MQL and flood coolant were exceptionally large, MQL
milling would likely be better than milling with flood coolant even in rough milling at high cutting speeds.
Rahman et al. (21) underpinned that MQL is preferable to flood coolant in slotting of an alloy steel, ASSAB 718HH steel of
hardness 35 HRC, under a mild cutting condition that the cutting speed was 75 m min1, the axial depth of cut was 0.35 mm, and
the feed rates were less than 0.02 mm tooth1, with an insert of the end mill of a P20 grade uncoated cemented carbide, where the
amount of MQL and the flow rate of water-soluble coolant were 8.5 ml h1 and 42 l min1, respectively. Indeed, their conclusion is
consistent with the correlation between the cutting speed and the rate of thermal damage shown in Figure 24, considering that the
tool life will be extended longer in MQL end milling than with flood coolant as long as the cutting condition is not extremely severe.
There are some practical cases reported in the literature in which the tool lives were successfully extended by MQL milling in
comparison with milling with flood coolant, and, interestingly, the actual axial depths of cut and the feed rates therein adopted were
smaller than those expected from the discussion above about the rate of thermal damage; for example, they are 0.35 mm and 0.01–
0.02 mm tooth1 for end milling an alloy steel (21), 3.0 mm and 0.1 mm tooth1 for end milling an aluminum alloy (22), 2.0 mm
and 0.2 mm tooth1 for face milling an aluminum-silicon alloy (25), 0.5 mm and 0.025–0.1 mm tooth1 for end milling a tita-
nium alloy (26), 0.6 mm and 0.1–0.2 mm tooth1 for end milling a hardened die steel of 41 HRC (27), 2.0 mm and
0.01 mm tooth1 for end milling a hardened die steel of 62 HRC (28), 0.2 mm and 0.14 mm tooth1 for face milling stainless steel
(29), 0.2 mm and 0.01–0.03 mm tooth1 for ball end milling hardened steel (30), 12 mm and 0.03 mm tooth1 for end milling
stainless steel (31), and 0.2 mm and 0.04 mm tooth1 for helical milling a titanium alloy (32), respectively. Judging from these
examples, the combination of an axial depth of cut of 2 mm and feeds per tooth of 0.14 and 0.22 mm used in Ref. (33) was certainly
inappropriate for MQL face milling of plain carbon steel, although the suitable cutting condition for MQL milling depends not only
on these two parameters, but also on the radial depth of cut, tool material, tool geometries, cutting oil consumption, airflow rate,
and how to prepare and spray oil droplets as well.
The necessary amount of cutting oil for MQL milling varies depending on the work material, but it is usually less than 30 ml h1.
Specific values disclosed in the literature include 8.5 ml h1 for an alloy steel (21), 2.4–3.6 ml h1 (22) or 30 ml h1 (25) for
aluminum alloys, 10 ml h1 (27) or 6 ml h1 (28) for hardened steels, 20 ml h1 (29) or 17 ml h1 (31) for stainless steel, and
10 ml h1 (32) for a titanium alloy. Note that a large amount of cutting oil was exceptionally applied in Ref. (26) for end milling
a titanium alloy at 40 ml h1, and also in Ref. (30) for ball end milling a hardened steel at 120 ml h1.
López de Lacalle et al. (22) demonstrated that the spraying direction of oil mist towards the cutting point exerts a significant
effect on the tool wear of an HSS end mill in the down milling process of an aluminum alloy, 5083-H112, where the cutting speed
was 942 m min1, the feed rate was 0.1 mm tooth1, the radial depth of cut was 3 mm, and the axial depth of cut was 3 mm. The
amount of MQL was either 2.4 or 3.6 ml h1 and the spraying direction was either 45 or 135 from the feed direction, as shown in
Figure 25. Their result about the flank wear evolution against the cutting length is reproduced in Figure 25 for milling with MQL
along two different spraying directions and for milling with flood coolant. They reported that the flank wear was smallest for MQL
with the spraying direction of 135 . Note that the reported difference in the flank wear between spraying directions of 45 and 135
was considerably large. Still, it should be remarked that only a tiny amount of cutting oil, 2.4 or 3.6 ml h1, could secure
a remarkable reduction in the flank wear.
Readers are encouraged to remember that coated tools should ideally be used to enhance the cutting performance of milling and
hobbing with MQL. A coated tool usually has a much higher wear resistance at high cutting speeds at high temperatures but a lower
toughness than an uncoated tool. With a coated tool, MQL can deliver excellent performance up to higher cutting speeds. The

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Machining with Least Quantity Lubrication 271

Figure 25 Flank wear evolution against the cutting length for end milling with MQL with the spraying directions of 45 and 135 and milling with
flood coolant. Reproduced from López de Lacalle, L. N.; Angulo, C.; Lamikiz, A.; Sánchez, J. A. Experimental and Numerical Investigation of the
Effect of Spray Cutting Fluids in High Speed Milling. J. Mater. Process. Technol. 2006, 172, 11–15.

difference in the rate of thermal damages in MQL milling between coated and uncoated tools is schematically shown in Figure 26.
Since MQL exerts less thermal impact on the tool than flood coolant usually does, the difference in damages between coated and
uncoated tools due to temperature change is small and often negligible. Contrary to conventional wisdom that coated tools are
brittle at high temperature changes, some newly developed coated tools are much stronger against thermal impacts than
conventional uncoated tools. Besides, coated tools must be indispensable in continuous machining processes with MQL, including
MQL turning and MQL drilling, where thermal impact due to a change in cutting temperature is small. Note that thermal barrier
coatings developed for coated tools (34) would compensate the low cooling rate of MQL machining, in which compressed airflow
plays a pivotal role in cooling. Various types of advanced coatings with both a high hardness and a high oxidation temperature
providing a higher heat resistance have so far been introduced into end mills, especially for milling hardened steel. A high heat
resistance of coating allows the coated tool to be used in high-speed end milling of hard materials at high cutting temperature under
dry or nearly dry conditions.
Liao et al. (27) reported that MQL was more preferable to flood coolant under all the cutting conditions tested for high-speed
end milling of a hardened die steel, NAK81 of hardness 41 HRC, with an indexable carbide insert coated with TiAlN and TiN.
The amount of MQL oil (5 cSt at 40  C) was 10 ml h1, which was sprayed with compressed air at the pressure of 0.45 MPa, while

Figure 26 The effect of coating of a tool on the thermal damages to a tool due to temperature and temperature change, TDq and TDDq, in
MQL milling.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
272 Machining with Least Quantity Lubrication

water-soluble coolant was applied to the end mill at the flow rate of 20 l min1. The cutting speeds tested were 150, 200, and
250 m min1, and the feed rates were 0.10, 0.15, and 0.20 mm tooth1. As for the axial depth of cut, it is generally recommended to
combine a large feed rate with a small axial depth of cut for milling hardened steel to minimize the damage to the coated end mill,
hence the axial depth of cut was set in their study at only 0.6 mm. As the tool rejection criteria, the maximum flank wear of
VBmax ¼ 0.1 mm and the width of notch wear of VN ¼ 0.1 mm were used. Figure 27 shows a result from Ref. (27) about the tool
lives in terms of the cutting length measured at the cutting speed of 200 m min1 alone.
Reference (27) also revealed that the average resultant cutting force measured with a piezoelectric dynamometer became smallest
for wet machining and largest for dry machining, and that the force required for MQL milling was only a little smaller than that
required for dry milling. Despite the fact that the cutting force in MQL cutting is larger than in wet cutting, the tool life in MQL
cutting was about twice longer than that in wet cutting. As a matter of fact, when flood coolant is applied, an extraordinarily large
number of thermal cracks will appear along the cutting edge as is seen in the scanning electron micrographs (SEMs) in Figure 28.
There are no SEMs of a cutting edge used in MQL machining available, but the number of cracks on a cutting edge in MQL cutting
must be much smaller, to the best knowledge of the author. This can be interpreted that the thermal impact on the cutting edge in
wet end milling of hardened steel is much larger than the mechanical impact even at a low feed rate and a low cutting speed. This is
the reason why MQL is highly recommended for end milling hardened steel. Note that although the tool life in MQL cutting is
generally monotonically diminished with an increasing cutting speed, it does not monotonically behave toward an increasing feed
rate. Indeed, in Ref. (27), the tool life takes its maximum at the feed rate of 0.15 mm tooth1 when the cutting speed is 150 and
200 m min1. It is widely known in end milling hardened steel that the longest tool life is often attained not only at a higher feed
rate but also at a higher cutting speed. This phenomenon was elucidated in Ref. (35) in relation to chip formation.
Reference (28) reported that end milling of the AISI D2 cold-worked die steel with a hardness of as high as 61 HRC with a coated,
two-flute flat end mill of 2 mm in diameter also bore quite similar relationships between tool lives for dry, MQL, and flood coolant
conditions to those reported for NAK81 by Liao et al. (27). The coating materials tested for the end mill were Ti0.75Al0.25N and
Ti0.69Al0.23Si0.08N, and the cutting speed was 75 m min1, viz., the rotational speed of the tool was 12 000 min1. The axial depth of
cut was 2 mm, the radial depth of cut was 0.02 mm, and the feed rate was 0.01 mm tooth1. The amount of MQL was 10 ml h1,
and the air pressure for spraying MQL was 0.5 MPa. Note that even in dry end milling, compressed air was also to be blown from
nozzles to the cutting point. It was then spotted in Ref. (28) that the combination of MQL and the Ti0.69Al0.23Si0.08N coating
exhibited the longest tool life in length.
Regarding the surface finish, the effect of MQL milling on the surface finish is not as straightforward as that on the tool life.
Indeed, Thepsonthi et al. (30) obtained opposite findings to those shown in Figure 29. According to the comparative result of high-
speed end milling of die steel shown in Figure 29 (27), MQL is better in surface finish than flood coolant at the cutting speed of
200 m min1 or more, while it becomes worse at the cutting speed of 150 m min1.
Thepsonthi et al. conducted high-speed slot cutting of the ASSAB DF3 hardened tool steel of a hardness of 51 HRC using a two-
flute ball end mill coated with TiAlN at the cutting speed of 125, 150, or 175 m min1 with the depth of cut of 0.2 mm at the feed
rate of 0.01, 0.02, or 0.03 mm tooth1. Water-miscible coolant of 10% concentration was applied from five nozzles around the tool
at the rate of 7.0 l min1 under the flood coolant condition. The cutting oil was sprayed by pulse-jets at the pulse rate of
400 pulse min1 under the pressure of 20 MPa and at the delivery rate of 120 ml h1 under the MQL condition. The direction of
spraying was set right along the feed direction. Figure 30 (30) shows the surface roughness and flank wear against the cutting speed
at the feed rate of 0.01 mm tooth1 with the cutting length of 6 m. The surface finish gets worse under the MQL condition as the
cutting speed increases, while it is rather improved with increasing cutting speed under the flood coolant condition. Note that these

Figure 27 Tool life as a function of feed rate and lubricating method at cutting speed 200 m min1. Reproduced from Liao, Y. S.; Lin, H. M.; Chen,
Y. C. Feasibility Study of the Minimum Quantity Lubrication in High-Speed End Milling of NAK80 Hardened Steel by Coated Carbide Tool. Int. J. Mach.
Tool Manuf. 2007, 47, 1667–1676.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Machining with Least Quantity Lubrication 273

Figure 28 SEM micrographs of thermal cracks and facets of breakage observed near the cutting edge for the cutting length of (a) 1.8 and (b)
1.9 m. Cutting conditions: flood cooling, cutting speed is 250 m min1, and feed rate is 0.15 mm per tooth. Reproduced from Liao, Y. S.; Lin, H. M.;
Chen, Y. C. Feasibility Study of the Minimum Quantity Lubrication in High-Speed End Milling of NAK80 Hardened Steel by Coated Carbide Tool. Int. J.
Mach. Tool Manuf. 2007, 47, 1667–1676.

Figure 29 The surface roughness (Ra) against the feed rate under three different lubrication conditions at the cutting speeds of (a) 150, (b) 200,
and (c) 250 m min1. Reproduced from Liao, Y. S.; Lin, H. M.; Chen, Y. C. Feasibility Study of the Minimum Quantity Lubrication in High-Speed End
Milling of NAK80 Hardened Steel by Coated Carbide Tool. Int. J. Mach. Tool Manuf. 2007, 47, 1667–1676.

changes with cutting speed are opposite to those obtained in Figure 29. Note also that the flank wear behaves differently from the
way the surface finish behaves against changes in the cutting speed. The flank wear in MQL milling always remains minor up to the
cutting speed of 175 m min1, due probably to the large amount of cutting oil of 120 ml h1.

11.11.3.3 Turning
In turning, except interrupted turning, it is not possible to apply oil mist directly to the cutting edge engaged in machining because
a chip being formed and the surface just machined are always in contact with the tool rake face and the flank wear land, respectively.
Hence, MQL turning is not as effective in cooling the cutting edge and lubricating tool faces as MQL milling. Cutting oil, having
both a large permeation depth into a capillary and high lubricity, should be used for MQL turning.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
274 Machining with Least Quantity Lubrication

Figure 30 The (a) surface roughness and (b) flank wear against the cutting speed at the feed rate of 0.01 mm tooth1 with the cutting length of
6 m. Reproduced from Thepsonthi, T.; Hamdi, M.; Mitsui, K. Investigation into Minimal-Cutting-Fluid Application in High-speed Milling of Hardened
Steel Using Carbide Mills. Int. J. Mach. Tool Manuf. 2009, 49, 156–162.

The amount of oil necessary to obtain good performance in MQL turning falls into two zones, i.e., a zone of a small amount
of oil including 6.0–9.6 ml h1 for alloy steel (36) and 7.0 ml h1 for carbon steel (37) and another zone of a large amount of
oil but still MQL including 60 ml h1 for alloy steel (38), 50 and 100 ml h1 for aluminum alloy (39), and 100 ml h1 for alloy
steel (40). It is worth noticing that the spraying of oil mist from the side of the flank face as shown in Figure 31 could lead to
the reduction of cutting oil to an amount less than 10 ml h1 (36,37). This can be interpreted that the lubrication between the
worn flank face and the machined surface from the side of the flank face is effective for extending tool life and improving the
surface finish.
As explained above, MQL turning does not always show better performances in extending tool life and improving surface finish
than wet machining or dry machining. MQL cannot be applied to turning hardened steel with a polycrystalline cBN (PCBN) tool
even if oil mist is sprayed from the side of the flank face (41). This fact reminds us that the dry condition is best at reducing the
damage of a PCBN tool during hard turning since a PCBN tool is more sensitive to thermal impact than a coated carbide tool. MQL
provides marginal improvement for turning stainless steel with a coated tool at oil flow rate 10 ml h1 (42) and turning hardened
steel with a ceramic tool at oil flow rate 10 ml h1 (43). It shows a little negative effect in machining plain carbon steel at oil flow
rate 12.5 ml h1 (44). Even in interrupted turning aluminum alloy at oil flow rate 30 ml h1 it exhibits a sufficient lubricating effect
only under light-loaded cutting conditions (45). From a viewpoint of reducing environmental load, MQL can be an alternative to
flood coolant.
There are several methods for further improving the performance of MQL turning: (1) the spraying of oil mist from the side of
the flank face, already explained above, (2) the shortening of the distance from the nozzle to the tool tip, (3) the spraying of oil mist
at higher air pressures through a small nozzle, (4) the spraying of wet oil mist without conversion from a dry mist to a wet one, and
(5) the enhancement of the cooling ability of MQL using chilled air and water droplets.

Figure 31 The application of oil mist from the side of the flank face. (a) Turning (reproduced from Wakabayashi, T.; Sato, H.; Inasaki I. Turning
Using Extremely Small Amount of Cutting Fluids. JSME Int. J. (Series C) 1998, 41, 143–148); (b) Grooving (reproduced from Obikawa, T.; Kamata, Y.;
Shinozuka, J. High Speed Grooving with Applying MQL. Int. J. Mach. Tool Manuf. 2006, 46, 1854–1861).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Machining with Least Quantity Lubrication 275

The compressed air carrying oil droplets expands with the distance from the air nozzle ln. According to Figure 1 the diameter of
white fog Dm increases linearly with the distance from the air nozzle. Thus, the flow rate of cutting oil through a unit area wp would
be inversely proportional to the square of distance from a nozzle and can be formulated by
D0 2 D0 2 1
wp ¼ wp0 ¼ wp0  2 ¼ wp0 
ln 2
2 (22)
Dm D0 þ kp ln 1 þ kp
D0
where wp0 is the flow rate of cutting oil through a unit area at the exit of the nozzle, D0 is the diameter of the nozzle, and kp is
constant. Equation [22] implies that the nozzle should be as close to the tool tip as possible for the purpose of the concentrated
deposition of oil droplets to the cutting zone. Effect of the shortening of the distance from the nozzle to the tool tip to 4.8 mm on
the flank wear was verified in finish turning Inconel 718 (46). The spraying of oil mist obliquely to the cutting edge is also effective
in reducing flank wear (46) since the oblique spraying of oil mist makes the air and oil droplets move along the cutting edge without
causing air stagnation.
The effect of the pressure of the compressed air on the corner wear and the flank wear of a coated tool in MQL grooving plain
carbon steel is demonstrated in Figure 32 (37). Increase in the pressure from 0.3 MPa to 0.7 MPa reduces the tool wear by about
half. When the oil mist is sprayed with compressed air at a high value of pressure, the diameter of the nozzle should be reduced so
that the flow rate of compressed air would not exceed the limited value of a compressor. Reduction in the diameter increases the
flow rate of cutting oil through a unit area at the exit of the nozzle, wp0 in eqn [22], leading to more preferable cutting performance.
In various MQL turning operations there may be the optimum pressure for given cutting conditions and oil flow rate. It was found
to be 0.4 MPa for finish turning Inconel 718 (47).
Like the spindle for MQL milling and MQL drilling shown in Figure 22, it is also preferable in MQL turning to spray wet oil mist
without conversion from a dry mist to a wet one. Such a setup for MQL turning is shown in Figure 33 (37). Wet mist is generated by
blowing dispensed oil with the compressed air in a small mixing chamber of the tool holder. The distance from the chamber to the

(a) (b)
0.30 0.30
MQL 0.7 MPa MQL 0.7 MPa
0.25 MQL 0.5 MPa 0.25 MQL 0.5 MPa
Corner wear (mm)

MQL 0.3 MPa MQL 0.3 MPa

Flank wear (mm)

Wet Wet
0.20 0.20

0.15 0.15

0.10 0.10

0.05 0.05

0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000
Cutting length (m) Cutting length (m)

Figure 32 The corner and flank wear against cutting length in MQL grooving at different pressures of compressed air. (a) Corner wear; (b) Flank
wear. Reproduced from Obikawa, T.; Kamata, Y.; Shinozuka, J. High Speed Grooving with Applying MQL. Int. J. Mach. Tool Manuf. 2006, 46,
1854–1861.

Figure 33 A setup for MQL turning without conversion from a dry mist to a wet one. Reproduced from Obikawa, T.; Kamata, Y.; Shinozuka, J. High
Speed Grooving with Applying MQL. Int. J. Mach. Tool Manuf. 2006, 46, 1854–1861.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
276 Machining with Least Quantity Lubrication

tool tip is about 30 mm. The oil droplets generated grow in size during passage through a nozzle of 1.1 mm in diameter before being
sprayed. The nozzle is opened on the flank face and directed to the cutting point. High cutting performances partly described above
were obtained using this setup (37,46).
With regard to the enhancement of the cooling ability of oil mist, chilled air at temperature 20  C and at a flow rate of
120 l min1 improved the cutting performances of MQL finish turning of Inconel 718 in tool life and surface finish (4). The
consumption rate of cutting oil was as much as 90 ml h1, probably because the oil mist was sprayed from the side of the rake face
to the cutting zone. Replacement of oil droplets with water droplets covered with oil film, called oil on water, which also enhances
the cooling ability of MQL, can reduce the cutting forces and prevent the work material of aluminum alloy from adhering to the tool
even under severe cutting conditions (45). The spraying of simultaneously dispensed oil and water with the compressed air can
easily generate water droplets covered with oil film using a triple wall pipe as shown in Figure 34 (45). A large amount of water at
a flow rate of 3000 ml h1 (50 ml min1) in addition to a small amount of oil at a flow rate of 30 ml h1 is necessary to obtain the
acceptable cutting performance.

11.11.3.4 Drilling
Drilling is a continuous cutting operation like turning. In addition, chips produced must be conveyed from the cutting position to
the outside of the hole. The adhesion between the chip and the drill very often causes the buildup of the work material on the tool
face under dry condition since the sliding of the chip softened at high temperature rapidly increases its real contact area against the
tool face through the mechanism of junction growth (48) if the pressure between them is raised by irregular chip formation, changes
in the shape and morphology of the produced chip, chip flow direction quite different from the direction of the flutes, etc. A certain
amount of the buildup may result in the severe chip clogging leading to the failure of a drill. Thus, the cooling and lubrication of the
chips and the tool faces of a drill are very important to prevent the work material from adhering to the tool and flute faces. Thus,
MQL drilling is not as effective as not only MQL milling but also MQL turning, unless cooling and lubrication in MQL drilling are
not enough for the combination of the work and tool materials and adopted cutting conditions.
Some remarkable results, e.g., such that the number of holes made with a coated carbide drill can be increased by a factor
of 3.8 by adopting MQL at a flow rate of 10 ml h1 instead of flood coolant in drilling plain carbon steel, as shown in
Figure 35, were demonstrated by HORKOS Corp (8). The drilling experiment was performed using a horizontal machining
center with a spindle shown in Figure 22. It is definitely important to horizontally drill holes and immediately evacuate
produced chips by the compressed air supplied through the oil holes of a drill so as to avoid unnecessary troubles that may
arise from the pile-up of the produced chips in the drill flutes when using a vertical machining center. Since the droplets
generated with a wet mist generator built in a spindle are relatively large, the amount of oil dispensed can be reduced by
almost stopping the stray mist.

Figure 34 Method for generating water droplets covered with oil film. Reproduced from Itoigawa, F.; Childs, T. H. C.; Nakamura, T.; Belluco, W.
Effects and Mechanisms in Minimal Quantity Lubrication Machining of an Aluminum Alloy. Wear 2006, 260, 339–344.

Figure 35 Tool lives in the number of holes attained with MQL and flood coolant. Reproduced from http://www.horkos.co.jp/english/products/
imql01.php. Courtesy of HORKOS Corp.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Machining with Least Quantity Lubrication 277

In drilling, the tool life depends on the chip control significantly as well as cutting speed, feed rate, and lubrication and
cooling conditions. Nevertheless, it is difficult to control chip formation including the flow direction, the up and side curls,
the thickness, and the length of a forming chip. A possible way to efficiently evacuate forming chips without flood coolant is
to adjust their shape to a helix flute to a certain extent before they cooled down to have a large stiffness if sprayed oil droplets
can prevent softened chips from adhering to the tool face. This mechanism may be applied to MQL drilling of plain carbon
steel showing the satisfactory result in Figure 35. Broken chips with an appropriate length often seen in drilling cast iron are
also desirable for the efficient chip evacuation and the extension of tool life. However, since the adhesion of many of work
materials to a drill is not negligible, the current goal of MQL drilling of these materials is to achieve the cutting performance
comparable to drilling with flood coolant. When chips produced in MQL drilling cannot be efficiently evacuated from
a drilling hole, pecking is often effective to extend tool life, e.g., MQL drilling of aluminum–silicon alloy containing 6%
silicon with a DLC coated drill (49) and MQL drilling of hardened plastic mold steel (AISI P20 steel) of 36–38 HRC with
a coated carbide drill (50).
The quality of drilled holes and tool wear in MQL drilling of aluminum–silicon alloy containing 7% silicon was investi-
gated by Braga et al. (51). The oil flow rate and air pressure were 10 ml h1 and 0.45 MPa, respectively. Two kinds of drills, i.e.
K10 drills and diamond-coated drills were selected, where the tool face of the latter drill was somewhat rough according to its
photomicrograph. The measurements of diameter, roughness, taper, and roundness proved that the quality of holes drilled
with MQL was either similar to or better than those with flood coolant. The flank wear measured at feed length 20 m was
almost the same for K10 and diamond-coated drills under MQL conditions and a K10 drill under the flood coolant condition,
while it was the smallest for a diamond-coated drill under flood coolant. Apparently, MQL had a negative effect on the flank
wear of diamond-coated drills. Here, it should be pointed out that the face of diamond-coated drill was rough. Hanyu et al.
(52) presented evidence that the surface roughness of the tool face of a diamond-coated drill dramatically changes the tool life
in drilling aluminum–silicon alloy containing 12% silicon as shown in Figure 36. A finely crystallized diamond-coated drill
with a smooth surface has a tool life more than 4 times longer than a diamond-coated drill with a rough surface when the oil
flow rate is 15 ml h1. The rough tool face of traditional diamond coating was covered with aluminum alloy after MQL drilling
2000 holes as shown in Figure 37(a). In contrast, no adhesion of aluminum alloy was observed on the smooth tool face of
finely crystallized diamond coating after MQL drilling 9216 holes when the chipping of the cutting edge occurred as shown in
Figure 37(b).
The effect of the property of MQL on the life of a small drill in high-speed drilling deep holes of plain carbon steel was
investigated for selecting MQL suitable to the particular work material and particular cutting conditions (53). Three types of MQL
selected were synthetic ester, alcohol added synthetic ester, and water added oil-free synthetic lubricant, which are superior in
lubrication, lubrication (primary) and cooling (secondary), and cooling, respectively. They were applied to the drilling point from
the mouth of a deep hole at an oil flow rate of 18 ml h1. From the tool life experiments, MQL with a high cooling capability was
found to be more advantageous in extending tool life than that with a high lubrication ability. The water added lubricant with low
viscosity has a high permeability. Hence, it can penetrate into the drilling point even in a high-speed drilling process. This implies
that the lubricant with a high permeability would be advantageous in high-speed drilling processes. In drilling aluminum–silicon
alloy containing 6% silicon and cast magnesium alloy AM60, it was proved that water is a promising MQL agent (54,55). When the
mist of distilled water is applied at the flow rate of 30 ml h1, the adhesion of the work materials is negligible on the face of an HSS

Figure 36 Tool lives of three different drills in drilling aluminum–silicon alloy (JIS ADC12) with air blow and MQL. Reproduced from Hanyu, H.;
Kamiya, S.; Murakami, Y.; Saka, M. Dry and Semi-Dry Machining Using Finely Crystallized Diamond Coating Cutting Tools. Surf. Coat. Technol. 2003,
173–174, 992–995.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
278 Machining with Least Quantity Lubrication

Figure 37 Tool faces of drills after MQL drilling of aluminum–silicon alloy. (a) Carbide drill with rough surface of traditional diamond coating after
drilling 2000 holes; (b) Carbide drill with smooth surface of finely crystallized diamond coating after drilling 9216 holes. Reproduced from Hanyu, H.,
Kamiya, S., Murakami, Y. and Saka, M. Dry and Semi-Dry Machining Using Finely Crystallized Diamond Coating Cutting Tools. Surf. Coat. Technol.
2003, 173–174, 992–995.

drill with nonhydrogenated DLC coating. It should be noted that when only an uncoated HSS drill is used, the best MQL is not
distilled water, but a fatty acid-based MQL fluid with extreme pressure additives (56). Hence, the selection of a drill is quite
important, especially in MQL drilling since the affinity between the tool and work materials has a direct effect on the cutting
performance.
As described above, MQL is not always the best lubrication/cooling condition for drilling. It is regrettable that MQL did not
exhibit the best performance for drilling austempered ductile iron (57) and that the best quality of a drilled hole was not obtained
by MQL in high-speed drilling of hardened steel of a hardness of 53 HRC (58).

11.11.3.5 Micro-MQL Machining and Other Future-Oriented MQLs

In the near future, oil consumption in non-heavy duty MQL machining may be reduced to the level of tens to hundreds of
microliters per hour. This is about one-hundredth to one-thousandth of the oil consumption applied to ordinary MQL machining
processes. Such a machining method is called micro-MQL machining. The first attempt was conducted at an oil flow rate of
0.6 ml h1 by Wakabayashi et al. (36) in 1998. Obikawa et al. (46) conducted finish turning of Inconel 718 using micro-MQL.
Recently, micro-MQL is applied to various micro-mechanical machining processes. For these applications, an oil mist generator for
micro-MQL shown in Figure 38 has been developed and already available. The maximum oil flow rate is only 1 ml h1. An array of
small holes drilled with micro-MQL is shown in Figure 39, in which 1225 holes of 0.28 mm in diameter are drilled through an
aluminum plate of 1.0 mm thick without any trouble.
Quite recently, new types of MQL were proposed, and their effects on the cutting performance were demonstrated to improve the
lubricity of MQL and enhance the machinability. This is because MQL machining is not always the best machining process; its
application is limited to machining under rather mild cutting conditions. Nam et al. (59) applied two kinds of MQL to microdrilling
with a twist drill of 200 mm in diameter. They are vegetable oil and paraffin oil containing different volume percent of nanodiamond
particles of 30 nm in size. The optimum concentration of nanodiamond particles was found to be 1 vol% for paraffin oil and 2 vol%
for vegetable oil. Park et al. (60) added solid lubricant to vegetable oil and applied it to MQL end milling of plain carbon steel with
a coated carbide ball end mill. They selected exfoliated graphite nanoplatelets (xGnP) consisting of small stacks of graphene 1–
15 nm thick because it is an excellent solid lubricant and xGnP added oil exhibits an excellent wettability. Cutting experiments
at an axial depth of cut of 1 mm, a feed rate of 0.28–0.36 mm tooth1, and an oil flow rate of 90 ml h1 showed that the
concentration of only 0.1 wt% of xGnP can enhance the cutting performance.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Machining with Least Quantity Lubrication 279

Figure 38 Oil mist generator for micro-MQL. Courtesy of Fuji BC Engineering Co., Ltd.

Figure 39 Small holes drilled with micro-MQL. Courtesy of Fuji BC Engineering Co., Ltd.

11.11.4 Concluding Remarks

In this chapter, MQL machining, one of the ECMTs, was first introduced; its environmental and cost adaptabilities were explained
based on data reported by Japanese companies. Then, the mechanisms of lubricating the tool–chip and tool–work interfaces by
means of oil mist were theoretically described in detail from the viewpoints of gas-phase and liquid-phase lubrication. The
conversion of dry mist to wet mist using an orifice was explained as a key process for enhancing the lubricity of oil mist when a dry
mist is generated to deliver cutting oil for long distance. Next, MQL machining technologies were reviewed with regard to milling,
turning, and drilling. In the review, the characteristic of MQL machining was emphasized for the appropriate application of MQL
machining to various work materials. Finally, machining with micro-MQL, an oil mist generator for micro-MQL, and new types of
MQL containing solid lubricants and its effect on drilling and ball end milling was briefly introduced. Theoretical description in this
chapter will be helpful when something new is required for MQL machining.

Acknowledgment

The author would like to thank Dr. Murakami at the Institute of Industrial Science, the University of Tokyo for reviewing the
manuscript.

References

1. Klocke, F.; Eisenblatter, G. Dry Cutting. CIRP Ann. 1997, 46, 519–526.
2. http://www.fuji-bc.com/english/ebtool/index.html.
3. Shokrani, A.; Dhokia, V.; Newman, S. T. Environmentally Conscious Machining of Difficult-to-Machine Materials with Regard to Cutting Fluids. Int. J. Mach. Tool Manuf. 2012,
57, 83–101.
4. Su, Y.; He, N.; Li, L.; Iqbal, A.; Xiao, M. H.; Xu, S.; Qiu, B. G. Refrigerated Cooling Air Cutting of Difficult-to-Cut Materials. Int. J. Mach. Tool Manuf. 2007, 47, 927–933.
5. Pusavec, F.; Kramar, D.; Krajnik, P.; Kopac, J. Transitioning to Sustainable Production – Part II: Evaluation of Sustainable Machining Technologies. J. Cleaner Prod. 2010, 18,
1211–1221.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
280 Machining with Least Quantity Lubrication

6. Press Release from the Energy Conservation Center, Japan: The Minister Prize of Economy, Trade and Industry in Awarding of Successful Case of Energy Conservation in
Factory and Building in 2002 (in Japanese).
7. Ikawa, S.; Iwatsubo, M.; Miyazaki, K. To This Point It Has Come. Dry Machining. Challenge to Dry Machining for Environmental Correspondence. Mech. Eng. 1999, 47 (5), 18–
22 (in Japanese).
8. http://www.horkos.co.jp/english/products/imql01.php.
9. Eisentraeger, A.; Schmidt, M.; Murrenhoff, H.; Dott, W.; Hahn, S. Biodegradability Testing of Synthetic Ester Lubricants – Effects of Additives and Usage. Chemosphere 2002,
48, 89–96.
10. http://www.eneosmotoroil.com/files/industrial/MQL_Bulletin.pdf.
11. Nagata, M.; Morita, H.; Yamada, T.; Iwasa, M. Approach to MQL Cutting by Using High Lubrication Cutting Fluid. In Proceedings of the Third International Conference on
Leading Edge Manufacturing in 21st Century, Nagoya; 2005; pp 127–132.
12. Wakabayashi, T.; Atsuta, T.; Tsukuda, A.; Sembongi, N.; Shibata, J.; Suda, S. Cutting Performance of Oxygen-Including Compounds in MQL Machining of Aluminum. Key Eng.
Mater. 2012, 523–524, 967–972.
13. Williams, J. A.; Tabor, D. The Role of Lubricants in Machining. Wear 1977, 43, 275–292.
14. Williams, J. A. The Action of Lubricants in Metal Cutting. J. Mech. Eng. Sci. 1977, 19, 202–212.
15. Wakabayashi, T.; Williams, J. A.; Hutchings, I. M. The Kinetics of Gas Phase Lubrication in the Orthogonal Machining of an Aluminum Alloy. J. Eng. Tribol. 1995, 209,
131–136.
16. Obikawa, T.; Asano, Y.; Kamata, Y. Computer Fluid Dynamics Analysis for Efficient Spraying of Oil Mist in Finish-Turning of Inconel 718. Int. J. Mach. Tool Manuf. 2009, 49,
971–978.
17. Childs, T. H. C.; Maekawa, K.; Obikawa, T.; Yamane, Y. Metal Machining: Theory and Application; Arnold: London, 2000; p 75.
18. Obikawa, T.; Yamaguchi, M.; Funai, K.; Kamata, Y.; Yamada, S. Air Jet Assisted Machining of Nickel-Base Superalloy. Int. J. Mach. Tool Manuf. 2012, 61, 20–26.
19. Trela, M.; Zembik, J.; Durkiewicz, B. Droplet Deposition on a Flat Plate from an Air/Water Turbulent Mist Flow. Int. J. Multiphase Flow 1982, 8, 227–238.
20. Veysey, J., II; Goldenfeld, N. Simple Viscous Flows: From Boundary Layers to the Renormalization Group. Rev. Mod. Phys. 2007, 79, 883–927.
21. Rahman, M.; Kumar, A. S.; Salam, M. U. Experimental Evaluation on the Effect of Minimal Quantities of Lubricant in Milling. Int. J. Mach. Tool Manuf. 2002, 42, 539–547.
22. López de Lacalle, L. N.; Angulo, C.; Lamikiz, A.; Sánchez, J. A. Experimental and Numerical Investigation of the Effect of Spray Cutting Fluids in High Speed Milling. J. Mater.
Process. Technol. 2006, 172, 11–15.
23. Aoyama, T. Development of a Mixture Supply System for Machining with Minimal Quantity Lubrication. CIRP Ann. 2002, 51, 289–292.
24. Paschall, T. W.; Gaynor, P. A. Oil Mist Lubricated Bearings. IEEE Ind. Appl. Mag. 2005, 11, 25–31.
25. Kishawy, H. A.; Dumitrescu, M.; Ng, E. G.; Elbestawi, M. A. Effect of Coolant Strategy on Tool Performance, Chip Morphology and Surface Quality during High-Speed Machining
of A356 Aluminum Alloy. Int. J. Mach. Tool Manuf. 2005, 45, 219–227.
26. Sun, J.; Wong, Y. S.; Rahman, M.; Wang, Z. G.; Neo, K. S.; Tan, C. H.; Onozuka, H. Effects of Coolant Supply Methods and Cutting. Conditions on Tool Life in End Milling
Titanium Alloy. Mach. Sci. Technol. 2006, 10, 355–370.
27. Liao, Y. S.; Lin, H. M.; Chen, Y. C. Feasibility Study of the Minimum Quantity Lubrication in High-Speed End Milling of NAK80 Hardened Steel by Coated Carbide Tool. Int. J.
Mach. Tool Manuf. 2007, 47, 1667–1676.
28. Kang, M. C.; Kim, K. H.; Shin, S. H.; Jang, S. H.; Park, J. H.; Kim, C. Effect of the Minimum Quantity Lubrication in High-Speed End Milling of AISI D2 Cold-Worked Die Steel (62
HRC) by Coated Carbide Tools. Surf. Coat. Technol. 2008, 202, 5621–5624.
29. Bruni, C.; d’Apolito, L.; Forcellese, A.; Gabrielli, F.; Simoncini, M. Surface Roughness Modelling in Finish Face Milling under MQL and Dry Cutting Conditions. Int. J. Mater. Form.
2008, 1 (Suppl. 1), 503–506.
30. Thepsonthi, T.; Hamdi, M.; Mitsui, K. Investigation into Minimal-Cutting-Fluid Application in High-Speed Milling of Hardened Steel Using Carbide Mills. Int. J. Mach. Tool Manuf.
2009, 49, 156–162.
31. Safian, S.; Hisyam, M. A.; Aman, S. Evaluation of Vegetable Oil as an Alternative Cutting Lubricant when End Milling Martensitic Stainless Steel Using Uncoated Carbide Tool.
J. Adv. Manuf. Technol. 2009, 3, 49–55.
32. Qin, X.; Gui, L.; Li, H.; Rong, B.; Wang, D.; Zhang, H.; Zuo, G. Feasibility Study on the Minimum Quantity Lubrication in High-Speed Helical Milling of Ti–6Al–4V. J. Adv. Mech.
Des. Syst. Manuf. 2012, 6, 1222–1233.
33. Da Silva, R. B.; Vieira, J. M.; Cardoso, R. N.; Carvalho, H. C.; Costa, E. S.; Machado, A. R.; De Avila, R. F. Tool Wear Analysis in Milling of Medium Carbon Steel with Coated
Cemented Carbide Inserts Using Different Machining Lubrication/Cooling Systems. Wear 2011, 271, 2459–2465.
34. Fox-Rabinovich, G. S.; Yamamoto, K.; Beake, B. D.; Gershman, I. S.; Kovalev, A. I.; Veldhuis, S. C.; Aguirre, M. H.; Dosbaeva, G.; Endrino, J. L. Hierarchical Adaptive Nano-
structured PVD Coatings for Extreme Tribological Applications: the Quest for Nonequilibrium States and Emergent Behavior. Sci. Technol. Adv. Mater. 2012, 13. AN. 043001.
35. Obikawa, T.; Anzai, M.; Egawa, T.; Narutaki, N.; Shintani, K.; Takeoka, E. High Speed Machining: A Review from a Viewpoint of Chip Formation. Adv. Mater. Res. 2011, 188,
578–583.
36. Wakabayashi, T.; Sato, H.; Inasaki, I. Turning Using Extremely Small Amount of Cutting Fluids. JSME Int. J. (Series C) 1998, 41, 143–148.
37. Obikawa, T.; Kamata, Y.; Shinozuka, J. High Speed Grooving with Applying MQL. Int. J. Mach. Tool Manuf. 2006, 46, 1854–1861.
38. Dhar, N. R.; Kamruzzaman, M.; Ahmed, M. Effect of Minimum Quantity Lubrication (MQL) on Tool Wear and Surface Roughness in Turning AISI-4340 Steel. J. Mater. Process.
Technol. 2006, 172, 299–304.
39. Sreejith, P. S. Machining of 6061 Aluminium Alloy with MQL, Dry and Flooded Lubricant Conditions. Mater. Lett. 2008, 62, 276–278.
40. Khan, M. M. A.; Mithu, M. A. H.; Dhar, N. R. Effects of Minimum Quantity Lubrication on Turning AISI 9310 Alloy Steel Using Vegetable Oil-Based Cutting Fluid. J. Mater.
Process. Technol. 2009, 209, 5573–5583.
41. Diniz, A. E.; Ferreira, J. R.; Filho, F. T. Influence of Refrigeration/Lubrication Condition on SAE 52100 Hardened Steel Turning at Several Cutting Speeds. Int. J Mach. Tool
Manuf. 2003, 43, 317–326.
42. Bruni, C.; Forcellese, A.; Gabrielli, F.; Simoncini, M. Effect of the Lubrication-Cooling Technique, Insert Technology and Machine Bed Material on the Workpart Surface Finish
and Tool Wear in Finish Turning of AISI 420B. Int. J. Mach. Tool Manuf. 2006, 46, 1547–1554.
43. Bruni, C.; Forcellese, A.; Gabrielli, F.; Simoncini, M. Hard Turning of an Alloy Steel on a Machine Tool with a Polymer Concrete Bed. J. Mater. Process. Technol. 2008, 202,
493–499.
44. Li, K. M.; Liang, S. Y. Performance Profiling of Minimum Quantity Lubrication in Machining. Int. J. Adv. Manuf. Technol. 2007, 35, 226–233.
45. Itoigawa, F.; Childs, T. H. C.; Nakamura, T.; Belluco, W. Effects and Mechanisms in Minimal Quantity Lubrication Machining of an Aluminum Alloy. Wear 2006, 260, 339–344.
46. Obikawa, T.; Kamata, Y.; Asano, Y.; Nakayama, K.; Otieno, A. W. Micro-Liter Lubrication Machining of Inconel 718. Int. J. Mach. Tool Manuf. 2008, 48, 1605–1612.
47. Kamata, Y.; Obikawa, T. High Speed MQL Finish-Turning of Inconel 718 with Different Coated Tools. J. Mater. Process. Technol. 2007, 192, 281–286.
48. Bowden, F. P.; Tabor, D. The Friction and Lubrication of Solids; Oxford University Press: Oxford, 1950; Vol. 1.
49. Dosbaeva, J.; Fox-Rabinovich, G.; Dasch, J.; Veldhuis, S. Enhancement of Wet- and MQL-Based Machining of Automotive Alloys Using Cutting Tools with DLC/Polymer Surface
Treatments. J. Mater. Eng. Perform. 2008, 17, 346–351.
50. Zeilmann, R. P.; Nicola, G. L.; Vacaro, T.; Teixeira, C. R.; Heiler, R. Implications of the Reduction of Cutting Fluid in Drilling AISI P20 Steel with Carbide Tools. Int. J. Adv. Manuf.
Technol. 2012, 58, 431–441.
51. Braga, D. U.; Diniz, A. E.; Miranda, G. W. A.; Coppini, N. L. Using a Minimum Quantity of Lubricant (MQL) and a Diamond Coated Tool in the Drilling of Aluminum–Silicon Alloys.
J. Mater. Process. Technol. 2002, 122, 127–138.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Machining with Least Quantity Lubrication 281

52. Hanyu, H.; Kamiya, S.; Murakami, Y.; Saka, M. Dry and Semi-Dry Machining Using Finely Crystallized Diamond Coating Cutting Tools. Surf. Coat. Technol. 2003, 173–174,
992–995.
53. Heinemann, R.; Hinduja, S.; Barrow, G.; Petuelli, G. Effect of MQL on the Tool Life of Small Twist Drills in Deep-Hole Drilling. Int. J Mach. Tool Manuf. 2006, 46, 1–6.
54. Bhowmick, S.; Alpas, A. T. Minimum Quantity Lubrication Drilling of Aluminium–Silicon Alloys in Water Using Diamond-Like Carbon Coated Drills. Int. J. Mach. Tool Manuf.
2008, 48, 1429–1443.
55. Bhowmick, S.; Alpas, A. T. The Role of Diamond-Like Carbon Coated Drills on Minimum Quantity Lubrication Drilling of Magnesium Alloys. Surf. Coat. Technol. 2011, 205,
5302–5311.
56. Bhowmick, S.; Lukitsch, M. J.; Alpas, A. T. Dry and Minimum Quantity Lubrication Drilling of Cast Magnesium Alloy (AM60). Int. J. Mach. Tool Manuf. 2010, 50, 444–457.
57. Meena, A.; El Mansori, M. Study of Dry and Minimum Quantity Lubrication Drilling of Novel Austempered Ductile Iron (ADI) for Automotive Applications. Wear 2011, 271,
2412–2416.
58. Brandao, L. C.; Neves, F. O.; Nocelli, G. C. Evaluation of Hole Quality in Hardened Steel with High-Speed Drilling Using Different Cooling Systems. Adv. Mech. Eng. 2011. Article
ID 746535, 7 pages.
59. Nam, J. S.; Lee, P. H.; Lee, S. W. Experimental Characterization of Micro-Drilling Process Using Nanofluid Minimum Quantity Lubrication. Int. J. Mach. Tool Manuf. 2011, 51,
649–652.
60. Park, K. H.; Ewald, B.; Kwon, P. Y. Effect of Nano-Enhanced Lubricant in Minimum Quantity Lubrication Balling Milling. J. Tribol. 2011, 133, 031803.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
This page intentionally left blank

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 11.12 Burr Formation in Machining Processes: A Review
IA Choudhury, University of Malaya, Kuala Lumpur, Malaysia
SA Lawal, University of Malaya, Kuala Lumpur, Malaysia and Federal University of Technology, Minna, Nigeria
 2014 Elsevier Ltd. All rights reserved.

11.12.1 Introduction 283

11.12.2 Burr Formation in Turning Process 284
11.12.3 Burr Formation in Drilling Process 288
11.12.4 Burr Formation in Milling Process 292
11.12.5 Conclusion and Future Research Direction 293
References 294

11.12.1 Introduction

The control and removal of burr have become one of the most important economic factors in machining processes and as a result
have been the focus of research in machining operations in the last five decades (1). Burrs, which are small pieces of deformed
material left on the edges of the workpiece, are found in most machining operations. It brings various challenges such as worsening
the dimensional accuracy and surface finish, (32) reducing the cutting performance and life of the cutting tools, and potentially
causing accidents to workers and consumers during usage. Burr formation is a complicated process, and its formation involves large
plastic deformation of work material, while the type of burrs and their characteristics depend on the type of machining process, the
process parameters, tool property, tool geometry, tool edge configuration, coolant, and workpiece material properties (2). The word
‘burr’ in machining process has been defined by many authorities. Schafer (3) described burr as the part of a workpiece that is
produced through manufacturing processes on an edge or a surface and which lies outside the desired geometry. The ISO 13715 (4)
defines the edges of a workpiece as burred if it has an overhang greater than zero, as shown in Figure 1. Ko and Dornfeld (5) defined
burr as an “undesirable projection of material formed as the result of plastic flow from a cutting or shearing operation.” One of the
first researchers to study burr formation at an academic level was Gillespie (6), whose definition of burr is limited to cutting and
shearing processes; hence, a burr produced by those operations includes “all the material extending past the theoretical intersection
of two surfaces, which surround the burr.” The reference in that case is the theoretical intersection of the two surfaces and not the
desired surface and, in addition, Gillespie’s definition includes burrs that lie inside the theoretical intersection as shown in Figure 2.
Beier (7) gave a comprehensive definition of burr as a body created on a workpiece surface during the manufacturing of a workpiece

Figure 1 Definition of burrs according to ISO 13175. Reproduced from International Standard ISO 13715. Technical Drawings – Edges of Undefined
Shape – Vocabulary and Indications, 2000.

Figure 2 Examples of burr definition according to Gillespie. Reproduced from Gillespie, L. K. The Battle of the Burr: New Strategies and New Tricks.
Manuf. Eng. 1996, 116 (2), 69–78.

Comprehensive Materials Processing, Volume 11 http://dx.doi.org/10.1016/B978-0-08-096532-1.01117-1 283

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
284 Burr Formation in Machining Processes: A Review

that extends over the intended and actual workpiece surface and has a slight volume in comparison with the workpiece, undesired,
but to some extended, unavoidable.
The first researcher to investigate chip formation in cutting process and burr formation mechanism was Pekelharing (8), and his
works described burrs formation in punching process, while the first fundamental work on burr formation mechanisms was
published by Gillespie and Blotter (9). Gillespie (10) presented an analytical model that illustrated burr formation mechanisms and
predicted burr properties. The results of this model were compared with experimental observations. The understanding of the
mechanisms behind burr formation had helped researchers focus on deburring. Advanced technology for deburring becomes more
important in order to improve productivity in machining of parts (11). Deburring has become a serious problem that needs urgent
attention as the formation of burr edge fractures during machining means change of the geometry of products or parts. When the
deburring of a precision part is not considered until the final stages of manufacturing, the potential loss due to any failure in the
selection, planning, or execution of the edge-finishing process is great (12). The cost of deburring these components may contribute
as much as 30% to the cost of finished parts (13). The selection of capable deburring and finishing processes for precision
components is highly dependent on knowledge of burr properties. Burr size, shape, and location as well as the allowable surface
finish are the primary factors in the selection of deburring process. Burr properties are influenced by part design and process
planning decisions. To classify whether a particular burr property is influenced primarily by the design stage or the manufacturing
stages requires burr formation data, burr formation models, and burr formation mechanism identification (12). Burr formation
during machining process is therefore a phenomenon that is undesirable and at the same time unavoidable. However, there are
many methods that have been suggested to either minimize burr formation or remove burrs.
Burr can be classified into rollover, Poisson, tear, and cutoff based on the mechanism of formation (9). It is sometimes classified as
backward flow, sideway flow, forward flow, and leaned burr by the burrs formation direction (14) and also as entrance burr, side burr,
and exit burr by the location of burr formation (9,14). However, two burr types have been identified to cause serious problems in
practice, these are rollover burrs and Poisson burrs. Ko and Dornfeld (5,11) investigated the burr formation model for orthogonal cutting
and their concern was the rollover burr mechanism, the forward flow, and exit burr. In another classification by Kishimoto (15), two
types of burr, primary and secondary burrs, were identified. He claimed that through proper selection of cutting conditions and tool
geometry, the rollover burr will be separated at its thinnest portion and only a small burr remained on the edge of the machined part.
Hence, the former normal burr was named a primary burr and the later one a secondary burr, which is the material remaining after the
breakage of the primary burr. Beier (7) described a secondary burr as material that remains on the edge of a part after a deburring process.
Some of the efforts in the past few decades have been devoted to study the mechanism of burr formation in machining processes
without thoroughly exhausting all the factors that could affect burr formation as identified by Yu Long and Changsheng Guo (2).
Kim et al. (16) studied formation in drilling process, and they found that shapes and sizes of drilling burrs depended on the process
parameters. They equally observed that drilling burrs have uniform shape for most materials when the feed or cutting speed is low,
and when material is ductile, the burrs are elongated, which results in a large burr height and burr volume. But if material is brittle in
nature, catastrophic fracture makes irregular-shaped burrs when the feed and speed are increased (17). Pande and Relekar (18)
described experimental investigations for reducing the burr formation while drilling through-holes in metals using uncoated
standard twist drills. Stein and Dornfeld (19) presented a study on the burr height, thickness, and geometry observed in the drilling
of 0.91-mm-diameter through-holes in stainless steel 304L. The mechanism of burr formation in face milling is similar to that in
drilling (20). Milling burrs are created mainly when the milling cutter exists at the edge of workpiece (21). Efforts by Park and
Dornfeld (22) and Min et al. (17) to study burr formation with FEM are yet to yield positive results due to the inability to simulate
all four stages of burr formation as a result of material model limitation (2).
Available literature on burr formation in machining processes seems to be silent on the critical role played by different types of
coolant applications, especially vegetable oil–based coolants, in determining the quality of surface finish of workpiece material. Most of
the literature on burr formation focused on tool geometry, tool edge configuration, and workpiece material properties. In order to do
justice to the causes of burr formation in machining process, it is necessary to take the temperature dependency of material properties into
account when explaining burr formation phenomena (23). Again, it has been established that the chemical composition and mechanical
properties of the work material, the tool, and the cutting fluid type are of vital importance in determining process performance and
finished surface quality (24). Experimental results have confirmed that coolants based on vegetable oils show better performance than
mineral oils during drilling of AISI 316L austenitic stainless steel using conventional high-speed steel (HSS) tools. Since the combi-
nations of all the input parameters, including coolants, have effects on the surface integrity of the workpiece during machining process,
the need to investigate the effect of different types of coolants on burr formation mechanism during machining processes is long overdue.
The application of vegetable oil–based cutting fluids has shown that better results were obtained for the output parameters compared to
when conventional or dry cutting were employed (25–28). This chapter therefore presents a review on burr formations in turning,
drilling, and milling processes and highlights the need to study the effect of different coolants on burr formation mechanism.

11.12.2 Burr Formation in Turning Process

Burrs occurring in turning processes are Poisson burrs, which are formed when the cutting edge of a tool extends beyond the
workpiece edge, as shown in Figure 3. A rollover burr can be formed if the cutting tool passes over a groove or cutting is interrupted
due to other geometric features of the workpiece. Gillespie (29) observed that, in the turning process, most burrs are created as
rollover burrs at the side of the workpiece when the tool exits from cutting.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Burr Formation in Machining Processes: A Review 285

Figure 3 Schematic of Poisson, tear, and rollover burrs. Reproduced from Gillespie, L. K.; Blotter, P. T. The Formation and Properties of Machining
Burrs. Trans. ASME J. Eng. Ind. 1976, 98, 66–74.

Pavel et al. (30) investigated the effect of tool wear on surface finish for a case of continuous and interrupted hard turning using
two types of workpieces for continuous and interrupted cuts, respectively. The continuous surface was a camshaft 155 mm long with
a 28.6 mm diameter. The material was AISI 1117 steel hardened to 62  1 Rockwell hardness C scale (HRC), which is usually used
where a combination of good machinability and more uniform response to heat treatment is needed. The second workpieces for
longitudinal interruptions were shafts having 10 splines. The surface to be machined was 63.8 mm long and had a 34.9 mm
diameter. The spline shafts were made of AISI 1137 steel having a medium hardness of 48  1 HRC and widely used for parts where
a large amount of machining is necessary or where threads, splines, or other operations offer special tooling problems. Amborite
DBC50 and Amborite DBN45 tools specially designed for turning operations were used in this investigation. The following four
types of tests were run and the surface finish was observed in parallel with tool wear. All the cutting parameters and setups were in
accordance with regimes that are used in practice without cutting fluid.
1. Initial experiments – cutting regime: depth of cut (ap ¼ 0.178 mm), cutting speed (v ¼ 125 m min1), and feed rate
(f ¼ 0.15 mm rev1).
2. Replica of initial experiments: first tests were replicated to verify and confirm the initial findings and avoided the possible
outliers.
3. Higher speed experiments: depth of cut (ap ¼ 0.178 mm), cutting speed (v ¼ 175 m min1), and feed rate (f ¼ 0.15 mm rev1).
4. Production-run simulation: depth of cut (ap ¼ 0.178 mm), cutting speed (v ¼ 125 m min1), and feed rate
(f ¼ 0.102 mm rev1).
One of the effects of tool wear on surface finish in interrupted cutting was an improvement on surface finish with tool wear;
however, a negative effect was observed in terms of burr formation. It was noticed that the actual experiments provided significant
burrs when tool wear reached relatively high levels (VBmax > 0.15 mm). Figure 4(a) and 4(b) shows a spline edge when hard
turning was performed with a fresh tool and when flank wear had a value of approximately 0.17 mm at the same edge, respectively.
One of the observations made in this study was that special care should be given to burr formation during interrupted cutting, to
avoid damage to adjacent surfaces.
The results obtained for this experiment could give different results entirely if the application of coolant was considered.
Different coolants have a way of affecting the performances of machining process, and it is believed that if coolant was applied, it
will likely alter the results.
In another development, an experimental study of the burr formation mechanism in feed direction was conducted by Toropov
et al. (31). Influence of tool angles and workpiece angles, as well as other cutting conditions, on burr dimension was considered. The
experiments on burr formation were carried out on a computer numerical control (CNC) turning machine tool. The experimental
setup, tool, and burr geometry are shown in Figure 5. While K10 grade of tungsten carbide–cobalt alloy was chosen as a cutting tool
material in turning of aluminum alloy Al6061-T6. Tables 1 and 2 present tool geometry and cutting conditions used in the
experiments, respectively.
The burr height (h) and burr thickness (b) were measured after every experiment using a laser measurement system (32). Table 3
presents cutting conditions used in a special experiment executed to allow observation of the burr formation process.
Figure 6 shows the influence of the lead angle (4) on the burr dimensions for cutting conditions given in Table 3. For small lead
angles, the burr formation is probably most related to sideward (or Poisson) burr.
Again, the influence of depth of cut on burr height shows that for a lead angle of 16 , the burr height is independent of depth of
cut, whereas for a lead angle of 32  or 47  , the burr height increased proportionate to the depth of cut. They observed that clearance
angle of the tool does not have any significant influence on burr dimensions, but increase in rake angle does cause considerable
reduction of burr thickness and height. One of the conclusions reached by the authors is that the mechanism of burr formation in
feed direction when cutting aluminum alloy Al6061-T6 using a sharp tool depends essentially on tool geometry, workpiece angle,

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
286 Burr Formation in Machining Processes: A Review

Figure 4 Burr formation in interrupted hard turning: (a) fresh tool, (b) worn tool. Reproduced from Pavel, R.; Marinescu, I.; Deis, M.; Pillar, J. Effect of
Tool Wear on Surface Finish for a Case of Continuous and Interrupted Hard Turning. J. Mater. Process. Technol. 2005, 170, 341–349.

and feed. The mechanism was determined mainly by the stress state in the chip formation zone, though stresses on the tool
clearance face have a very slight influence on the burr formation. The increased tool rake angle led to a favorable change of the stress
state in the chip formation zone, which resulted in a considerable reduction of burr dimensions.
The authors considered the influence of tool angles and workpiece angles to study burr formation in turning aluminum alloy. In
spite of significant success in studying burr formation in feed direction, it is still unclear about the mechanism behind it. The
information available on the cutting conditions shows that the experiment was conducted in dry cutting mode. Since one of the

Figure 5 Experimental setup, tool geometry, and final burr dimensions. Reproduced from Toropov, A.; Ko, S.-L.; Kim, B.-K. Experimental Study of
Burrs Formed in Feed Direction when Turning Aluminum Alloy Al6061-T6. J. Mach. Tools Manuf. 2005, 45, 1015–1022.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Burr Formation in Machining Processes: A Review 287

Table 1 Tool geometry used in experiments

Inclination of major End cutting End relief

Rake angle a (  ) Clearance angle ac (  ) Lead angle 4 (  ) cutting edge (  ) edge angle (  ) angle (  )

5, 0, þ5, þ10, þ20 5, 10, 15, 20 16, 32, 47, 66, 81 0 5 5

Reproduced from Toropov, A.; Ko, S.-L.; Kim, B.-K. Experimental Study of Burrs Formed in Feed Direction when Turning Aluminum Alloy
Al6061-T6. J. Mach. Tools Manuf. 2005, 45, 1015–1022.

Table 2 Cutting conditions used in experiments

Cutting speed v (m min1) Feed rate f (mm rev1) Depth of cut t (mm) Workpiece angle W (  )

800 0.05, 0.1, 0.15, 0.2 0.5, 1.0, 1.5, 2.0 90, 109, 118, 133, 147

Reproduced from Toropov, A.; Ko, S.-L.; Kim, B.-K. Experimental Study of Burrs Formed in Feed Direction when Turning Aluminum Alloy
Al6061-T6. J. Mach. Tools Manuf. 2005, 45, 1015–1022.

Table 3 Constant cutting conditions used in experiments

Cutting speed v Feed rate f Depth of Workpiece Rake Clearance Inclination of major End cutting End relief
(m min1) (mm rev1) cut (mm) angle W (  ) angle a ( ) angle ac (  ) cutting edge (  ) edge angle (  ) angle (  )

800 0.1 1 90 0 10 5 5 5

Reproduced from Toropov, A.; Ko, S.-L.; Kim, B.-K. Experimental Study of Burrs Formed in Feed Direction when Turning Aluminum Alloy Al6061-T6. J. Mach. Tools Manuf.
2005, 45, 1015–1022.

Figure 6 Burr dimensions vs tool lead angle (cutting conditions as in Table 3). Reproduced from Toropov, A.; Ko, S.-L.; Kim, B.-K. Experimental Study
of Burrs Formed in Feed Direction when Turning Aluminum Alloy Al6061-T6. J. Mach. Tools Manuf. 2005, 45, 1015–1022.

functions of coolant is to cool at high temperature, it is possible that if coolant was employed in this research, a different response
would have been obtained on the burr formation parameters.
Ma et al. (33) examined the suppression of burrs in turning with ultrasonic elliptical vibration cutting using theoretical models of the
stresses of deformation zone on the workpiece edge in burr formation in ultrasonic elliptical vibration cutting based on three-
dimensional cutting model. The theoretical model was clarified experimentally and observed that friction between the tool rake face
and the chip is reduced or reversed by the elliptical vibration (34–36). The cutting conditions for the experiment are shown in Table 4.
The heights of the feed-direction burrs measured in three cutting methods are shown in Figure 7. It can be understood that the
heights of burrs generated in both conventional vibration cutting and elliptical vibration cutting were reduced and became smaller
and smaller with the increase of maximum vibration speed to cutting speed ratio. Based on theoretical analysis, the authors believed
that the pushing stress and bending stress of deformation zone on the workpiece edge in burr formation for two vibration cutting
methods are reduced due to the separating characteristics between the rake face of the tool and the chip, which resulted in decreased
heights of the burrs. The authors observed that both the theoretical analysis and experimental results proved that burrs can be
effectively suppressed by ultrasonic elliptical vibration cutting.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
288 Burr Formation in Machining Processes: A Review

Table 4 Cutting conditions

Work material Aluminum (52S)

Size 30 mm  200 mm
Tool insert and its geometry Carbide (rake angle: 5 )
Clearance angle: 12
Approach angle: 65
Nose radius: 0.1 mm
Cutting conditions Speed: 3.94–18.3 m min1
Feed rate: 0.025 mm rev1
Depth of cut: 0.05 mm
Vibration conditions Locus: circle (radius: 3.5 mm)
Linear (amplitude: 3.5 mm)
Resonant frequency: 18.66 kHz

Reproduced from Ma, C.; Shamoto, E.; Moriwaki, T.; Zhang, Y.; Wang, L. Suppression of Burrs in Turning with
Ultrasonic Elliptical Vibration Cutting. Int. J. Mach. Tools Manuf. 2005, 45, 1295–1300.

Figure 7 Height of burrs in three cutting methods. Reproduced from Ma, C.; Shamoto, E.; Moriwaki, T.; Zhang, Y.; Wang, L. Suppression of Burrs in
Turning with Ultrasonic Elliptical Vibration Cutting. Int. J. Mach. Tools Manuf. 2005, 45, 1295–1300.

It has been understood experimentally and theoretically that the friction between the tool rake and chip is reduced or reversed by
the elliptical vibration. It is therefore suggested that an alternative or control experiment with the use of vegetable oil–based coolant,
which has the ability to improve friction between tool rake and chip, be used to study the burr formation under these conditions.

11.12.3 Burr Formation in Drilling Process

Drilling is the most popular in machining, and burrs are formed in every machining process as a result of plastic deformation of the
work material. Burrs are formed when a drill enters and exits the hole (37). In drilling, the burr that forms at the entrance of the hole
can be as a result of tearing: a bending action followed by clean shearing or lateral extrusion. The burr that is formed when a sharp
drill exits the workpiece is a Poisson burr resulting from rubbing at the margins of the drill, and when a normal or worn-out drill
exits the uncut chip rolls, it results in a rollover burr (29). Serious problems in deburring occur on the exit stage when burrs formed
are much larger or when the exit burr is formed inside a cavity or inside a crossing hole, because there are no tools available for
deburring (37,38) and sometimes, deburring is not possible.
Ko et al. (39) examined the effect of drill’s geometry on burr formation. In their study, the need to use a drill with varying
geometry, i.e., step angle and point angle, was emphasized. Two types of drills with a cutting speed of 35 m min1 and five feed rates
at 50, 100, 150, 200, and 250 mm min1 were used for SM45C alloy steel. The two types of drills were conventional carbide drill

Table 5 Specification of drills for burr formation experiment

Drill Point angle, Ɵ1 ( ) Diameter, D1 (mm) Step angle, Ɵ2 ( ) Step diameter, D2 (mm) Step length, L (mm)

Conventional drill 140 10 0 0 0

Step drill 140 10 130 8 2
140 10 75 8 2
140 10 60 8 2
140 10 40 8 2

Reproduced from Ko, S.-L.; Chang, J.-E.; Yang, G.-E. Burr Minimizing Scheme in Drilling. J. Mater. Process. Technol. 2003, 140, 237–242.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Burr Formation in Machining Processes: A Review 289

with a 140 angle and a step drill designed to contain two different cutting edges. Each cutting edge of the step drill had a specific
angle and diameters of Ɵ1 and Ɵ2 and D1 and D2, respectively, and a step distance between edges was L. Table 5 presents the
geometrical specification of the conventional drill with a 140 point angle and the step drills to compare burr formations.
The front cutting edge with 8.0 mm diameter and 140 angle performed the drilling. The step edge with a 75 step angle and 10.0 mm
diameter removed the remaining part that resulted in a 10.0 mm hole. This experiment was conducted without using any coolant.
Figures 8 and 9 show the burr formation classification in drilling process and the burrs formed using two kinds of drill in this study,
respectively. It was observed that the burr formed by the conventional drill had uniform shape (type B burr in Figure 8). The type B
burr was formed using a step drill that first cut through the front edge, which was similar to conventional drilling. The second drill that
cut through the step edges removed the burr formed during the first cutting and produced very small burrs, as shown in Figure 9. The
cap remained with a 250 mm min1 feed rate. This cap produced during the first drilling was attached to the burr formed in the second
drilling. It was equally observed that, in the step drilling, only very tiny burrs that can be easily removed were formed. A laser sensor
was used to measure burr geometry. The average burr heights were represented and compared in Figure 10. The burr height in
conventional drills started at 0.14–0.31 mm and increased with the feed rate, which was larger than those ranging from 0.07 to
0.21 mm in step drills. It was noted that it was possible to effectively use the step drill in minimizing burr formation in this study.
This experiment was carried out without the use of coolant because the authors wanted to have a clear view of the burr formation
process. The use of coolant could still be adopted as there are many types of coolant applications in machining processes. The
application of coolant in this type of experiment could help to understand burr minimization in drilling better.
Dornfeld et al. (41) investigated the effects of tool geometry as well as process conditions on the drill burr formation using
titanium alloy (Ti–6Al–4V) material. Drilling was done with solid carbide tools with and without coolant and high-speed cobalt
drills without coolants. For the dry cutting, two different types of carbide drills were used with the following geometries: (a) two-
flute drill 6.35 mm diameter and 118 point angle and (b) three-flute drill of 6.35 mm diameter, 150 point angle, both with 25
helix angle. The cutting conditions involved were cutting speed, 36.6 and 42.7 m min1, and three levels of feed rate, 0.0254,

Figure 8 Classification of burr formation in drilling. Reproduced from Ko, S.-L.; Chang, J.-E.; Yang, G.-E. Burr Minimizing Scheme in Drilling. J. Mater.
Process. Technol. 2003, 140, 237–242.

Figure 9 Burr formation in each drilling operation for SM45C. Reproduced from Ko, S.-L.; Chang, J.-E.; Yang, G.-E. Burr Minimizing Scheme in
Drilling. J. Mater. Process. Technol. 2003, 140, 237–242.

Figure 10 Burr height in each drilling operation for SM45C. Reproduced from Ko, S.-L.; Chang, J.-E.; Yang, G.-E. Burr Minimizing Scheme in Drilling.
J. Mater. Process. Technol. 2003, 140, 237–242.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
290 Burr Formation in Machining Processes: A Review

Table 6 Tool geometry for wet cutting

Group Point style Point angle ( ) Helix angle ( ) Lip relief angle ( )

1 Split point 135 35 12

2 Split point 130 35 12
3 Split point 118 35 12
4 Split point 135 35 10
5 Split point 135 35 14
6 Split point 135 30 12
7 Helical point 135 30 12

Reproduced from Dornfeld, D. A.; Kim, J. S.; Dechow, H.; Hewson, J.; Chen, L. J. Drilling Burr Formation in Titanium Alloy,
Ti-6Al-4V. CIRP Ann. 1999, 48 (1), 73–76.

0.0508, and 0.0762 mm rev1 were selected for each types of drill, hence a total of 12 cutting conditions (42). For the wet cutting
conditions, two sets of drilling experiments were conducted with coolant. The first experiment was to determine the influence of the
tool geometry on the burr formation, and the second experiment was designed to determine the influence of the cutting conditions
such as feed rate and cutting speed on burr formation. Cobalt high-speed steel drills with diameter 10 mm and various tool
geometries as shown in Table 6 were used.
All the experiments were conducted on Ti–6Al–4V plates of 125 mm  100 mm  6 mm using a CNC-milling machine. The
following observations were made by the authors:
1. Four types of burr formations, uniform burr, lean back burr, roll back burr, and roll back burr with widened exit, were seen in dry
cutting of Ti–6Al–4V.
2. Roll back burr due to thermal effects was observed in dry cutting with relatively high feed rates and cutting speeds. This was
confirmed by comparison with burrs in wet cutting with reduced thermal effects.
3. Ring formation burr was observed in wet cutting and is an intermediate type between plain uniform burr without attachment
and a burr with a drill cap formation.
4. Geometry of the drill greatly affects burr formation; helical point drill produced smaller burrs than split point drill, larger helix
angle and increasing point angle both reduced burr height and thickness.
In this study, an attempt was made to compare the effects of coolant on burr formation during drilling of Ti–6Al–4V alloy. It is
clearly depicted in Figure 11 that the use of coolant can improve the burr formation during any machining processes (43). It
confirmed that the need to take the temperature dependency of material properties into account when explaining burr formation
phenomena is very significant.
Lin and Shyu (44) described an experimental investigation of improvement of tool life and exit burr using variable feeds when
drilling stainless steel with coated drills. The experiment was conducted on a CNC machine using an austenitic stainless steel plates

Figure 11 Burrs observed in drilling Ti–6Al–4V. Reproduced from Dechow, H. Influence of the Tool on Hole Quality when Drilling Ti-6Al-4V Including
the Aspects of Reaming. Diploma-Thesis Study, LMA, University of California, Berkeley.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Burr Formation in Machining Processes: A Review 291

SUS 304 of 150 mm  100 mm  15 mm with four types of twist drills (TiN, TiCN, CrN, and TiALN) coating standard HSS drills.
The diameter of the drill and the cutting speed employed was 8 mm and 25.2 m min1, respectively. The thickness of all the coatings
on the drills was 3 mm. A chuck was used to hold the drills and a water-soluble coolant was supplied at a rate of 5.01 L min1. The
authors observed that the burr formation was severe in the exit zone when drilling stainless steel; they suggested that it may be due
to the high toughness of the stainless steel. Figure 12 indicates the exit burr height for the holes for constant and variable feed
machining with a TiN-coated drill. It can be seen that the maximum burr height occurred with constant feed machining.
The amplitude of variation a ¼ 0.8 represents a lower feed at the exit zone and a higher feed at the middle zone. However, it does
not produce the lower burr height, as seen from this figure. This is due to the increased outer corner wear of the drill at the higher
feed machining in the middle zone. The burr height is smallest when the amplitude of the variation a ¼ 0.6. Figure 13 shows the
burr height versus holes for the four different coated drills. The TiN-coated drill showed the smallest burr height while the

Figure 12 Burr height vs holes for constant and variable feed machining Ti-coated drill. Reproduced from Lin, T.-R.; Shyu, R.-F. Improvement of Tool
Life and Exit Burr Using Variable Feeds when Drilling Stainless Steel with Coated Drills. Int. J. Adv. Manuf. Technol. 2000, 16, 308–313.

Figure 13 Burr height vs holes for variable feed machining when the amplitude of variation a ¼ 0.6. Reproduced from Lin, T.-R.; Shyu, R.-F. Improvement
of Tool Life and Exit Burr Using Variable Feeds when Drilling Stainless Steel with Coated Drills. Int. J. Adv. Manuf. Technol. 2000, 16, 308–313.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
292 Burr Formation in Machining Processes: A Review

TiALN-coated drill exhibited the largest burr height. It was also found that the burr height increases suddenly during the drilling of
the last holes for most of the coated drills.
It was observed that the variable feed machining was superior to constant feed machining with respect to tool life or burr height
and the amplitude of variation of feed a ¼ 0.6 is optimum for maximum tool life or minimum burr height.
The authors employed a water-soluble coolant supplied at a rate of 5.01 L min1; this is the only information about the
application of coolant in this experiment, it was never considered during the analysis of results because a control experiment was
not set up. To determine the effect of coolant in this type of experimental setup, a control experiment is necessary. This will help to
understand the role played by the application of coolant in the setup.

11.12.4 Burr Formation in Milling Process

The type of burr in milling has been described by Chern (21) as highly dependent on the in-plane exit angle. He observed five types
of burrs in milling process: (1) knife-type burr, (2) wave-type burr, (3) curl-type burr, (4) edge breakout burr, and (5) secondary
burr. Olvera and Barow (45) studied the influence of cutting parameters on the formation of burrs in square shoulder face milling.
Exit burr in the cutting direction, exit burr in the feed direction, and burr formed at the top edge were discussed through their
extensive experimental research. Another study of burr formation was done by Lin (46). He conducted a series of single-tooth face
milling tests on stainless steel to study burr formation and tool chipping. He found that the burr formation is closely related to the
chipping depth of tool edge. Chern (47) observed burr formation and edge breakout on the workpiece exit edge in orthogonal
cutting and in face milling of aluminum alloys. Kishimoto et al. (48) conducted face milling experiments in normalized carbon steel
S45C to investigate the burr formation in connection with cutting conditions and tool geometry. In their tests, two types of burrs
were found and named (1) primary burr and (2) secondary burr. The primary burr is the rollover burr produced on the tool exit
edge. The burr thickness was found to vary from minimum burr thickness to maximum burr thickness along the length of the burr.
Chern (21) experimentally examined burr formation mechanism in face milling of aluminum alloys. The experiment involved
the use of three types of aluminum alloys (Al1100, Al2024-T4, and Al6061-T6) with a fly milling cutter (tool bit) made of high-
speed steel. The cutting speed, which had been found to have insignificant influence on burr formation (47,49), was fixed for all the
tests. Tables 7 and 8 show the cutting conditions and tool geometry.
The following were observed: (1) in-plane exit angle strongly influences the geometry of the burrs in face milling and the five
types of burrs were created in the experiments on aluminum alloys; (2) wave-type burr was created when in-plane exit angle
approximates 90 ; wave-type burr will increase the difficulty of deburring due to its complexity of geometric shape and larger
thickness and thus should be avoided; and (3) formation of the secondary burr was dominated by the depth of cut, with some
influence of increase in the feed rate. The value of the critical depth of cut for the secondary burr increased with in-plane exit angle
and fracture strain of the workpiece.
This study involved the burr formation in the face milling process by investigating the influence of cutting conditions on burr
formation in face milling of aluminum alloys. The cutting conditions show that the experiment was conducted without the use of
coolant. It is likely that the application of coolant in this study will alter the results of the experiment, since the temperature of the
material will definitely change under coolant condition.
Heisel et al. (50) examined burr formation in milling with minimum quantity lubrication (MQL). The tests were conducted on
an EX-CELL-O XHC241 machine center using a single channel unit of Lubrix as the MQL system. Ecocut Mikro plus 82 developed

Table 7 Cutting conditions

Cutting speed 2.48 m s1 (fixed)

Depth of cut 0.25–2.00 mm
Feed rate 0.03–0.46 mm per tooth
In-plane exit angle 30–165
Cutting fluid Air

Reproduced from Olvera, O.; Barrow, G. An Experimental Study of Burr Formation in

Square Shoulder Face Milling. Int. J. Mach. Tools Manuf. 1996, 36 (9), 1005–1020.

Table 8 Tool geometry

Geometry Unit

Corner angle 30

Axial relief angle 15
End cutting edge angle 10
Axial rake angle 15
Radial rake angle 0
Nose radius 1 mm

Reproduced from Olvera, O.; Barrow, G. An Experimental Study of Burr Formation in

Square Shoulder Face Milling. Int. J. Mach. Tools Manuf. 1996, 36 (9), 1005–1020.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Burr Formation in Machining Processes: A Review 293

Figure 14 Burr formation. Reproduced from Aramcharoen, A.; Mativenga, P. T.; Yang, S.; Cooke, K. E.; Teer, D. G. Evaluation and Selection of Hard
Coatings for Micro-milling of Hardened Tool Steel. Int. J. Mach. Tools Manuf. 2008, 48, 1578–1584.

especially for MQL machining and based on special fatty alcohols was used as lubricant. A face milling cutter and an angle milling
cutter were used as test tools. The CVD-coated indexable inserts have a layer structure of TiCN þ Al2O3 (þTiN) and a cutting edge
length of b ¼ 1.2 mm. The tool cutting edge angle is k ¼ 90 for both milling cutters. The heat-treatable steel C45E was used as
reference material. The face milling cutter has a diameter of d ¼ 50 mm, a helix angle of d ¼ 12 , and six indexable inserts. It can be
used for a cutting speed of up to vc ¼ 250 m min1 and a depth of cut of ap ¼ 11 mm. The angle milling cutter has a diameter (d) of
25 mm, a helix angle of d ¼ 8 , and three indexable inserts. It can be used for the same parameters as the face milling cutter. The
indexable inserts used were identical for both tools and merely differed in corner radius, which was rS ¼ 0.4, 0.8, and 1.2 mm for the
tests. The cutting speed was vc ¼ 225 m min1 for the comparative tests, and the feed per tooth (fz) was 0.11 mm rev.1. For further
tests, the cutting speed was varied in the range from vc ¼ 150 to 225 m min1, and the feed per tooth was varied in the range between
fz ¼ 0.05 and 0.11 mm. The tests were performed at a constant depth of cut of ap ¼ 3 mm. In addition to that, the width of cut ae was
varied. Concerning the face milling cutter, the milling was conducted in the middle of the workpiece with a width of cut of ae ¼ 12.5,
25, and 37.5 mm. Regarding the angle milling cutter, widths of cut of ae ¼ 6.25, 12.5, 18.75, and 23.5 mm were investigated.
The results obtained show that the burr value increases in the machining with minimum quantity lubrication compared to dry
machining, but does not change when varying the minimum quantity. A variation in cutting speed at constant feed showed no
considerable influence on burr formation. However, when varying the feed per tooth, the exit burr curve of the lateral face shifts
toward higher values in dry machining compared with minimum quantity lubrication. Regarding angle milling cutters, investi-
gations into the influence of corner radius revealed that the burr value increases with growing corner radius. In face milling, it can be
detected that the burr value decreases with increasing corner radius. The supply of the fluid to the cutting region is another
parameter that was varied within the framework of these investigations. In this connection, the supply of the fluid through an
external nozzle proved to be disadvantageous. The burr values of the exit burrs were higher than those of the internal supply and dry
machining.
The results of this experiment show that it is not only the application of coolant that can have an effect on the machining output,
but also the method of applications such as conventional method (flooding), high-pressure coolant, and MQL will in one way or
the other affect the burr formation mechanism in any machining processes.
Aramcharoen et al. (51) evaluated some selected hard coatings for micromilling of hardened tool steels. The experiment
involved the use of hardened H13 tool steel (45HRC) as workpiece material and a cutting tool made from ultrafine tungsten carbide
grain structure with a two-flute flat microend mill. The tools were coated using (CFUBMSIP) technology (52) in TiN, CrN, TiCN,
TiAlN, and CrTiAlN coatings (34) and the evaluation of coating performance carried out at cutting conditions. Cutting parameters
were a spindle speed of 30 000 rpm, depth of cut of 20 mm, maximum undeformed chip thickness of 5 mm, and feed rate of
300 mm min1. After machining, tool wear condition was evaluated using the SEM.
One of the observations made by the authors, which is relevant to this chapter is that, in general, most coatings led to reduced
burr size compared to the uncoated tools, as shown in Figure 14.
Thus both a sharp cutting edge (as reported in the literature) and thin, high-performance tool coatings are essential in reducing
burr size. The use of a coating (with good adhesion) provides some protection from chipping for the cutting edge and also slows
down along cutting edge radius. Otherwise, increased edge radius results in more negative rake angles, which promote a plowing
mechanism for material removal and burr formation. This experiment was conducted without the use of coolant; these results could
be improved with the application of coolants.

11.12.5 Conclusion and Future Research Direction

Types of machining process, the process parameters, tool property, tool geometry, tool edge configuration, coolant, and workpiece
material properties are reported to be responsible for burr formations in machining processes. The influence of the tool lead angle
(Ɵ) on the burr dimension has been established. It was observed that the smaller the lead angle (Ɵ), the increased probability that

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
294 Burr Formation in Machining Processes: A Review

sideward or Poisson burr would be formed during turning of aluminum alloy Al6061-T6 with a K10 grade tungsten carbide–cobalt
alloy tool. The use of an ultrasonic elliptical vibration system during machining of aluminum 52S with a carbide tool was found to
be capable of suppressing burr formation during turning.
The type of drills used in drilling process was found to have an effect on the burr height. For instance, the burr height in
a conventional drill started at 0.14–0.31 mm and increased with feed rate, which was found to be larger than those in the range of
0.07–0.21 mm in step drills while drilling SM45C alloy material. It was equally observed that geometry of the drill greatly affects
burr formation; a helical point drill produced smaller burrs than a split point drill, and larger helix angle and increasing point angle
both reduced burr height and thickness during drilling of Ti–6Al–4V material with cobalt high-speed drills. Burr formation in face
milling process by investigating the influence of cutting conditions on burr formation in face milling of aluminum alloys shows that
the value of the critical depth of cut for the secondary burr increased with in-plane exit angle and fracture strain of the workpiece.
The role of different coolants and method of application on burr formation has not been thoroughly investigated. Since burr
formation is a complicated process and its formation involves large plastic deformation of work material, any factor that affects the
temperature behavior of the material must be investigated. For instance, ring formation burr was observed in wet cutting and
was an intermediate type between plain uniform burr without attachment and a burr with a drill cap formation during drilling of
Ti–6Al–4V material with cobalt high-speed drills. This affirmed the need for research work to be focused on the effect of coolant on
burr formation mechanism.

References

1. Auchi, J. C.; Dornfeld, D.; Arrazola, P. J.; Franke, V.; Leitz, L.; Min, S. Burrs-Analysis, Control and Removal. CIRP Ann. Manuf. Technol. 2009, 58, 519–542.
2. Long, Yu; Guo, Changsheng. Finite Element Modeling of Burr Formation in Orthogonal Cutting. Mach. Sci. Technol. Int. J. 2012, 16 (3), 321–336.
3. Schäfer, F. Deburring; Krausskopfverlag: Mainz, 1975 (in German).
4. International Standard ISO 13715. Technical Drawings – Edges of Undefined Shape – Vocabulary and Indications, 2000.
5. Ko, S. L.; Dornfeld, D. A. A Study on Burr Formation Mechanism. Trans. ASME J. Eng. Mater. Technol. 1991, 113 (1), 75–87.
6. Gillespie, L. K. The Battle of the Burr: New Strategies and New Tricks. Manuf. Eng. 1996, 116 (2), 69–78.
7. Beier, H. M. Handbuch Entgrattechnik: Wegweiser zur Gratminimierung und Gratbeseitigung für Konstruktion und Fertigung; Hanser Verlag, 1999.
8. Pekelharing, A. J. Why and How Does the Chip Curl and Break. CIRP Ann. 1964, 12 (3), 144–147.
9. Gillespie, L. K.; Blotter, P. T. The Formation and Properties of Machining Burrs. Trans. ASME J. Eng. Ind. 1976, 98, 66–74.
10. Gillespie, L. K. Quantitative Approach to vibratory deburring effectiveness. SME Technical paper MRR75–11, 1975.
11. Ko., S. L.; Dornfeld, D. A. Analysis and Modeling of Burr Formation and Breakout in Metal. In Mechanics and Deburring and Surface Finishing Process; Stango, R. J.,
Fitzpatrick, P. R., Eds.; American Society for Mechanical Engineering: New York, 1989; pp 79–91.
12. Leopold, J.; Schmidt, G.; Hoyer, K.; Freitag, A. Modeling and Simulation of Burr Formation – State of the Art and Future Trends. In Proceedings of the 8th International
Workshop on Modeling of Machining Operations, Chemnitz, Germany, 2005; pp 73–83.
13. Gillespie, L. K. Deburring Precision Miniature Parts. Precis. Eng. 1979, 1 (4), 189–198.
14. Nakayama, K.; Arai, M. Burr Formation in Metal Cutting. CIRP Ann. 1987, 36 (1), 33–36.
15. Kishimoto, W.; Miyake, T.; Yamamoto, A.; Yamanaka, K.; Takano, K. Study of Burr Formation in Face Milling. Bull. Jpn. Soc. Precis. Eng. 1981, 15 (1), 51–53.
16. Kim, J.; Min, S.; Dornfeld, D. A. Optimization and Control of Drilling Burr Formation of AISI 304 L and AISI 4118 Based on Drilling Burr Control Charts. Int. J. Mach. Tools Manuf.
2001, 41, 923–936.
17. Min, S.; Dornfeld, D. A.; Kim, J.; Shyu, B. Finite Element Modeling of Burr Formation in Metal Cutting; Consortium on Deburring and Edge Finishing, Laboratory for
Manufacturing and Sustainability, UC: Berkeley, 2007.
18. Pande, S. S.; Relekar, H. P. Investigations on Reducing Burr Formation in Drilling. Int. J. Mach. Tools Des. Res. 1986, 26 (3), 339–348.
19. Stein, J. M.; Dornfeld, D. A. Burr Formation in Drilling Miniature Holes. CIRP Ann. 1997, 46 (1), 63–66.
20. Min, S.; Dornfeld, D. A. Technology Assessment on Current Advanced Research Projects in Burr Formation and Deburring, Report; The Association for Manufacturing
Technology, 2004.
21. Chern, G. L. Experimental Observation and Analysis of Burr Formation Mechanism in Face Milling of Aluminum Alloys. Int. J. Mach. Tools Manuf. 2006, 46, 1517–1525.
22. Park, I. W.; Dornfeld, D. A. A Study on Burr Formation Processes Using the Finite Element Method, Part I. Transaction of ASME. J. Eng. Mater. Technol. 2000, 122, 221–228.
23. Link, R. Gradtbildung und strategien zur gratreduzierung bei der zerspanung mit geometrisch bestimmter schneide. Ph.D. dissertation, RWTH Aachen, Germany, 1992.
24. Kajdas, C. Additives for Metalworking Lubricants - a Review. Lubr. Sci. 1989, 1 (4), 385–409.
25. Belluco, W.; De Chiffre, L. Surface Integrity and Part Accuracy in Reaming and Tapping Stainless Steel with New Vegetable Based Cutting Oils. Tribol. Int. 2002, 35, 865–870.
26. Belluco, W.; De Chiffre, L. Testing of Vegetable Based Cutting Fluids by Hole Making Operations. Lubr. Eng. 2001, 57 (1), 12–16.
27. Belluco, W.; De Chiffre, L. Performance Evaluation of Vegetable – Based Oils in Drilling Austenitic Stainless Steel. J. Mater. Process. Technol. 2004, 148, 171–176.
28. Huseyin, M. Cetin; Ozcelik, Babur; Kuram, Emel; Demirbas, Erhan Evaluation of Vegetable Based Cutting Fluids with Extreme Pressure and Cutting Parameters in Turning of AISI
304L by Taguchi Method. J. Cleaner. Prod. 2011, 19, 2049–2056.
29. Gillespie, L. K. Deburring and Edge Finishing Handbook; Society of Manufacturing Engineers: New York, 1999.
30. Pavel, Radu; Marinescu, Ioan; Deis, Mick; Pillar, Jim Effect of Tool Wear on Surface Finish for a Case of Continuous and Interrupted Hard Turning. J. Mater. Process. Technol.
2005, 170, 341–349.
31. Toropov, Andrey; Ko, Sung-Lim; Kim, Byung-Kwon Experimental Study of Burrs Formed in Feed Direction when Turning Aluminum Alloy Al6061-T6. J. Mach. Tools Manuf.
2005, 45, 1015–1022.
32. Ko, S.-L.; Lee, J.-K. Development of Burr Measurement System Using Laser and Its Application. In Proceedings of the Sixth International Conference Precision Surface
Finishing and Deburring Technology; St. Petersburg, 21–23 Sept 2000. pp 109–123.
33. Ma, Chunxiang; Shamoto, E.; Moriwaki, T.; Zhang, Yonghong; Wang, Lijiang Suppression of Burrs in Turning with Ultrasonic Elliptical Vibration Cutting. Int. J. Mach. Tools
Manuf. 2005, 45, 1295–1300.
34. Shamoto, E.; Moriwaki, T. Elliptical Vibration Cutting. CIRP Ann. 1994, 43, 35–38.
35. Shamoto, E.; Morimoto, Y.; Moriwaki, T. Elliptical Vibration Cutting (The Second Report). J. Jpn. Soc. Precis. Eng. 1999, 65, 411–417.
36. Ma, C.; Shamoto, E.; Moriwaki, T.; Wang, L. Study of Machine Accuracy in Ultrasonic Elliptical Vibration Cutting. Int. J. Mach. Tools Manuf. 2004, 44, 1305–1310.
37. Stein, J. M. Burr Formation in Precision Drilling of Stainless Steel. Ph.D. Dissertation, University of California, Berkeley, 1995.
38. Stein, J. M. Influence of Workpiece Exit Angle on Burr Formation in Drilling Intersection Holes. Trans. NAMRI/SME 1996, 24, 39–44.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Burr Formation in Machining Processes: A Review 295

39. Ko, Sung-Lim; Chang, Jae-Eun; Yang, Gyun-Eui Burr Minimizing Scheme in Drilling. J. Mater. Process. Technol. 2003, 140, 237–242.
40. Ko, S. L.; Lee, J.-K. Analysis on Burr Formation in Drilling with New Concept Drill. J. Mater. Process. Technol. 2001, 113, 392–398.
41. Dornfeld, D. A.; Kim, J. S.; Dechow, H.; Hewson, J.; Chen, L. J. Drilling Burr Formation in Titanium Alloy, Ti-6Al-4V. CIRP Ann. 1999, 48 (1), 73–76.
42. Communication with Chen, L. J.; El-Wardany, T. of United Technology Research Centre.
43. Dechow, H. Influence of the Tool on Hole Quality when Drilling Ti-6Al-4V Including the Aspects of Reaming. Diploma-Thesis Study, LMA, University of California, Berkeley.
44. Lin, T.-R.; Shyu, R.-F. Improvement of Tool Life and Exit Burr Using Variable Feeds when Drilling Stainless Steel with Coated Drills. Int. J. Adv. Manuf. Technol. 2000, 16,
308–313.
45. Olvera, O.; Barrow, G. An Experimental Study of Burr Formation in Square Shoulder Face Milling. Int. J. Mach. Tools Manuf. 1996, 36 (9), 1005–1020.
46 Lin, T. S. Experimental Study of Burr Formation and Tool Chipping in the Face Milling of Stainless Steel. J. Mater. Process. Technol. 2000, 108 (1), 12–20.
47. Chern, G. L. Analysis of Burr Formation and Breakout in Metal Cutting, Ph.D. Thesis, University of California, Berkeley, CA, 1993.
48. Kishimoto, W.; Miyake, T.; Yamamoto, A.; Yamanaka, K.; Takano, K. Study of Burr Formation in Face Milling. Bull. JSPE 1981, 15 (1), 51–52.
49. Chern, G. L.; Dornfeld, D. A. Burr/Breakout Models Development and Experimental Verification. J. Eng. Mater. Technol. 1996, 118, 201–206.
50. Heisel, U.; Schaal, M.; Wolf, G. Burr Formation in Milling with Minimum Quantity Lubrication. J. Prod. Eng. Res. Dev. 2009, 3, 23–30.
51. Aramcharoen, A.; Mativenga, P. T.; Yang, S.; Cooke, K. E.; Teer, D. G. Evaluation and Selection of Hard Coatings for Micro-milling of Hardened Tool Steel. Int. J. Mach. Tools
Manuf. 2008, 48, 1578–1584.
52. Teer, D. G. Magnetron Sputter Ion Plating. U.S. Patent No. 5556519, 17 September, 1996.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
This page intentionally left blank

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 11.13 Heat-Assisted Machining
AKM Nurul Amin, International Islamic University Malaysia, Kuala Lumpur, Malaysia
TL Ginta, Universiti Teknologi, PETRONAS, Malaysia
 2014 Elsevier Ltd. All rights reserved.

11.13.1 Introduction 298

11.13.2 Principles of Hot Machining 299
11.13.3 Brief Overview of PEM 299
11.13.3.1 Background 299
11.13.3.2 Principle 300
11.13.3.3 Advantages of Plasma Arc Hot Machining 300
11.13.4 Brief Overview of Laser-Assisted Machining (LAM) 301
11.13.4.1 Principle of LAM Application in Turning 301
11.13.4.2 Principle of LAM Application in Milling 302
11.13.4.3 Benefits of LAM 302
11.13.5 Effect of Heating Temperature (in the Furnace Heating Method) on the Machinability of Materials 303
11.13.5.1 Experimental Methods 303
11.13.5.2 Effect of Heating Temperature on Chip-Tool Contact Length and Chip Shrinkage 303
11.13.5.3 Effect of Heating Temperature on Cutting Force Components 305
11.13.5.4 Effect of Heating Temperature on Chip Serration 305
11.13.5.5 Effect of Heating Temperature on Tool Wear 306
11.13.5.5.1 Effect of Heating Temperature on Tool Wear during Turning of BT-3-1 307
11.13.6 Induction Heat-Assisted Machining Applied to End Milling 310
11.13.6.1 Method of Heating and Process Parameter Control 310
11.13.6.2 Induction Heating Equipment 310
11.13.6.3 Cutting Force Measurement 310
11.13.6.4 Data Acquisition System for Vibration/Chatter and Temperature Monitoring 310
11.13.6.5 Cutting Tools and Workpiece Materials 311
11.13.6.5.1 Cutting Tools 311
11.13.6.5.2 Workpiece Materials 312
11.13.6.5.3 Temperature Control for Preheating Experiments 312
11.13.7 Benefits of Induction Heat-Assisted End Milling of Stainless Steel AISI 304 313
11.13.7.1 Benefits in Terms of Chatter Suppression in the Frequency Range from 0 to 10 000 Hz 313
11.13.7.1.1 At Cutting Condition ( VC ¼ 90 m s1, DC ¼ 2 mm, and Zt ¼ 0.2 mm per tooth) 313
11.13.7.1.2 Benefits of Cutting Condition ( VC ¼ 150 m s1, DC ¼ 2 mm, and Zt ¼ 0.2 mm per tooth) 313
11.13.7.2 Benefits in Terms of Surface Roughness Reduction 316
11.13.7.2.1 Benefits at Cutting Condition: VC ¼ 90 m s1, DC ¼ 2 mm, and Zt ¼ 0.2 mm per tooth and case IV
(VC ¼ 150 m s1, DC ¼ 2 mm, and Zt ¼ 0.2 mm per tooth) 316
11.13.7.3 Benefits in Terms of Tool Wear Reduction 316
11.13.8 Benefits of Induction Heat-Assisted End Milling of Titanium Alloy – Ti–6Al–4V 318
11.13.8.1 Effect on Vibration/Chatter 318
11.13.8.2 Effect on Cutting Force 319
11.13.8.3 Effect on Tool Wear and Tool Wear Morphology 320
11.13.8.4 Tool Wear Morphology 322
11.13.8.5 Surface Roughness and Surface Integrity 322
11.13.8.6 Surface Integrity and the Underlying Layer 325
11.13.8.6.1 Surface Integrity 325
11.13.8.6.2 The Underlying Layer 326
11.13.8.7 Chips Morphology 327
11.13.8.8 Discussion 327
11.13.8.9 Summary 329
11.13.9 Conclusion 330
References 330

Comprehensive Materials Processing, Volume 11 http://dx.doi.org/10.1016/B978-0-08-096532-1.01118-3 297

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
298 Heat-Assisted Machining

11.13.1 Introduction

Heat-assisted machining or hot machining, as it is more popularly known, has emerged as an alternative method of machining
providing improved machinability of difficult-to-cut metals and alloys. This technique of machining has been under consid-
eration since the late nineteenth century. It was observed that metals tend to deform more easily when heated, thus enhancing
their machinability. Hot machining found widespread application in the manufacture of engineering components in the late
twentieth century, a century after it was first introduced. The principle behind hot machining is to increase the difference in
hardness of the cutting tool and workpiece, leading also to the reduction in the component forces, improvement in surface finish,
and longer tool life.
The manufacturing industry has explored various heating techniques such as electric current, arc, high-frequency induction, laser
beam, electron beam, and plasma jet heating. However, all of these techniques are suitable for some but unsuitable for other specific
operations. For instance, electron beam heating, requiring a vacuum for operation, was found to be expensive for machining
applications, though the process can be successfully applied for cutting thick layer of metal with minimum metal losses and high
quality of the generated surfaces. Laser heating, on the other hand, was found to be suitable for both cutting and machining
applications. Laser is suitable for metals and nonmetals, such as ceramics. Laser can be customized for turning as well as milling,
including micromilling applications because of the high flexibility of the focused beam. But when intended to be applied for bulk
milling, such as end milling with a large-diameter tool intended for rough machining, the laser beam needs to be defocused to
a wider area that requires a high-power laser to heat up the removable layer of material. Furthermore, the heating efficiency of the
laser and the reflectivity of the laser beam are some of the additional problems in laser-assisted machining (LAM). Apart from that,
the high costs of high-powered lasers (for example, a 1.5 kW CO2 laser costs more than $150 000) and the large power consumption
slowed down implementation of LAM (1).
Plasma-enhanced machining (PEM) has emerged as an economical alternative to LAM and has received significant attention in
Germany. Machining experiments performed using PCBN tool inserts in turning ceramics showed that relief face tool wear was
reduced by 40%. PEM was also used in turning extremely hard metals with tungsten carbide inserts. Although PEM is found to
improve the machining performance, there is no applicable protection to avoid the heat effect on the cutting edge. Furthermore, the
notch wear of the tool is another problem associated with PEM (2). PEM has yet another limitation: It is not suitable for milling
applications where the feed rate is low and intense heat would lead to melting of the work material.
Heat retained in the steel rolls after the hot rolling operation was used to achieve low cutting forces and high tool life.
Researchers used the even furnace heating method to study the influence of heating temperature on various criteria of machinability
of work materials. In the early 1980s, Talantov et al. (3) and Amin and Talantov (4) studied the influence of the furnace heating
method of workpiece on the machinability of carbon steel, stainless steel, and titanium alloy. They found substantial reduction in
the tool wear rate and chatter when heating was employed. However, the researchers used the furnace method only to demonstrate
the effect of heating the work material on the machinability of the materials since the furnace heating method is definitely not
recommended for application in production.
Induction heating is a long method that is very efficient in terms of heat efficiency, and the cleanliness of the operation itself is
very conducive for surface-hardening operation of metals. The industry has seen many diversified applications of induction heat-
assisted quenching operation wide included in the manufacturing transfer lines as the final operation on parts after the machining
operations. The short heating time practically leads to no rusting of the part, and thus it does not call for grinding needs. Amin and
Abdelgadir showed the creativity of using induction heating for online heating of a thin layer of the workpiece immediately prior to
the end milling process to improve the machinability of the material (5). Amin and his team conducted many research studies on
heat-assisted machining using the new heating method. Amin and Abdelgadir (6) showed that induction heat-assisted machining
was able to suppress chatter by 98% and to increase tool life severalfold in the case of steel. Amin et al. (7) conducted research on the
influence of heating temperature (using the same heating method) on the performance of circular carbide inserts in the end milling
of carbon steel. They concluded that the online workpiece heating could help in substantially reducing the acceleration amplitude of
chatter, as well as in appreciably lowering average surface roughness and tool wear values. In another work, Amin et al. (8) used
induction heat-assisted machining method to improve the machinability of Ti–6Al–4V alloy. They conducted the research on the
use of online heating during machining of Ti–6Al–4V. This method of heating was found to result in a reduction in amplitude of
chip serration and the formation of thicker chips leading to an increase in chip-tool contact length. Reduction of chatter amplitudes
at resonance frequencies was due to the formation of stable chips and higher damping capacity from the side of the preheated
workpiece. As a result, uniform tool wear with almost 2.5 times the reduction of average flank wear values was achieved. While the
surface roughness was found to vary from 0.5 to 1.3 mm, with an increase in the cutting speed range from 40 to 160 m min1 during
normal room temperature machining, the same for heat-assisted machining was found to be approximately 0.2 mm in the same
cutting speed range. The surface roughness value thus achieved after heat-assisted machining was low enough to suggest the
elimination of the fine grinding and rough polishing operations.
Therefore, it can be concluded that induction heat-assisted machining is a good alternative to consider in vertical milling
operations of difficult-to-cut metals and alloys. However, as mentioned earlier, this method is not suitable for turning opera-
tions, though some researchers have attempted to customize it for turning operations as well to designing the coil to be wrapped
around the workpiece. However, this design is not universal enough to be recommended for all turning operations: The last part
of the workpiece cannot be machined using such an arrangement of the coil because it would be obstructed by the rotation of
the spindle.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Heat-Assisted Machining 299

11.13.2 Principles of Hot Machining

One way to improve the machinability of high-strength materials is through application of heat to the workpiece either immediately
prior to or during machining. Highly intense local input of thermal energy by an appropriate heat source, as indicated in the
previous section, drastically reduces the material strength when heated above some specific temperatures, depending on the
properties of the work material. The softened metal is then easily removed by conventional cutting processes, such as turning and
milling. Successful application of heat-assisted machining requires a localized and controlled heating of the material within the
machining zone directly in front of the tool. A decisive precondition for success of hot machining is the decrease of material strength
at elevated temperatures (9). According to investigations on conventional heating methods, a sufficient energy density on the
workpiece is essential for the efficiency and feasibility of hot machining since otherwise the adequate material plastification could
not be guaranteed due to fast heat conduction in milling (10) and heat convection in the case of turning operation due to the
rotation of the workpiece at high speed. Investigations on the temperature-dependent material behavior revealed a significant loss of
material strength above a certain temperature level for high-strength metal alloys (Figure 1). The application of intense laser
radiation as a thermal tool was found to be one of the suitable heating methods in hot machining (11,12). Induction heating is
another low-cost and clean method that is very suitable for milling operations (13) and arc heating in the case of turning (14).
The advantages of hot (heat-assisted) machining processes are reduced cutting forces, low tool wear, reduced chatter, and
improved surface finish (14). The reduction of cutting and feed forces is associated with low shear yield strength of the work material
at the primary and secondary shear planes. During hot machining, the plasticity of the work material increases, leading to increased
chip-tool contact length, which plays a role in reducing the normal stress acting on the tool. Heating also reduces the intensity of
chip serration, facilitating lower fluctuations of the cutting forces which lowers the dynamic stresses applied to the tool, thereby
facilitating lower tool wear in heat-assisted machining.
Reduction of the intensity of chatter is associated with the increased internal damping of the system due to the increased
plasticity of the work material, which absorbs the energy of the vibrating tool and dampens the vibration. Depending on the
material properties, various degrees of damping are achieved during heat-assisted machining. The machined surface is improved for
two reasons: lower amplitude of chatter and improved flow of the material over the rake and flank surfaces of the tool forming the
two new surfaces. In machining, surface finish is known to be improved as a result of increases in cutting speed since higher cutting
temperatures are generated during machining at higher cutting speeds, facilitating easy shearing and flow of the chip and machined
surfaces as it also occurs in heat-assisted machining.

11.13.3 Brief Overview of PEM

11.13.3.1 Background
Many feasibility studies have been conducted over the past years with the use of intense heat to locally soften the work material prior
to machining in an attempt to improve the machinability of difficult-to-cut materials. PEM is one of the most promising techniques
that has been developed because it can provide the necessary heat intensities required to soften the workpiece, especially in turning,
where high intensity is an essential requirement. Despite the earlier sign of its potentials, the high cost of high-power lasers

Figure 1 Temperature-dependent mechanical properties of high-strength metal alloys. Reproduced from Brecher, C.; Rosen, C. J.; Emonts, M.
Laser-Assisted Milling of Advanced Materials. Phys. Procedia 2010, 5, 259–272.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
300 Heat-Assisted Machining

prevented LAM from being widely adopted by industry. Plasma-assisted machining is found to be an economic alternative to LAM.
PEM can offer comparable heating rates to lasers at much lower cost. For adopting PEM for industrial applications, its characteristics
must be fully understood in order to achieve optimal machining conditions.
The ability to generate and transfer the desired degree of localized heating to the workpiece is critical to the success of PEM.
Workpiece temperature plays an important role in the chip formation during the metal-cutting process as it affects the material
deformation. The large amount of energy generated due to the bulk deformation and friction is almost exclusively converted to
thermal energy, leading to high chip and tool-cutting temperatures. Temperature in the workpiece is especially important when
thermally enhanced machining is used. The effects of externally applied heat sources on the temperature distribution of the
workpiece must be known. Peak temperatures must be known so that thermal damage is prevented or minimized in the workpiece
surface, and the temperature must be known at the cutting point to control the process.

11.13.3.2 Principle
By applying a plasma arc to soften a workpiece zone just in front of the cutting tool, very-high-energy densities are realized and the
heat is confined. A plasma arc consists of a high-velocity, high-temperature stream of ionized gas capable of supporting a high-
current, low-voltage electric arc. A plasma torch produces this phenomenon by having a tungsten electrode centrally placed
within a water-cooled copper nozzle. A gas stream is fed down the annulus between these, the gas being ionized by a high-frequency
discharge between the copper nozzle and the central electrode. This is followed by a low-current pilot arc and then by a high-current
main arc. The arc characteristics and reliability of arc striking are improved with the balanced geometry of the nozzle orifice (15).
Heating in PEM is provided by direct current (dc), transferred arcs that generate thermal or equilibrium plasmas. Thermal arc
plasma generators consist of a thoriated tungsten cathode and a cooled nozzle through which the plasma gas flows (Figure 2). The
nozzle serves as anode when used with nonconducting workpieces, but, with conducting materials, the arc is transferred to the
workpiece, which works as the anode. Fiber-optic radiation thermometers are capable of measuring surface temperatures with
a desired range and target emissivity from up to 1.0, with the probe head providing a suitable target size that can be positioned on
the cutting surface between the plasma nozzle and the cutting tool. With highly localized energy available at low gas-flow rates,
transfer arcs are well suited to PEM of electrically conducting superalloys, with typical plasma peak temperature reaching 16 000 K.
Carl et al. (15), in their experimental evaluation of PEM performed on a PEM system, conducted turning using a plasma heating
system. The plasma torch was fitted with a copper nozzle of 3.18 mm orifice diameter. Thoriated tungsten cathodes with a 20
included angle were used throughout the experiments. Various measurements were performed in-process and offline the perfor-
mance of hot turning, applying the given heating method.

11.13.3.3 Advantages of Plasma Arc Hot Machining

Carl et al. (15) found that the benefits of PEM are demonstrated through the reduction of cutting forces and improved surface
roughness over a wide range of cutting conditions. Madhavulu and Ahmed (16) observed in their work that the major advantages of
the plasma-assisted turning process are increased metal removal rates, lower spindle power requirement, and possibility of
machining hard and tough metals even when fully hardened and heat treated. The results from their tests conducted on stainless
steel indicated that hot machining leads to a 1.8 times gain in metal removal rate and a 1.67 times improvement in tool life. They
observed, however, that, in hot machining the sum of the power required by the machine tool and the plasma heating system is far
larger than the power required for the machine tool in the conventional machining process. But from their cost analysis, they found
that the cost of power was much smaller than the savings in tooling cost and the overall machining cost due to lower lead time and
higher productivity.

Figure 2 Details of a plasma arc (transferred arc) generator. Reproduced from Sun, S.; Harris, J.; Brandt, M. Parametric Investigation of Laser-Assisted
Machining of Commercially Pure Titanium. Adv. Eng. Mater. 2008, 10, 565–572.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Heat-Assisted Machining 301

11.13.4 Brief Overview of Laser-Assisted Machining (LAM)

Laser-assisted processing of material is one of the emerging fields in advanced manufacturing. The advantages that make the lasers
increasingly attractive in industrial production include coherence, focus ability, very-high-power intensity, power-shaping capa-
bility, and ease of automation with in-process sensing (17). It also offers the potential to realize innovative design with high
flexibility, a high processing speed, and good quality in many manufacturing processes. The capital investment may be higher, but
this is offset by the benefits gained in many applications.

11.13.4.1 Principle of LAM Application in Turning

Application of laser beam heating in turning operations is relatively simple due to the stationary nature of the cutting tool (18). The
variables in the laser beam are its position, spot size, incident angle, and tool to beam distance. The laser beam is placed normal to
the workpiece surface in most of the reported work on laser-assisted turning operations (19–23), as indicated in Figure 3. This
arrangement is sometimes preferred since machining is easy and the laser equipment does not come close to the machining area, but
the heating efficiency of the machined surface is poor in this case and is not high enough deep for cut. The other alternative used by
many researchers (24–26) is to place the laser gun incident normal to the cutting (chamfer) surface (as shown in Figure 4). The laser
spot size is required to fully cover the chamfer surface in order to achieve uniform reduction in the cutting forces in the x-, y-, and z-
directions (26). However, even partial coverage of the chamfer surface by the laser beam close to the machined surface can
dramatically reduce tool wear (27).
Shin and colleagues (28,29) used multiple distributed laser units, simultaneously heating both the unmachined surface and the
chamfer surface (Figure 5) to create the desired temperature distribution through the depth of cut in the workpiece, which is

Figure 3 LAM setup with the laser beam normal to the workpiece surface. Reproduced from Wu, J. F.; Guu, Y. B. Laser Assisted Machining Methods and
Device, US Patent 7,002,100 B2, 2004.

Figure 4 The relative position of the laser beam, workpiece, and cutting tool in laser-assisted turning operation: (a) End-view; (b) Side view. Reproduced
from Sun, S.; Harris, J.; Brandt, M. Parametric Investigation of Laser-Assisted Machining of Commercially Pure Titanium. Adv. Eng. Mater. 2008, 10, 565–572.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
302 Heat-Assisted Machining

Figure 5 Laser-assisted turning utilizing two laser beams. Reproduced from Shin, Y.C. Laser Assisted Machining Process with Distributed Lasers.
US Patent 0062920A1, 2007.

reported to result in longer tool life. The position of the laser beam relative to the tool is critical. Tool–beam distance, along with
cutting speed, determines the time interval between the laser heating and machining operation and hence the temperature
distribution at the cutting zone. It is found that the larger reduction in the cutting forces is achieved with the laser spot positioned
closer to the cutting tool when cutting hardened steel (30), commercially pure titanium (26), and high chromium white cast iron
(31). However, if the tool–beam is too close to the tool, machining problems may result (32). Therefore, the tool must be kept at
a minimum distance from the laser beam.

11.13.4.2 Principle of LAM Application in Milling

Milling integration of the laser beam with a milling machine as the tool is rotating is a complicated task. Generally, the beam can be
arranged separately from the tool or integrated with the spindle. For surface milling, the easiest way is to set the beam in front of the
tool in the feed direction as shown in Figure 6. The limited spot size of the external heat source limits the ability to cover the width
of cut by a single spot in most applications. Therefore, the beam can only heat part of the cut width. This could be the middle
(position 1) (33) or tool entry point (position 2), or both entry and exit points (position 2 and position 3) by two beams (34,35).
A high-power laser, multiple beams, or a line beam are required to cover the width of the cutting zone (region 4) (35–38). The
dynamic impact on the cutting tool, as the rotating tool intermittently engages and disengages with the workpiece at the entry and
exit points, respectively, causes significant vibration and ultimately tool fracture during milling operations. As a result, laser beam
heating at the entry point is more significant for longer tool life and a reduction in chatter. Rajagopal et al., (25) experimenting with
a 14 kW CO2 laser on titanium and Inconel 718, showed that beam location is important during LAM. Various coatings were
studied, including India ink, silicon carbide, and potassium silicate, with varying degrees of successful absorptivity enhancement.

11.13.4.3 Benefits of LAM

Komenduri et al. (39) conducted research on LAM of Ti–6Al–4V and Inconel 718 using an Nd–YAG pulse laser, which has the
advantage of providing higher absorptivity for metals due to its shorter wavelength. To overcome the intermittent energy problem,
two different techniques were applied. First, Komenduri used the laser as a primary heat source at a high pulse rate and observed

Figure 6 Illustration of the integration of the beam with the cutting tool for surface milling. Reproduced from Sun, S.; Harris, J.; Brandt, M. Parametric
Investigation of Laser-Assisted Machining of Commercially Pure Titanium. Adv. Eng. Mater. 2008, 10, 565–572.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Heat-Assisted Machining 303

a 49% reduction in cutting forces and chip fragmentation during LAM of Inconel 718. For Ti–6Al–4V forces dropped by 30% in air
and 60% with oxygen assistance. Second, he used the laser to precondition the workpiece before machining and obtained a 28%
reduction in forces for Ti–6Al–4V. However, it lacked any detailed analysis, no improvements in tool wear were reported, and
surface roughness was observed.
Several researchers have investigated advanced ceramics in a thermally assisted turning operation. For example, Copley (1), who
applied a fillister-shaped removal mode to three-dimensional machining of Si3N4, pointed to the relation between the fillister shape
and laser polarization. His results show that the strength of the Si3N4 after machining is about 30% greater than that produced by
conventional diamond grinding. Copley also proposed the application of a slanted laser to heat materials at the same time that
a single-point cutting tool is removing the material by lathe. This method did increase the material removal rate and tool life but was
thought to suit only metal materials. In their search for a new ceramics cutting technique, Uehara and Takeshita (12) used hot
machining to cut mullite and Si3N4 ceramics, measuring the outcomes in terms of cutting force, chip shape, surface roughness, and
tool wear. The first successful use of LAM in machine ceramic material was by Konig and Zaboklicki (40), who obtained continuous
chips for hot-pressed silicon nitride when its temperature exceeded 1200 1C. They applied LAM to both turning and milling ceramics
and composites, and reported low cutting force, small tool wear, and a high material removal rate as characteristics. They found the
surface roughness to be equal to that of a ground surface with Ra less than 0.5 mm. However, by not examining material-removal
mechanisms and subsurface damage, this experiment neglected to collect information essential to the LAM process, namely, the effects
of laser energy on workpiece temperature and the impact of temperature on tool wear. Chryssolouris (41) tested the application of
LAM to general metals, heating the workpiece surface with a continuous laser beam before removing the material with a cutting tool.
This experiment found a reduction in tool wear, with a cost reduction of 60–80% over the conventional grinding method.

11.13.5 Effect of Heating Temperature (in the Furnace Heating Method) on the Machinability of Materials
11.13.5.1 Experimental Methods
Furnace heating is not a practical heating method. Nevertheless, Amin (42) used this method in his study to investigate the influence
of workpiece heating on chip serration, coefficient of chip shrinkage, chip-tool contact length, cutting force, and tool wear in
turning. An induction furnace was used for the purpose. First, the workpiece was heated to a temperature approximately 50 C
above the required temperature to compensate for the loss of heat during the transfer of the workpiece from the furnace to the
machine and clamping. Second, an auxiliary cutting tool, moving ahead of the main tool by 2–3 mm, was employed to remove the
cooled outer skin (1–1.5 mm) of the workpiece, as shown in Figure 7. The inner temperature of the workpiece was measured using
a thermocouple at the start and end of the experiment. The variation in the workpiece temperature was controlled within 50  C of
the designed temperatures. Depth of cut and feed were maintained at 2 mm and 0.467 mm per rotation, respectively. Investigations
were conducted for heat-resistant steel IJ 481 (Russian Grade) and titanium alloy–BT-3-1 (Russian Grade).
In order to study the frozen chip, the drop tool apparatus shown in Figure 8 was used. Optical microscopes were used to study
the chip-tool samples and tool wear, and a 3-axis dynamometer was used for force measurement.

11.13.5.2 Effect of Heating Temperature on Chip-Tool Contact Length and Chip Shrinkage
The effects of heating temperature on the chip shrinkage coefficient and morphology of chip were studied. It was observed that with
the increase in the heating temperature the coefficient of chip shrinkage and the chip-tool contact length increase at the beginning
up to a particular temperature, specific to the optimum temperature of preheating, and then decline with further increase in

Figure 7 Arrangement of cutting after preheating in furnace. Reproduced from Amin, A. K. M. N. Investigation of the Laws Governing the Formation
of Chatter during Metal Cutting Processes and their Influence on Tool Wear. Ph.D. Thesis (in Russian), Volgograd Polytechnic Institute, Volgograd,
Russia, 1982.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
304 Heat-Assisted Machining

Figure 8 Drop tool apparatus for lathe application. Reproduced from Amin, A. K. M. N. Investigation of the Laws Governing the Formation of Chatter
during Metal Cutting Processes and their Influence on Tool Wear. Ph.D. Thesis (in Russian), Volgograd Polytechnic Institute, Volgograd, Russia, 1982.

temperature (Figure 9). The optimum temperatures with the two values at the maximum were found to be 570 K (300  C) and
770 K (500  C) for heat-resistant steel IJ 481 and titanium alloy–BT-3-1, respectively (Figure 9). The increase in chip-tool contact
length and coefficient of chip shrinkage is related to the increase in the ductility of the work material. However, as the heating
temperature is further increased (beyond the optimum temperature), the temperature at the chip-tool interface increases, which
leads to lower yield strength of the material at the flow zone. As a result, lower amounts of work are done at the rake face of the tool
due to chip movement, resulting in a shift of the shear plane angle to higher values according to the minimum energy theory of
metal cutting (43). At higher values of the shear angle, the chip starts to become thinner and the chip-tool contact length is reduced.

Figure 9 Dependence of chip-tool contact length, c and coefficient of chip shrinkage, x on workpiece heating temperature. Work materials: Heat-
resistant steel IJ 481 (Russian Grade) and titanium alloy–BT-3-1 – (Russian Grade). Tool material: BK 8 (uncoated WC–Co tool with 8% carbon).
Reproduced from Amin, A. K. M. N. Investigation of the Laws Governing the Formation of Chatter during Metal Cutting Processes and their Influence on
Tool Wear. Ph.D. Thesis (in Russian), Volgograd Polytechnic Institute, Volgograd, Russia, 1982.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Heat-Assisted Machining 305

11.13.5.3 Effect of Heating Temperature on Cutting Force Components

The effect of workpiece heating temperature on the horizontal components of cutting force was investigated for both of the work
materials. Figures 10(a)–10(c) shows the plot of the upper and lower levels of in-force components (Pz, Py and Px) versus heating
temperature, qheat for titanium alloy BT-6, and Figures 11(a)–11(c) shows the same thing for heat-resistant steel IJ 481. As shown
in Figures 10(b), 10(c), 11(b), and 11(c), the upper and lower levels of the horizontal components of cutting force Py and Px attain
their maximum values at a particular temperature (570 K and 770 K for steel IJ 481 and titanium alloy–BT-3-1, respectively),
specific to the material. The same phenomenon was observed in the cases of chip-tool contact length and coefficient of chip
shrinkage. However, as also observed in Figures 10(a) and 11(a), in the case of the vertical (tangential) component of cutting force,
Fz, a similar increase in cutting force at the specific temperature was not observed. On the contrary, a continuous decrease in the
value of this component of cutting force with the increase in temperature was recorded.
It was further observed that the amplitude of vibration of cutting force (distance between the upper and lower boundaries of the
cutting force curves) is very low at the optimum preheating temperature. The relatively high value of the horizontal cutting force
components at the optimum temperature is related to the increased length of contact of the chip with the tool and the value of the
coefficient of chip shrinkage, as discussed in the previous section.

11.13.5.4 Effect of Heating Temperature on Chip Serration

The chip root specimens were studied for room temperature and heat-assisted machining conditions. Two sample chip roots
specimens for BT-3 are shown in Figures 10(a) and 10(b). Under the room temperature condition, the titanium alloy chips are

Figure 10 Dependence of horizontal components cutting force Pz, Py, and Px on workpiece heating temperature. Work materials: titanium alloy–BT-3-1 –
(Russian Grade). Tool material: BK 8 (uncoated WC–Co tool with 8% carbon). Reproduced from Amin, A. K. M. N. Investigation of the Laws Governing the
Formation of Chatter during Metal Cutting Processes and their Influence on Tool Wear. Ph.D. Thesis (in Russian), Volgograd Polytechnic Institute,
Volgograd, Russia, 1982.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
306 Heat-Assisted Machining

Figure 11 Dependence of horizontal components cutting force Pz, Py, and Px on workpiece heating temperature. Work materials: Heat-resistant steel
IJ 481 (Russian Grade). Tool material: BK 8 (uncoated WC–Co tool with 8% carbon). Reproduced from Amin, A. K. M. N. Investigation of the
Laws Governing the Formation of Chatter during Metal Cutting Processes and their Influence on Tool Wear. Ph.D. Thesis (in Russian), Volgograd
Polytechnic Institute, Volgograd, Russia, 1982.

highly serrated. But at the critical heating temperature (Figures 7, 8(b), and 8(c)), the chip serration is greatly reduced. Chip
serration amplitude was greatly reduced, and almost continuous chips were formed at the critical temperature in the case of tita-
nium alloy–BT-3-1, though titanium chips are entirely serrated in machining under room temperature condition (Figure 12(a)).
However, in the case of heat-resistant steel IJ 481, chip serration could be entirely eliminated (42).

11.13.5.5 Effect of Heating Temperature on Tool Wear

The effect of heating temperature on average flank wear is shown in Figures 13(a) and 13(b) for titanium alloy–BT-3-1 and heat-
resistant steel IJ 481, respectively. These two figures show that flank wear is lowest at the corresponding critical heating
temperatures for the two materials (770 K for BT-3-1 and 570 for IJ 481). Hence the critical temperature can be termed the
optimum heating temperature.
Reduction of average tool wear at this temperature is related to lower stresses of the tool as a result of increased chip-tool
contact length and less intense chatter due to the formation of more stable chips. In addition, the initial wear is minimal at
the optimum heating temperature for both materials due to the effect of lower dynamic stresses on the tool and the lower yield
strength on the work material at the higher heating temperatures. However, at the temperature above the optimum value, initial
wear along with the average value of the flank wear increases due to higher rates of diffusion wear at the elevated temperature.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Heat-Assisted Machining 307

Figure 12 Influence of workpiece heating on chip serration. Work materials: titanium alloy–BT-3-1 – (Russian Grade). Tool material: BK 8
(uncoated WC–Co tool with 8% carbon). Reproduced from Amin, A. K. M. N. Investigation of the Laws Governing the Formation of Chatter during
Metal Cutting Processes and their Influence on Tool Wear. Ph.D. Thesis (in Russian), Volgograd Polytechnic Institute, Volgograd, Russia, 1982.

The reduction in flank wear observed after equal amounts of machining time is approximately five times in the case of BT-3-1 and
4.5 times in the case of IJ 481.

11.13.5.5.1 Effect of Heating Temperature on Tool Wear during Turning of BT-3-1

Views of the tool rake and flank faces after 360 s of machining under dry room temperature conditions, at the optimum cutting
temperature and a higher temperature, are shown in Figures 8, 12, 13, and 14 for titanium alloy–BT-3-1. Figures 8–12 shows that
under room temperature conditions, the wear intensity is high and failure of the tool is caused mainly by brittle failure modes. It
may be related to high chip serration amplitude and dynamic loading on a small chip-tool contact area, specific for titanium
alloys. Figures 14(a) and 14(b) demonstrated that the contact length in room temperature machining was only 0.5 mm
(approximately).
At the optimum temperature, the tool wear was very nominal at both the flank and the rake faces (Figure 15). The contact length
is approximately doubled compared to that under room temperature conditions. The low-wear intensity and uniform wear of the
rake and flank surfaces of the tool were due to the beneficial role of heat-assisted machining in terms of lower chatter amplitude,
lower dynamic loading on the tool, and increased chip-tool contact length.
At the 1070 K (z800  C) temperature, tool wear was very smooth at both the flank and rake faces (Figure 16) due to the
diffusion mechanism of wear. However, the wear intensity was much higher and the chip-tool contact length much smaller
compared to that at optimum temperature conditions. This relatively low and uniform wear compared to that at room temperature
condition is again due to the beneficial role of heat-assisted machining in terms of lower chatter amplitude and lower dynamic
loading on the tool.
It may therefore be concluded that there is a particular heating temperature for a given work material, the so-called
optimum heating temperature, which when applied during machining improves the overall machinability of the work
material, in terms of cutting force, tool wear/tool life, and chatter. Heating above the optimum temperature leads to higher
tool wear/lower tool life due to excessive heat generation and higher rate of diffusion wear and superficial plastic deformation
mechanisms.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
308 Heat-Assisted Machining

Figure 13 Dependence of average tool flank wear on cutting time for different workpiece heating temperatures: (a) For titanium alloy–BT-3-1; (b) Heat-
resistant steel IJ 481. Tool material: BK 8 (WC–Co tool with 8% carbon). Reproduced from Amin, A. K. M. N. Investigation of the Laws Governing the
Formation of Chatter during Metal Cutting Processes and their Influence on Tool Wear. Ph.D. Thesis (in Russian), Volgograd Polytechnic Institute,
Volgograd, Russia, 1982.

Figure 14 Views of the rake face (a) and the flank surface (b) of the tool for titanium alloy–BT-3-1 – (Russian Grade) after 360 s of machining under
dry room temperature condition. Tool material: BK 8 (WC–Co tool with 8% carbon). Reproduced from Amin, A. K. M. N. Investigation of the Laws
Governing the Formation of Chatter during Metal Cutting Processes and their Influence on Tool Wear. Ph.D. Thesis (in Russian), Volgograd Polytechnic
Institute, Volgograd, Russia, 1982.
www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Heat-Assisted Machining 309

Figure 15 Views of the rake face (a) and the flank surface (b) of the tool for titanium alloy–BT-3-1 – (Russian Grade) after 360 s of machining under heat-
assisted condition at the optimum heating temperature of 770 K. Tool material: BK 8 (WC–Co tool with 8% carbon). Reproduced from Amin, A. K. M. N.
Investigation of the Laws Governing the Formation of Chatter during Metal Cutting Processes and their Influence on Tool Wear. Ph.D. Thesis
(in Russian), Volgograd Polytechnic Institute, Volgograd, Russia, 1982.

Figure 16 Views of the rake face (a) and the flank surface (b) of the tool for titanium alloy–BT-3-1 – (Russian Grade) after 360 s of machining under heat
assisted condition at the highest applied heating temperature of 1070 K. Tool material: BK 8 (WC–Co tool with 8% carbon). Reproduced from
Amin, A. K. M. N. Investigation of the Laws Governing the Formation of Chatter during Metal Cutting Processes and their Influence on Tool Wear.
Ph.D. Thesis (in Russian), Volgograd Polytechnic Institute, Volgograd, Russia, 1982.
www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
310 Heat-Assisted Machining

11.13.6 Induction Heat-Assisted Machining Applied to End Milling

11.13.6.1 Method of Heating and Process Parameter Control
End milling tests were conducted on a Vertical Machining Center (VMC ZPS, Model: MCFV 1060 LR). Figure 17 shows a block
diagram of the experimental setup.

11.13.6.2 Induction Heating Equipment

Induction heaters provide alternating electric current to an electric coil (the induction coil), as a result of which the induction coil
becomes the electrical (heat) source that induces a high-frequency alternating electrical current into the workpiece to be heated. No
contact is required between the workpiece and the induction coil acting as the heat source, and the heat is restricted to localized
areas or surface zones immediately adjacent to the coil. This happens because the alternating current (ac) in an induction coil has an
invisible force field (or magnetic flux) around it. When the coil is placed next to the tool approximately 5 mm above the workpiece
surface, the lines of force concentrate in the air gap between the coil and the workpiece. The induction coil actually functions as
a primary transformer, with the workpiece to be heated becoming the secondary transformer. The force field surrounding the
induction coil induces an equal and opposing alternating electric current in the workpiece, with the workpiece then heating due to
the resistance to the flow of this induced high-frequency alternating electric current. The rate of heating the workpiece is dependent
on the frequency and intensity of the induced current, the specific heat of the material, the magnetic permeability of the material,
and the resistance of the material to the flow of current. The induced currents are sometimes referred to as eddy currents, with the
highest intensity current being produced within the area of the intense magnetic fields.
To heat the workpiece, a Portable Transistor Induction Heating Machine SP-25AB (25 kW capacity) and GP-35AB (25 kW capacity)
were used for steel and titanium alloy Ti–6Al–4V, respectively. In the present study, a portable high-frequency inducting heating
equipment was used for preheating the work material just prior to machining with the heating coil placed ahead of the cutting tool.
This high-frequency current with alternative polarity generates eddy currents in the surface layer of the workpiece to heat up the layer.
This process has numerous advantages over other heating methods since the generated electric current is simpler to control than
other processes. The heating system consists of three major components (Figure 18): high frequency transformer (Invertors),
Matching Box (Transformer and Condenser), and cooling unit (designed for industrial use).

11.13.6.3 Cutting Force Measurement

The Kistler Rotating Cutting Force Dynamometer was used for measuring cutting forces. The complete measuring system consists of
rotor (type 9125A), stator (type 5235), connecting cable (type 1500A37), and signal conditioner (5267A1/A2). The setup for the
cutting force measurement is presented in Figure 19. The computer software used is Kistler DynoWare (type: 2825D1-2, version
2.31), which is a universal and operator-friendly software. The instrument can measure two components of cutting force: thrust
force for along the z-axis (Fz) in N and cutting torque along the z-axis (Mz) in Nm.

11.13.6.4 Data Acquisition System for Vibration/Chatter and Temperature Monitoring

The data acquisition system includes vibration sensors, thermocouples, and a transducer. The transducer was connected to the
computer via an interface card. Figure 15 shows a block diagram of the experimental setup. A photograph of the data acquisition

Figure 17 Experimental setup. Reproduced from Turnad, L. G. Improvement of Machinability of Titanium Alloy Ti-6Al-4V through Workpiece
Preheating. Ph.D. Thesis, Manufacturing and Materials Engineering Department, IIUM, Malaysia, 2009.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Heat-Assisted Machining 311

Figure 18 Portable transistor induction heating machine GP-30AB. (a) High-frequency transformer (Invertors); (b) Matching box (transformer and
condenser); (c) Cooling unit (specially designed). Reproduced from Turnad, L. G. Improvement of Machinability of Titanium Alloy Ti-6Al-4V through
Workpiece Preheating. Ph.D. Thesis, Manufacturing and Materials Engineering Department, IIUM, Malaysia, 2009.

Figure 19 Cutting force measurement, vibration sensor, and coil induction heating setup 9. Reproduced from Turnad, L. G. Improvement of
Machinability of Titanium Alloy Ti-6Al-4V through Workpiece Preheating. Ph.D. Thesis, Manufacturing and Materials Engineering Department, IIUM,
Malaysia, 2009.

system is shown in Figure 20. As with any data acquisition system, the hardware of the system needs to be programmed by software.
A customized version of LABVIEW software called DASYLab5.6 was used. DASYLab5.6 stands for Data Acquisition System Labo-
ratory. In the data acquisition program the data are sent as data blocks via connections (data channels) between the single modules,
so DASYLab works in a block-dependent mode. Each module output has a 64-kbyte buffer to hold the processed data.
The sensor attached to the vibrating system sends the signal to the computer through the signal conditioning module and the
interface card. Throughout this research, DASYLab5.6 software was used to collect, simulate, and analyze the data (vibration signal
and temperature).

11.13.6.5 Cutting Tools and Workpiece Materials

11.13.6.5.1 Cutting Tools
The inserts chosen for this study were Sandvik uncoated tungsten carbide-cobalt (WC–Co), inserts of the shape shown in Figure 21.
The shape and geometry of the cutting tools are illustrated in Table 1. The inserts were fitted on a 20-mm-diameter Sandvik End
mills tool holder.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
312 Heat-Assisted Machining

Figure 20 Data acquisition system. Reproduced from Amin, A. K. M. N.; Abdelgadir, M. The Effect of Preheating of Work Material on Improved
Machinability of Materials. In Proceedings of the International Conference AIMTDR, Ranchi, India, 2002.

Figure 21 Insert shape and geometry. Reproduced from Main Catalogue, Sandvik Coromat, 2006.

Table 1 Insert geometry values

L iW d1 s bS rε ano

11 6.8 2.8 3.59 1.2 0.8 21

Reproduced from Main Catalogue, Sandvik Coromat, 2006.

11.13.6.5.2 Workpiece Materials

Results of improvement in the machinability of two different work materials due to the application of induction heat-assisted
machining are presented in this section. The materials are stainless steel AISI 304 and titanium alloy Ti–6Al–4V. The composition of
stainless steel AISI 304 is: 0.08 C, 19 Cr, 9 Ni, and 2.0 Mn. The other material, Ti–6Al–4V, has a þ b phases, and its composition is:
5.5–6.76 Al, 3.5–4.5 V, balance Ti.

11.13.6.5.3 Temperature Control for Preheating Experiments

In the induction heating system used during experiments, there was no online measurement of surface temperature during the
machining process. In order to control the surface temperature, it was imperative to set the relationship of current input with
temperature for different feed rates (traveling speed of the table). Preliminary experiments were performed to establish such
a relationship. Figure 22 presents the experimental setup for determining the current–temperature–feed relationship.
The distance between the coil and workpiece was kept constant at 5 mm and was also maintained throughout all the
experimental work. The sensing rod of a thermocouple was attached to the workpiece surface. The input current was set to

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Heat-Assisted Machining 313

Figure 22 Experimental setup for determining the current–temperature–feed relationships. Reproduced from Turnad, L. G. Improvement of
Machinability of Titanium Alloy Ti-6Al-4V through Workpiece Preheating. Ph.D. Thesis, Manufacturing and Materials Engineering Department, IIUM,
Malaysia, 2009.

a particular value to produce a desired level of heating temperature on the workpiece surface. The workpiece was then passed
with a specific feed rate, so that the sensing rod of thermocouple remained just below the coil. The surface temperature
reading was taken, when the coil just passed the sensing rod of the thermocouple. The same trial was repeated two to three
times for every experiment, and the average temperature was considered. Thus, for a particular current and feed value, the
surface temperature was determined. The results of temperature control experiments for the feed are developed in heat-
assisted machining. By using the regression method, characteristic equations of current as a function of temperature were
developed at the 95% confidence interval. Those equations were then utilized to set the current value for a particular heating
temperature. The input currents were set to reach the desired level of surface temperature during the preheated machining
experiment.

11.13.7 Benefits of Induction Heat-Assisted End Milling of Stainless Steel AISI 304

These experiments were conducted using an induction heating system of relatively low-power capacity (25 KVA) for machining
applications on work materials requiring low heating temperature. As discussed in the previous section, the heating temperature was
maintained within 25  C at various table feed rates.
The influence of heating on three main aspects of machining, namely, chatter, surface roughness, and tool wear, were considered
for two different cutting speeds (90 m min1 and 150 m min1) at a constant depth of cut of 2 mm and two feed values of 0.1 mm
per tooth and 0.2 mm per tooth. Sample data for two conditions with the higher values of feed and depth of cut (90 m min1 and
150 m min1, and DC ¼ 2 mm and Zt ¼ 0.2 mm per tooth) are presented below.

11.13.7.1 Benefits in Terms of Chatter Suppression in the Frequency Range from 0 to 10 000 Hz
11.13.7.1.1 At Cutting Condition ( VC ¼ 90 m s1, DC ¼ 2 mm, and Zt ¼ 0.2 mm per tooth)
Figure 23(a) and 23(b) shows the acceleration plots for the amplitudes when machining at room temperature and under heat-
assisted machining conditions, respectively, in the frequency range of 0–5000, and Figure 24 summarizes the reduction in
magnitude of the acceleration amplitudes for this condition.
The experiments found that an almost free vibration free-state was obtained since a reduction up to 98% was observed in the
entire frequency range.

11.13.7.1.2 Benefits of Cutting Condition ( VC ¼ 150 m s1, DC ¼ 2 mm, and Zt ¼ 0.2 mm per tooth)
Figures 25(a) and 25(b) shows the acceleration plots for the amplitudes when machining at room temperature and under heat-
assisted machining conditions, respectively, in the frequency range of 0–5000, and Figure 26 compares the maximum acceleration
amplitude peaks at different frequency intervals in this range under room temperature and heat-assisted machining of stainless steel
at 275  C.
Based on the results of these the experiments, almost free vibration free state was obtained since a reduction up to 98% was
observed in the entire frequency range. Vibration results obtained under room temperature and heat-assisted machining in all the
frequency ranges in the entire range of investigated parameters are shown in Table 2. As can be seen, the chatter amplitude in these
cases is reduced up to almost 10 times.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
314 Heat-Assisted Machining

Figure 23 FFT plot for acceleration amplitudes under: (a) Room temperature condition; (b) Heat-assisted condition at 300  C. * Conversion factor
1  103 V ¼ 9.80665 m s2. Reproduced from Amin, A. K. M. N.; Abdelgadir, M. The Effect of Preheating of Work Material on Improved Machinability of
Materials. In Proceedings of the International Conference AIMTDR, Ranchi, India, 2002.

Impact of Heat-Assisted Machining on Chatter

Acceleration Amplitude, m2 xE-2

Acceleration Amplitude
6
RT HA 4.81
5
4
3
2 1.37 1.18 1.18 1.47
1 0.42 0.15
0.13 0.17 0.12
0
50–100 101–500 501–1000 1001–2000 2001–5000
Frequency Ranges, Hz

Figure 24 Comparison of maximum acceleration amplitude peaks at different frequency intervals in the range 0–5000 Hz under room temperature and
heat-assisted machining of stainless steel at 300  C.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Heat-Assisted Machining 315

Figure 25 FFT plot for acceleration amplitudes under: (a) Room temperature condition; (b) Under heat-assisted condition at 265  C. * Conversion
factor 1  103 V ¼ 9.80665 m s2. Reproduced from Amin, A. K. M. N.; Abdelgadir, M. The Effect of Preheating of Work Material on Improved
Machinability of Materials. In Proceedings of the International Conference AIMTDR, Ranchi, India, 2002.

Impact of Heat-Assisted Machining on Chatter

Acceleration Amplitude, m2 xE-2

Acceleration Amplitude
18 16.48
16
RT HA
14
12
10
8
5.59 6.18
6
3.24
4
1.47 0.93
2 0.07 0.53 0.33
0.13
0
50–100 101–500 501–1000 1001–2000 2001–5000
Frequency Ranges, Hz

Figure 26 Comparison of maximum acceleration amplitude peaks at different frequency intervals in the range 0–5000 Hz under room temperature
and heat-assisted machining of stainless steel at 265  C.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
316 Heat-Assisted Machining

Table 2 Vibration results obtained under room temperature and heat-assisted machining in all the frequency ranges in the entire
range of investigated parameters in end milling of stainless steel

Frequency range
Cutting Feed Depth of
speed m s1 per tooth cut (mm) Temp. (  C) 50–100 100–500 500–1000 1000–2000 2000–5000

90 0.1 1 R.T. Frequency 0.03923 0.17652 0.08826 0.10787 0.20594

90 369.1 Amplitudes (m s2) 0.00883 0.01961 0.02059 0.01765 0.05492
90 2 R.T. 0.03923 0.15691 0.07845 0.12749 0.29420
90 369.1 0.00588 0.02059 0.01275 0.01275 0.07747
90 0.2 1 R.T. 0.05884 0.22555 0.25497 0.07845 0.19613
90 300.6 0.00785 0.02844 0.00883 0.00883 0.01177
90 2 R.T. 0.13729 0.11768 0.11768 0.14710 0.48053
90 300.6 0.01275 0.01667 0.04217 0.01177 0.04511
120 1 R.T. 0.01275 0.01667 0.04217 0.01177 0.04511
120 325.6 0.00686 0.04805 0.01765 0.01079 0.04511
120 0.1 2 R.T. 0.03923 0.16671 0.08826 0.10787 0.14710
120 325.6 0.00686 0.08041 0.02157 0.00883 0.01667
120 0.2 1 R.T. 0.05884 0.21575 0.11768 0.10787 0.60801
120 290.6 0.00588 0.01275 0.01569 0.00588 0.00588
120 2 R.T. 0.11768 0.12749 0.19613 0.22555 0.86299
120 290.6 0.00588 0.06963 0.02550 0.01667 0.02648
150 0.1 1 R.T. 0.09807 0.08826 0.15691 0.12749 1.12776
150 309.3 0.00686 0.04707 0.03334 0.00981 0.01667
150 2 R.T. 0.04903 0.12749 0.08826 0.10787 0.61782
150 309.3 0.01471 0.04609 0.03236 0.05884 0.03923
150 1 R.T. 0.09807 0.57859 0.30401 0.42169 0.59821
150 274.5 0.00686 0.00981 0.02059 0.00785 0.01569
150 0.2 2 R.T. 0.14710 0.55898 0.61782 0.32362 1.64752
150 274.5 0.00686 0.09316 0.05296 0.01275 0.03334
Reproduced from Amin, A. K. M. N.; Abdelgadir, M. The Effect of Preheating of Work Material on Improved Machinability of Materials. In Proceedings of the
International Conference AIMTDR, Ranchi, India, 2002.

11.13.7.2 Benefits in Terms of Surface Roughness Reduction

11.13.7.2.1 Benefits at Cutting Condition: VC ¼ 90 m s1, DC ¼ 2 mm, and Zt ¼ 0.2 mm per tooth and case IV
(VC ¼ 150 m s1, DC ¼ 2 mm, and Zt ¼ 0.2 mm per tooth)
Table 3 summarizes the values of the Ra, Ry, and Rz for both machining conditions. As is evident from Table xx, surface
roughness – Ra, Ry, and Rz are lowered by 28%, 23%, and 21%, respectively, in heat-assisted machining compared to room
temperature conventional machining at 90 m min1 at the given feed and depth of cut. The reductions in the Ra, Ry, and Rz are
by 32%, 48%, and 58%, respectively, at 90 m min1 at the given feed and depth of cut due to the introduction of heat-assisted
machining.
The surface roughness results obtained under room temperature and heat-assisted machining in the entire range of investigated
cutting parameters are shown in Table 4. It can be observed that the surface roughness values of Ra, Ry, and Rz are reduced
substantially in all the cases.

11.13.7.3 Benefits in Terms of Tool Wear Reduction

Figures 27(a) and 27(b) shows the flank wear after room temperature and heat-assisted machining, respectively, at 90 m min1,
and Figures 28(a) and 28(b) shows the flank wear after room temperature and heat-assisted machining, respectively, at
150 m min1. It is clear that there is substantial reduction in flank wear due to the introduction of heat-assisted machining at these
two cutting speeds at the given feed and depth of cut.

Table 3 Comparison of Ra, Ry, and Rz values for the two machining conditions at 90 m min1 and 150 m min1

Property Normal machining Heat-assisted machining (%) reduction at Normal machining Heat-assisted machining (%) reduction at
measured at V ¼ 90 m min1 at V ¼ 90 m min1 V ¼ 90 m min1 at V ¼ 150 m min1 at V ¼ 150 m min1 V ¼ 150 m min1

Ra (mm) 1.84 1.32 28 1.74 1.19 32

Ry (mm) 7.5 5.8 23 9.4 4.9 48
Rz (mm) 6.3 5.0 21 9.2 3.9 58
Reproduced from Amin, A. K. M. N.; Abdelgadir, M. The Effect of Preheating of Work Material on Improved Machinability of Materials. In Proceedings of the International
Conference AIMTDR, Ranchi, India, 2002.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Heat-Assisted Machining 317

Table 4 Surface roughness results obtained under room temperature and heat-assisted machining in the entire range of investigated cutting
parameters in the end milling of stainless steel

Percent reduction
1
Cutting speed (m min ) Feed per tooth (mm) Depth of cut (mm) Temp. (mm) Ra Ry Rz Ra Ry Rz

90 0.1 1 R.T. 0.68 2.70 2.50 42.65 14.81 24.00

369.1 0.39 2.30 1.90
2 R.T. 0.89 4.40 3.50 11.24 18.18 17.14
369.1 0.79 3.60 2.90
0.2 1 R.T. 1.35 6.80 6.20 17.04 23.53 17.74
300.6 1.12 5.20 5.10
2 R.T. 1.84 7.50 6.30 28.26 22.67 20.63
300.6 1.32 5.80 5.00
120 0.1 1 R.T. 0.43 2.50 2.00 25.58 24.00 25.00
325.6 0.32 1.90 1.50
2 R.T. 0.88 3.60 3.00 26.14 5.88 6.67
325.6 0.65 3.40 2.80
0.2 1 R.T. 1.12 5.00 4.20 16.96 18.00 9.52
290.6 0.93 4.10 3.80
2 R.T. 1.72 9.50 8.90 22.67 38.95 46.07
290.6 1.33 5.80 4.80
150 0.1 1 R.T. 1.10 5.00 3.10 39.09 24.00 22.58
309.3 0.67 3.80 2.40
2 R.T. 0.96 3.60 3.00 18.75 5.56 6.67
309.3 0.78 3.40 2.80
0.2 1 R.T. 1.56 5.60 4.90 23.72 25.00 20.41
274.5 1.19 4.20 3.90
2 R.T. 1.74 9.40 9.20 31.61 47.87 57.61
274.5 1.19 4.90 3.90
Reproduced from Amin, A. K. M. N.; Abdelgadir, M. The Effect of Preheating of Work Material on Improved Machinability of Materials. In Proceedings of the International
Conference AIMTDR, Ranchi, India, 2002.

Figure 27 Flank wear, stainless steel for machining: (a) At room temperature; (b) Under heat assistance at 275  C at 90 m min1, depth of cut 2 mm,
and feed 0.2 mm per tooth. Reproduced from Amin, A. K. M. N.; Abdelgadir, M. The Effect of Preheating of Work Material on Improved Machinability
of Materials. In Proceedings of the International Conference AIMTDR, Ranchi, India, 2002.

Figure 28 Flank wear, stainless steel for machining: (a) At room temperature; (b) Under heat assistance at 265  C at 150 m min1, depth of cut 2 mm,
and feed 0.2 mm per tooth. Reproduced from Amin, A. K. M. N.; Abdelgadir, M. The Effect of Preheating of Work Material on Improved Machinability
of Materials. In Proceedings of the International Conference AIMTDR, Ranchi, India, 2002.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
318 Heat-Assisted Machining

11.13.8 Benefits of Induction Heat-Assisted End Milling of Titanium Alloy – Ti–6Al–4V

This chapter presents results obtained through various experiments and investigations to assess the effect of heat-assisted machining
on the improvement of machinability in the end milling of Ti–6Al–4V alloy with uncoated WC–Co inserts, followed by relevant
discussions. These experiments were conducted using an induction heating system of higher power capacity (35 kW) since titanium
requires higher heating temperature compared to carbon and stainless steel. The major machinability criteria that have been crit-
ically analyzed as functions of primary machining variables as well as preheating temperatures are as follows:
l Tool life and tool wear morphology
l Cutting forces
l Vibration/chatter
l Surface roughness and surface integrity
l Chips morphology

The influence of heating on different aspects of machinability, namely, cutting force, chatter, surface roughness, tool wear, metal
removal rate, chip formation, and surface integrity of the machined part, was considered. The design of experiments and the cutting
conditions for this part are shown in Section 11.13.7.

11.13.8.1 Effect on Vibration/Chatter

Vibration/chatter in machining is an undesirable but unavoidable phenomenon. The term is defined as the self-excited violent
relative dynamic motion between the cutting tool and workpiece. In the current study, Fast Fourier Transformation (FFT) analysis of
the time-domain vibration signals, captured by the accelerometer mounted on the spindle, was conducted using the DASYLab 5.5
software. FFT for four different conditions, at room temperature, and for three different temperatures of heat application are
presented for comparison in Figures 29(a–d). The machining parameters used in these experiments were cutting
speed ¼ 70 m min1, feed ¼ 0.088 mm per tooth, axial depth of cut ¼ 1 mm. These diagrams show that heat-assisted machining has
a great influence on vibration and chatter in end milling of titanium alloy Ti–6Al–4V. As can be observed from the figures, there are
four main peaks of amplitude in the range of 0–12 500 Hz. These peaks are eventually used to analyze the effects of preheating on
the amplitude of chatter acceleration.
The maximum acceleration amplitudes for room temperature and heat-assisted machining at different heating temperatures are
presented in Table 5 along with the percentage of reductions at the respective heating temperatures compared to room temperature
amplitudes. Reduction of acceleration of amplitude was found to range from 25 to 50%, 42.9 to 66.7%, and 64.3 to 88.2% at 315,
450, and 650  C, respectively.
The reductions of the peak values are systematically calculated, and the effects of heating temperature on the amplitudes are
plotted as bar charts in Figure 30. It is observed clearly that the amplitude of all four peaks in the frequency range from 0 to
12 000 Hz experience drastic reduction due to heat-assisted machining. The reduction is found to increase with the heating
temperature, with the highest value at the highest heating temperature of 650  C.
The results therefore suggest that heat-assisted machining can be employed successfully to suppress vibration/chatter in the
machining of titanium alloy Ti–6Al–4V. Reduction of chatter during cutting due to heat-assisted machining decreases the dynamic
loading on the tool tips and plays a favorable role in reducing tool wear and improving surface finish.

Figure 29 FFT output of end milling with uncoated WC–Co: (a) At room temperature; (b) Preheating at 315  C; (c) Preheating at 450  C; (d) Preheating
at 650  C (cutting speed ¼ 70 m min1, axial DOC ¼ 1 mm, feed ¼ 0.088 mm per tooth). Reproduced from Turnad, L. G. Improvement of Machinability of
Titanium Alloy Ti-6Al-4V through Workpiece Preheating. Ph.D. Thesis, Manufacturing and Materials Engineering Department, IIUM, Malaysia, 2009.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Heat-Assisted Machining 319

Table 5 Acceleration amplitudes of vibration and the percentage of reduction (inserts: uncoated WC–Co)

Maximum acceleration amplitude (m s2) Amplitude reductions (%)

   
Frequency range (Hz) R.T. 315 C 450 C 650 C 315 C 450  C 650  C

0–2500 0.006 0.003 0.002 0.001 50 66.7 83.3

2500–5000 0.012 0.009 0.005 0.003 25.2 58.3 75
7500–10 000 0.014 0.010 0.008 0.005 28.6 42.9 64.3
10 000–12 500 0.017 0.012 0.007 0.002 29.4 58.8 88.2
Reproduced from Turnad, L. G. Improvement of Machinability of Titanium Alloy Ti-6Al-4V through Workpiece Preheating. Ph.D. Thesis, Manufacturing and Materials Engi-
neering Department, IIUM, Malaysia, 2009.

Figure 30 The effect of heating temperature on acceleration amplitude of vibration (cutting speed ¼ 70 m min1, axial DOC ¼ 1 mm, feed ¼ 0.088 mm
per tooth, insert: uncoated WC–Co). Reproduced from Turnad, L. G. Improvement of Machinability of Titanium Alloy Ti-6Al-4V through Workpiece
Preheating. Ph.D. Thesis, Manufacturing and Materials Engineering Department, IIUM, Malaysia, 2009.

11.13.8.2 Effect on Cutting Force

This section presents a comparison of cutting force under these two conditions and discusses in detail the effects of heating
temperature, cutting speed, as well as feed on cutting force. The resultant cutting forces recorded from the experiments were used for
comparison.
The effects of heating temperature and cutting speed on the resultant cutting force are presented in Figure 31. The figure affirms
that an increase in heating temperature as well as cutting speed leads to reduction of the resultant cutting force. For instance, the
resultant cutting force is found to be reduced by 44.5 and 40.6% at 30.6 and 160 m min1, respectively, due to heating at only
450  C (the optimum heating temperature being 650  C). On the other hand, an increase in cutting speed from 30.6 to
160 m min1 reduces the resultant cutting force by 36.8 and 32.5% for room temperature and heat-assisted machining respectively.
The effect of heat-assisted machining conducted using three different feed values was investigated. Figure 32 presents the results
of this investigation. Cutting force is also found to drastically rise with the increase of feed. However, the reduction in cutting force
due to heat-assisted machining is clearly indicated in the figure for all three feed values applied in the experiments.

Figure 31 Combined effects of cutting speed and heat assistance (at 450  C) on the resultant cutting force (insert: uncoated WC–Co). (Feed ¼ 0.088 mm per
tooth and axial depth of cut ¼ 1 mm). Reproduced from Turnad, L. G. Improvement of Machinability of Titanium Alloy Ti-6Al-4V through Workpiece
Preheating. Ph.D. Thesis, Manufacturing and Materials Engineering Department, IIUM, Malaysia, 2009.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
320 Heat-Assisted Machining

Figure 32 Effects of feed and heat assistance (at 450  C) on the resultant cutting force (insert: uncoated WC–Co). (Cutting speed ¼ 70.1 m min1 and
axial depth of cut ¼ 1 mm). Reproduced from Turnad, L. G. Improvement of Machinability of Titanium Alloy Ti-6Al-4V through Workpiece Preheating.
Ph.D. Thesis, Manufacturing and Materials Engineering Department, IIUM, Malaysia, 2009.

As shown in Figure 32, the cutting force increases by 51.4 and 81.1% when feed is increased from 0.05 to 0.15 mm per tooth for
room temperature and heat-assisted conditions, respectively. The reduction in the resultant force by 53.6 and 44.5% at the feed of
0.05 and 0.088 mm per tooth, respectively, was achieved by heating the workpiece at only 450  C is applied. This is attributable to
the drop in the yield strength of the workpiece material at the elevated temperature during the heat-assisted end milling, which
eventually reduces the normal and the shear stresses acting on the tool and thus contributes to lower the cutting forces.
The effect of heating temperature on the resultant force was investigated for a given set of cutting parameters: Cutting
speed ¼ 70 m min1, axial DOC ¼ 1 mm, feed ¼ 0.088 mm per tooth. Figure 33 illustrates the effect of heating temperature on
cutting force. The resultant cutting force decreases sharply from 331.5 N at room temperature to 177.6 N at 315  C. With further
increase in heating temperature, the reduction is more gradual. The resultant force is reduced to only 134 N at 650  C.

11.13.8.3 Effect on Tool Wear and Tool Wear Morphology

Figures 34 and 35 compare tool life and volume of metal removal (VMR) per tool life, respectively, in the cutting speed range from
30 m min1 up to 160 m min1 for the given insert (WC–Co). Tool life and volume of metal removal per tool life at heat-assisted
machining are much higher compared to room temperature machining, though these two parameters are found to decline with the
increase in cutting speed, especially beyond 70 m min1. Tool life and VMR are insignificant at the highest cutting speed of
160 m min1. The maximum benefit of heat-assisted machining can be derived at the cutting speed of 70 m min1, since the VMR is
the same as at the lower speed of 30 m min1 and application of this speed would ensure higher productivity. At this cutting speed,
tool life is increased by 205.6% and VMR by 214.3% compared to the corresponding room temperature condition.
The effects of heating temperature on tool life and tool life improvement at constant cutting speed of 70 m min1 are presented
in Figure 36. Interestingly, the heating of work material substantially improves tool life, and as a result, enhances the VMR.
However, at lower heating temperature (e.g., 315  C), there is insignificant improvement in tool life. But as the temperature was

Figure 33 Effect of heating temperature on the resultant of cutting force (cutting speed ¼ 70 m min1, axial DOC ¼ 1 mm, feed ¼ 0.088 mm per tooth,
insert: uncoated WC–Co). Reproduced from Turnad, L. G. Improvement of Machinability of Titanium Alloy Ti-6Al-4V through Workpiece Preheating. Ph.D.
Thesis, Manufacturing and Materials Engineering Department, IIUM, Malaysia, 2009.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Heat-Assisted Machining 321

Tool Life versus Cutting Speed for Room Temperature and

Heat-Assisted Machining
140
127.8
120
RT HA

Tool Life, min

100
80
55
60
44
40
18 18
20
15
0
0 50 100 150 200
Cutting Speed, m min–1

Figure 34 Comparison of tool life under room temperature and heat-assisted machining (cutting speed ¼ 70 m min1, axial DOC ¼ 1 mm,
feed ¼ 0.088 mm per tooth, insert: uncoated WC–Co).

VMR versus Cutting Speed for Room Temperature and

Heat-Assisted Machining
140
VMR per tool Edge. cm3

120 109.7 110

RT HA
100
80
60
37.7 35
40
34.9
20
22
0
0 50 100 150 200
Cutting Speed, m min–1

Figure 35 Comparison of volume of metal removal per tool life under room temperature and heat-assisted machining (cutting speed ¼ 70 m min1,
axial DOC ¼ 1 mm, feed ¼ 0.088 mm per tooth, insert: uncoated WC–Co).

Figure 36 Effects of heating temperature on (a) Tool life; (b) Tool life improvement (cutting speed ¼ 70 m min1, axial DOC ¼ 1 mm, feed ¼ 0.088 mm
per tooth, insert: uncoated WC–Co). Reproduced from Turnad, L. G. Improvement of Machinability of Titanium Alloy Ti-6Al-4V through Workpiece
Preheating. Ph.D. Thesis, Manufacturing and Materials Engineering Department, IIUM, Malaysia, 2009.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
322 Heat-Assisted Machining

Figure 37 Effects of feed on tool life (cutting speed ¼ 70 m min1, axial DOC ¼ 1 mm, insert: uncoated WC–Co). Reproduced from Turnad, L. G.
Improvement of Machinability of Titanium Alloy Ti-6Al-4V through Workpiece Preheating. Ph.D. Thesis, Manufacturing and Materials Engineering
Department, IIUM, Malaysia, 2009.

increased, a significant improvement in tool life was achieved. This is related to the earlier results showing that heating of work
material suppressed chatter/vibration and lowered the cutting force. With heating at 650  C, tool life is increased by more than
270%. Hence 650  C may be considered as the optimum heating temperature for the given work material (Ti–6Al–4V). The findings
confirm that induction coil heat-assisted end milling is a prospective method in enhancing the tool life.
The effect of heat-assisted machining for three different feed values are illustrated in Figure 37. The benefit of heat-assisted
machining conducted at 450  C is maximum at the lowest feed of 0.05 m per tooth, and at the higher feed rate the effect is lower.
Perhaps higher heating temperature of up to 650  C is required to heat up the thicker undeformed chip thickness in order to derive
the full benefit of heat-assisted machining.

11.13.8.4 Tool Wear Morphology

The scanning electron microscope (SEM) was used in the investigation of tool wear morphology. The effects of cutting speed, feed,
axial depth of cut, and heating temperature on the tool wear profile were thoroughly investigated. The findings of these investi-
gations are presented in this section.
Figure 38 compares the SEM pictures of worn tools after heat-assisted machining at different temperatures. Heat-assisted
machining guarantees superior tool life in the end milling of Ti–6Al–4V, as previously exemplified in Section 8.3. This fact is also
verified and substantiated by analyzing and comparing the tool wear for both room temperature and heat-assisted machining
conditions. The experimental runs with the same cutting speed, feed, axial depth of cut but with different temperatures have been
essentially selected for this purpose.
Figure 38 shows that flank wear is reduced to different degrees after heating to different temperatures compared to that under
room temperature conditions. Nonuniform flank wear and attrition type wear are found to be more prevalent in room temperature,
and heat-assisted machining applying low heating temperature of 315  C and diffusion wear are the most prominent in other two
cases of heat-assisted machining, with higher heating temperatures of 450 and 650  C. One cause of this appreciable reduction of
tool wear is related to the reduction of acceleration amplitude of vibration/chatter due to heat-assisted machining (6). Among other
reasons for lower tool wear is the lower cutting force during heat-assisted machining and reduction of dynamic loads acting on the
tool. Softer workpiece with reduced shear strength as a result of heating effectively helps reduce the stress acting on the tool, and it is
responsible for reducing tool wear, leading to increased tool life. Heat-assisted machining reduces the strain hardening and flow
stress of material, which also contribute to increased tool life.
Figures 39(a–d) shows the SEM views of chip-tool contact length for room temperature and heat-assisted machining. As seen in
these figures, heating temperature markedly increases the chip-tool contact length. Figure 40 illustrates the effect of heating
temperature on chip-tool contact length. It can be observed from Figure 40 that the contact length is increased from 185 mm at room
temperature to 280 mm at 650  C of preheating; that is, chip-tool contact length is increased by 51.4%.
Longer chip-tool contact length in heat-assisted machining is beneficial in two ways: first, it shifts the hot spot away from the
cutting edge, and second, the normal stress acting on the tool becomes lower (4). Heating also results in lower frictional forces
acting on the rake face of the tool because of the lower shear strength of the flow zone materials owing to the higher temperature of
the zone. As a result, dynamic loads that act on the tool are substantially lower for heat-assisted machining compared to room
temperature machining. All of these aspects add to the benefits of heat-assisted machining.

11.13.8.5 Surface Roughness and Surface Integrity

This section presents a comparison of surface roughness obtained under room temperature and heat-assisted machining conditions.
In these investigations, the heating temperature was only varied keeping the machining conditions same for both machining

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Heat-Assisted Machining 323

Figure 38 SEM views of worn tools at: (a) Room temperature (tool life: 17.1 min); (b) Heating at 315  C (tool life: 19.2 min); (c) Heating at 450  C
(tool life: 23.5 min); (d) Heating at 650  C (tool life: 55.7 min) (cutting speed ¼ 70 m min1, axial DOC ¼ 1 mm, feed ¼ 0.088 mm per tooth, insert:
uncoated WC–Co). Reproduced from Turnad, L. G. Improvement of Machinability of Titanium Alloy Ti-6Al-4V through Workpiece Preheating. Ph.D.
Thesis, Manufacturing and Materials Engineering Department, IIUM, Malaysia, 2009.

methods. The surface integrity and subsurface alternation have also been thoroughly investigated by employing SEM and Vickers
microhardness testing. Microhardness was measured to observe the distributions of the hardness beneath the machined surface and
to determine the effect of heating on subsurface hardness and microstructure. The changes in microstructure were investigated using
a SEM. The effects of heating on the subsurface alternations are discussed in this section. Prior to investigating surface integrity,
samples were prepared by cutting the work specimen perpendicular to machined surfaces using electro discharge wire cutting. The
samples were then mounted using hot mounting, and grounded using silicon carbide abrasive papers, followed by polishing with
alumina solutions, and finally etched with 10% HF, 5% HNO3, 85% H2O solutions. Microhardness was measured along the depth
(perpendicular to the machined surface), at an interval of 0.01 mm from the top surface up and continued up to a depth of 0.5 mm.
The dependence of surface roughness on heating temperature is shown in Figure 41. As the figure illustrates, surface roughness is
sharply reduced from the room temperature value of 0.25–0.19 mm when hot machined at 315  C. Surface roughness value is found
to slightly increase as the heating temperature is further increased up to 650  C. Suppressed vibration/chatter amplitude during
heat-assisted machining and higher heating temperature at the chip-tool interface facilitating easier chip flow are the main factors
contributing to reducing surface roughness.
Increasing trends toward surface roughness values at higher heating temperature may be attributable to the higher chemical
reactivity of the work material, which encourages welding between chips and tools to form a pile-up of material similar to built-up
edge (BUE), which is more prominent in the case of preheated machining, as shown in Figures 42(a–d). Similar to room
temperature machining, the presence of BUE can be observed in most of the cutting conditions under preheated machining. This is
attributed to the high chemical reactivity between the chip and cutting tool. At a cutting speed of 70 m min1 (Figures 42(a) and
42(b)), the BUE is very stable, but at 126 and 160 m min1 it is unstable and scattered (Figures 42(c) and 42(d)). As the preheating
temperature is increased, the chemical reactivity drastically increases. The BUE is developed on the rake face and in the flank face,
similar to cutting at room temperature. Essentially, the stable BUE developed helps to protect the tool from wear by acting as
a shield on the tool tip. Unstable BUE, on the other hand, causes a higher rate of wear and adversely affects the surface finish.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
324 Heat-Assisted Machining

Figure 39 SEM views of chip-tool contact length (cutting speed ¼ 70 m min1, axial DOC ¼ 1 mm, feed ¼ 0.088 mm per tooth, insert: uncoated
WC–Co). Reproduced from Turnad, L. G. Improvement of Machinability of Titanium Alloy Ti-6Al-4V through Workpiece Preheating. Ph.D. Thesis,
Manufacturing and Materials Engineering Department, IIUM, Malaysia, 2009.

Figure 40 Chip-tool contact length versus preheating temperature (cutting speed ¼ 70 m min1, axial DOC ¼ 1 mm, feed ¼ 0.088 mm per tooth, insert:
uncoated WC–Co). Reproduced from Turnad, L. G. Improvement of Machinability of Titanium Alloy Ti-6Al-4V through Workpiece Preheating.
Ph.D. Thesis, Manufacturing and Materials Engineering Department, IIUM, Malaysia, 2009.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Heat-Assisted Machining 325

Figure 41 Effects of heating temperature on surface roughness (cutting speed ¼ 70 m min1, axial DOC ¼ 1 mm, f ¼ 0.088 mm per tooth, insert:
uncoated WC–Co). Reproduced from Turnad, L. G. Improvement of Machinability of Titanium Alloy Ti-6Al-4V through Workpiece Preheating.
Ph.D. Thesis, Manufacturing and Materials Engineering Department, IIUM, Malaysia, 2009.

Figure 42 SEM views of built-up edge under preheated conditions: (a) V ¼ 70, fz ¼ 0.05, d ¼ 1, Temp ¼ 315; (b) V ¼ 70, fz ¼ 0.088, d ¼ 1,
Temp ¼ 450; (c) V ¼ 126, fz ¼ 0.06, d ¼ 1, Temp ¼ 580; (d) V ¼ 160, fz ¼ 0.088, d ¼ 1, Temp ¼ 450 (insert: uncoated WC–Co). Reproduced from
Turnad, L. G. Improvement of Machinability of Titanium Alloy Ti-6Al-4V through Workpiece Preheating. Ph.D. Thesis, Manufacturing and Materials
Engineering Department, IIUM, Malaysia, 2009.

11.13.8.6 Surface Integrity and the Underlying Layer

11.13.8.6.1 Surface Integrity
The surface integrity of machined surfaces has been investigated by employing the SEM. The objective of the investigation is to
identify the presence of defects, such as cracks, droplets of BUE, or chip fragments on machined surfaces, especially at higher heating
temperature, to explore the causes of increased surface roughness when the heating temperature is increased beyond 315  C, as
discussed in the previous subsection. In Figures 43(a) and 43(b), the presence of some chip fragments and BUE on the surface for
both room temperature and heat-assisted machining is clearly seen. High temperature generated during cutting of Ti–6Al–4V is

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
326 Heat-Assisted Machining

Figure 43 SEM views of machined surface in end milling at: (a) Room temperature; (b) Heating at 650  C (cutting speed ¼ 70 m min1, axial
DOC ¼ 1 mm, feed ¼ 0.088 mm per tooth, insert: uncoated WC–Co). Reproduced from Turnad, L. G. Improvement of Machinability of Titanium Alloy
Ti-6Al-4V through Workpiece Preheating. Ph.D. Thesis, Manufacturing and Materials Engineering Department, IIUM, Malaysia, 2009.

mainly responsible for the buildup of chip fragments on the machined surface, which finally depreciates the surface finish.
Application of high-pressure air blowing directed to the machining zone during operation could eliminate the formation of these
depositions and thereby improve the surface finish further.

11.13.8.6.2 The Underlying Layer

For this investigation the distribution of the microhardness below the surface layer was recorded under different cutting conditions.
Apart from that, the microstructure of the layer below the surface was also investigated. Distributions of microhardness measured up
to certain depths beneath the surface after end milling with uncoated WC–Co inserts under room temperature and two heating
temperature are informatively presented in Figure 44. The measurements were taken at a length interval of 0.01 mm starting from
the top surface up to a depth of 0.5 mm beneath the surface. According to the hardness, the surface layer is divided into three zones:
the heat-affected zone (Zone I), the strain-hardened zone (Zone II), and the base material (Zone III).
Microhardness values in Zone I are relatively lower than those of the strain-hardening zone (Zone I) and the base materials
(Zone III) as shown in Figure 44. It may be related to the fact that the heat that is generated during cutting leads to softening of
a thin layer below the surface due to annealing. As the heating temperature increases, the driving force for annealing obviously
increases, which results in lower microhardness. For instance, the distribution of microhardness after end milling at heating
temperature of 650  C is slightly lower compared to those under room temperature and heat-assisted machining at 315  C, as
shown in Figure 44. The figure also shows that Zone I reaches the range of up to w0.15 mm below the surface. In the zone more
than 0.15 mm from the surface, the effect of heating substantially decreases, which is inadequate as a driving force for annealing
arrangement. Due to the plastic deformation, the microhardness in this zone (Zone II) is also relatively high and is then
considered as a strain-hardening zone. Moreover, as the distance from the surface increases, the strain-hardening effect caused by
shearing and plastic deformation gradually decreases, and as a result, the distribution hardness returns to the initial value (base
metal). In the case of end milling with uncoated WC–Co inserts, the distance of the strain-hardening zone is found to be up to
0.4 mm beneath the surface.
Under room temperature machining, the heat generated during cutting has less effect on the microstructure alteration. No
appreciable grain growth occurs in this condition, as presented in Figure 45(b) compared to the original microstructure
(Figure 45(a)). When heat-assisted machining is applied at 650  C, there is a slight grain growth as shown in Figure 45(c). In
metallurgical terms, an increase in the size of grains decreases the hardness of the materials, as also observed at the beginning of this
subsection.

Figure 44 Microhardness value beneath the machined surface in end milling with uncoated WC–Co inserts (cutting speed ¼ 70 m min1, axial
DOC ¼ 1 mm, feed ¼ 0.088 mm per tooth). Reproduced from Turnad, L. G. Improvement of Machinability of Titanium Alloy Ti-6Al-4V through Workpiece
Preheating. Ph.D. Thesis, Manufacturing and Materials Engineering Department, IIUM, Malaysia, 2009.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Heat-Assisted Machining 327

Figure 45 Microstructure alterations in end milling: (a) The base metal; (b) Room temperature; (c) Heating at 650  C (Insert: uncoated WC–Co, cutting
speed ¼ 70 m min1, axial DOC ¼ 1 mm, feed ¼ 0.088 mm per tooth). Reproduced from Turnad, L. G. Improvement of Machinability of Titanium Alloy
Ti-6Al-4V through Workpiece Preheating. Ph.D. Thesis, Manufacturing and Materials Engineering Department, IIUM, Malaysia, 2009.

11.13.8.7 Chips Morphology

Chips formed during room temperature and heat-assisted machining conditions were analyzed carefully in order to determine the
effects of heating on the phenomenon such as chip serration, chip shrinkage coefficient, and chip forming frequency. Figures 46(a–d)
presents the SEM views of the chips formed at different heating temperatures. The figures are arranged serially with increasing
temperature. It can be observed from the SEM views of chips that there are chip serrations running across the whole width of the chips
in all the investigated cases. These teeth are termed primary serrated teeth. There are also marks of grouping of several primary serrated
elements at the upper free edge and in some cases at the lower tool nose-side edge of the chip. These larger coagulated elements are
known as secondary serrated teeth. This section presents the effect of online heating temperature on the primary serration of chip.
Frequency of the chip serration was calculated using the parameters cutting speed, coefficient of chip shrinkage, magnification of
the SEM micrograph, and numbers of serrated elements (number of serrated teeth). The chip shrinkage coefficient is shown in
Figure 47 as a function of preheating (heating) temperature.
Figures 48(a–c) presents a cross-section of chips produced during end milling at a particular set of cutting parameters (cutting
speed ¼ 70 m min1, fz ¼ 0.088 mm per tooth, axial DOC ¼ 1 mm, insert: uncoated WC–Co) conducted under three different cutting
conditions: room temperature and at two heating temperatures (450  C and 650  C). Figure 46 shows that the chips become more
stable with the increase in heating temperature. The chips are almost continuous, practically without serration at the heat-assisted
temperature of 650  C (Figure 46(c)). A lower peak-to-valley ratio of the chip was found during employing heating. The peak-to-
valley ratio decreases with the increase in heating temperature and comes to almost 1 at the highest heating temperature of 650  C.
Figure 49 shows the peak-to-valley ratio versus heating temperature for chips produced at the cutting speed of 70 m min1,
feed ¼ 0.088 mm per tooth, and axial DOC ¼ 1 mm using uncoated WC–Co inserts (44).

11.13.8.8 Discussion
The results of experiments presented in this section clearly show that heat-assisted machining appreciably increased the tool life and
VMR. For instance, at a cutting speed of 30.6 m min1, 44 min of tool life, and 37.7 cm3 of VMR were achieved in room temperature

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
328 Heat-Assisted Machining

Figure 46 Chip produced in various runs at different temperature in end milling with uncoated WC–Co inserts for the same cutting condition at: (a)
Room temperature; (b) 315  C; (c) 450  C; (d) 650  C (cutting speed ¼ 70 m min1, axial DOC ¼ 1 mm, feed ¼ 0.088 mm per tooth). Reproduced from
Turnad, L. G. Improvement of Machinability of Titanium Alloy Ti-6Al-4V through Workpiece Preheating. Ph.D. Thesis, Manufacturing and Materials
Engineering Department, IIUM, Malaysia, 2009.

Figure 47 Chip shrinkage coefficients versus heating temperature (cutting speed ¼ 70 m min1, feed ¼ 0.088 mm per tooth, axial DOC ¼ 1 mm, insert:
uncoated WC–Co). Reproduced from Turnad, L. G. Improvement of Machinability of Titanium Alloy Ti-6Al-4V through Workpiece Preheating. Ph.D.
Thesis, Manufacturing and Materials Engineering Department, IIUM, Malaysia, 2009.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Heat-Assisted Machining 329

Figure 48 Cross-section of chips produced in machining using the same set of parameters but at different temperature: (a) Room temperature; (b)
450  C; (c) 650  C (cutting speed ¼ 70 m min1, fz ¼ 0.088 mm per tooth, axial DOC ¼ 1 mm, insert: uncoated WC–Co).

Figure 49 Peak to valley ratio versus heating temperature (cutting speed ¼ 70 m min1, feed ¼ 0.088 mm per tooth, and axial DOC ¼ 1 mm, insert:
uncoated WC–Co). Reproduced from Turnad, L. G. Improvement of Machinability of Titanium Alloy Ti-6Al-4V through Workpiece Preheating.
Ph.D. Thesis, Manufacturing and Materials Engineering Department, IIUM, Malaysia, 2009.

machining, while, 127.8 min in tool life and 109.7 cm3 in VMR were achieved in heat-assisted machining at 650  C under the same
cutting speed. The improvement of the tool life may be explained from several perspectives. Induction heating substantially
decreases the resultant cutting force. Lower cutting force is attributed to the lower yield strength of the workpiece material at the
elevated temperature. Thus, in heat-assisted machining, the shear stress reduces and contributes to lower the cutting forces. Heat-
assisted machining also helps to ease the chip flow, which reduces frictional forces in the rake and flank faces of the tool. For
instance, a decrease in cutting force by approximately 60% was recorded during heat-assisted machining at 650  C, with uncoated
WC–Co inserts at the cutting speed of 70 m min1. This result indicates the overwhelming benefits of using induction heating.
Suppression of chatter/vibration during heat-assisted machining is another advantage. The results of the experiments on the
effects of heating temperature on chatter/vibration during cutting show that noticeable suppressions of chatter/vibration take place
when heating is employed. For instance, a chatter/vibration was suppressed up to 88.2% during heat-assisted machining with
uncoated WC–Co inserts at heating temperature of 650  C. It substantiates the idea that heat-assisted machining can be utilized to
eliminate chatter in machining. A reduced chatter/vibration eventually reduces the bouncing effect on the tool tip, and conse-
quently, reduces the tool wear rates.
An increased chip-tool contact length due to the heating is also attributable to an increase in tool life. For instance, increased
chip-tool contact length in 2.3 times was achieved when employing heating at temperature of 450  C. Longer chip-tool contact
length decreases the shear stress and the temperature close to tool tip. Those various factors play favorable roles and eventually
reduce tool wear, and, consequently, increase tool life.

11.13.8.9 Summary
The investigation on the machinability of uncoated WC–Co and PCD inserts in the end milling of Ti–6Al–4V alloy has been
extensively conducted under room temperature and heat-assisted conditions. The machinability assessment provides a comparison
between room temperature and heat-assisted conditions. The evaluation includes tool life and tool wear morphology, surface
roughness and surface integrity, cutting force, acceleration amplitude of vibration/chatter, chip-tool contact length, and chip
morphology. Heat-assisted machining is found to help substantially reduce the tool wear rate, as well as increase tool life and
volume of metal removal in the machining of Ti–6Al–4V alloy. This notable improvement in terms of tool life and volume of metal
removal are explained as follows. Heat-assisted machining helps substantially soften the workpiece prior to cutting, which decreases

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
330 Heat-Assisted Machining

the resultant cutting force. Lower cutting force and longer chip-tool contact length reduces the stresses acting on the cutting edge,
and consequently decreases the tool wear rate. Suppression of vibration/chatter was also responsible for increasing tool life during
heat-assisted machining as a result of lower dynamic stresses acting on the tool. Heat-assisted machining also significantly
contributes to increase the chip-tool contact length. This aspect is important because of the shifting of the hot spot away from the
cutting edge. The stresses acting on the tool are lowered, which leads to longer tool life. Heat-assisted machining also leads to
a reduction in the chip serration frequency. A thin and long continuous chip produced during heat-assisted machining substantially
helps reduce the cutting force, lowering the cutting pressure and cutting temperature.
Heat-assisted machining is also found to eliminate the effects of strain hardening during cutting. Results from microhardness
measurement beneath the machined surface convincingly proved that the strain hardening caused by room temperature machining
was sufficiently reduced by employing heating. In the subsurface investigation, three main zones beneath the top surface in cutting
this alloy – heat-affected zone, strained hardening zone, and base metal zone – were identified. These zones are highly affected by
the temperature employed during cutting.
Lastly, heat-assisted machining was found to appreciably improve the surface roughness, which eliminates the need for grinding
and polishing operations. As a conclusion, heat-assisted machining with induction heating has been successfully proved to be an
alternative method in increasing tool life, enhancing the volume of metal removal, and achieving a good surface finish in the end
milling of titanium alloy Ti–6Al–4V with uncoated WC–Co. Thus, the improved machinability of titanium alloy Ti–6Al–4V through
workpiece heating has been successfully established.
Finally, in all the cutting conditions and applications, including turning and end milling of the different work materials,
application of heat-assisted machining has been proved to be a more beneficial method for improving the machinability of plain-
carbon steel, heat-resistant and stainless steel, titanium alloys, and even hardened steels and alloys.

11.13.9 Conclusion

For successful application of heat-assisted or hot machining, reduction of strength of the material layer being removed needs to be
achieved through localized and controlled heating of the work material within the machining zone. The heat is generally applied
online during the machining process, but it can also be applied immediately prior to machining. The optimum heating temperature,
at which maximum benefit of heat-assisted machining can be derived, is different for different materials and is close to the
temperature at which the strength of the materials starts to decline abruptly.
Machinability enhancement of difficult-to-cut metals and alloys, as a result of application of external heat, is manifested in
substantial reduction of cutting force and chatter amplitude and improved tool life, surface finish, and chip flow over the tool.
Increases in chip-tool contact length, reduction of chip serration, and chatter amplitude achieved in heat-assisted machining facilitate
reduced dynamic loading on the tool and lower intensity of tool wear and subsequently higher tool life. Lower surface roughness
values attained in heat-assisted machining provide opportunities to cut down some of the inline finishing operations, such as grinding
and rough polishing, which apparently would result in reduced machining lead time and consequently lower machining cost.
Subsurface investigations revealed that the strain-hardened layer produced in conventional room temperature machining is
completely or partially eliminated during heat-assisted machining and would make it easier to remove the subsequent material
layer and reduce the notch wear related to the strain-hardened layer.

References

1. Copley, S. M. Laser Shaping of Materials, Lasers in Materials Processing. In Conference Proceedings of the American Society for Metals, Los Angeles, CA, 1983, pp 82–92.
2. Konig, W.; Wagemann, A. Machining of Ceramic Components: Process-Technological Potentials. In Machining of Advanced Materials, NIST Special Publication, 1990, Vol. 847,
pp 3–16.
3. Talantov, N. V.; Amin, A. K. M. N.; Chereomushnikov, N. P. Temperature Deformation Laws of Chatter Formation during Metal Cutting Process. In Abstracts of the papers
presented at the 5th Soviet National Conference, Teplophysika Technologichieskikh Processov, Volgograd, 1980, p 92.
4. Amin, A. K. M. N.; Talantov, N. V. Influence of the Instability of Chip Formation and Preheating of Work on Tool Life in Machining High Temperature Resistant Steel and Titanium
Alloys. Mech. Eng. Res. Bull. 1986, 9, 52–62.
5. Amin, A. K. M. N.; Abdelgadir, M. Control of Chatter Arising during Machining on Vertical Machining Center and Improvement of Tool Life and Surface Finish of the Machined
Part by Preheating of the Workpiece by Induction Heating. Malaysia Patent MY-126082-A, 1982.
6. Amin, A. K. M. N.; Abdelgadir, M. The Effect of Preheating of Work Material on Chatter during End Milling of Medium Carbon Steel Performed on a Vertical Machining Center.
J. Mater. Sci. Eng. Trans. ASME, USA 2003, 125, 667–680.
7. Amin, A. K. M. N.; Abraham, I.; Khairusshima, N. Performance of Circular Carbide Inserts in End Milling of Carbon Steel Under Preheating Condition. In Proceedings of the
ICAMT, 2004, pp 65–73.
8. Amin, A. K. M. N.; Abdelgadir, M.; Kamaruddin, K. Effect of Workpiece Preheating on Machinability of Titanium Alloy. In Proceedings of the International Conference ICAMT,
2004, pp 145–152.
9. Klocke, F.; Zaboklicki, A. Innovation in der Umformtechnik – Laserunterstützte Verfahren. In Blech Rohre Profile, 1995, pp 383–343.
10. Gesellschaft, V. D. I.; Produktionstechnik, A. D. B. Materialbearbeitung mit dem Laserstrahl im Geräte- und Maschinenbau; VDI-Verlag: Düsseldorf, 1990.
11. Brecher, C.; Rosen, C. J.; Emonts, M. Laser-Assisted Milling of Advanced Materials. Phys. Procedia 2010, 5, 259–272.
12. Uehara, K.; Takeshita, H. Cutting Ceramics with a Technique of Hot Machining. Ann. CIRP 1986, 35 (1), 55–58.
13. Amin, A. K. M. N.; Dolah, S.; Mahmud, M.; Lajis, M. A. Effects of Workpiece Preheating on Surface Roughness, Chatter and Tool Performance during End Milling of AISI D2 Tool
Steel. J. Mater. Process. Technol. 2008, 201, 466–470.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Heat-Assisted Machining 331

14. Kok, P. Y.; Amin, A. K. M. N.; Faris, A. I. Performance of Preheating, Cryogenic Cooling and Combined Approaches in Improving Machinability of Stainless Steel in Turning. In
Proceedings of International Conference on Mechanical Engineering – ICME, Dhaka, 2005, 34(1–5).
15. Leshock, Carl, E; Kim, Jin-Nam; Shin, Yung, C. Plasma Enhanced Machining of Inconel 718: Modeling of Workpiece Temperature with Plasma Heating and Experimental
Results. Int. J. Mach. Tools Manuf. 2001, 41, 877–897.
16. Madhavulu, G.; Ahmed, B. Hot Machining Process for Improved Metal Removal Rates in Turning Operations. J. Mater. Process. Technol. 1994, 44, 199–206.
17. Steen, W. Laser Material Processing, 3rd ed.; Springer: New York, 2003.
18. Sun, S.; Brandt, M.; Dargusch, M. S. Thermally Enhanced Machining of Hard-to-Machine Materials – A Review. Int. J. Mach. Tools Manuf. 2010, 50, 2010, 663–680.
19. Dumitrescu, P.; Koshy, P.; Stenekes, J.; Elbestawi, M. A. High-Power Diode Laser Assisted Hard Turning AISI D2 Tool Steel. Int. J. Mach. Tools Manuf. 2006, 46, 2009–2016.
20. Rozzi, J. C.; Pfefferkorn, F. E.; Incropera, F. P.; Shin, Y. C. Transient Thermal Response of Rotating Cylindrical Silicon Nitride Workpiece Subjected to Translating Laser Heat
Source, Part I: Comparison of Surface Temperature Measurements with Theoretical Results. J. Heat Transfer Trans. ASME 1998, 120, 899–906.
21. Pfefferkorn, F. E.; Incropera, F. P.; Shin, Y. C. Heat Transfer Model of Semi-Transparent Ceramics Undergoing Laser-Assisted Machining. Int. J. Heat Mass Transfer 2005, 48,
1999–2012.
22. Wu, X.; Zhang, H.; Wang, Y. Three-Dimensional Thermal Analysis for Laser Assisted Machining of Ceramics Using FEA. In Proceedings of SPIE, Fourth International Symposium
on Advanced Optical Manufacturing and Testing Technologies: Advanced Optical Manufacturing Technologies, Vol. 7282, 2009, pp 72822I-1–72822I-5.
23. Wu, J. F.; Guu, Y. B. Laser Assisted Machining Methods and Device, US Patent 7,002,100 B2, 2004.
24. Weck, M.; Zeppelin, W. V.; Hermanns, C. Laser – A Tool for Turning Centers. In Laser Assisted Net Shape Engineering, Proceedings of the LANE’94. Vol. 1; Geiger, M.,
Vollertsen, F., Eds.; Meisenbach: Bamberg, 1994; pp 427–437.
25. Rajagopal, S.; Plankenhorn, D. J.; Hill, V. L. Machining Aerospace Alloys with the Aid of a 15kW Laser. J. Appl. Metalwork. 1982, 2, 170–184.
26. Sun, S.; Harris, J.; Brandt, M. Parametric Investigation of Laser-Assisted Machining of Commercially Pure Titanium. Adv. Eng. Mater. 2008, 10, 565–572.
27. Anderson, M.; Patwa, R.; Shin, Y. C. Laser-Assisted Machining of Inconel 718 with an Economic Analysis. Int. J. Mach. Tools Manuf. 2006, 46, 1879–1891.
28. Ding, H.; Shin, Y. C. Laser-Assisted Machining of Hardened Steel Parts with Surface Integrity Analysis. Int. J. Mach. Tools Manuf. 2010, 50, 106–114.
29. Shin, Y.C. Laser Assisted Machining Process with Distributed Lasers. US Patent 0062920A1, 2007.
30. Salem, W. B.; Marot, G.; Moisan, A.; Longuemard, J. P. Laser Assisted Turning during Finishing Operation Applied to Hardened Steels and Inconel 718. In Laser Assisted Net
Shape Engineering, Proceedings of the LANE’94. Vol. 1; Geiger, M., Vollertsen, F., Eds.; Meisenbach: Bamberg, 1994; pp 455–464. Proceedings of the LANE’94.
31. Armitage, K. Laser Assisted Machining of High Chromium White Cast Iron. Thesis for Master of Engineering, Swinburne University of Technology, 2005.
32. Walter, J.; Ritzi, R. Increasing Tool Life by Laser Assisted Turning. In Laser Assisted Net Shape Engineering 4, Proceedings of the LANE 2004, Vol. 2; Geiger, M., Otto, A., Eds.;
Meisenbach: Bamberg, 2004; pp 1157–1164.
33. Yang, B.; Lei, S. Laser-Assisted Milling of Silicon Nitride Ceramic: A Machinability Study. Int. J. Mechatron. Manuf. Syst. 2008, 1, 116–130.
34. Lacalle, L. N. L. D.; Sanchez, J. A.; Lamikiz, A.; Celaya, A. Plasma Assisted Milling Heat-Resistant Superalloys. J. Manuf. Sci. Eng. Trans. ASME 2004, 126, 274–285.
35. Thomas, T.; Vigneau, J. O. Laser-Assisted Milling Process. US Patent 5,906,459, 1999.
36. Konig, W. K.; Zaboklicki, A. K. Laser-Assisted Hot Machining of Ceramics and Composite Materials. In International Conference on Machining of Advanced Materials, NIST
Special Publication, Gaithersburg, MD, 1993, Vol. 847, pp 455–463.
37. Demmer, A.; Bausch, S.; Groll, K. Perspectives for Laser-Assisted Machining: Cost-Effective Processing of Difficult-to Machine Materials. Ind. Laser User 2005, 38–41.
38. Tian, Y.; Wu, B.; Anderson, M.; Shin, Y. C. Laser-Assisted Milling of Silicon Nitride and Inconel 718. J. Manuf. Sci. Eng. Trans. ASME 2008, 130, 031013-1–031013-9.
39. Komanduri, R.; Flom, D. G.; Lee, M. Highlights of DARPA Advanced Machining Research Program. In American Society of Mechanical Engineers, Production Engineering
Division (Publication) PED 12, 1984; pp 15–36.
40. Konig, W.; Zaboklicki, U. Laser-Assisted Hot Machining of Ceramics and Composite Materials. In Machining of Advanced Materials, NIST Special Publication, 1993, Vol. 847,
pp 455–463.
41. Chryssolouris, G.; Anifantis, N.; Karagiannis, S. Laser Assisted Machining: An Overview, 75th anniversary Issue. J. Manuf. Sci. Eng. Trans. ASME 1997, 119, 766–769.
42. Amin, A. K. M. N. Investigation of the Laws Governing the Formation of Chatter during Metal Cutting Processes and their Influence on Tool Wear. Ph.D. Thesis (in Russian),
Volgograd Polytechnic Institute, Volgograd, Russia, 1982.
43. Trent, E. M. Metal Cutting, 3rd ed.; Butterworths Publishing Company: UK, 2000.
44. Turnad, L. G. Improvement of Machinability of Titanium Alloy Ti-6Al-4V through Workpiece Preheating. Ph.D. Thesis, Manufacturing and Materials Engineering Department,
IIUM, Malaysia, 2009.
45. Amin, A. K. M. N.; Abdelgadir, M. The Effect of Preheating of Work Material on Improved Machinability of Materials. In Proceedings of the International Conference AIMTDR,
Ranchi, India, 2002.
46. Main Catalogue, Sandvik Coromat, 2006.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
This page intentionally left blank

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 11.14 Micro-Electrical Discharge Machining (Micro-EDM): Processes,
Varieties, and Applications
MP Jahan, Western Kentucky University, Bowling Green, KY, USA
M Rahman and YS Wong, National University of Singapore, Singapore
 2014 Elsevier Ltd. All rights reserved.

11.14.1 Introduction 334

11.14.2 Brief Overview of Micro-EDM 335
11.14.2.1 Working Principle 335
11.14.2.2 Sparking and Gap Phenomena of Micro-EDM 335
11.14.2.3 Differences between Macro- and Micro-EDM 336
11.14.3 Micro-EDM System Components 336
11.14.3.1 Pulse Generator or Discharging Unit 336
11.14.3.1.1 Transistor-Type Pulse Generator 336
11.14.3.1.2 RC-Type Pulse Generator 337
11.14.3.1.3 Transistor-Type Isopulse Generator: Modification of Transistor-Type 337
11.14.3.1.4 Capacity-Coupled Pulse Generator: Modification of RC-Type 338
11.14.3.2 Servo Control System or Gap Control Unit 338
11.14.3.3 Dielectric Circulation System or Flushing Unit 339
11.14.4 Micro-EDM Process Control Parameters 340
11.14.4.1 Electrical Parameters 340
11.14.4.1.1 Discharge Energy 340
11.14.4.1.2 Breakdown, Gap, Open-Circuit, and Discharge Voltage 340
11.14.4.1.3 Peak Current 341
11.14.4.1.4 Average Current 341
11.14.4.1.5 Pulse Duration 341
11.14.4.1.6 Pulse Interval 341
11.14.4.1.7 Duty Factor 342
11.14.4.1.8 Pulse Frequency 342
11.14.4.1.9 Pulse Waveform Shapes and Types 342
11.14.4.1.10 Polarity 343
11.14.4.2 Nonelectrical Parameters: Materials Parameters 343
11.14.4.2.1 Electrode Materials 343
11.14.4.2.2 Wire Electrode Materials for Micro-Wire Electro-Discharge Machining 344
11.14.4.2.3 Workpiece Materials 344
11.14.4.2.4 Dielectric Materials 344
11.14.4.3 Motion and Gap Control Parameters 345
11.14.4.3.1 Servo Feed and EDM Speed 345
11.14.4.3.2 Electrode Rotation 345
11.14.4.3.3 Workpiece Rotation 345
11.14.4.3.4 Tool and Electrode Geometry, Design, and Shape 345
11.14.4.3.5 Wire Tension for Micro-WEDM 346
11.14.4.3.6 Wire Speed and Wire Vibration 346
11.14.4.3.7 Tool Electrode and Workpiece Vibration 346
11.14.4.3.8 Types of Dielectric Flushing Process 346
11.14.4.3.9 Flushing Pressure 346
11.14.5 Micro-EDM Performance Measurement Parameters 346
11.14.5.1 Micro-EDM Erosion Efficiency 346
11.14.5.2 Material Removal Rate 347
11.14.5.3 Electrode Wear Ratio 347
11.14.5.4 Surface Roughness 347
11.14.5.5 Surface Integrity 347
11.14.5.6 Spark Gap/Gap Width/Overcut 348
11.14.6 Micro-EDM Process Varieties and Related Applications 348
11.14.6.1 Die-Sinking Micro-EDM 348
11.14.6.1.1 Working Principle 348
11.14.6.1.2 Application Examples of Die-Sinking Micro-EDM 348
11.14.6.2 Microwire Electro-Discharge Machining (Micro-WEDM) 348

Comprehensive Materials Processing, Volume 11 http://dx.doi.org/10.1016/B978-0-08-096532-1.01107-9 333

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
334 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications

11.14.6.2.1 Working Principle 348

11.14.6.2.2 Application Examples of Micro-WEDM 349
11.14.6.3 Milling Micro-EDM 351
11.14.6.3.1 Working Principle 351
11.14.6.3.2 Application Examples of Milling Micro-EDM 351
11.14.6.4 Dry and Near-Dry Micro-EDM 351
11.14.6.4.1 Working Principle 351
11.14.6.4.2 Application Examples of Dry and Near Dry Micro-EDM 353
11.14.6.5 Micro-EDM Drilling 355
11.14.6.5.1 Working Principle 355
11.14.6.5.2 Application Examples of Micro-EDM Drilling 355
11.14.6.6 Microelectro-Discharge Grinding Process 356
11.14.6.6.1 Working Principles of Different Micro-EDG Processes 356
11.14.6.6.2 Application Examples of Different Micro-EDG Processes 358
11.14.6.7 Planetary or Orbital Micro-EDM 361
11.14.6.7.1 Working Principle 361
11.14.6.7.2 Application Examples of Planetary or Orbital Micro-EDM 362
11.14.6.8 Reverse Micro-EDM 362
11.14.6.8.1 Working Principle 362
11.14.6.8.2 Application Examples of Reverse Micro-EDM 363
11.14.7 Advances in Micro-EDM: Innovative Applications 363
11.14.7.1 Deburring of Microfeatures by Micro-EDM 363
11.14.7.2 Micro-EDM as a Deposition Process 365
11.14.7.3 Spherical Probe Fabrication Using WEDG Combined with Single Discharge 365
11.14.7.4 Micro-EDM of Electrically Nonconductive Ceramic 366
11.14.7.5 3D Microstructuring in CNT Forests Using Dry Micro-EDM 366
11.14.7.6 Batch Mode Microstructuring Using Micro-EDM Combined with LIGA 367
11.14.7.7 Simple and Complex Microrod Fabrication Using Self-Drilled Holes 367
11.14.8 Summary 368
References 368

11.14.1 Introduction

In recent years, the trend in miniaturization of products is persistent in the areas of information technology, biotechnology, and
environmental and medical industries. The demand for microproducts and components has been rapidly increasing in electronics,
optics, medicine, biotechnology, automotive, communications, and avionics industries (1,2). Micromachining is the key sup-
porting technology that has to be developed to meet the challenges posed by the requirements of product miniaturization and
industrial realization of nanotechnology (3).
Micromachining can be carried out by techniques based on mechanical processes such as turning, drilling, milling, and grinding;
electrophysical and chemical processes such as electro-discharge machining (EDM) and electrochemical machining (ECM); and
energy-beam machining such as laser, electron, and focus ion beam. Beam-based micromachining using laser beam, electron beam,
or ion beams has some limitations due to poor control of 3D structures, low material removal rate, and low aspect ratio. The
mechanical micromachining processes, where the tools are in direct mechanical contact with the workpiece, usually have higher
material removal rate and are capable of generating 2D and 3D microfeatures on a wide range of materials (4). However, the
problem associated with the mechanical micromachining processes are difficulty or inability of machining hard and difficult-to-cut
materials like cemented carbides, ceramics, nickel, and titanium-based superalloys (5), deformation and/or breakage of smaller
parts or tools during the miniaturization due to the contact force between the tool and workpiece (6), more frequent tool failure
during machining hard materials (5), and lack of mass production capabilities (7).
In order to overcome the technical difficulties in conventional micromachining processes and high costs associated with the
elevated hardness and intrinsic brittleness, nonconventional electrophysical and chemical micromachining processes are increas-
ingly attempted for the machining of difficult-to-cut materials, particularly for applications where dimensional accuracy with
complex geometries are primary requirements. Among the electrophysical and chemical micromachining processes, micro-electro-
discharge machining (micro-EDM) and micro-electro-chemical machining (micro-ECM) are the only methods capable of
machining electrically conductive materials irrespective of their hardness and strength (8). However, the problem associated with
the micro-ECM process is that it generates a resistant oxide layer on the machined metal surface promoting a very slow material
removal rate. Due to a lower material removal rate and ability to provide a shiny surface, so far micro-ECM has been reported to be
used for a finishing purpose only for the machining of hard materials (9). As a result, micro-EDM has become one of the important
and cost-effective nonconventional methods of machining extremely hard and brittle materials (10). The major advantage of EDM

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications 335

or micro-EDM over the conventional machining process is that it is an electrothermal process of removing metal regardless of
hardness where the force between the workpiece and tool is negligible. Thus, the error caused by the tool deformation due to force is
almost zero (11). Furthermore, there are no chatters, mechanical stress, and vibration problem during the machining, as there is no
direct contact between the electrode and the workpiece (12). The noncontact machining process has been endlessly evolving from
a mere tool and die-making process to a microscale application machining. Micro-EDM has similar characteristics as EDM except
that the size of the tool, discharge energy, and axis movement resolutions are in micron level (13).
The EDM was first started in the USSR in the 1940s, when two Soviet husband and wife scientists, Doctors B.R. and N.I.
Lazarenko, first applied it to a machine for stock removal (12). The die-sinking process was defined as early as in the 1940s (12) with
the advent of the pulse generators, planetary and orbital motion techniques, computer numerical control (CNC), and the adaptive
control mechanism. The evolution of the wire-EDM in the 1970s (14) was due to powerful generators, new wire tool electrodes,
better mechanical concepts, improved machine intelligence, and better flushing. The early demonstration of micro-EDM was done
in 1968 by Kurafuji and Masuzawa (15), who achieved drilling of a minute hole of several microns in a 50-mm thick carbide plate.
Later in 1985, Masuzawa et al. invented the wire electrodischarge grinding (WEDG) technique (16), making micro-EDM capable of
machining microelectrode down to several microns on-machine.

11.14.2 Brief Overview of Micro-EDM

11.14.2.1 Working Principle
Micro-Electrical Discharge Machining is the process of machining electrically conductive materials by using precisely controlled
sparks that occur between an electrode and a workpiece in the presence of a dielectric fluid (17). Micro-EDM is based on the erosion
of electrically conductive materials through the series of spatially discrete high-frequency electrical discharges (sparks) between the
tool and the workpiece (18). Figure 1 illustrates that each spark occurs between the closest points of the electrode and the workpiece
(17). The spark removes material from both the electrode and workpiece, which increases the sparking gap (distance between the
electrode and the workpiece) at that point. This causes the next spark to occur at the next-closest point between the electrode and
workpiece. As micro-EDM is a thermal process where material is removed by heat. Every discharge (or spark) melts a small amount
of material from both of the electrodes. Part of this material is removed by the dielectric fluid and the remaining amount solidifies on
the surface of the electrodes. The net result is that each discharge leaves a small crater on both the workpiece and tool electrode (19).

11.14.2.2 Sparking and Gap Phenomena of Micro-EDM

The sparking phenomena during micro-EDM can be separated into three important phases named as preparation phase for ignition,
discharge phase, and interval phase between discharges (20). When the gap voltage is applied, an electric field or energy column is
created. The strength of the energy column increases with the decrease of the distance between the electrode and workpiece. The
point of least resistance between the tool electrode and workpiece determine the discharge locations. As the electrode approaches the
workpiece, the electrical field eventually breaks down the insulating properties of the dielectric fluid. The resistance of the dielectric
fluid decreases because of the heating due to the electric field and later due to the liberation of debris particles removed as a result of
sparking. The electric field results in the large amount of current flow through the gap helping in material removal. The material
removal mainly occurs by melting and evaporation of both the workpiece and electrode material in addition to the dielectric. This
sparking phenomenon produces metal debris particles as well as sheath of gasses composed of hydrogen, carbon, and various
oxides. In order for the micro-EDM process to be stable, the proper selection of pulse duration and pulse interval is necessary. During
the pulse interval, the current is switched off and the heat source is thereby eliminated and the sheath of vapor around the spark

Figure 1 Representation of sparking and gap phenomena in EDM; (a) model of EDM gap phenomena and (b) discharge phenomena in EDM gap.
Reproduced from Kunieda, M.; Lauwers, B.; Rajurkar, K. P.; Schumacher, B. M. Advancing EDM through Fundamental Insight into the Process.
Ann. CIRP 2005, 54, 599–622.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
336 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications

implodes. The collapse of the vapor sheath creates a void or vacuum and draws in fresh dielectric fluid to flush away debris and cool
the area. In addition, during the pulse interval, the reionization occurs which provides a favorable condition for the next spark.
Figure 1 illustrates the phase of the electrical discharges, sparking, and gap phenomena during the EDM process (21).

11.14.2.3 Differences between Macro- and Micro-EDM

The physical principle of the micro-EDM process is essentially similar to that of the EDM process with some differences to be
discussed here. The machining process is driven by an assigned and controlled gap, voltage, energy, and frequency of discharge.
High frequencies (>200 Hz) and small energies (106–107 J) for every discharge (40–100 V) are required to obtain high accuracy
and good surface qualities (roughness of about 0.1 mm) (22). Even though micro-EDM is based on the same physical principle of
spark erosion, it is not merely an adoption of the EDM process for machining at micron level. There are significant differences in the
size of the tool used, fabrication method of micro-sized tools, the power supply of discharge energy, movement resolution of
machine tools’ axes, gap control, and flushing techniques, and also in the processing technique (3,13,23,24). For example, micro-
EDM milling, WEDG, and repetitive pattern transfer are commonly employed in and more specific to the micro-EDM process. Some
other differences between macro- and micro-EDM are listed below:
l The most important difference between micro-EDM and EDM (for both wire and die-sinking EDM) is the dimension of the
plasma channel radius that arises during the spark: in conventional EDM is much smaller than the electrode but the size is
comparable for micro-EDM (25).
l Smaller electrodes (micro-WEDG and micro-block electro-discharge grinding (BEDG) can produce electrodes as small as Ø5 mm
and thin wires can be <Ø20 mm) used in the micro-EDM process present a limited heat conduction and low mass to dissipate
the spark heat. Excessive spark energy can produce the wire rupture (or electrode burn in die-sinking micro-EDM), being the
maximum applicable energy limited by this fact in micro-EDM (24,26).
l Together with the energy effects, the flushing pressure acting on the electrode greatly varies in micro-EDM with respect to the
conventional EDM process: the electrode pressure area is smaller but the electrode stiffness is lower, increasing the risk of
electrode breakage or tool deflection. The debris removal is more difficult in micro-EDM because the gap is smaller, the dielectric
viscosity is high, and the pressure drop in microvolumes is higher (26).
l In the conventional EDM, the higher precision can only be achieved if electrode vibrations and wear are controlled. On the other
hand, the precision and accuracy of the final products are much higher in micro-EDM (26).
l For each discharge, the electrode wear in micro-EDM is proportionally higher than in conventional EDM. The electrode is
softened, depending on the section reduction on the spark energy (27).
l In micro-EDM, the maximum peak energy must be limited to control the unit removal rate per spark (UR (24,27)) and use small
electrodes and wires. Due to lower peak energy, the crater size is also much smaller in micro-EDM. Figure 2 shows a comparison
of single crater sizes generated during conventional EDM and micro-EDM.

11.14.3 Micro-EDM System Components

11.14.3.1 Pulse Generator or Discharging Unit
11.14.3.1.1 Transistor-Type Pulse Generator
The transistor-type pulse generator is widely used in conventional EDM as it provides a higher removal rate due to its high discharge
frequency. The pulse duration and discharge current can arbitrarily be changed depending on the machining characteristics required.
A series of resistances and transistors are connected in parallel between the direct current power supply and the discharge gap. The
discharge current proportionally increases to the number of transistors, which is switched on at the same time. The switching
ON–OFF of the gate control circuit is operated by the FET. In order to generate a single pulse, gap voltage is monitored to detect the
occurrence of discharge and after preset discharge duration, the FET is switched off. However, there is a delay in signal transmission

Figure 2 Comparison between crater dimensions in (a) conventional EDM (left) and (b) micro-EDM (right). Reproduced from Uhlmann, E.;
Piltz, S.; Doll, U. Machining of Micro/Miniature Dies and Moulds by Electrical Discharge Machining – Recent Development. J. Mater. Process.
Technol. 2005, 167, 488–493.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications 337

from the occurrence of discharge to the switching off of the FET due to the time constants in the voltage attenuation circuit, pulse
control circuit, and insulating circuit and gate drive circuit for the FET (29).
The applications of the transistor-type generator in micro-EDM were first studied by Masuzawa and Fujino (30), and they re-
ported on successfully generating a discharge pulse used for rough machining. Nakazawa et al. (31) and Hara et al. (32) also
conducted studies on the development of the transistor-type generator for micro-EDM and reported that it was difficult for them to
make sure that electrical breakdown occurs whenever open voltage is applied because the discharge delay time is not always shorter
than the pulse duration (29).
One of the major advantages of the transistor-type pulse generator is that the discharge process can be easily controlled by
detecting the discharge state in the gap in the transistor-type pulse generator. If the transistor type is used, it takes at least several tens
of nanoseconds for the discharge current to diminish to zero after detecting the occurrence of discharge because the electric circuit
for detecting the occurrence of discharge, the circuit for generating an output signal to switch off the power transistor, and the power
transistor itself have a certain amount of delay time. Hence, it is difficult to keep the constant discharge duration shorter than several
tens of ns using the transistor-type pulse generator (29).

11.14.3.1.2 RC-Type Pulse Generator

The RC-type pulse generator was the first type used for EDM, and it is still used in finishing and micromachining because the
conventional transistor pulse generators do not produce a constant-energy pulse that is sufficiently short (21,29). In an RC or
relaxation type circuit, discharge pulse duration is dominated by the capacitance of the capacitor and the inductance of the wire
connecting the capacitor to the workpiece and the tool (7). The frequency of discharge (discharge repetition rate) depends upon the
charging time, which is decided by the resistor (R) used in the circuit. Therefore, R should not be made very low because arcing
phenomenon can occur instead of sparking and a critical resistance is desirable that will prevent arcing (23). Discharge energy is
determined by the used capacitance and by the stray capacitance that exists between the electric feeders, tool electrode holder and
work table, and between the tool electrode and workpiece. This means the minimum discharge energy per pulse is determined by
the stray capacitance. Hence, all the charge stored in the stray capacitance is discharged to the working gap together with the charge
stored in the capacitor wired to the circuit. In the final finishing, when minimum discharge energy is necessary, the capacitor is not
wired and machining is conducted with the stray capacitance only (7). It can easily generate pulses with high peak current values and
short duration, allowing efficient and accurate material removal, and meanwhile achieving the required surface quality. Finally,
pulse conditions with shorter discharge duration and higher peak current provide better surface roughness due to a smaller
discharge crater (21). Figure 3 shows the schematic representation of basic transistor- and RC-type pulse generators.
However, machining using the RC pulse generator usually has an extremely low removal rate from its low discharge frequency
due to the time needed to charge the capacitor. In addition, a uniform surface finish becomes difficult to obtain because the
discharge energy varies depending on the electrical charge stored in the capacitor before dielectric breakdown. The RC pulse
generator has no way to control the pulse interval. Moreover, thermal damage can easily occur on the workpiece if the dielectric
strength is not recovered after the previous discharge and the current continues to flow through the same plasma channel in the gap
without charging the capacitor (29).

11.14.3.1.3 Transistor-Type Isopulse Generator: Modification of Transistor-Type

The major problem of a transistor-type pulse generator is that the delay time in the transmission of detected signals is long. In order
to shorten this delay time, the transistor-type isopulse generator for micro-EDM shown in Figure 4(a) was developed (29). Rough
and semifinishing can be carried out by short-circuiting points P1 and P2, and finishing by inserting the supplementary circuit. In the
case of roughing and semifinishing, the discharge current is cut off by FET1. The pulse current is monitored to detect discharge
instead of monitoring the gap voltage. In addition, as the output of the current sensor is lower than 5 V, it can directly be input into
the pulse control circuit, eliminating the use of the voltage attenuation circuit. As a result, the developed delay time in the new
transistor-type isopulse generator has been shortened to a considerable extent, and the minimum pulse duration of about 80 ns was
obtained. In order to further shorten the pulse duration, it is necessary to shut off the discharge current without the use of a pulse
control circuit and gate drive circuit. The circuit encircled by the dotted line shown in Figure 4(a) was therefore developed. To
immediately shut off the discharge current when discharge occurs in the gap, this circuit is activated by disconnecting points P1 and
P2 and inserting the circuit between them. When discharge occurs in the gap, Tr1, Tr2, and Tr4 are turned on due to the discharge

Figure 3 Schematic representation of basic circuit diagram of (a) transistor-type and (b) RC-type pulse generator.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
338 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications

Figure 4 (a) Basic circuit diagram of transistor-type isopulse generator for micro-EDM (reproduced from Han, F.; Yamada, Y.; Kawakami, T.;
Kunieda, M. Improvement of Machining Characteristics of Micro-EDM Using Transistor Type Isopulse Generator and Servo Feed Control. Precis. Eng.
2004, 28, 378–385) and (b) capacity-coupled pulse generator developed for nano-EDM (reproduced from Kunieda, M.; Hayasaka, A.; Yang, X. D.;
Sano, S.; Araie, I. Study on Nano EDM Using Capacity Coupled Pulse Generator. Ann. CIRP 2007, 56 (1), 213–216).

current, resulting in FET2 going off. Since FET2 is turned off, the discharge is stopped and the discharge current becomes zero. Hence,
Tr1, Tr2, and Tr4 are turned off and FET2 is turned on, autonomously initializing the circuit.

11.14.3.1.4 Capacity-Coupled Pulse Generator: Modification of RC-Type

During the micro-EDM using RC-type pulse generator, the minimum electric discharge energy per pulse is determined by the stray
capacitance, indicating a limit to miniaturization (33). However, machining of microrods smaller than 1.0 mm in diameter has been
found difficult because discharge crater diameters smaller than 2.0 mm cannot be obtained using the conventional RC-type pulse
generator as stray capacitance cannot be completely eliminated (34,35). Therefore, a new pulse generator using the capacity coupling
method was developed (36). With this method, since electric feeding can be carried out without contact with the tool electrode, the
influence of the stray capacitance in the circuit can be eliminated, thereby realizing discharge craters of nanometer order.
As shown in Figure 4(b), the feeding electrode, tool electrode, and workpiece are coupled by capacity in series. In the figure, C1 is
the capacitance of the feeding gap between the feeding electrode and tool electrode, and C2 is the capacitance of the working gap
between tool electrode and workpiece. A pulse voltage V is applied between the feeding electrode and workpiece with a constant
pulse duration. The internal electric resistance of the pulse generator is R0. A definite distance is set between the feeding electrode and
tool electrode, in order to ensure that discharge does not occur and there is no conduction of electric charge between them. When the
voltage of the pulse generator becomes V, the capacitances of both the feeding gap and working gap, C1 and C2, are charged. In the
working gap, the tool electrode and workpiece are respectively charged positive and negative, creating a high electric field. Accord-
ingly, discharge occurs and electrons are conducted from the workpiece to the tool electrode. Since discharge duration is significantly
short, not longer than several tens of nanoseconds, dielectric breakdown strength of the working gap is immediately recovered.

11.14.3.2 Servo Control System or Gap Control Unit

In the micro-EDM process, to ensure stable and efficient machining, the gap condition between the electrode and workpiece must be
maintained at the desired state by servo-controlled electrode feed mechanism. The accurate detection of discharge states and a stable
servo feed control method are the key technologies that are the prerequisite and guarantee of high-efficiency and high-stability
processing (37). A stable gap control system also enables better dimensional accuracy of micromachined features (7).
During the micro-EDM, the discharging gap between the tool electrode and workpiece is several microns or even less (38) and
therefore requires design of special servo controls compared to conventional EDM. The microcontroller is the most important
module of a servo control system, which sends the control digital signal, translates analog to digital signal by an A/D module, and
drives the motor by the PWM module (38). To every microfeeding, shift detection result must be fed back to the microcontroller, to
adjust the output shift in order for precise control of the servo control system. The servo control can be operated based on different
algorithms or principles such as predicting the gap distance and offsetting the tool position, ignition delay time, average gap voltage,
the average delay time, etc. (7).
Ignition delay time (td) is an important indicator of the isolation condition of the discharge gap. Larger gap width causes longer
ignition delays, resulting in a higher average voltage. Tool feed speed increases when the measured average gap voltage is higher than
the preset servo reference voltage and vice versa (21). On the contrary, the feed speed decreases or the electrode is retracted when the
average gap voltage is lower than the servo reference voltage, which is the case for smaller gap widths resulting in a smaller ignition
delay. Thus, short circuits caused by debris particles and humps of discharge craters can be avoided. Also, quick changes in the
working surface area, when tool electrode shapes are complicated, does not result in hazardous machining. In some cases, the

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications 339

Figure 5 Principle of servo control system based on ignition delay. Reproduced from Kunieda, M.; Lauwers, B.; Rajurkar, K. P.;
Schumacher, B. M. Advancing EDM through Fundamental Insight into the Process. Ann. CIRP 2005, 54, 599–622.

Figure 6 Schematic of advanced servo feed control system. Reproduced from Kunieda, M.; Hayasaka, A.; Yang, X. D.; Sano, S.; Araie, I. Study on
Nano EDM Using Capacity Coupled Pulse Generator. Ann. CIRP 2007, 56 (1), 213–216.

average ignition delay time is used in place of the average gap voltage to monitor the gap width (39). In other attempts, gap
monitoring circuits were developed to identify the states and ratios of gap open, normal discharge, transient arcing, harmful arcing,
and short circuit (40,41). These ratios were used as input parameters for online EDM control based on various control strategies. The
servo feed control shown in Figure 5 keeps the working gap at a proper width.
Figure 6 shows the schematic representation of an advanced servo control system for high precision micro- and nano-EDM (36).
In an advanced servo control system for micro-EDM, the servo can be controlled by three methods: servo feed control of piezo table,
servo feed control of z-axis of the machine tool, and cooperative control of both piezo table and z-axis (36). With the servo feed
control method of piezo table, the average working gap voltage was compared with the reference servo voltage, and then the
difference value was amplified for the input to the piezo drive. The piezo table is proportionally displaced according to the input
voltage, thereby achieving servo control of the gap width. With the servo feed control method of the z-axis, the average working
voltage was also compared with the reference servo voltage, and then the difference value was input into the z-axis control drive
equipment. With the cooperative control method, the two controls described above were used together. Although the response of
the piezo table was faster than the z-axis of the machine, the movable range of about 18 mm was short. The cooperative control can
therefore ensure fast response in a broader moving range.

11.14.3.3 Dielectric Circulation System or Flushing Unit

The dielectric circulation system is the integral part of the micro-EDM system. The dielectric system consists of dielectric fluid,
dielectric reservoir, pump to supply the dielectric fluid into the work tank, filter to remove the debris particles from the dielectric and
to ensure recirculation of fresh dielectric to the machining zone, pipe and nozzles to supply the dielectric into the gap between the
workpiece and the electrode, and a flushing pressure control valve to maintain/control the flushing pressure of the dielectric.
Dielectric fluid is the most important component of the flushing unit, providing a known electrical barrier between the electrode
and workpiece, and acting as a means for removal of spark debris from the spark gap. The major functions of the dielectric
circulation system or flushing unit are as follows:
l To distribute the dielectric flow through the spark gap to remove gaseous and solid debris generated during EDM
l To introduce fresh and clean dielectric fluid to the cut
l To flush away the chips or metal particles generated in the spark gap
l To maintain the dielectric temperature well below its flash point
l To act as a cooler for cooling the electrode and workpiece

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
340 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications

11.14.4 Micro-EDM Process Control Parameters

11.14.4.1 Electrical Parameters
11.14.4.1.1 Discharge Energy
The discharge energy is the most important parameter responsible for the machining during the micro-EDM process and it is mainly
composed of other operating parameters and can represent the micro-EDM performance as a single parameter. The discharge energy
for micro-EDM is calculated based on the type of pulse generator. The discharge energy of a pulse generator can be determined from
its electrical and discharge parameters. The higher the discharge energy, the higher the material removal rate (MRR) will be. However,
the relative electrode wear also increases and surface finish deteriorates with the increase of discharge energy.
In the transistor-type pulse generator, when the transistors are switched on, the open voltage ue is applied between the tool electrode
and workpiece, but discharge does not immediately occur but occurs after the ignition delay time. After the dielectric breakdown,
a discharge current, ie, is passed through the gap. The gate control circuit keeps the transistors on for the discharge duration, te, after the
dielectric breakdown, resulting in a uniform discharge crater size. Then, after the fixed discharge interval, to, the transistors are again
switched on and open voltage is applied between the electrodes. The discharge energy per single pulse q is expressed as (21),
q ¼ ue  ie  te [1]
where ue is discharge voltage, ie is discharge current, and te is pulse duration.
For the RC-type, the charging time of capacitor (C) is considered as off-time or pulse interval, whereas the discharging time is
considered as pulse on-time. One important characteristic of an RC-type pulse generator is that the breakdown or discharging
voltage (V) is lower than the charging voltage, therefore, sometimes discharging starts before the capacitor is fully charged (29),
which creates nonuniform discharge energy. The peak current is the amount of current reached before starting discharging. The
discharge energy per single pulse q is expressed as (27),
q ¼ ð1=2ÞCV 2 [2]
where C is the capacitance used for machining and V is the discharging voltage.
Figure 7 shows the voltage and current characteristics for transistor- and RC-type pulse generators for ideal conditions. The actual
voltage and characteristic curves are shown in Figure 8.

11.14.4.1.2 Breakdown, Gap, Open-Circuit, and Discharge Voltage

Breakdown voltage is the threshold voltage at which the initiation of breakdown occurs. However, before current can flow, the open
gap voltage increases until it has created an ionization path through the dielectric. Once the current starts to flow, voltage drops and
stabilizes at the working gap level. The voltage between the gap of the electrode and workpiece is known as gap voltage. The applied
voltage determines the total energy of the spark. Higher voltage settings increase the gap, which improves the flushing conditions
and helps to stabilize the machining and increase MRR. But at the same time, higher voltage will also contribute to poor surface

Figure 7 Ideal voltage-time (top) and current-time (bottom) characteristics curve/waveform for (a) transistor-type and (b) RC-type pulse
generators. Reproduced from McGeough, J. A. Advanced Methods of Machining, 1st ed.; Chapman and Hall: USA, 1988. ISBN: 0-412-31970-5 and
Jahan, M. P.; Wong, Y. S.; Rahman, M. A Study on the Quality Micro-Hole Machining of Tungsten Carbide by Micro-EDM Process Using Transistor
and RC-Type Pulse Generator. J. Mater. Process. Technol. 2009, 209 (4), 1706–1716.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications 341

Figure 8 The actual voltage and current characteristics curve for (a) transistor-type and (b) RC-type pulse generators. Reproduced from
McGeough, J. A. Advanced Methods of Machining, 1st ed.; Chapman and Hall: USA, 1988. ISBN: 0-412-31970-5.

roughness. Discharge voltage is the voltage at which discharging occurs, thus machining happens and is dependent on the spark gap
and breakdown strength of the dielectric.

11.14.4.1.3 Peak Current

The term ‘peak current’ is often used to indicate the highest current during the machining. The higher the peak current, the larger the
discharge energy is. During each on-time pulse, the current increases until it reaches a preset level, which is expressed as the peak
current. Higher currents will improve MRR, but at the cost of surface finish and tool wear. The peak current mainly determines the
amount of power used in discharge machining, measured in units of amperage, and is the most important machining parameter in
micro-EDM.

11.14.4.1.4 Average Current

Average current is the average of the amperage in the spark gap measured over a complete cycle. The average current is maintained all
over the machining period and thus considered as effective discharge current. It is calculated by multiplying peak current by duty
factor.
Average current ðAÞ [ Duty factor ð%Þ 3 Peak current ðAÞ

11.14.4.1.5 Pulse Duration

Pulse duration is the period of time the current is allowed to flow per cycle during the micro-EDM process. The discharge energy is
really controlled by the peak current and the length of the pulse on-time. It is the ‘work’ part of the spark cycle, when the current
flows and work is done only during this time. Material removal is directly proportional to the amount of energy applied during this
time. With a longer period of spark duration, the resulting craters will be broader and deeper; therefore, the surface finish will be
rougher. Shorter spark duration, on the other hand, helps to obtain a fine surface finish. However, excessive pulse duration can be
counter-productive (44). Extended pulse duration also allows more heat to sink into the workpiece and spread, which means the
recast layer will be larger and the heat affected zone will be deeper. Longer pulse duration can also restrict electrodes from machining
by making the machining process unstable and producing frequent arcing and short-circuiting, thus resulting in more electrode
wear.

11.14.4.1.6 Pulse Interval

Pulse interval is the period of time between two successive pulse durations when the discharge is turned off. Pulse off-time is the
duration of the rest or pause required for reionization of the dielectric. This time allows the molten material to solidify and to be
washed out of the spark gap. If the pulse off-time is too short, it will cause sparks to be unstable, and then more short-circuiting will
occur. On the other hand, a higher pulse off-time results in higher machining time, but it can provide stability required to
successfully EDM a given application. When the pulse off-time is insufficient as compared to on-time, it will cause erratic cycling and
retraction of the advancing servo motor, slowing down the operation. Pulse interval will affect the speed and stability of the cut. In
theory, the shorter the interval, the faster the machining operation will be. However, pulse interval must be greater than the
deionization time to prevent continued sparking at one point (45).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
342 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications

11.14.4.1.7 Duty Factor

Duty factor is a percentage of the pulse duration relative to the total cycle time. It is a measure of efficiency and is calculated by
dividing the on time by the total cycle time. Generally, a higher duty factor means increased cutting efficiency. It is calculated in
percentage by dividing pulse duration by the total cycle time (on-time þ off-time).
Pulse duration ðmsÞ
Duty factor ð%Þ ¼  100
Total cycle time ðmsÞ
11.14.4.1.8 Pulse Frequency
Pulse frequency is the number of cycles produced across the gap in 1 s. The higher the frequency, finer is the surface finish that can be
obtained. With an increase of number of cycles per second, the length of the on-time decreases. Short on-times remove very little
material and create smaller craters. This produces a smoother surface finish with less thermal damage to the workpiece. Pulse
frequency is calculated by dividing 1000 by the total cycle time (on-time þ off-time) in microseconds (44).
1000
Pulse frequency ðkHzÞ ¼
Total cycle time ðmsÞ

11.14.4.1.9 Pulse Waveform Shapes and Types

As can be seen from Figure 7(a), the ideal pulse shapes for the transistor-type pulse generator are rectangular. However, the pulse
shape can be changed to trapezoidal, triangular, etc. It has been reported by Bruyn (46) that using trapezoidal shaped pulses instead
of rectangular pulses can improve the performance by reducing the electrode wear ratio. In addition, different types of pulses, such
as open circuit pulses, normal discharge pulses, arcing pulses, and short-circuit pulses, have a significant effect on the micro-EDM
performance and necessarily depends on the stability of the micro-EDM process. Arc and short-circuit pulses are considered as
harmful pulses and normal pulses are the most desired type of pulses. The objective of any successful micro-EDM process is to
increase the percentage of normal discharge pulses and reduce the percentage of ineffective pulses. A different pulse discrimination
system or adaptive control system can be used to calculate and control different percentages of pulses, thus controlling the
machining stability and performance of micro-EDM (8,47,48). Figures 9 and 10 show different types of pulses in micro-EDM for
transistor and RC-type pulse generators, respectively. In addition, different shapes of pulses are presented in Figure 11.

Figure 9 Different types of pulses for a typical transistor-type pulse generator. Reproduced from Snoeys, R.; Dauw, D.; Jennes, M. Survey of Adaptive
Control and Detection Systems. Ann. CIRP 1982, 31 (2), 483–489.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications 343

Figure 10 Different types of pulses for a typical RC-type pulse generator. Reproduced from Yeo, S. H.; Aligiri, E.; Tan, P. C.; Zarepour, H. An
Adaptive Speed Control System for Micro Electro Discharge Machining. In CPl 181, Third Manufacturing Engineering Society International
Conference; Segui, V. J., Reig, M. J., Eds.; American Institute of Physics, 2009; pp 61–72.

Figure 11 Different shaped pulse waveforms (voltage – time characteristic curve): (a) RC-type relaxation pulse generator, (b) rotary impulse
generator with rectifier, (c) transistor-type electronic pulse generator, and (d) modified transistor-type pulse electronic pulse generator (combination
of R–C circuit with transistor-type pulse generator). Reproduced from Module 9 Non-Conventional Machining – Lesson 39, Electro Discharge
Machining. Version 2 ME, IIT, Kharagpur.

11.14.4.1.10 Polarity
Generally, during the micro-EDM process, electrons are emitted from the cathode and move toward the anode. After reaching the
anode, the electrons strike the anode surface to cause metal ion to be removed from the anode material. Therefore, it is the anode
that losses more weight due to more material removal from its surface. This is the more common reason for getting a high material
removal rate when the workpiece is anode and the electrode is used as a cathode (43,51). During the micro-EDM or EDM,
a convention of straight and reverse polarity is used. Straight polarity is known as the polarity in which the electrode is usually
a cathode (). Here, work surface energy can be controlled by controlling the current so that anode drop energy provides proper
wear and the desired surface finish. On the other hand, the polarity is called reverse polarity, in which electrode anode (þ) and work
cathode (), and in which rough cut higher cutting rates can be obtained with virtually no electrode wear.

11.14.4.2 Nonelectrical Parameters: Materials Parameters

11.14.4.2.1 Electrode Materials
Since micro-EDM is a thermal process, the influence of thermal properties of electrode materials on the performance of micro-EDM
is significant. When the heat flux from the arc column is equal, higher heat conductivity results in lower temperature on the electrode
surface (21). Hence, materials with higher heat conductivity are suitable as tool electrodes. Materials with higher melting and

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
344 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications

Table 1 Common electrode and wire materials used in micro-EDM

No. Electrode material Wear ratio Material removal rate Fabrication Cost Application

1 Copper Low High on rough range Easy High On all metals

2 Brass High High only on finishing Easy Low On all metals
3 Tungsten Lowest Low Difficult High Small holes are drilled
4 Tungsten copper alloy Low Low Difficult High Used higher accuracy work
5 Cast iron Low Low Easy Low Used on few materials
6 Steel High Low Easy Low Used for finishing work
7 Zinc-based alloy High High on rough range Easy die casted High On all metals
8 Copper graphite Low High Difficult High On all metals

Reproduced from Pandey, P. C.; Shan, H. S. Modern Machining Process; Tata McGraw-Hill Publishing Company Ltd, 1999, 84–113.

boiling points are also suitable as tool electrodes. The important properties of the electrode materials influencing the micro-EDM
process include thermal and electrical conductivity, melting and boiling temperature, and specific heat of the materials (52). The
common electrode and wire materials used in micro-EDM are listed in Table 1 (53).

11.14.4.2.2 Wire Electrode Materials for Micro-Wire Electro-Discharge Machining

The performance of the micro-wire electro-discharge machining (WEDM), i.e., the MRR, wire breakage, and speed of the machining
are influenced by electrical properties, thermal properties, as well as tensile strength of the wire materials. The wire materials should
possess a high discharge capacity, low electrical resistance, high tensile strength at high temp, and reasonably low cost. In addition,
thermal characteristics of the wire materials like specific heat, heat conductivity, and melting and boiling points should also be taken
into consideration when selecting proper wire materials for micro-WEDM. Table 2 presents the wire electrode materials, diameters,
and application in WEDM process (54).

11.14.4.2.3 Workpiece Materials

The workpiece materials should have electrical conductivity enough to be machinable by micro-EDM. The EDM machinability of
a workpiece material depends on the thermal and conductivity, specific heat, and melting and evaporation points (55).

11.14.4.2.4 Dielectric Materials

During the micro-EDM process as the machining zone is immersed in the dielectric media, the properties of dielectric-like
chemical compositions, viscosity, dielectric strengths, and cooling rates, etc. play an important role in micro-EDM process.
Moreover, the dielectric fluid also serves several functions such as to flush the debris from the machined zone and act as
a coolant. The higher the flash point temperature and dielectric strength, the safer and finer degree of controlled sparking
during micro-EDM. The lower the viscosity of the dielectric fluid provides better accuracy and finishing during micro-EDM.
Lower specific gravity and colorless dielectric is more desirable for better performance (56). The micro-EDM performance,
especially the tool wear ratio (TWR) and MRR are affected by the dielectric materials and the method of dielectric flushing
directly influences the final surface finish. Table 3 presents a comparative performance analysis of three commonly used
dielectric materials (5).

Table 2 Wire electrode as tool in WEDM

Application based on electrode wire diameter

Wire diameter (mm) Application

25–50 Intricate openings

70–100 Tight radius, slots, and holes
100–300 Internal and external features

Application based on electrode material

Wire material Application

Brass All metals, holes

Copper All metals, holes
Tungsten All metals (especially refractory metals),
small slots, or holes
Copper–Tungsten All metals, carbide slots, thin slots
Steel Nonferrous, holes
Molybdenum Refractory, holes

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications 345

Table 3 Micro-EDM performance of hydrocarbon oil, deionized water, and oxygen as dielectric

Hydrocarbon oil as dielectric Deionized water as dielectric Oxygen as dielectric

No electrolytic damage Electrolysis occurs, which may cause No electrolytic damage

electrolytic damage
More susceptible to Susceptible to corrosion and rusting due No corrosion but may suffer from
thermal damage and microcracking to electrolysis rusting due to oxidation
Machined surface becomes Less hardening of machined surface as Hardening may occur
harder and brittle water also acts as coolant
Lower surface roughness Comparatively poorer surface Intermediate or poor surface
and shiny surface finish finish
Lower machining speed Comparatively faster machining than Higher or intermediate machining
compared to water as dielectric oil-based dielectric speed
Less electrode wear Severe electrode wear Lowest electrode wear
Better process stability Intermediate process stability Poor process stability
Lower operating and Comparatively higher operating Lower maintenance and operating
maintenance cost than water and maintenance costs costs
Mainly used in die-sinking EDM Mainly used for wire EDM Mainly for milling EDM

11.14.4.3 Motion and Gap Control Parameters

11.14.4.3.1 Servo Feed and EDM Speed
The main purpose of the servo feed control is to maintain proper spark gap or gap width during the machining in addition to
making the process more stable by minimizing the open circuit, arcing, and short-circuiting during machining. The servo feed rate
increases when the measured average gap voltage is higher than the preset servo reference voltage and vice versa (21).
The ‘EDM gap control speed’ used in the micro-EDM process is similar to the cutting feed rate of conventional machining. The
EDM speed controls the speed at which the electrode approaches the workpiece, as in conventional machining the feed rate
determines the cutting speed. To maintain continuous spark in between the electrode and workpiece, the speed of the z-axis is
controlled based on the following equation (57):
 
Fz ¼ k sgn Vgap  Vth

where, Fz is the z-axis feed rate (servo feed rate), Vgap is the gap voltage between the electrode and workpiece, Vth is the threshold
value for the gap control, and ‘k’ is a control parameter that determines the speed of the micro-EDM gap control. This k has been
defined as the ‘EDM gap control speed’ parameter.
Higher EDM speed is necessary for higher MRR. However, too high a value of EDM speed will only create an unfavorable
machining condition by frequent short circuiting, causing delay rather than increasing MRR, and increasing the TWR at the same
time. Therefore, optimal selection of servo feed rate and EDM speed is necessary for stable and improved micro-EDM performance.

11.14.4.3.2 Electrode Rotation

The performance of the micro-EDM process can be significantly enhanced by introducing the rotary motion of the tool electrode.
The electrode rotation improves the flushing process in micro-EDM, which contributes in improving the overall dimensional
accuracy and surface finish. With the increase of electrode rotational speed, the tangential velocities of the electrode increases, which
promotes the disturbance of the dielectric (58). The increased flow speed of the dielectric helps to depart the debris from the
machined zone, thus facilitating further material removal from the workpiece. The relative electrode wears decrease with the
increase of electrode rotational speed.

11.14.4.3.3 Workpiece Rotation

In addition to electrode rotation, the rotation of the workpiece was used for improving the dielectric flushing. Guu and Hocheng
(59) provided a workpiece rotary motion to improve the circulation of the dielectric fluid in the spark gap and temperature
distribution of the workpiece yielding improved MRR and surface roughness (SR). On the other hand, Kunieda and Masuzawa (60)
proposed a horizontal EDM (HEDM) process where change in the basic construction in addition to the rotary motion of the
workpiece offered an accessible evacuation of debris improving the erosion efficiency and accuracy of the sparking process. HEDM
has also been experimented in the micromachining of small parts (61,62).

11.14.4.3.4 Tool and Electrode Geometry, Design, and Shape

Tool electrode geometry is concerned with the shape of the tool electrodes, i.e., square, rectangle, cylindrical, circular, etc. It is found
that shape of the electrode effects the electrode wear ratio (EWR). The tool having less aspect ratio gave a higher value of EWR
(63,64). In addition, the electrode design and shape can certainly improve the flushing condition and overall performance of the
micro-EDM. Improved flushing of debris has been reported by using a single-side notch electrode compared to a cylindrical
electrode (58). Using a helical micro-tool electrode for micro-EDM combined with ultrasonic vibration can substantially reduce the
EDM gap, taper, and machining time for deep microhole drilling (65).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
346 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications

11.14.4.3.5 Wire Tension for Micro-WEDM

The amount of wire tension affects the dynamic stability condition of the micro-WEDM process. The deflection of the wire happens
due to different kind of forces working on it, such as electromagnetic force, flushing pressure, and pressure of the spark (54,66). If
tension is less, there is a greater chance of wire bending and also inaccuracy in machining. Because of continuous motion of the wire,
if proper tension is not maintained, there could be high vibration at the machining area. This can cause undesirable gap width,
excessive short circuit, and even wire breakage. Too high a wire tension again can cause the wire to often break.

11.14.4.3.6 Wire Speed and Wire Vibration

Wire speed is the velocity of the wire at which it moves across the workpiece during machining. The speed of the wire should not be
too high so as to reduce the usage of the wire. At very slow speeds, the wire tends to break more often, since the same region gets
eroded more, reducing the tensile strength of the wire. During the micro-WEDM, the external forces and vibration should be
eliminated as it can reduce the accuracy of the machined feature. The sources of wire vibration during the micro-WEDM are the axial
tension on the wire, electro-static force produced by the electric field between the workpiece and the wire electrode, electro-dynamic
and explosion force caused by spark discharge, and the damping force caused by the dielectric medium (54,67).

11.14.4.3.7 Tool Electrode and Workpiece Vibration

In recent years, the application of tool electrode and workpiece vibration has been found to significantly improve the micro-EDM
performance, especially in micro-EDM drilling of deep holes. The working gap between the electrode and the workpiece changes
due to the forward and backward movements of tool electrode or workpiece toward each other. As a result, the pressure of the
dielectric liquid inside the spark gap continuously changes. When the working gap becomes smaller, the dielectric liquid is forced
out of the gap and when the gap between the tool and workpiece increases again, fresh dielectric is drawn into the gap that will
eventually improve the overall flushing process. The higher efficiency gained by the application of tool or workpiece vibration is
mainly attributed to the improvement in dielectric circulation, which facilitates the debris removal and the creation of a continuous
pressure change between the electrode and the workpiece, as an enhancement of molten metal ejection from the surface of the
workpiece (68,69). The tool and workpiece vibration can be ultrasonic (70) or a low frequency vibration (68,69). It was reported
that employing an ultrasonic vibration can distinctly improve the machining quality and the efficiency (70,71). Moreover, a low
frequency workpiece vibration was also found to significantly improve the micro-EDM drilling performance by reducing the
number of ineffective pulses, decreasing the electrode wear ratio, and increasing the MRR and aspect ratio of the microholes
(68,69,72).

11.14.4.3.8 Types of Dielectric Flushing Process

Flushing removes eroded particles from the gap for efficient cutting and improved surface finish of machined material. Flushing
also enables fresh dielectric oil flow into the gap and cools both the electrode and the workpiece. The two important types of
flushing are the injection, or pressure flushing, and suction flushing. The jet pressure flushing through the hole of the electrode
usually uses higher flushing pressure, and side pressure flushing usually uses lower flushing pressure. Jet flushing was found to
be more effective for higher aspect ratio microholes, whereas side flushing was found to be useful in other applications and
more commonly used in micro-EDM. Both the pressure and suction flushing was found to significantly reduce the electrode
wear rate during the micro-EDM drilling and produces larger gap width at the exit compared to the inlet (73). A special rotary
electrode movement to enhance the pumping action of dielectric fluid during the lifting motion has been applied (74). Orbiting
of the tool or workpiece has also been found to assist flushing and improve machining conditions (75). In addition, the
flushing direction can have significant influence on the machining performance. Flushing from one direction can cause
increased density of debris particles in the downstream, resulting in uneven distribution of gap width deteriorating the
machining accuracy (76). Therefore, sometimes flushing from both sides, alternate flushing, and sweeping flushing is
preferable.

11.14.4.3.9 Flushing Pressure

During micro-EDM to maintain stable machining, it is critical to flush debris particles and cool the working gap in order to prevent
the localization and concentration of discharge locations (21). High flushing pressure can improve the overall flushing mechanism;
improve machining stability and MRR during micro-EDM, especially in micro-EDM drilling. However, very high pressure can
increase position error in addition to reducing dimensional accuracy due to deflection of the thin electrode used in micro-EDM.
Pressure or suction flushing through holes in the electrode or workpiece remains one of the most efficient flushing methods at
least if those holes have to be provided anyway or does not harm the workpiece.

11.14.5 Micro-EDM Performance Measurement Parameters

11.14.5.1 Micro-EDM Erosion Efficiency
Although the micro-EDM performance is mostly evaluated by MRR, EWR and surface roughness, a new performance characteristic
termed as micro-EDM ‘erosion efficiency’ was calculated by Wong et al. (23) to present the effectiveness of micro-EDM. During the
micro-EDM, a part of supplied energy from the spark forms the microcrater, which determines erosion efficiency (h), i.e., the ratio of

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications 347

the actual energy (Ee) used to erode the microcrater to the supplied energy in the gap (Eg). The remaining energy supplied between
the gap will be equal to the energy lost in the anode, dielectric medium, etc.
Ee Actural erosion energy
h¼ ¼
Eg Supplied erosion energy

When the energy is supplied, the erosion of material can first occur by melting and/or then by vaporizing as the heat is being
conducted to the electrode. The average efficiency of erosion, when estimated to be primarily due to melting or evaporation alone, is
found to be up to an order of magnitude higher at lower-energy discharges than that at higher-energy discharges.

11.14.5.2 Material Removal Rate

The material removal rate is defined as the amount of material removed from the workpiece per unit time. The material removal rate
can be calculated from the volume of material removal or from the weight difference before and after machining. It is an indication
of how fast or slow the machining rate is and an important performance parameter in micro-EDM, as this is usually a very slow
process. Higher machining productivity must also be achieved with a desired accuracy and surface finish. The MRR greatly depends
on the process parameters. A higher value of discharging voltage, peak current, pulse duration, duty cycle, and lower values of pulse
interval can result in higher MRR. In addition to these electrical parameters, other nonelectrical parameters and material properties
have significant influence on MRR.

11.14.5.3 Electrode Wear Ratio

The electrode wear ratio is the most commonly used terminology for getting information about electrode wear and is defined as the
volume of metal lost from the tool electrode divided by the volume of metal removed from the workpiece. High tool wear rates result
in inaccurate machining and considerably add to the expense since the tool electrode itself must be first accurately machined. To
reduce the influence of the electrode wear, it is necessary either to feed electrode larger than the workpiece thickness in the case of
making through-holes, or to prepare several electrodes for roughing and finishing in the present state of technology (11). Tool life is
an important concern in micro-EDM. During micro-EDM, the pulse condition with shorter discharge duration and lower peak current
brings about both lower tool electrode wear ratio and better surface roughness. It is seen that the tool wear characteristics are
associated with material properties, especially the boiling point (11). The volumetric wear ratio of the electrode becomes small for the
electrode material with high boiling point, high melting point, and high thermal conductivity, which is independent of the workpiece
materials (11,52). Corner wear of electrode relates to diffusion of heat and is more obvious in lower thermal conductivity electrodes.
The wear of the electrode is also related to such factors as the distribution of discharge power between both electrodes and the
thermodynamic constants of materials. The increase gap voltage, peak current, and pulse duration usually results in increased
electrode wear ratio. However, proper/optimum selections of parameters are necessary for controlled/minimized electrode wear.

11.14.5.4 Surface Roughness

During each electrical discharge, intense heat is generated that causes local melting or even evaporation of the workpiece material.
With each discharge a crater is formed on the workpiece. Some of the molten material is produced by the discharge, is carried away
by the dielectric circulation, and the remaining melt resolidifies to form an undulating terrain. In micro-EDM, the surface topog-
raphy and roughness are largely determined by eroded crater size and crater uniformity relating to discharge pulse energy (77,78).
Low open circuit voltage, shorter pulse duration with higher frequency, and enough higher pulse interval produces small craters and
hence less surface roughness (79). The surface roughness slightly increases with the increase of peak current value up to a certain
level and then vigorously increases with any increase in peak current (80). When compared against different dielectric fluid pressure,
surface roughness shows a slightly decreasing trend with increasing pressure. Again, depending on the electrical and thermal
properties of the workpiece and electrode material, the surface roughness varies (52).

11.14.5.5 Surface Integrity

The surface integrity during the micro-EDM mainly represents the subsurface quality, which is characterized by the presence of white
layer, surface cracks, and residual stress (81). The machined workpiece surface integrity also includes the microstructures,
composition, and hardness, under a wide range of machining conditions (82). The crack formation during micro-EDM is attributed
to the presence of thermal and tensile stresses within the machined component (83). Thermal stress is produced when the electrode
discharges bombard the surface of the sample during the machining process, and tensile stress is generated as a result of temperature
gradient at a different regime of the surface within the sample (83). The rapid cooling rate at the end of the discharges during EDM
causes surface defects such as cracks, plastic deformation, and the build-up of residual stresses (82). It has been confirmed from
extensive studies that there is an intrinsic relationship between the micro-EDM energy, integrity of the machined surface, and
mechanical properties of the machined surface. An increase in pulse discharge energy results in a thicker white layer and more
surface defects (84). In addition, the residual stress also increases if higher pulse energy is used for machining, resulting in more
numbers of microcracks beneath the surface (82). An increase in pulse current, pulse duration, or pulse voltage can impair the
quality and integrity of the surface (85). The inferior quality and integrity of the machined surface is associated with poor fracture

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
348 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications

strength, transverse rupture strength, fatigue strength, and other poor mechanical properties (86,87). In addition to pulse energy, the
quality and integrity of the machined surface is also influenced by the electrode material used for machining (81), flushing pressure
(82), type of dielectric fluid (84), amount of carbon or migrated materials (81), and overall, the workpiece materials.

11.14.5.6 Spark Gap/Gap Width/Overcut

For the die-sinking micro-EDM, the spark gap is defined as the distance between the electrode and the workpiece. For micro-WEDM,
the spark gap is more commonly addressed as kerf width or gap width and is measured by half or the value obtained by subtracting the
tool/wire diameter from the diameter of the machined periphery. This spark gap or kerf width affects the ability to achieve dimensional
accuracy and surface finishes. The lower and consistent in size the gap is, the more predictable the resulting dimension and machining
accuracy will be. During micro-EDM, in order to achieve micro-features the spark gap should be very small (88). However, minimizing
the spark gap is a real challenge in micro-EDM because minimizing the gap between the electrode and workpiece also tends to make
the process unstable (89). It has been found that the spark gap has a proportional relationship with the gap voltage (90). However, the
spark gap and surface roughness are also influenced by pulse on time. The peak current and the applied energy also influences the spark
gap. Therefore, the main parameters affecting spark gap were identified as open circuit voltage, peak current, and pulse on time.

11.14.6 Micro-EDM Process Varieties and Related Applications

11.14.6.1 Die-Sinking Micro-EDM
11.14.6.1.1 Working Principle
Die-sinking micro-EDM is the earliest and most common type of micro-EDM process. In die-sinking micro-EDM, an electrode with
desired micro-features is employed on the workpiece to produce corresponding mirror images. In die-sinking micro-EDM, the tool
electrode has the complementary form of the finished workpiece and literally sinks into the workpiece. The servo controller
monitors the gap conditions (voltage and current) and synchronously controls the different axes to machine the mirror image of the
tool. Figure 12 shows the schematic representation of the die-sinking micro-EDM (91).

11.14.6.1.2 Application Examples of Die-Sinking Micro-EDM

11.14.6.1.2.1 Fabrication of micromolds and microstructures
Micro die-sinking using microstructured form electrodes is mainly used for the manufacture of replication tools for microinjection
molding or hot embossing, where micromechanical parts can be produced in large numbers. High wear resistant composites based on
refractory materials like tungsten–copper or cemented carbide are preferred in micro die-sinking EDM. Conventional flushing strategies
cannot be used in micro die-sinking EDM, since flushing through the electrodes is not possible because of their small dimensions.
Together with the extremely small gap width, the poor flushing conditions put great demands on the feed control in die-sinking micro-
EDM (28). Figure 13 shows different micro die-sinking electrodes and fabricated microstructures using a complex shaped electrode.

11.14.6.2 Microwire Electro-Discharge Machining (Micro-WEDM)

11.14.6.2.1 Working Principle
In micro-WEDM, a continuously traveling microwire is used to cut through a conductive workpiece according to the programmed
path. The basic mechanism of micro-WEDM is the same as micro-EDM and the material removed as a result of a series of electric

Figure 12 Schematic showing the principle of die-sinking micro-EDM. Reproduced from Bleys, P.; Kruth, J. P.; Lauwers, B. Sensing and
Compensation of Tool Wear in Milling EDM. J. Mater. Process. Technol. 2004, 149, 139–146.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications 349

Figure 13 Micro die-sinking electrode for micromixing device made of fine-grained graphite machined (reproduced from Uhlmann, E.; Piltz, S.;
Doll, U. Machining of Micro/Miniature Dies and Moulds by Electrical Discharge Machining – Recent Development. J. Mater. Process. Technol. 2005,
167, 488–493), (b) fabricated microgear electrodes, and (c) gear-array microstructures by die-sinking micro-EDM. Reproduced from Tong, H.; Li, Y.;
Wang, Y. Experimental Research on Vibration Assisted EDM of Micro-Structures with Non-Circular Cross-Section. J. Mater. Process. Technol. 2008,
208 (1–3), 289–298.

Figure 14 Schematic showing the principle of micro-WEDM. Reproduced from Kunieda, M.; Lauwers, B.; Rajurkar, K. P.; Schumacher, B. M.
Advancing EDM through Fundamental Insight into the Process. Ann. CIRP 2005, 54, 599–622.

sparks between the workpiece and wire electrode. Since wire orientation can be changed by controlling the horizontal position of
the upper wire guide relative to the lower guide, all types of ruled surfaces can be cut by the wire-WEDM process (21). Figure 14
represents the schematic representation of sinking, milling, and wire micro-EDM.

11.14.6.2.2 Application Examples of Micro-WEDM

11.14.6.2.2.1 Micro-structuring using micro-WEDM
Micro-WEDM is the most commonly used process for the direct production of micromechanical devices and for microstamping
tools (28). Axisymmetric products can be machined by the microwire EDM method with a rotating workpiece system (93). The
capability of microwire EDM is fully exploited in the machining of a complex Chinese pagoda (1.25 mm  1.75 mm) shown in
Figure 15(a) (28,94,95). A characteristic for microwire electrical discharge machining is the application of the smallest wire elec-
trodes with wire diameters of 0.1 mm down to 0.02 mm, in order to machine the smallest cutting widths or inner corners in
microstructures. In combination with high precision machining systems and super-finishing spark generators, minimum slit widths
of 0.04 at machining accuracies in the range of 0.001 mm and aspect ratios of more than 100 can be achieved using micro-WEDM
(28). Structures with a width of 15 mm can be reproducibly machined by microwire electrical discharge machining with surface
roughness of less than 0.1 mm (28). Figure 15 shows different types of microstructures produced by micro-WEDM.

11.14.6.2.2.2 Fabrication of arrays of microelectrode and microholes

The micro-WEDM technique is very effective in the fabrication of in situ arrays of microfeatures. It is comparatively easier to control
the path and movement of wire in fabricating a large number of microelectrodes array unlike other micro-EDM based processes. The
fabrication of microelectrode arrays by micro-WEDM and their application in upward batch micro-EDM drilling for fabrication of
microholes was developed (96). In order to drill the microhole arrays in an efficient manner, a miniature of high-aspect-ratio
microstructure arrays with a (10  10) squared microbatch of electrodes is fabricated first. The microelectrode arrays are made of
tungsten carbide with a diameter of 800 mm cut with a horizontal wire from top to bottom (Figure 16(a) and 16(b)). To provide an
electrode with an adequate length during discharging, each micropillar is fabricated to have a length greater than 700 mm. Thereafter,
the arrays of microelectrodes are used to fabricate arrays of microholes using upward batch micro-EDM (Figure 16(c) and 16(d)).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
350 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications

Figure 15 Microstructuring by micro-WEDM. (a) Pagoda machined by microwire EDM (reproduced from Liao, Y.-S.; Chen, S.-T.; Lin, C.-S.
Development of a High Precision Tabletop Versatile CNC Wire-EDM for Making Intricate Micro Parts. J. Micromech. Microeng. 2005, 15 (2), 245–253),
(b) microwire electrical discharge machined gear with module 0.1; diameter of the applied wire electrode 30 mm (reproduced from Uhlmann, E.;
Piltz, S.; Doll, U. Machining of Micro/Miniature Dies and Moulds by Electrical Discharge Machining - Recent Development. J. Mater. Process. Technol. 2005,
167, 488–493), and (c) steel gear wheel made by micro-WEDM (reproduced from Schoth, A.; Forster, R.; Menz, W. Micro Wire EDM for High Aspect Ratio
3D Microstructuring of Ceramics and Metals. Microsyst. Technol. 2005, 11 (4–5), 250–253).

Figure 16 (a) Schematic illustration of microelectrode arrays using micro-WEDM, (b) illustration of upward batch micro-EDM, (c) fabricated arrays
of microelectrodes, (d) pattern transfer by upward batch micro-EDM. Reproduced from Chen, S.-T. Fabrication of High-Density Micro Holes by
Upward Batch Micro EDM. J. Micromech. Microeng. 2008, 18, 085002 (9 pp).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications 351

Figure 17 Schematic showing the principle of milling micro-EDM. Reproduced from Bleys, P.; Kruth, J. P.; Lauwers, B. Sensing and Compensation
of Tool Wear in Milling EDM. J. Mater. Process. Technol. 2004, 149, 139–146.

11.14.6.3 Milling Micro-EDM

11.14.6.3.1 Working Principle
Milling micro-EDM is a comparatively newer process, which eliminates the need for complex-shape electrodes usually required in
die-sinking. In this process, usually tubular or cylindrical microelectrodes are employed to produce the desired complex shape by
scanning. A cylindrical electrode rotates around its axis (Z-axis) with the scanning movements in X and Y directions. The contour of
a particular layer is specified in the part program of CNC. However, electrode compensation is an important factor to consider as
electrode length is reduced after scanning every layer. Figure 17 shows the schematic representation of milling micro-EDM (91).

11.14.6.3.2 Application Examples of Milling Micro-EDM

11.14.6.3.2.1 Fabrication of 3D micromolds, cavities, and dies
In recent years, micro-EDM milling is found to be a great alternative of conventional die-sinking micro-EDM for the fabrication
of 3D micromolds and dies. Micromolds with widely spread microstructures are very difficult to fabricate using micro-WEDM or
die-sinking micro-EDM due to their dimensions. Micro-EDM milling is mainly used when large and complex geometries are
required. Micro-EDM milling can be used, in which a path-controlled multiaxis feed motion is performed between rotating tool
electrode and workpiece. The use of geometrically simple rotating electrodes significantly decreases effort and costs for electrode
production. Either commercially available microelectrodes can be used or microelectrodes can be machined on-machine by the
microelectro-discharge grinding (EDG) process. During micro-EDM milling of a 3D microstructure, a specific CAD/CAM system is
needed for generating a tool path and related machining process. When a 3D microstructure is machined, it is necessary to
compensate the worn length of the electrode. Compensation of electrode wear can be obtained through evaluating the electrode
wear ratio. However, when a 3D cavity has an irregular geometry and the surface is very hard to describe by the general mathematical
equation, the evaluation of the electrode wear ratio becomes very difficult. Furthermore, the machining environment is not uniform
in the various part of the workpiece with complicated cavity. Therefore, the online measurement of the electrode wear is necessary
for maintaining the machining accuracy. Until now, many approaches for milling EDM and micro-EDM have been taken including
electrode wear prediction, sensing, modeling, and compensation. Some of the important approaches for wear compensation of
milling micro-EDM are the linear wear compensation method (LCM) (28,97), uniform wear method (UWM), combination of the
linear compensation and the uniform wear method (CLU) (98), on-line measurement compensation method, general wear
compensation method (101) etc. Figures 18–21 show microstructures fabricated by milling micro-EDM with different wear
compensation techniques of electrode (100–104).

11.14.6.3.2.2 Fabrication of fine features using micro-EDM milling

Micro-EDM milling can fabricate fine microfeatures with sharp and smooth edges and thus can replace the need for setting up
a micro-WEDM attachment and can save machining time. Moreover, these microfeatures with special shape can be fabricated more
easily compared to die-sinking where the electrodes need to fabricate. Figure 22 shows various fine features with sharp and burr-free
edges fabricated using micro-EDM milling (105).

11.14.6.4 Dry and Near-Dry Micro-EDM

11.14.6.4.1 Working Principle
Dry or near-dry micro-EDM is a recent variant of micro-EDM that has been found to decrease the pollution caused by the use of
liquid dielectric, which leads to the production of vapors during machining and adds cost to manage the waste, thus suitable for
green manufacturing. The dry or near-dry micro-EDM can be the die-sinking type or milling type. The dry micro-EDM uses oxygen or
air as the dielectric medium for micro-EDM (106). On the other hand, the near-dry micro-EDM uses liquid–gas mixture or mist as

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
352 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications

Figure 18 Micro-EDM milling with linear compensation method: (a) microcavity in hot forming tool steel using simple electrode of 100 mm
(reproduced from Uhlmann, E.; Piltz, S.; Doll, U. Machining of Micro/Miniature Dies and Moulds by Electrical Discharge Machining - Recent
Development. J. Mater. Process. Technol. 2005, 167, 488–493), (b) microsize pyramid (150 mm (L)  140 mm (W)  50 mm (H)) by micro-EDM
milling (reproduced from Asad, A. B. M. A.; Masaki, T.; Rahman, M.; Lim, H. S.; Wong, Y. S. Tool-Based Micro-Machining. J. Mater. Process.
Technol. 2007, 192–193, 204–211), and (c) small pyramid (L: 25 mm, W: 25 mm, H: 35 mm, step size 7 mm) by micro-EDM milling (reproduced
from Asad, A. B. M. A.; Masaki, T.; Rahman, M.; Lim, H. S.; Wong, Y. S. Tool-Based Micro-Machining. J. Mater. Process. Technol. 2007, 192–193,
204–211).

Figure 19 Micro-EDM milling with uniform wear compensation method: (a) 3D micromold of a car (reproduced from Yu, Z. Y.; Masuzawa, T.;
Fujino, M. Micro- EDM for Three-Dimensional Cavities – Development of Uniform Wear Method. Ann. CIRP 1998, 47 (1), 169–172), (b) 3D micro-
EDM using CAD/CAM (reproduced from Rajurkar, K. P.; Yu, Z. Y. 3D Micro-EDM Using CAD/CAM. Ann. CIRP 2000, 49 (1), 127–130), (c) 1/8 ball in
a square cavity (reproduced from Narasimhan, J.; Yu, Z. Y.; Rajurkar, K. P. Tool Wear Compensation and Path Generation in Micro and Macro
EDM. Trans. NAMRI/SME 2004, 32, 151–158).

Figure 20 (a) Simple cavity and (b) complex 3D cavity by the CLU method (combination of the LCM and the UWM). Reproduced from Yu, H.-L.;
Luan, J.-J.; Li, J.-Z.; Zhang, Y.-S.; Yu, Z.-Y.; Guo, D.-M. A New Electrode Wear Compensation Method for Improving Performance in 3D Micro EDM
Milling. J. Micromech. Microeng. 2010, 20 (5), 055011

the dielectric (107). The mist dielectric can be a mixture of gas medium (air/nitrogen) and liquid (water/kerosene) (107). In this
process, usually a thin walled tubular electrode is used through which high-pressure gas or air is supplied to the machining zone
(Figure 23). The role of the gas is to act as dielectric, remove the debris from the gap, and cooling of the interelectrode gap. Although
the dry and near dry micro-EDM has been found to be a variant of die-sinking or milling micro-EDM with gas medium of dielectric,
there is significant differences in plasma characterization, gap control, or even the material removal mechanism (108,109).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications 353

Figure 21 (a) Microtriangle cavity machined by using on-machine measurement (image processing) (reproduced from Yan, M.-T.; Huang, K.-Y.;
Lo, C.-Y. A Study on Electrode Wear Sensing and Compensation in Micro-EDM Using Machine Vision System. Int. J. Adv. Manuf. Technol. 2009, 42,
1065–1073), (b) 3D free form machining by on-machine measurement (longitudinal electrode wear) (reproduced from Zhao, W.; Yang, Y.; Wang, Z.;
Zhang, Y. A CAD/CAM System for Micro-ED-Milling of Small 3D Freeform Cavity. J. Mater. Process. Technol. 2004, 149, 573–578), and (c) platinum
hemisphere machined by micro-EDM milling with general wear compensation method (reproduced from Hang, G.; Caol, G.; Wang, Z.; Tang, J.;
Wang, Z.; Zhao, W. Micro-EDM Milling of Micro Platinum Hemisphere. In Proceedings of the 1st IEEE International Conference on Nano/Micro
Engineered and Molecular Systems, January 18–21, Zhuhai, China, 2006, pp 579–584).

Figure 22 Fine features (through) fabricated by micro-EDM milling; (a) microslots of 30 mm width (reproduced from Asad, A. B. M. A.; Masaki, T.;
Rahman, M.; Lim, H. S.; Wong, Y. S. Tool-Based Micro-Machining. J. Mater. Process. Technol. 2007, 192–193, 204–211), (b) micro spinneret of
12 mm width (reproduced from Asad, A. B. M. A.; Masaki, T.; Rahman, M.; Lim, H. S.; Wong, Y. S. Tool-Based Micro-Machining. J. Mater. Process.
Technol. 2007, 192 -193, 204–211), (c) two 10 mm slots with 2.5 mm thick separating wall on a 50 mm thick SUS 304 stainless steel (reproduced
from Asad, A. B. M. A.; Masaki, T.; Rahman, M.; Lim, H. S.; Wong, Y. S. Tool-Based Micro-Machining. J. Mater. Process. Technol. 2007, 192–193,
204–211), (d) microflower machined on 50 mm stainless steel plate (reproduced from Asad, A. B. M. A.; Masaki, T.; Rahman, M.; Lim, H. S.;
Wong, Y. S. Tool-Based Micro-Machining. J. Mater. Process. Technol. 2007, 192–193, 204–211), and (e) vacuum microgripper micronozzle machined
on alumina ceramic composite by micro-EDM milling process (reproduced from 105. Modica, F.; Ferraris, E.; Trotta, G.; Fassi, I.; Reynaerts, D. Fabrication
of Micro-Nozzles via m-EDM Process. In Proceedings of International Conference on Advances in Materials and Processing Technologies (AMPT 2010);
Chinesta, F., Chastel, Y., Mansori, M. E., Eds.; American Institute of Physics, 2010; pp 1261–1266. CP1315).

11.14.6.4.2 Application Examples of Dry and Near Dry Micro-EDM

11.14.6.4.2.1 Microstructuring by dry micro-EDM
The most common use of dry micro-EDM was found to be dry micro-EDM milling for the fabrication of 3D microfeatures and slots.
Because dry micro-EDM uses pipe electrodes with the gas dielectric flowing inside the pipe, the movement of the pipe is necessary for
better flushing, therefore, being used more in milling. In addition, dry micro-EDM using smaller pipe electrodes are found to provide
improved performance in the contouring of round surface compared to oil EDM milling and sinking in terms of reduced electrode wear
ratio and machining speed. Figure 24(a) and 24(b) show the fabrication of spherical surfaces and slots with fine edges using milling dry
micro-EDM (106,110). The machining of ‘Y-shaped microfeature’ and microhole are shown in Figure 24(c) and 24(d), respectively (110).

11.14.6.4.2.2 Near mirror finish machining by near dry micro-EDM

The application of near-dry EDM milling as a finishing process has been reported to achieve a mirrorlike surface finish (107,111).
A liquid–gas mist mixture is the dielectric medium delivered through a rotating tubular electrode in the near-dry EDM milling. Near-

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
354 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications

Figure 23 (a) Principle of the dry micro-EDM using high pressure gas/oxygen flow through the tube electrode (reproduced from Zhang, Q. H.;
Zhang, J. H.; Deng, J. X.; Qin, Y.; Niu, Z. W. Ultrasonic Vibration Electrical Discharge Machining in Gas. J. Mater. Process. Technol. 2002, 129,
135–138) and (b) principle of dry milling micro-EDM (reproduced from Tao, J.; Shih, A. J.; Ni, J. Near-Dry EDM Milling of Mirror-Like Surface
Finish. Int. J. Elect. Mach. 2008, 13, 29–33s).

Figure 24 (a) Machining of spherical surface by dry micro-EDM milling (reproduced from Yu, Z.; Jun, T.; Kunieda, M. Dry EDM of Cemented
Carbide. J. Mater. Process. Technol. 2004, 149, 353–357), (b) machining of fine slots using CuW pipe electrode by dry micro-EDM milling, (c) micro-
feature machined by dry micro-EDM contouring, and (d) microhole machined by dry micro-EDM drilling (reproduced from Yu, Z.; Takahashi, J.;
Nakajima, N.; Sano, S.; Karato, K.; Kunieda, M. Feasibility of 3-D Surface Machining by Dry EDM. Int. J. Elect. Mach. 2005, 10, 15–20).

dry EDM exhibits the advantage of good machining stability and smooth surface finish at low discharge energy input. The near dry
EDM milling has been performed to machine pockets with tubular electrode. Kerosene with air mist has been used as dielectric and
copper infiltrated graphite electrode is used for machining of pockets. The mirrorlike surface finish showing the reflection of the
electrode tip can be observed in the image. Figure 25 shows the examples of mirror surface produced by near dry micro-EDM
milling (107,111).

Figure 25 (a) Illustration of the mirrorlike machined surface (reproduced from Tao, J.; Shih, A. J.; Ni, J. Near-Dry EDM Milling of Mirror-Like
Surface Finish. Int. J. Elect. Mach. 2008, 13, 29s–33s) and (b) flat surface machining using a number of machining steps with different discharge
setup parameters in near dry micro-EDM milling (reproduced from Tao, J. Investigation of Dry and Near-Dry Electrical Discharge Milling Processes.
Ph.D. Thesis, The University of Michigan, 2008).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications 355

11.14.6.5 Micro-EDM Drilling

11.14.6.5.1 Working Principle
In micro-EDM drilling, microelectrodes are used to ‘drill’ microholes in the workpiece. However, the problem with deep small-hole
drilling by micro-EDM is that forming and clamping the long electrode are difficult. Therefore, the microelectrodes are used to
fabricate on-machine with high accuracy using different micro-EDG process shown in figure. This also solves the clamping problem,
because the electrode is clamped before the forming process and never reclamped until the hole is machined (112). Figure 26 shows
the consecutive process of on-machine electrode fabrication, measurement of microelectrode, and micro-EDM drilling with
fabricated electrode (113).

11.14.6.5.2 Application Examples of Micro-EDM Drilling

11.14.6.5.2.1 Fabrication of smaller and higher aspect ratio microholes
Microholes are the most basic features that can be machined by micro-EDM drilling process. Microholes are found in various
applications like fuel injection nozzles, spinneret holes, standard defects for testing material, biomedical filters, and so on
(13). Micro-EDM has been found inevitable in the fabrication of high-aspect-ratio microholes in difficult-to-cut materials,
where a conventional microdrilling process was found difficult. However, the microelectrodes have to be fabricated on-
machine using any of the micro-electrodischarge grinding processes in order to maintain high accuracy and reduce clamp-
ing and position error. Figure 27 shows some examples of very fine microelectrodes and microholes obtained by micro-EDM
drilling.
In most of the cases, micro-EDM drilling is vertical drilling by the assistance of an electric spark. Besides vertical type micro-EDM
drilling, horizontal type drilling was also established in order to easily remove the debris and improve the flushing conditions,
especially for deep-hole drilling (112). Using the developed horizontal system 50 mm diameter microholes with 10 times the aspect
ratio was fabricated (Figure 28).

11.14.6.5.2.2 Micro-EDM drilling of noncircular holes

The micro-EDM drilling process can be used for machining different noncircular holes such as triangular, rectangular, hexagonal
shapes, and so on. The machining of different shapes can be done either by incorporating the planetary or orbital movement or
fabricating the desired shaped electrode for the micro-EDM drilling process. The process of using planetary electrode movement is

Figure 26 Micro-EDM drilling process: (a) on-machine electrode fabrication by BEDG, (b) on-machine measure by laser, and (c) drilling of
high-aspect-ratio microholes. Reproduced from 113. Lim, H. S.; Wong, Y. S.; Rahman, M.; Lee, E. M. K. A Study on the Machining of
High-Aspect Ratio Micro-Structures Using Micro EDM. J. Mater. Process. Technol. 2003, 140, 318–325.

Figure 27 (a) Smallest hole of B 6.5 mm machined on 50 mm stainless steel plate (Reproduced from Asad, A. B. M. A.; Masaki, T.; Rahman, M.; Lim, H.
S.; Wong, Y. S. Tool-Based Micro-Machining. J. Mater. Process. Technol. 2007, 192–193, 204–211), (b) B 5 mm hole machined on 15 mm thick workpiece,
and (c) B 7 mm hole machined on copper workpiece (reproduced from Egashira, K.; Mizutani, K. EDM at Low Open-Circuit Voltage. Int. J. Electromach.
2005, 10, 21–26).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
356 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications

Figure 28 (a) Top view of B 50 mm hole with aspect ratio 10, (b) appearance at the entrance edge of the microhole, and (c) cross-section of
microhole along the center line (reproduced from Masuzawa, T.; Tsukamoto, J.; Fujino, M. Drilling of Deep Micro-Holes by EDM. Ann. CIRP 1989,
38 (1), 195–198).

Figure 29 Fabrication of noncircular microholes using micro-EDM: (a) triangular hole on stainless steel plate (reproduced from Rahman, M.;
Lim, H. S.; Neo, K. S.; Kumar, A. S.; Wong, Y. S.; Li, X. P. Tool-Based Nanofinishing and Micromachining. J. Mater. Process. Technol. 2007, 185
(1–3), 2–16), (b) rectangular hole machined on SUS 304 steel (reproduced from Rahman, M.; Asad, A. B. M. A.; Masaki, T.; Saleh, T.; Wong, Y. S.;
Kumar, A. S. A Multiprocess Machine Tool for Compound Micromachining. Int. J. Mach. Tools Manuf. 2010, 50 (4), 344–356), and (c) hexagonal
microholes on a copper strip (reproduced from Chern, G.-L.; Wang, S.-D. Punching of Noncircular Micro-Holes and Development of Micro-Forming.
Precis. Eng. 2007, 31 (3), 210–217).

termed as planetary or orbital EDM and will be discussed separately. The electrode can be turned into different shapes on the
machine using microelectro-discharge grinding processes or can be obtained in different shapes. The machining process and
parameters are similar to the micro-EDM drilling process. However, the flushing of debris may be difficult during the machining of
noncircular holes as electrode rotation cannot be used due to anisometric geometry. Therefore, special consideration might be
needed to improve the flushing of debris and machining stability. Figure 29(a) shows a triangular hole machined on a stainless steel
plate (114). To achieve the shape, a 500 mm tungsten electrode was machined to the triangular shape first using the WEDG process
followed by the micro-EDM drilling process. Figure 29(b) shows the fabrication of rectangular shape microhole fabricated on SUS
304 steel using a modified square electrode (3). Figure 29(c) shows the fabrication of the successive hexagonal microholes punched
on a copper strip using a micro-EDM process (115).

11.14.6.6 Microelectro-Discharge Grinding Process

11.14.6.6.1 Working Principles of Different Micro-EDG Processes
11.14.6.6.1.1 Block micro-EDG
One of the most commonly used micro-EDM variant is the microelectro-discharge grinding (micro-EDG). During micro-EDM, to
fabricate a microelectrode on-machine from an electrode thicker than the required one, the micro-EDG process with a sacrificial
electrode is used. Block micro-EDG is a simple process requiring a precise sacrificial rectangular block with high wear resistance (WC
was used in this study due to its high resistance to wear) and a commercially available electrode. However, one important thing is
the alignment of the block respective to the electrode. It is very important that the sacrificial block should be properly aligned
(within an accuracy of 2 mm) in order to avoid electrodes being more tapered, thus reducing dimensional accuracy. It has been
found that due to wear of the sacrificial block also, the diameter of the fabricated electrode is sometimes difficult to predict.
Therefore, an on-machine camera with a measuring unit is installed in the setup to measure on-machine. In this method, the block
is used as a cutting electrode and a cylindrical rod is used as the workpiece in the EDM process. The microelectrode that needs to be
machined is fed against the conductive block. The machining is carried out at different conditions by applying a controlled electric
spark and by forcing the dielectric medium to flow through the spark gap between the block and the rod. Figure 30 shows the
schematic diagram of the block micro-EDG process (6).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications 357

Figure 30 Schematic showing the principle of stationary block micro-EDG process: (a) at the beginning of the process and (b) the fabricated
micro-electrode. Reproduced from Jahan, M. P.; Rahman, M.; Wong, Y. S.; Fuhua, L. On-Machine Fabrication of High-Aspect-Ratio Micro-Electrodes
and Application in Vibration-Assisted Micro-Electrodischarge Drilling of Tungsten Carbide. Proc. Inst. Mech. Eng., Part B 2010, 224 (5), 795–814.

11.14.6.6.1.2 Moving block micro-EDG process

Although, the process is simple in mechanism, it needs several important considerations to fabricate successful, dimensionally accurate,
and very high aspect ratio microelectrodes. The major problem of block micro-EDG is the taperness of the fabricated microelectrodes.
Therefore, a new and modified block micro-EDG process was developed using either the scanning movement of the electrode or block
in addition to downward movement. The introduction of scanning motion significantly reduces the taperness of the microelectrodes.
The reason for taperness in stationary block is because, at the start of machining, more surface area is exposed to the spark zone, which
decreases with the block wear. However, if the electrode is moved along the block, then the wear of the block is uniformly distributed,
which reduces the taperness of the microelectrodes. Figure 31 shows the Block-mEDM process with moving electrode (6).

11.14.6.6.1.3 Micro-wire electrodischarge grinding

Micro-WEDG is a microfabrication process that uses electrical discharges in a dielectric fluid to erode material from conductive
microwires and produce microshafts. The discharges occur across a small gap (w2 mm) filled with dielectric oil. The workpiece is
held vertically in a mandrel that rotates at several thousand revolutions per minute and its position is slowly fed in the z direction.
The wire is supported on a wire guide and its position is controlled in the x- and y-directions. Each electrical discharge erodes
material from the workpiece and the anode wire. To prevent discharges from worn regions of the anode wire, the wire travels at

Figure 31 Schematic showing the principle of moving block micro-EDG: (a) at the beginning of the process and (b) the fabricated micro-electrode.
Reproduced from Jahan, M. P.; Rahman, M.; Wong, Y. S.; Fuhua, L. On-Machine Fabrication of High-Aspect-Ratio Micro-Electrodes and Application
in Vibration-Assisted Micro-Electrodischarge Drilling of Tungsten Carbide. Proc. Inst. Mech. Eng., Part B 2010, 224 (5), 795–814.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
358 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications

Figure 32 Schematic showing the principle of (a) microwire electro-discharge grinding (micro-WEDG) (reproduced from Song, K. Y.; Chung, D. K.;
Park, M. S.; Chu, C. N. Micro-Electrical Discharge Drilling of Tungsten Carbide Using Deionized Water. J. Micromech. Microeng. 2009, 19, 045006
(10 pp) and (b) micro-EDG using rotating disk as sacrificial electrode (reproduced from Lim, H. S.; Wong, Y. S.; Rahman, M.; Lee, E. M. K. A Study
on the Machining of High-Aspect Ratio Micro-Structures Using Micro EDM. J. Mater. Process. Technol. 2003, 140, 318–325).

a fixed traveling speed and is fed from a reel and take-up system. The micro-WEDG was first invented by Masuzawa (16) in 1985
described as a means to manufacture microcylindrical electrodes and has made a great change in the miniaturization, as micro-
WEDG was found to fabricate very thin micro-EDM electrodes with a very high aspect ratio. Figure 32 shows the schematic
representation of micro-WEDG (116) and EDG using a sacrificial rotating disk (113).
In another similar process, a rotating disk can replace the moving wire and reduces the chance of wire breakage during the
machining process. However, the use of a rotating disk involves a rather complicated setup, although it provides good shape
accuracy (113).

11.14.6.6.2 Application Examples of Different Micro-EDG Processes

11.14.6.6.2.1 In situ fabrication of micro-EDM tool by different micro-EDG processes
One of the most common applications of the micro-EDG process is the fabrication of very long and thin electrodes on the machine
before using it for the micro-EDM process. During micro-EDM, it is not recommended to change the microelectrode during
machining because it incurs inaccuracy due to the change in setup or reclamping of the microelectrode. From an electrode thicker
than the required diameter, a cylindrical electrode is fabricated by the EDG process using a sacrificial electrode. Different setup and
trajectory control of the sacrificial electrode can be used in this process, such as using a ‘stationary block,’ ‘rotating disk,’ ‘wire EDG
(WEDG)’ and moving BEDG, etc. Figure 10 shows the fabricated microelectrodes using stationary BEDG, moving BEDG and micro-
WEDG, and rotating disk EDG process. It has been reported that, among the various micro-EDM techniques for on-machine
fabrication, micro-WEDG, and moving BEDG can produce dimensionally more accurate microelectrodes with better surface
finish. However, microelectrodes with the lowest diameter of 4.3 mm were obtained by the micro-WEDG process. Figure 33 shows
the fabrication of fine micro-EDM tools/rods using different micro-EDG processes (117).
In addition to fabricating circular micro-EDM electrodes, the micro-EDG processes can be used for machining noncircular
shaped electrodes. The same micro-WEDG process can be used for fabricating microelectrodes with different regular cross-sections,
such as circular, triangular, polygonal, etc. The rotating electrode in micro-WEDG is controlled to move in the directions of X, Y, and
Z to obtain the desired shaped electrode. Figure 34 shows samples of microelectrodes fabricated by the micro-WEDG process (114).

11.14.6.6.2.2 Fabrication of cutting tool for conventional micromachining processes

11.14.6.6.2.2.1 Ultrasharp m-turning tool fabrication by m-EDM
Commercially available poly crystalline diamond (PCD) inserts, designed for light finishing cut, has a relatively large tool nose
radius, e.g., 100 mm (Figure 35(a)). This tool nose resolves the cutting force on the shaft into two components, namely Fx and Fy, as
can be seen in Figure 35(a). The Fy component of the cutting force does the actual cutting while the Fx component causes deflection
of the microshaft. A commercially available PCD insert can be modified using micro-EDG to achieve a very sharp cutting edge, so as
to significantly reduce the Fx component of the cutting force, which is illustrated in Figure 35(b). Thus, this makes it possible to
achieve a straight shaft at a much smaller diameter. A comparison of the microshafts fabricated with round tool nose and modified
tool nose is shown in Figure 35(d).

11.14.6.6.2.2.2 Fabrication of milling tool

The use of the micro-WEDG for the production of milling tools has several advantages. The geometry can be changed quite easily
and the potential of scaling down the size of the milling tools is very high. In comparison to other contactless machining tech-
nologies, micro-EDM has an acceptable machining time and the resulting costs for the machining are tolerable in comparison to

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications 359

Figure 33 On-machine fabricated microelectrodes obtained by (a) 44.5 mm CuW electrode by stationary BEDG (reproduced from Asad, A. B. M. A.;
Masaki, T.; Rahman, M.; Lim, H. S.; Wong, Y. S. Tool-Based Micro-Machining. J. Mater. Process. Technol. 2007, 192–193, 204–211), (b) 45 mm W
electrode by moving BEDG process (reproduced from Asad, A. B. M. A.; Masaki, T.; Rahman, M.; Lim, H. S.; Wong, Y. S. Tool-Based Micro-
Machining. J. Mater. Process. Technol. 2007, 192–193, 204–211), (c) 10 mm electrode fabricated by micro-WEDG process reproduced from
Kim, Y. T.; Park, S. J.; Lee, S. J. Micro/Meso-Scale Shapes Machining by Micro EDM Process. Int. J. Precis. Eng. Manuf. 2005, 6 (2), 5–11, and
(d) 4.3 mm diameter shaft by micro-WEDG process (reproduced from Han, F.; Yamada, Y.; Kawakami, T.; Kunieda, M. Investigations on Feasibility of
Sub Micrometer Order Manufacturing Using Micro-EDM. In Proceedings of American Society of Precision Engineering (ASPE) Annual Meeting; 2003, Vol. 30,
pp 551–554).

Figure 34 Sample micro-electrodes of different cross sections fabricated by micro-WEDG: (a) cylindrical electrode with a diameter of B 30 mm,
(b) conical electrode, (c) triangular electrode, and (d) cross-section of the triangular electrode featuring a very sharp corner. Reproduced from
Rahman, M.; Lim, H. S.; Neo, K. S.; Kumar, A. S.; Wong, Y. S.; Li, X. P. Tool-Based Nanofinishing and Micromachining. J. Mater. Process. Technol.
2007, 185 (1–3), 2–16.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
360 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications

Figure 35 (a) Resolution of cutting force components (Fx and Fy) in commercial PCD cutting tool and micro-EDG modified tool, (b) modified
turning tool with block mEDG, (c) fabricated 33 mm microelectrode by conventional m-turning with modified tool, and (d) micro-motor shafts by
m-turning with round tool nose and modified tool. Reproduced from Rahman, M.; Lim, H. S.; Neo, K. S.; Kumar, A. S.; Wong, Y. S.; Li, X. P.
Tool-Based Nanofinishing and Micromachining. J. Mater. Process. Technol. 2007, 185 (1–3), 2–16.

machining with an ion beam. An advantage of using the micro-EDM for the milling process is the prevention of inaccuracy by
rechucking processes (118). Figure 36(a) and 36(b) shows the fabricated milling tool and the machined slot generated during
milling. Moreover, fabrication of microend mill with a single flute by micro-WEDG process for milling has been reported by Morgan
et al. (119), for milling soft materials like brass and aluminum. Figure 36(c) and 36(d) show a micrograph of a 100-mm diameter
tungsten carbide microtool and fabricated microslot. Tungsten carbide was chosen as the tool material because of its high hardness
and low wear rate. Three-fourths of the cylinder was removed to provide a single cutting edge, and then a 45 slice was also removed
from the nose of the tool to provide clearance for various micromilling applications. Microtools fabricated by WEDG can be used to
remove material by mechanical cutting, rather than with electrical discharges, to achieve better surface finishes and higher MRR.

Figure 36 (a) 100 mm diameter milling tools in tungsten carbide fabricated by the micro-WEDG process, (b) surface finish and edge of the slot
machined by a fabricated milling tool, (reproduced from Fleischer, J.; Masuzawa, T.; Schmidt, J.; Knoll, M. New Applications for Micro-EDM.
J. Mater. Process. Technol. 2004, 149, 246–249), (c) 100 mm diameter tungsten carbide microtools by micro-WEDG, and (d) micrograph of groove
machined in AA3003 aluminum (reproduced from 119. Morgan, C. J.; Vallance, R. R.; Marsh, E. R. Micro-Machining and Micro-Grinding with Tools
Fabricated by Micro Electro-Discharge Machining. Int. J. Nanomanuf. 2006, 1 (2), 242–258).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications 361

Figure 37 (a) Fabrication of 150 mm m-grinding tool in situ by the m-EDG process, (b) machining of ‘NUS’ (slot width 150 mm  depth 50 mm)
on BK-7 glass by the m-grinding process using the fabricated grinding tool, and (c) surface of the slot machined by the m-grinding process on
BK-7 glass. Reproduced from Asad, A. B. M. A.; Masaki, T.; Rahman, M.; Lim, H. S.; Wong, Y. S. Tool-Based Micro-Machining. J. Mater. Process.
Technol. 2007, 192–193, 204–211).

Figure 38 (a) Schematic representation of fabrication of the single microdisk using micro-WEDG mechanism, (b) schematic representation for
cutting the arrays of microslit using the MRDE, and (c) photograph of series pattern microdisk. Reproduced from Kuo, C.-L.; Huang, J.-D. Fabrication
of Series-Pattern Micro-Disk Electrode and Its Application in Machining Micro-Slit of Less than 10 mm. Int. J. Mach. Tools Manuf. 2004, 44, 545–553.

11.14.6.6.2.2.3 Fabrication of microgrinding tools

Another important application of micro-EDM is the fabrication of microgrinding tools using PCD or tungsten carbide materials.
PCD, which can be shaped with WEDG, is emerging as a tool material for microgrinding hard and brittle materials. The cobalt
binder provides an electrically conductive network that can be removed with EDM. The diamond cutting edges are exposed as the
discharges erode away the cobalt binder. Figure 37(a) shows a cylindrical PCD-bonded tool with 100 mm diameter cutting tip
prepared for fabrication of a 100-mm wide slot on glass. Figure 37(b) shows ‘NUS’ written using m-grinding process with the
grinding tool. Figure 37(c) shows surface quality generated by the m-grinding process (97). The fabricated slot shows a very fine and
smooth surface that is comparable to the surface obtained from ductile mode cutting of glass in macroscale.

11.14.6.6.2.3 Fabrication of series-pattern microdisk electrode by micro-WEDG

Besides arrays of microholes, fabricating series-pattern microdisk electrodes by micro-WEDG and its application in machining
microslit by micro-EDM was developed (120). First, a raw pin clamped at the mandrel and horizontally rotated on the bearing
surface is machined into the required length and diameter. Then, the wire traveling on the wire guide can be fed toward the radius
direction of pin up to the required depth. The disk thickness is dominated by a tracking path of wire electrode. The fabricated single
microrotating disk electrode (MRDE) can also act as a tool electrode for fabricating arrays of microslit. The MRDE can then be fed
into the workpiece from the surface to the appropriate depth. A straight line with a microslit can therefore be formed by means of
moving the Y-stage carrying the workpiece and array of microdisk are generated by micro-EDM (Figure 38).

11.14.6.7 Planetary or Orbital Micro-EDM

11.14.6.7.1 Working Principle
The planetary or orbital micro-EDM using the movement of the electrode was found to be very useful particularly in the drilling of
microholes, as flushing is more difficult for a thin electrode (121). A common problem during the die-sinking micro-EDM or micro-
EDM drilling is the debris accumulation, which becomes worse during machining high-aspect-ratio microstructures. Adding
a relative motion between the electrode and workpiece, other than the electrode feeding motion, produces a wide clearance between
them for fluid circulation and then reduces debris concentration, resulting in a high material removal rate, low electrode wear ratio,
and higher machining accuracy. This results in less wear of the bottom edges of the tool and, therefore, minimizes undesirable

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
362 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications

Figure 39 (a) Schematic representation of planetary micro-EDM for a circular microhole (reproduced from Egashira, K.; Taniguchi, T.; Hanajima, S.
Planetary EDM of Micro Holes. Int. J. Elect. Mach. 2006, 11, 15–18) and (b) planetary movement for a noncircular hole (reproduced from Yu, Z. Y.;
Rajurkar, K. P.; Shen, H. High Aspect Ratio and Complex Shaped Blind Micro Holes by Micro EDM. Ann. CIRP 2002, 51 (1), 359–362).

tapering and waviness at the bottom surface of the blind microhole (122). The tool path depends on the complexity of the feature to
be machined. Besides improving the material removal rate and decreasing tool electrode wear, another advantage of orbital micro-
EDM in drilling of microholes is that it can fabricate microholes of different diameters with the same electrode size by changing the
orbit radius (121). Figure 39(a) and 39(b) represent the schematic of planetary micro-EDM for round microhole and square
microhole, respectively.

11.14.6.7.2 Application Examples of Planetary or Orbital Micro-EDM

11.14.6.7.2.1 Machining of circular and noncircular through/blind microholes
One of the major advantages of orbital EDM is that the one electrode can be used for machining of various shaped holes of various
diameters. Figure shows machining of circular- and oval-shaped microholes with an 18 mm diameter electrode. Two different size
circular holes have been machined using planetary motion diameters of 8 mm and 16 mm orbital, respectively, as shown in
Figure 40(a) and 40(b). Figure 40(c) shows machining of an elliptical hole with oval movement of the electrode.
An advantage of the orbital EDM is that it is very useful during machining of blind and noncircular microholes. As there is no
rotary motion of the tools during machining of blind holes, flushing of debris particles becomes difficult. Therefore, the machining
becomes unstable resulting in sparking on those debris particles and creating arcing and short circuiting. At this point, introducing
the planetary movement at the electrode helps in flushing out debris by continually changing the spark gap. As a result, a higher
aspect ratio of noncircular blind holes can be achieved. Figure 41(a)–(c) show the machining of blind triangular, square, and
pentagonal micro-holes with sharp corners and edges. The orbital parameters used for machining these noncircular blind holes are
a total electrode feed of 100 mm and offset of tool paths by 15–20 mm have been machined.

11.14.6.8 Reverse Micro-EDM

11.14.6.8.1 Working Principle
The reverse micro-EDM is being accomplished by reversing the polarity so that the material erodes from the larger diameter rod. The
reverse micro-EDM comprises different steps including fabrication of a single microelectrode using the micro-EDG process,
fabrication of arrays of microholes that will act as negative electrodes during reverse EDM, and finally fabrication of multiple
electrodes using reverse micro-EDM. The polarity of the electrode and sacrificial workpiece is interchanged during reverse micro-
EDM, so that the electrode can be extruded among the arrays of microholes. In this process, the electrode is considered as

Figure 40 Microholes drilled with orbital micro-EDM using an 18 mm electrode: (a) circular hole with orbital path diameter of 8 mm, (b) circular
hole with orbital diameter of 16 mm, and (c) elliptical hole with oval planetary motion path (for all cases the electrode feed length is 100 mm).
Reproduced from Egashira, K.; Taniguchi, T.; Hanajima, S. Planetary EDM of Micro Holes. Int. J. Elect. Mach. 2006, 11, 15–18.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications 363

Figure 41 Machining of (a) triangular, (b) square, and (c) pentagonal microholes using orbital micro-EDM with total electrode feed of 100 mm and
offset of tool paths by 15–20 mm. Reproduced from Yu, Z. Y.; Rajurkar, K. P.; Shen, H. High Aspect Ratio and Complex Shaped Blind Micro Holes by
Micro EDM. Ann. CIRP 2002, 51 (1), 359–362.

a workpiece and is fed down to the holes of the metal plate to be machined by electrical discharge, which occurs within machining
gap. However, the regions that correspond to the holes are not machined. Finally, microelectrodes are machined as many as the
holes. The principle of reverse micro-EDM is schematically presented in Figure 42 (123).

11.14.6.8.2 Application Examples of Reverse Micro-EDM

11.14.6.8.2.1 Batch mode microelectrode fabrication
In recent years, for the fabrication of multiple electrodes or arrays of microelectrodes, a newer technique named reverse micro-EDM
has been found to be very useful. In this process, the electrode is considered as a workpiece and is fed down to the holes of the metal
plate to be machined by electrical discharge, which occurs within the machining gap. However, the regions that correspond to the
holes are not machined. Finally, microelectrodes are machined as much as the holes. Figure 43 shows fabrication of a single
microelectrode using block micro-EDG, arrays of microholes using that electrode, and fabricated arrays of microelectrodes using
reverse micro-EDM (124). Figure 44 shows the arrays of microelectrodes fabricated by reverse micro-EDM (123).

11.14.7 Advances in Micro-EDM: Innovative Applications

11.14.7.1 Deburring of Microfeatures by Micro-EDM
Burrs are an undesired and unavoidable by-product of most conventional manufacturing processes, such as cutting, forming,
blanking, and shearing operations, because of the plastic flow of material. A burr is defined as a projection of undesired material
beyond the desired machined features (125). A novel application of micro-EDM for removing the smaller burrs generated during

Figure 42 Principle of reverse micro-EDM; (a) fabrication of microhole using normal micro-EDM and (b) fabrication of microelectrode using
reverse micro-EDM. Reproduced from Kim, B. H.; Park, B. J.; Chu, C. N. Fabrication of Multiple Electrodes by Reverse EDM and Their Application in
Micro ECM. J. Micromech. Microeng. 2006, 16 (4), 843–850.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
364 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications

Figure 43 Schematic representation of fabricating arrays of microelectrodes using reverse micro-EDM: (a) fabrication of single electrode by block
micro-EDG process, (b) Machining of arrays of micro-holes using fabricated electrode, and (c) Fabrication of arrays of microelectrodes using reverse
micro-EDM. Reproduced from Weiliang, Z.; Zhenlong, W.; Desheng, D. A New Micro-EDM Reverse Copying Technology for Microelectrode Array
Fabrication. In Proceedings of International Technology and Innovation Conference 2006 (ITIC 2006) (CP524) Hangzhou, China, 6–7 Nov. 2006,
pp 1633–1636.

Figure 44 (a) Plate electrodes for REDM and (b) 5  5 arrays of multiple electrodes (diameter 35 mm, length 1.5 mm) microelectrodes machined by
reverse micro-EDM. Reproduced from Kim, B. H.; Park, B. J.; Chu, C. N. Fabrication of Multiple Electrodes by Reverse EDM and Their Application in Micro
ECM. J. Micromech. Microeng. 2006, 16 (4), 843–850.

the micromilling process has been reported (125). Micro-EDM using low discharge energy and a small-diameter cylindrical tool is
introduced for deburring microfeatures. The proposed method selectively removes only the burrs near the tool with minimum
collateral damage of the machined features because the machining range can be controlled by the spark gap. As micro-EDM uses
little electrical power and a small-diameter cylindrical tool, the tool can easily access microscale features and remove small amounts
of material from the burrs as shown in Figure 45. Burrs with a height from 200 to 1000 mm and width from 80 to 200 mm were
successfully removed by the three consecutive steps of rough deburring, finish deburring, and edge finishing, with an error in height
of 1 mm or less and a deburred edge width from 50 to 60 mm.

Figure 45 (a) Concept of micro-EDM deburring: deburring on the top plane and side plane of the edge, (b) burrs generated after a microslot
machining process by micromilling, (c) finishing of edges (one of two edges) by micro-EDM deburring. Reproduced from Jeong, Y. H.; Yoo, B. H.;
Lee, H. U.; Min, B. K.; Cho, D. W.; Lee, S. J. Deburring Microfeatures Using Micro-EDM. J. Mater. Process. Technol. 2009, 209, 5399–5406.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications 365

Figure 46 Micro-EDM removal and deposition process in the same setup: (a) fabrication of tool electrode using block micro-EDM, (b) process of
depositing materials on the fabricated tool, and (c) machining of deposited rod using micro-EDM. Reproduced from Zilong, P.; Zhenlong, W.;
Yinghuaia, D.; Hui, C. Development of a Reversible Machining Method for Fabrication of Microstructures by Using Micro-EDM. J. Mater. Process.
Technol. 2010, 210, 129–136.

11.14.7.2 Micro-EDM as a Deposition Process

The micro-EDM process is being primarily considered as a material removal process. In a recent study, the micro-EDM process has
been used as a deposition technique for fabricating a cylindrical microrod or microelectrode (126). The research of EDM deposition
was first carried out by Hayakawa et al. (127). They reported that when the tool electrode was set as the anode and the EDM process
was carried out in gas, the deposition process can be realized (127). Zilong et al. (126) carried out the micro-EDM deposition
process using a microtool electrode, with a 200 mm diameter, obtained using the block mEDM process as shown in Figure 46(a).
With the Z axis vertical feeding, metal material from the tool electrode can be released by the discharge energy and deposited on the
workpiece surface to form a microstructure (Figure 46(b)). The diameter of the deposited microcylinder is almost the same as that
of the microtool electrode. The high tool electrode wear will be beneficial to the deposition process. So, the tool electrode is set as
the anode and the machining process is carried out in air, which can increase the tool electrode wear. The process mechanism of
mEDM deposition can be described as follows: medium ionization and forming of a discharge channel, the tool electrode material
melting or gasifying removal, the removed tool electrode material deposited on the cathode, and deionization. Figure 46 shows the
principle of the micro-EDM deposition process and machining of deposited electrode using micro-EDM in the same setup (126).
Figure 47 shows the deposited and machined electrodes (126).

11.14.7.3 Spherical Probe Fabrication Using WEDG Combined with Single Discharge
A novel application of micro-EDM combining WEDG technology with one-pulse electro-discharge (OPED) has been presented to
fabricate multi-microspherical probes using difficult-to-machine tungsten and tungsten carbide (128). The single- and

Figure 47 (a) Brass microcylinder fabricated by micro-EDM deposition process, (b) machining of brass cylinder to a square cylindrical column by
the micro-EDM milling process, (c) steel microcylinder fabricated by micro-EDM deposition, and (d) machining of 80 mm holes radial to
the deposited steel cylinder using micro-EDM drilling in the same setup. Reproduced from Zilong, P.; Zhenlong, W.; Yinghuaia, D.; Hui, C.
Development of a Reversible Machining Method for Fabrication of Microstructures by Using Micro-EDM. J. Mater. Process. Technol. 2010, 210, 129–136.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
366 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications

Figure 48 (a) Principle and steps of fabricating spherical probe, (b) fabricated single microprobe, and (c) multiprobe (reproduced from Sheu, D.-Y.
Multi-Spherical Probe Machining by EDM Combining WEDG Technology with One-Pulse Electro-Discharge. J. Mater. Process. Technol. 2004, 149,
597–603).

multispherical probe has been fabricated using different steps. First, a thin cylindrical and square rod of desired dimension is
fabricated using the two step micro-WEDG process. After the fabrication of the rod of desired shape and dimension, the setup is
converted in OPED system and a single pulse electric discharge is provided at the tip of the fabricated rod (Figure 48(a)). The
maintenance of appropriate gap distance and control of gap is important during the OPED process, as too small a gap distance can
break the microrod. Using the similar principle, multispherical probes can also be fabricated. WEDG can be used to fabricate cross-
section of desired type (e.g., three microrods or four microrods from a large electrode). After the fabrication of multiple micro-
electrodes or rods, single pulse electric discharges are provided to all of the microelectrodes. Figure 48(b) and 48(c) show
the Scanning Electron Microscope (SEM) image of the single and multispherical probes and a cross-section of fabricated micro-
electrodes. The smallest spherical probe with 10 mm electrode diameter and 15 mm sphere diameter has been reported using this
micro-EDM based novel process, which is the smallest in machining of all spherical probes reported by different machining
processes (128).

11.14.7.4 Micro-EDM of Electrically Nonconductive Ceramic

Although, micro-EDM is known as a process of machining conductive materials, a novel micro-EDM based process named the
‘assisted electrode method’ was invented for machining electrically nonconductive or insulating ceramic materials. In the assisted
electrode method, the nonconductive surface is covered with the electrical conductive layer during discharge, which acts as an
assisted electrode. The layer holds the electrical conductivity during discharge and results in machining from nonconductive ceramic
materials. The assisted electrode can be obtained by different processes: putting TiN layer using physical vapor deposition (PVD)
process (129), colloidal graphite solution (130), or silver varnish with 45.9% silver content (131). Figure 49 shows the machining
of a microhole using colloidal graphite solution and PVD TiN coating (132). It can be seen from Figure 49(b) that on the TiN
assisting electrode, a thermally affected zone was detected around the entrance area of the hole. On the other hand, the carbon back
layer on the EDMed surface using colloidal graphite solution left a less heat-affected zone around the entrance of the microhole
(Figure 49(a)).

11.14.7.5 3D Microstructuring in CNT Forests Using Dry Micro-EDM

Vertically aligned carbon nanotubes, so called carbon nano-tubes (CNT) forests, have important applications in micro-electro-
mechanical systems (MEMS) due to unique electrical, mechanical, thermal, and other properties. A novel application of micro-EDM

Figure 49 (a) Assisting electrode: EDMed surface colloidal graphite solution or carbon back layer, and (b) assisting electrode: PVD TiN.
Reproduced from Muttamara, A.; Fukuzawa, Y.; Mohri, N.; Tani, T. Probability of Precision Micro-Machining of Insulating Si3N4 Ceramics by EDM.
J. Mater. Process. Technol. 2003, 140, 243–247.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications 367

Figure 50 (a) Multilevel microchannel structures showing a 5 mm structure with an aspect ratio of 20, (b) 3D microstructure: Individual needlelike
structure, and (c) 4  4 arrays of needlelike structure. Rreproduced from Dahmardeh, M.; Nojeh, A.; Takahataa, K. Possible Mechanism in Dry
Micro-Electro-Discharge Machining of Carbon-Nanotube Forests: A Study of the Effect of Oxygen. J. Appl. Phys. 2011, 109, 093308.

for the fabrication of microstructure in CNT forests has been reported (133). The machining is done using the dry micro-EDM with
air or oxygen as dielectric fluid. An RC-type pulse generator has been used for the dry micro-EDM of CNT forests as it uses
nanosecond pulses of arc discharge for machining, thus, enabling precise control of discharge energy delivered to a CNT forest while
protecting the sample from overheating (23). The objective is to use high-frequency pulses of electrical discharge to locally machine
the nanotubes in order to create target shapes in a forest. With this approach, forest microstructures can be fabricated to have varying
shapes along their height, unachievable with conventional prepatterned chemical vapor deposition growth techniques. Micro-
machining of multilayer geometries as well as arrayed needlelike microstructures with angled surfaces can be fabricated using this
process (Figure 50). Microstructure with an aspect ratio of 20 with the smallest feature of 5 mm has been machined in forests
without disordering the vertical orientation of the nanotubes (Figure 50).

11.14.7.6 Batch Mode Microstructuring Using Micro-EDM Combined with LIGA

A novel technique of a batch mode microstructuring process has been demonstrated that uses a LIGA (a German acronym for
‘Lithographie, Galvanoformung, Abformung,’ in English (X-ray) Lithography, Electroplating, and Molding) process for fabricating
high-aspect-ratio batch mode microelectrodes and micro-EDM for machining of batch mode microholes using those electrodes
(134). The LIGA process uses X-ray lithography to form high aspect ratio molds for electroplated structures. Figure 51(a) shows an
example of a 20  20 array of high-aspect-ratio electrodes of electroplated copper with 20-mm diameter, 60-mm pitch, and 300-mm
structural height. The LIGA-fabricated array of 400 Cu electrodes with 20-mm diameter was used to machine through-holes in 50-mm
thick stainless steel using micro-EDM as shown in Figure 51(b). The machining time was about 5 min, which is 600 times less than
that required for serial machining by a single electrode. Arrayed electrodes of even complicated cross-section shapes, like hexagonal
and gear, can be fabricated using LIGA with a high aspect ratio.

11.14.7.7 Simple and Complex Microrod Fabrication Using Self-Drilled Holes

A novel method based on micro-EDM has been proposed by Yamazaki et al. (135) to machine microrods and holes without using
any special equipment like WEDG. First, the rod electrode is rotated and fed into a plate electrode to make a hole. The polarities of
the rod and plate electrodes are respectively selected as negative and positive during the machining of a microhole. After the rod

Figure 51 (a) 20  20 array of LIGA fabricated copper electrodes and (b) through-holes batch machined in 50-mm thick stainless steel using the
array. Reproduced from 134. Takahata, K.; Gianchandani, Y. B. Batch Mode Micro-Electro-Discharge Machining. IEEE/ASME J. Microelectromech.
Syst. 2002, 11, 102–110.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
368 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications

Figure 52 (a) Microrod of B 4 mm formed by self-drilled microhole and machined microhole of B 5 mm using the formed rod, (b) microelectrode
and microhole obtained by cross-shaped right-angled rod. Reproduced from Yamazaki, M.; Suzuki, T.; Mori, T.; Kunieda, M. EDM of Micro-Rods by
Self-Drilled Holes. J. Mater. Process. Technol. 2004, 149, 134–138.

electrode returns to its initial position, the rod electrode is moved from the center of the hole at a certain distance. The polarity of the
rod electrode is then reversed, and the rod electrode is fed into the plate electrode either with or without rotation. Positioning of the
rod electrode in reference to the tool electrode is not necessary because self-drilled holes are utilized as the tool electrodes.
Figure 52(a) shows a microrod of 4 mm in diameter formed by this method. Using this rod electrode as a tool electrode, a microhole
with a diameter of 5 mm was perforated on a copper plate. In Figure 52(b), a cross-shaped right-angled slot was formed using a rod
electrode. Positioning the rod electrode on each corner of the cross-shaped slot without a rotation, micro-EDM was performed to
form the cross-shaped rod with reversed polarity. Using this cross-shaped electrode, cross-shaped holes were formed on a plate.

11.14.8 Summary

This chapter presented a comprehensive overview on the micro-EDM process, its varieties, and important applications of each
variety of micro-EDM. The chapter started with a brief overview of the micro-EDM process including the physics of the process,
sparking and gap phenomena in micro-EDM, differences between macro- and micro-EDM, and a brief discussion on system
components of micro-EDM. An overview of different process parameters (electrical, nonelectrical, materials, and motion control
parameters, etc.) and performance measure parameters (micro-EDM erosion efficiency, material removal rate, electrode wear ratio,
surface roughness, etc.) has been presented with definition and terminology. A detailed section on different varieties of micro-EDM,
their working principle, and related important applications are also presented. Finally, some novel applications of micro-EDM
indicating the research advances and development in the area of micro-EDM have been presented.
Micro-EDM is a flexible micromanufacturing process offering a wide variety in processes, materials, and applications. Due to
its versatility, micro-EDM has great potential to fabricate a wide range of microstructures, mechanical components, and devices
for different areas of applications like automotive parts, fiber-optics and MEMS, aerospace, medical and biomedical appli-
cations, microelectronics, micromolds, and microtools. However, some of the major challenges of micro-EDM are slow
production rate, a postprocessing requirement to improve the surface finish, and low throughput. Therefore, the current research
trend is to solve the problems associated with micro-EDM either by developing newer hybrid machining processes or incor-
porating novel ideas to improve the existing process and creating new varieties. Therefore, the continuous improvements in the
micro-EDM research, innovation of new varieties of micro-EDM and development of micro-EDM based hybrid micromachining
technologies have the potential to complement the weakness of micro-EDM, thus, finally increasing the throughput and
productivity in micromanufacturing.

References

1. Alting, L.; Kimura, F.; Hansen, H. N.; Bissacco, G. Micro Engineering. Ann. CIRP 2003, 52 (2), 635–657.
2. Ehmann, K. F.; Bourell, D.; Culpepper, M. L.; Hodgson, T. J.; Kurfess, T. R.; Madou, M.; Rajurkar, K. P. International Assessment of Research and Development in
Micromanufacturing; World Technology Center (WTEC), Inc: Baltimore, Maryland, USA, 2005.
3. Rahman, M.; Asad, A. B. M. A.; Masaki, T.; Saleh, T.; Wong, Y. S.; Kumar, A. S. A Multiprocess Machine Tool for Compound Micromachining. Int. J. Mach. Tools Manuf.
2010, 50 (4), 344–356.
4. Liu, X.; DeVor, R. E.; Kapoor, S. G.; Ehmann, K. F. The Mechanics of Machining at the Microscale: Assessment of the Current State of the Science. J. Manuf. Sci. Eng. 2004,
126 (4), 666–678.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications 369

5. Jahan, M. P.; Rahman, M.; Wong, Y. S. A Review on the Conventional and Micro-Electrodischarge Machining of Tungsten Carbide. Int. J. Mach. Tools Manuf. 2011, 51 (12),
837–858.
6. Jahan, M. P.; Rahman, M.; Wong, Y. S.; Fuhua, L. On-Machine Fabrication of High-Aspect-Ratio Micro-Electrodes and Application in Vibration-Assisted Micro-Electrodischarge
Drilling of Tungsten Carbide. Proc. Inst. Mech. Eng., Part B 2010, 224 (5), 795–814.
7. Rajurkar, K. P.; Levy, G.; Malshe, A.; Sundaram, M. M.; McGeough, J.; Hu, X.; Resnick, R.; De Silva, A. Micro and Nano Machining by Electro-Physical and Chemical
Processes. Ann. CIRP 2006, 55 (2), 643–666.
8. Mahdavinejad, R. A.; Mahdavinejad, A. ED Machining of WC–Co. J. Mater. Process. Technol. 2005, 162–163, 637–643.
9. Masuzawa, T.; Kimura, M. Electrochemical Surface Finishing of Tungsten Carbide Alloy. Ann. CIRP 1991, 40, 199–202.
10. Guitrau, E. B. The EDM Handbook; Hanser Gardner Publications: Cincinnati, 1997.
11. Tsai, Y.-Y.; Masuzawa, T. An Index to Evaluate the Wear Resistance of the Electrode in Micro-EDM. J. Mater. Process. Technol. 2004, 149, 304–309.
12. Ho, K. H.; Newman, S. T. State of the Art Electrical Discharge Machining (EDM). Int. J. Mach. Tools Manuf. 2003, 43, 1287–1300.
13. Masuzawa, T. State of the Art Micromachining. Ann. CIRP 2000, 49, 473–488.
14. Ho, K. H.; Newman, S. T.; Rahimifard, S.; Allen, R. D. State of the Art Wire Electrical Discharge Machining. Int. J. Mach. Tools Manuf. 2004, 44, 1247–1259.
15. Kurafuji, H.; Masuzawa, T. Micro-EDM of Cemented Carbide Alloys. Jpn. Soc. Electromech. Eng. 1968, 2 (3), 1–16.
16. Masuzawaa, T.; Fujinoa, M.; Kobayashia, K.; Suzukib, T.; Kinoshita, N. Wire Electro-Discharge Grinding for Micro-Machining. CIRP Ann. Manuf. Technol. 1985, 34 (1),
431–434.
17. Jameson, E. C. Description and Development of Electrical Discharge Machining (EDM). In Electrical Discharge Machining; Society of Manufacturing Engineers: Dearborn,
Michigan, 2001; p 12.
18. Llanes, L.; Idanez, E.; Martinez, E.; Casas, B.; Esteve, J. Influence of Electrical Discharge Machining on the Sliding Contact Response of Cemented Carbides. Int. J. Refract.
Met. Hard Mater. 2001, 19, 35–40.
19. Allen, D. M.; Lecheheb, A. Microelectro-Discharge Machining of Ink Jet Nozzles: Optimum Selection of Material and Machining Parameters. J. Mater. Process. Technol. 1996,
58, 53–66.
20. Schumacher, B. M. After 60 Years of EDM the Discharge Process Remains Still Disputed. J. Mater. Process. Technol. 2004, 149, 376–381.
21. Kunieda, M.; Lauwers, B.; Rajurkar, K. P.; Schumacher, B. M. Advancing EDM through Fundamental Insight into the Process. Ann. CIRP 2005, 54, 599–622.
22. Gentili, E.; Tabaglio, L.; Aggogeri, F. Review on Micromachining Techniques, Courses and Lectures, 2005. www.dimgruppi.ing.unibs.it.
23. Wong, Y. S.; Rahman, M.; Lim, H. S.; Han, H.; Ravi, N. Investigation of Micro-EDM Material Removal Characteristics Using Single RC-Pulse Discharges. J. Mater. Process.
Technol. 2003, 140 (1–3), 303–307.
24. Rahman, M.; Asad, A. B. M. A.; Masaki, T.; Wong, Y. S.; Kumar, A. S. Integrated Hybrid Micro/Nano-Machining. In ASME 2007 International Manufacturing Science and
Engineering Conference (MSEC2007), October 15–18, 2007, Atlanta, Georgia, USA, Paper no. MSEC2007-31009, pp 197–209.
25. Katz, Z.; Tibbles, C. J. Analysis of Micro-Scale EDM Process. Int. J. Adv. Manuf. Technol. 2005, 25, 923–928.
26. Micro Manufacturing, Differences between Macro and Micro EDM, http://www.micromanufacturing.net/didactico/Desarollo/edm-eng/6-differences-between-macro-and-
micro-edm.
27. Masuzawa, T. Micro EDM. In Proceedings of the ISEM XIII; Fundación Tekniker, 2001; pp 3–19. ISBN: 932064-0-7.
28. Uhlmann, E.; Piltz, S.; Doll, U. Machining of Micro/Miniature Dies and Moulds by Electrical Discharge Machining – Recent Development. J. Mater. Process. Technol. 2005,
167, 488–493.
29. Han, F.; Yamada, Y.; Kawakami, T.; Kunieda, M. Improvement of Machining Characteristics of Micro-EDM Using Transistor Type Isopulse Generator and Servo Feed Control.
Precis. Eng. 2004, 28, 378–385.
30. Masuzawa, T; Fujino, M. Micro Pulse for EDM. In Proceedings of the Japan Society for Precision Engineering Autumn Conference, Japan, 1980, pp 140–142 (in Japanese).
31. Nakazawa, K.; Han, F.; Kunieda, M. Micro-EDM Using Transistor Type Pulse Generator. In Proceedings of the Japan Society for Precision Engineering Spring Conference,
Japan, 2000, p 259 (in Japanese).
32. Hara, S.; Nishioki, N. Ultra-High Speed Discharge Control for Micro Electric Discharge Machining. Initiatives of Precision Engineering of a Millennium; Kluwer Academic
Publisher, 2001; pp 194–198.
33. Kawakami, T.; Kunieda, M. Study on Factors Determining Limits of Minimum Machinable Size in Micro EDM. Ann. CIRP 2005, 54 (1), 167–170.
34. Egashira, K.; Mizutani, K. EDM at Low Open-Circuit Voltage. Int. J. Electromach. 2005, 10, 21–26.
35. Han, F.; Yamada, Y.; Kawakami, T.; Kunieda, M. Investigations on Feasibility of Sub Micrometer Order Manufacturing Using Micro-EDM. In Proceedings of American Society of
Precision Engineering (ASPE) Annual Meeting, 2003, Vol. 30, pp 551–554.
36. Kunieda, M.; Hayasaka, A.; Yang, X. D.; Sano, S.; Araie, I. Study on Nano EDM Using Capacity Coupled Pulse Generator. Ann. CIRP 2007, 56 (1), 213–216.
37. Zhang, L.; Jia, Z.; Liu, W.; Li, A. A Two-Stage Servo Feed Controller of Micro-EDM Based on Interval Type-2 Fuzzy Logic. Int. J. Adv. Manuf. Technol. 2012, 59 (5–8),
633–645.
38. Yang, G. H.; Liu, F.; Lin, H. B. Research on an Embedded Servo Control System of Micro-EDM. Appl. Mech. Mater. 2011, 120, 573–577.
39. Altpeter, F.; Perez, R. Relevant Topics in Wire Electrical Discharge Machining Control. J. Mater. Process. Technol. 2004, 149 (1–3), 147–151.
40. Rajurkar, K. P.; Wang, W. M. On-line Monitor and Control for Wire Breakage in WEDM. Ann. CIRP 1991, 40 (1), 219–222.
41. Snoeys, R.; Dauw, D.; Kruth, J. P. Improved Adaptive Control System for EDM Processes. Ann. CIRP 1980, 29 (1), 97–101.
42. McGeough, J. A. Advanced Methods of Machining, 1st ed.; Chapman and Hall: USA, 1988. ISBN: 0-412-31970-5.
43. Jahan, M. P.; Wong, Y. S.; Rahman, M. A Study on the Quality Micro-Hole Machining of Tungsten Carbide by Micro-EDM Process Using Transistor and RC-Type Pulse
Generator. J. Mater. Process. Technol. 2009, 209 (4), 1706–1716.
44. Kumar, S.; Singh, R.; Singh, T. P.; Sethi, B. L. Surface Modification by Electrical Discharge Machining: A Review. J. Mater. Process. Technol. 2009, 209, 3675–3687.
45. Fuller, J. E. Electrical Discharge Machining, Vol. 16; ASM Machining Handbook, 1996; pp 557–564.
46. Bruyn, H. E. D. Slope Control – A Great Improvement in Spark Erosion. Ann. CIRP 1968, 16, 183–186.
47. Yeo, S. H.; Aligiri, E.; Tan, P. C.; Zarepour, H. An Adaptive Speed Control System for Micro Electro Discharge Machining. In CPl 181, Third Manufacturing Engineering Society
International Conference; Segui, V. J., Reig, M. J., Eds.; American Institute of Physics, 2009; pp 61–72.
48. Liao, Y. S.; Chang, T. Y.; Chuang, T. J. An On-Line Monitoring System for a Micro Electrical Discharge Machining (Micro-EDM) Process. J. Micromech. Microeng. 2008, 18,
035009 (8 pp).
49. Snoeys, R.; Dauw, D.; Jennes, M. Survey of Adaptive Control and Detection Systems. Ann. CIRP 1982, 31 (2), 483–489.
50. Module 9 Non-Conventional Machining – Lesson 39, Electro Discharge Machining. Version 2 ME, IIT, Kharagpur.
51. Lee, S. H.; Li, X. P. Study of the Effect of Machining Parameters on the Machining Characteristics in Electrical Discharge Machining of Tungsten Carbide. J. Mater. Process.
Technol. 2001, 115, 344–355.
52. Jahan, M. P.; Wong, Y. S.; Rahman, M. A Study on the Fine-Finish Die-Sinking Micro-EDM of Tungsten Carbide Using Different Electrode Materials. J. Mater. Process.
Technol. 2009, 209, 3956–3967.
53. Pandey, P. C.; Shan, H. S. Modern Machining Process; Tata McGraw-Hill Publishing Company Ltd, 1999; pp 84–113.
54. Alam, S. M. Design, Development and Study of Optimal Parameters of a Micro Wire Electrical Discharge Machining (WEDM) Device. Master’s Thesis, National University of
Singapore, 2007.
55. Mahardika, M.; Tsujimoto, T.; Mitsui, K. A New Approach on the Determination of Ease of Machining by EDM Processes. Int. J. Mach. Tools Manuf. 2008, 48, 746–760.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
370 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications

56. Kibria, G.; Sarkar, B. R.; Pradhan, B. B.; Bhattacharyya, B. Comparative Study of Different Dielectrics for Micro-EDM Performance during Microhole Machining of Ti-6Al-4V
Alloy. Int. J. Adv. Manuf. Technol. 2010, 48, 557–570.
57. Jahan, M. P.; Wong, Y. S.; Rahman, M. Effect of Non-Electrical and Gap Control Parameters in the Micro-EDM of WC–Co. J. Mach. Form. Technol. 2009, 1–2, 51–78.
58. Yan, B. H.; Huang, F. Y.; Chow, H. M.; Tsai, J. Y. Micro-Hole Machining of Carbide by Electrical Discharge Machining. J. Mater. Process. Technol. 1999, 87, 139–145.
59. Guu, Y. H.; Hocheng, H. Effects of Workpiece Rotation on Machinability during Electrical Discharge Machining. Mater. Manuf. Process. 2001, 16 (1), 91–101.
60. Kunieda, M.; Masuzawa, T. A Fundamental Study on a Horizontal EDM. Ann. CIRP 1988, 37 (1), 187–190.
61. Masuzawa, T.; Okajima, K.; Taguchi, T. EDM-Lathe for Micromachining. Ann. CIRP 2002, 51 (1), 355–358.
62. Mohri, N.; Tsukamoto, J.; Fujino, M. Drilling of Deep Micro-Holes by EDM. Ann. CIRP 1989, 38 (1), 195–198.
63. Murali, M.; Yeo, S. H. A Novel Sparks Erosion Technique for the Fabrication of High Aspect Ratio Micro-Grooves. Microsyst. Technol. 2004, 10, 628–632.
64. Singh, S.; Maheswari, S.; Pandey, P. C. Optimization of Multiperformance Characteristics in Electrical Discharge Machining of Aluminium Matrix Composites (AMCs) Using
Taguchi DOE Methodology. Int. J. Manuf. Res. 2007, 2 (2), 138–163.
65. Hung, J. C.; Lin, J. K.; Yan, B. H.; Liu, H. S.; Ho, P. H. Using a Helical Micro-Tool in Micro-EDM Combined with Ultrasonic Vibration for Micro-Hole Machining. J. Micromech.
Microeng. 2006, 16, 2705–2713.
66. Hewidy, M. S.; Ebeid, S. J.; El-Taweel, T. A.; Youssef, A. H. Modelling the Performance of ECM Assisted by Low Frequency Vibrations. J. Mater. Process. Technol. 2007, 189,
466–472.
67. Guo, Z. N.; Lee, T. C.; Yue, T. M.; Lau, W. S. A Study of Ultrasonic Aided Wire Electrical Discharge Machining. J. Mater. Process. Technol. 1997, 63, 823–828.
68. Jahan, M. P.; Rahman, M.; Wong, Y. S. Evaluation of the Effectiveness of Low Frequency Workpiece Vibration in Deep-Hole Micro-EDM Drilling of Tungsten Carbide. J. Manuf.
Process. 2012, 14 (3), 343–359.
69. Jahan, M. P.; Saleh, T.; Rahman, M.; Wong, Y. S. Development, Modeling, and Experimental Investigation of Low Frequency Workpiece Vibration-Assisted Micro-EDM of
Tungsten Carbide. J. Manuf. Sci. Eng. 2010, 132 (5), 054503 (8 pp).
70. Huang, H.; Zhang, H.; Zhou, L.; Zheng, H. Y. Ultrasonic Vibration Assisted Electro-Discharge Machining of Microholes in Nitinol. J. Micromech. Microeng. 2003, 13, 693–700.
71. Kremer, D.; Lebrun, J. L.; Hosari, B.; Moisan, A. Effects of Ultrasonic Vibration on the Performances in EDM. Ann. CIRP 1989, 38 (1), 199–202.
72. Jahan, M. P.; Saleh, T.; Rahman, M.; Wong, Y. S. Study of Micro-EDM of Tungsten Carbide with Workpiece Vibration. Adv. Mater. Res. 2011, 264–265, 1056–1061.
73. Koenig, W.; Weill, R.; Wertheim, R.; Jutzler, W. I. The Flow Fields in the Working Gap with Electro-Discharge-Machining. Ann. CIRP 1977, 41 (1), 239–242.
74. Masuzawa, T.; Heuvelman, C. J. A Self-Flushing Method with Spark-Erosion Machining. Ann. CIRP 1983, 32 (1), 109–111.
75. Levy, G. N.; Ferroni, B. Planetary Spark Erosion – Applications and Optimization. In Proceedings of the 16th MTDR Conference, 1975, pp 291–297.
76. Masuzawa, T.; Cui, X. Improved Jet Flushing for EDM. Ann. ClRP 1992, 41 (1), 239–242.
77. Jahan, M. P.; Anwar, M. M.; Wong, Y. S.; Rahman, M. Nanofinishing of Hard Materials Using Micro-EDM. Proc. Inst. Mech. Eng., Part B 2009, 223, 1127–1142.
78. Jahan, M. P.; Rahman, M.; Wong, Y. S. Study on the Nano-Powder Mixed Sinking and Milling Micro-EDM of WC–Co. Int. J. Adv. Manuf. Technol. 2011, 53, 167–180.
79. Masuzawa, T.; Yamaguchi, M.; Fujino, M. Surface Finishing of Micropins. Ann. CIRP 2005, 54 (1), 171–174.
80. Hewidy, M. S.; El-Taweel, T. A.; El-Safty, M. F. Modelling the Machining Parameters of Wire Electrical Discharge Machining of Inconel 601 Using RSM. J. Mater. Process.
Technol. 2005, 169, 328–336.
81. Lee, H. T.; Hsu, F. C.; Tai, T. Y. Study of Surface Integrity Using the Small Area EDM Process with a Copper–Tungsten Electrode. Mater. Sci. Eng. A 2004, 364, 346–356.
82. Lee, L. C.; Lim, L. C.; Narayanan, V.; Venkatesh, V. C. Quantification of Surface Damage of Tool Steels after EDM. Int. J. Mach. Tools Manuf. 1988, 28, 359–372.
83. Lee, H. T.; Tai, T. Y. Relationship between EDM Parameters and Surface Crack Formation. J. Mater. Process. Technol. 2003, 142, 676–683.
84. Rebelo, J. C.; Dias, A. M.; Kremer, D.; Lebrun, J. L. Influence of EDM Pulse Energy on the Surface Integrity of Martensitic Steels. J. Mater. Process. Technol. 1998, 84,
90–96.
85. Guu, Y. H.; Hocheng, H.; Chou, C. Y.; Deng, C. S. Effect of Electrical Discharge Machining on Surface Characteristics and Machining Damage of AISI D2 Tool Steel. Mater. Sci.
Eng. A 2003, 358, 37–43.
86. Lenz, E.; Koenig, W.; Wertheim, R.; Katz, E. Cracking Behavior of Sintered Carbides during EDM. Ann. CIRP 1975, 24, 109–114.
87. Guu, Y. H.; Hocheng, H. Improvement of Fatigue Life of Electrical Discharge Machined AISI D2 Tool Steel by TiN Coating. Mater. Sci. Eng. A 2001, 318, 155–162.
88. Pham, D. T.; Dimov, S. S.; Bigot, S.; Ivanov, A.; Popov, K. Micro-EDM – Recent Developments and Research Issues. J. Mater. Process. Technol. 2004, 149, 50–57.
89. Jahan, M. P.; Wong, Y. S.; Rahman, M.; Asad, A. B. M. A. An Experimental Investigation on the Surface Characteristics of Tungsten Carbide for the Fine-Finish Die-Sinking
and Scanning Micro-EDM. Int. J. Abras. Technol. 2009, 2, 223–244.
90. Amorim, F. L.; Weingaertner, W. L. The Influence of Generator Actuation Mode and Process Parameters on the Performance of Finish EDM of a Tool Steel. J. Mater. Process.
Technol. 2005, 166 (3), 411–416.
91. Bleys, P.; Kruth, J. P.; Lauwers, B. Sensing and Compensation of Tool Wear in Milling EDM. J. Mater. Process. Technol. 2004, 149, 139–146.
92. Tong, H.; Li, Y.; Wang, Y. Experimental Research on Vibration Assisted EDM of Micro-Structures with Non-Circular Cross-Section. J. Mater. Process. Technol. 2008, 208 (1-3),
289–298.
93. Klocke, F.; Lung, D. Micro Contouring by EDM with Fine Wires. In Proceedings of the 13th International Symposium for Electromachining, Bilbao, Spain, 2001, pp 767–782.
94. Liao, Y.-S.; Chen, S.-T.; Lin, C.-S. Development of a High Precision Tabletop Versatile CNC Wire-EDM for Making Intricate Micro Parts. J. Micromech. Microeng. 2005, 15 (2),
245–253.
95. Schoth, A.; Forster, R.; Menz, W. Micro Wire EDM for High Aspect Ratio 3D Microstructuring of Ceramics and Metals. Microsyst. Technol. 2005, 11 (4–5), 250–253.
96. Chen, S.-T. Fabrication of High-Density Micro Holes by Upward Batch Micro EDM. J. Micromech. Microeng. 2008, 18, 085002 (9 pp).
97. Asad, A. B. M. A.; Masaki, T.; Rahman, M.; Lim, H. S.; Wong, Y. S. Tool-Based Micro-Machining. J. Mater. Process. Technol. 2007, 192–193, 204–211.
98. Yu, Z. Y.; Masuzawa, T.; Fujino, M. Micro- EDM for Three-Dimensional Cavities – Development of Uniform Wear Method. Ann. CIRP 1998, 47 (1), 169–172.
99. Rajurkar, K. P.; Yu, Z. Y. 3D Micro-EDM Using CAD/CAM. Ann. CIRP 2000, 49 (1), 127–130.
100. Narasimhan, J.; Yu, Z. Y.; Rajurkar, K. P. Tool Wear Compensation and Path Generation in Micro and Macro EDM. Trans. NAMRI/SME 2004, 32, 151–158.
101. Yu, H.-L.; Luan, J.-J.; Li, J.-Z.; Zhang, Y.-S.; Yu, Z.-Y.; Guo, D.-M. A New Electrode Wear Compensation Method for Improving Performance in 3D Micro EDM Milling.
J. Micromech. Microeng. 2010, 20 (5), 055011.
102. Yan, M.-T.; Huang, K.-Y.; Lo, C.-Y. A Study on Electrode Wear Sensing and Compensation in Micro-EDM Using Machine Vision System. Int. J. Adv. Manuf. Technol. 2009, 42,
1065–1073.
103. Zhao, W.; Yang, Y.; Wang, Z.; Zhang, Y. A CAD/CAM System for Micro-ED-Milling of Small 3D Freeform Cavity. J. Mater. Process. Technol. 2004, 149, 573–578.
104. Hang, G.; Caol, G.; Wang, Z.; Tang, J.; Wang, Z.; Zhao, W. Micro-EDM Milling of Micro Platinum Hemisphere. In Proceedings of the 1st IEEE International Conference on
Nano/Micro Engineered and Molecular Systems, January 18–21, Zhuhai, China, 2006, pp 579–584.
105. Modica, F.; Ferraris, E.; Trotta, G.; Fassi, I.; Reynaerts, D. Fabrication of Micro-Nozzles via m-EDM Process. In Proceedings of International Conference on Advances in
Materials and Processing Technologies (AMPT 2010); Chinesta, F., Chastel, Y., Mansori, M. E., Eds.; American Institute of Physics, 2010; pp 1261–1266. CP1315.
106. Yu, Z.; Jun, T.; Kunieda, M. Dry EDM of Cemented Carbide. J. Mater. Process. Technol. 2004, 149, 353–357.
107. Tao, J.; Shih, A. J.; Ni, J. Near-Dry EDM Milling of Mirror-Like Surface Finish. Int. J. Elect. Mach. 2008, 13, 29–33.
108. Subbu, S. K.; Karthikeyan, G.; Ramkumar, J.; Dhamodaran, S. Plasma Characterization of Dry m-EDM. Int. J. Adv. Manuf. Technol. 2011, 56 (1–4), 187–195.
109. Zhang, Q. H.; Zhang, J. H.; Deng, J. X.; Qin, Y.; Niu, Z. W. Ultrasonic Vibration Electrical Discharge Machining in Gas. J. Mater. Process. Technol. 2002, 129, 135–138.
110. Yu, Z.; Takahashi, J.; Nakajima, N.; Sano, S.; Karato, K.; Kunieda, M. Feasibility of 3-D Surface Machining by Dry EDM. Int. J. Elect. Mach. 2005, 10, 15–20.
111. Tao, J. Investigation of Dry and Near-Dry Electrical Discharge Milling Processes. Ph.D. Thesis, The University of Michigan, 2008.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications 371

112. Masuzawa, T.; Tsukamoto, J.; Fujino, M. Drilling of Deep Micro-Holes by EDM. Ann. CIRP 1989, 38 (1), 195–198.
113. Lim, H. S.; Wong, Y. S.; Rahman, M.; Lee, E. M. K. A Study on the Machining of High-Aspect Ratio Micro-Structures Using Micro EDM. J. Mater. Process. Technol. 2003, 140,
318–325.
114. Rahman, M.; Lim, H. S.; Neo, K. S.; Kumar, A. S.; Wong, Y. S.; Li, X. P. Tool-Based Nanofinishing and Micromachining. J. Mater. Process. Technol. 2007, 185 (1–3), 2–16.
115. Chern, G.-L.; Wang, S.-D. Punching of Noncircular Micro-Holes and Development of Micro-Forming. Precis. Eng. 2007, 31 (3), 210–217.
116. Song, K. Y.; Chung, D. K.; Park, M. S.; Chu, C. N. Micro-Electrical Discharge Drilling of Tungsten Carbide Using Deionized Water. J. Micromech. Microeng. 2009, 19, 045006
(10 pp).
117. Kim, Y. T.; Park, S. J.; Lee, S. J. Micro/Meso-Scale Shapes Machining by Micro EDM Process. Int. J. Precis. Eng. Manuf. 2005, 6 (2), 5–11.
118. Fleischer, J.; Masuzawa, T.; Schmidt, J.; Knoll, M. New Applications for Micro-EDM. J. Mater. Process. Technol. 2004, 149, 246–249.
119. Morgan, C. J.; Vallance, R. R.; Marsh, E. R. Micro-Machining and Micro-Grinding with Tools Fabricated by Micro Electro-Discharge Machining. Int. J. Nanomanuf. 2006, 1 (2),
242–258.
120. Kuo, C.-L.; Huang, J.-D. Fabrication of Series-Pattern Micro-Disk Electrode and Its Application in Machining Micro-Slit of Less than 10 mm. Int. J. Mach. Tools Manuf. 2004,
44, 545–553.
121. Egashira, K.; Taniguchi, T.; Hanajima, S. Planetary EDM of Micro Holes. Int. J. Elect. Mach. 2006, 11, 15–18.
122. Yu, Z. Y.; Rajurkar, K. P.; Shen, H. High Aspect Ratio and Complex Shaped Blind Micro Holes by Micro EDM. Ann. CIRP 2002, 51 (1), 359–362.
123. Kim, B. H.; Park, B. J.; Chu, C. N. Fabrication of Multiple Electrodes by Reverse EDM and Their Application in Micro ECM. J. Micromech. Microeng. 2006, 16 (4), 843–850.
124. Weiliang, Z.; Zhenlong, W.; Desheng, D. A New Micro-EDM Reverse Copying Technology for Microelectrode Array Fabrication. In Proceedings of International Technology and
Innovation Conference 2006 (ITIC 2006) (CP524) Hangzhou, China, 6–7 Nov. 2006, pp 1633–1636.
125. Jeong, Y. H.; Yoo, B. H.; Lee, H. U.; Min, B. K.; Cho, D. W.; Lee, S. J. Deburring Microfeatures Using Micro-EDM. J. Mater. Process. Technol. 2009, 209, 5399–5406.
126. Zilong, P.; Zhenlong, W.; Yinghuaia, D.; Hui, C. Development of a Reversible Machining Method for Fabrication of Microstructures by Using Micro-EDM. J. Mater. Process.
Technol. 2010, 210, 129–136.
127. Hayakawa, S.; Ori, R. I.; Itoigawa, F.; Nakamura, T.; Matsubara, T. Fabrication of Microstructure Using EDM Deposition. In International Symposium for Electromachining
(ISEM XIII), Bilbao, Spain, 2001, Vol. 13, pp 783–793.
128. Sheu, D.-Y. Multi-Spherical Probe Machining by EDM Combining WEDG Technology with One-Pulse Electro-Discharge. J. Mater. Process. Technol. 2004, 149, 597–603.
129. Fukuzawa, Y.; Mohri, N.; Tani, T. Electrical Discharge Machining Phenomena of Insulating Silicon Ceramics with an Assisting Electrode. Int. J. Elect. Mach. 1997, 2, 25–30.
130. Mohri, N.; Fukuzawa, Y.; Tani, T. Some Considerations to Machining Characteristics of Insulating Ceramics towards Practical Use in Industry. Ann. CIRP 2002, 51, 161–164.
131. Schubert, A.; Zeidler, H.; Wolf, N.; Hackert, M. Micro Electro Discharge Machining of Electrically Nonconductive Ceramics. In The 14th International ESAFORM Conference on
Material Forming: ESAFORM 2011, Vol. 1353; AIP Conference Proceedings, 2011; pp 1303–1308.
132. Muttamara, A.; Fukuzawa, Y.; Mohri, N.; Tani, T. Probability of Precision Micro-Machining of Insulating Si3N4 Ceramics by EDM. J. Mater. Process. Technol. 2003, 140,
243–247.
133. Dahmardeh, M.; Nojeh, A.; Takahataa, K. Possible Mechanism in Dry Micro-Electro-Discharge Machining of Carbon-Nanotube Forests: A Study of the Effect of Oxygen.
J. Appl. Phys. 2011, 109, 093308.
134. Takahata, K.; Gianchandani, Y. B. Batch Mode Micro-Electro-Discharge Machining. IEEE/ASME J. Microelectromech. Syst. 2002, 11, 102–110.
135. Yamazaki, M.; Suzuki, T.; Mori, T.; Kunieda, M. EDM of Micro-Rods by Self-Drilled Holes. J. Mater. Process. Technol. 2004, 149, 134–138.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
This page intentionally left blank

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 11.15 Electrochemical Micromachining
SS Joshi and D Marla, Indian Institute of Technology Bombay, Mumbai, India
 2014 Elsevier Ltd. All rights reserved.

11.15.1 Introduction 373

11.15.1.1 Historical Perspective 374
11.15.1.2 Advantages 374
11.15.1.3 Applications 375
11.15.2 Fundamentals of EMM 375
11.15.2.1 Basic Electrochemistry 375
11.15.2.1.1 Anode Reactions 377
11.15.2.1.2 Cathode Reactions 377
11.15.2.1.3 Reactions within the Bulk Electrolyte 377
11.15.2.2 Material Removal 377
11.15.2.3 Microscopic Analysis of EMM 378
11.15.2.4 Electric Double Layer 381
11.15.3 Pulsed Electrochemical Micromachining 381
11.15.4 Experimental Setup 383
11.15.4.1 Mechanical Machine Setup 385
11.15.4.2 Electrolyte Flow System 385
11.15.4.3 Power Supply 386
11.15.4.4 EMM Process Controller 386
11.15.5 Characteristics of EMM 387
11.15.5.1 Voltage 387
11.15.5.2 Pulse On-Time and Off-Time 387
11.15.5.3 Electrolyte 389
11.15.5.4 Interelectrode Gap 389
11.15.5.5 Tool Electrode 389
11.15.6 EMM-Based Processes 391
11.15.6.1 Drilling Using EMM 391
11.15.6.1.1 Electrochemical Drilling for Microholes 391
11.15.6.1.2 Shaped Tube Electrochemical Drilling 392
11.15.6.1.3 Capillary Drilling 393
11.15.6.1.4 Electrostream Drilling 393
11.15.6.1.5 Electrolytic Jet Drilling 393
11.15.6.1.6 Comparison of Drilling Processes 393
11.15.6.2 Wire-EMM 394
11.15.6.2.1 Microelectrochemical Milling 395
11.15.6.3 Electrochemical Polishing 396
11.15.6.4 Solid Electrochemical Micromachining 397
11.15.6.5 STM-Based Electrochemical Nanofabrication 398
11.15.6.6 EMM-Based Hybrid Processes 398
11.15.6.6.1 Laser-Assisted Jet Electrochemical Machining 398
11.15.6.6.2 Oxide Film Laser Lithography Using EMM 399
11.15.6.6.3 Microelectrochemical Discharge Machining 400
11.15.6.6.4 Ultrasonic-Assisted EMM 401
11.15.6.6.5 Microelectrochemical Grinding 401
11.15.7 Conclusions and Future Scope 401
References 402

11.15.1 Introduction

Electrochemical machining (ECM) is one of the latest innovations and a powerful technique in manufacturing industry. It has
farther reaching capabilities for metal removal than tradition mechanical tools and is in many ways a simpler method. This process
is based on controlled metal dissolution from a workpiece that constitutes the anode in an electrolytic cell. Essentially, this process
involves passing current between the two electrodes, between which the electrolyte flows and carries away the dissolved metal. Metal
is gently removed by dissolution atom by atom, rather than mechanically torn away as done in the traditional machining processes.

Comprehensive Materials Processing, Volume 11 http://dx.doi.org/10.1016/B978-0-08-096532-1.01108-0 373

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
374 Electrochemical Micromachining

Table 1 Comparison between ECM and EMM

Major machining characteristics Electrochemical machining (ECM) Electrochemical micromachining (EMM)

Voltage 10–30 V <10 V

Current 150–10 000 A <1 A
Current density 20–200 A cm2 75 A cm2
Power supply-DC Continuous/pulses Pulsed
Frequency Hz–kHz range kHz–MHz range
Electrolyte flow 10–60 m s1 <3 m s1
Electrolyte type Salt solution Natural salt or dilute acid/alkaline solution
Electrolyte temperature 24–65  C 37–50  C
Electrolyte concentration >20 g l1 <20 g l1
Size of the tool Large to medium (>1 mm) Micro (<1 mm)
Interelectrode gap 100–600 mm 5–50 mm
Operation Maskless Mask/maskless
Machining rate 0.2–10 mm min1 w5 mm min1
Side gap >20 mm <10 mm
Accuracy 0.1 mm 0.02–0.1 mm
Surface finish Good, 0.1–1.5 mm Excellent, 0.05–0.4 mm
Problems due to waste disposal/toxicity Low Low to moderate
Reproduced from Bhattacharyya, B.; Munda, J.; Malapati, M. Advancement in Electrochemical Micro-machining. Int. J. Mach. Tools Manuf. 2004, 44, 1577–
1589.

When the ECM process is adopted in micromachining range for manufacturing precision shapes, it is called electrochemical
micromachining (EMM). The term micromachining refers to material removal at small dimensions that range from several
micrometers to a millimeter (1–1000 mm) (1). However, due to the very small interelectrode gap (several micrometers) between the
anode and the cathode, the presence of an electric double layer (EDL), high current density, and flow through a narrow inter-
electrode gap makes the EMM process quite different from that of the electrochemical process usually applied at macroscales. Table
1 illustrates the general comparison between ECM and EMM (2).

11.15.1.1 Historical Perspective

The foundation for ECM was laid in the nineteenth century by Michael Faraday (1791–1867), one of the greatest researchers of all
time, when he discovered the principle of anodic metal dissolution in an electrochemical cell. During the early years after the
discovery by Michael Faraday, the technique was extensively used for electroplating and electropolishing (3).
However, ECM turns out to have been first proposed in 1929 by a Russian researcher, W. Gusseff (4). He filed a patent for an
ECM process with many features almost identical to the process as now practiced. Many years later, in 1941, Burgess (5) appeared
with a publication in the Electrochemical Society demonstrating the possibilities of the process. He drew attention to the magical
way in which material is steadily removed without deforming the workpiece, emphasizing its striking differences with the tradi-
tional machining process. Despite several useful attributes of the process, it was more than 10 years before it was possible to control
the process to be used in the industry.
In 1959, Anocut Engineering Company established the anodic metal machining technique as a commercially suitable technique.
One year later, this was followed by SIFCO (Steel Improvement and Forge Company) for commercial application (5). Anocut
Engineering Company played a major role in pioneering the process and in overcoming the technical problems peculiar to the
equipment used for it. Other firms entered the field, and through collaboration between users and machine tool makers, brought
about the rapid development of a wide range of electrochemical machines from 1960 through the rest of the decade. The technique
was applied in several ways as a machining technique in the 1960s and 1970s, particularly in the gas turbine industry (3). Over the
next few years, ECM received a great impetus. Its applications encompassed all fields and have become boundless. Nowadays,
several companies are using an ECM technique that can hardly be compared to the anodic dissolution technique it once was.
Using the latest technological advances, the process is also applied extensively for micromachining operations. The interelec-
trode gap size in EMM process is controlled in the range of a few micrometers. A pulsating power supply is used to produce
ultrashort pulses of the order of nanoseconds. The use of pulsed current enabled the process to be localized, thereby increasing the
accuracy of the process. Electrolyte regeneration using microfiltration has enabled cleaning of electrolyte to a parts per million level
and can therefore be reused indefinitely. All these features have made the ECM process viable for application for micromachining.
Currently, EMM has found wide application in micromechanics, microelectronics, and microsystem technologies because of its
advantages concerning complex shape generation, precision, surface quality, etc. (6).

11.15.1.2 Advantages
The EMM process offers numerous advantages over conventional machining techniques. Unlike the conventional machining
processes, EMM offers no burrs on the machined surfaces. EMM is a noncontact process, and due to this, upper layer deformations
on the workpiece and tool wear do not occur. Apart from conventional processes, some nonconventional processes like electric
discharge machining (EDM) and laser beam machining (LBM) are thermal processes that involve the formation of heat-affected

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrochemical Micromachining 375

zones and microcracks on the machined surface (7). In comparison, EMM does not produce thermal or mechanical stresses on the
machined surface. The surface finish of the resulting machined surface has an excellent quality. Unlike chemical-based machining
processes like chemical etching, which use toxic chemicals and generate hazardous waste, EMM is environmentally friendly and
does not pose any such hazards. One of the most important features of the EMM process is that it can be used to produce complex
three-dimensional (3D) shapes that are almost impossible to be fabricated using conventional techniques (8).
The process is also independent of the hardness of the workpiece material and the tool. This makes the process feasible to be
applied for machining of hard and tough alloys, even with soft electrodes. The material removal is precisely governed by the current
density and to a very less extent on the material properties. The material removal rate (MRR) in EMM is one of the highest among the
nonconventional processes like EDM, ultrasonic machining, etc. for micromachining applications. The process is also economically
feasible. The initial investments are very high, but the subsequent recurring costs are low. The tooling costs are much less as are the
running costs, because the power consumption is low. This makes the process more economical.
Some of the important advantages like high MRRs, good surface finish, better precision and control, environmental friendly
nature, and its applicability to wide range of materials have made EMM a very promising micromachining technology that can be
applied in wide areas of micromanufacturing.

11.15.1.3 Applications
During the early years after its commercial use, ECM was widely used in diverse industrial application in the fields of aerospace,
automobile, and electronics for manufacturing of various parts. ECM was looked upon as an alternative to conventional machining
operations for difficult-to-cut materials and for generating complex structures (9). With advances in machining accuracy, ECM was
later applied for manufacturing of very small mechanical parts and was effectively used for manufacturing of components in
electronic and precision industries (10). Over the last two decades, ECM was applied for machining at microscales by Philishave
(now Philips) for manufacturing of shave heads that have 140 mm slots with an accuracy of 10 mm over an area of 5 cm2 (11). EMM
has a wide application in aerospace and aircraft industries. One of the most widely used application is in the production of
microholes in turbine blades for generating cooling effects (12). Drilling using EMM is one of the widely used techniques because of
the acceptable limits of the hole taper. Electrochemical polishing (ECM) offers a very good surface finish and is exploited in
a number of applications that require extremely high surface finish. One such application is the surface finishing of print bands,
which is an essential requirement in the print band manufacturing process (2). Electrochemical deburring is widely used for
removing burrs left by other operations with virtually no damage to the workpiece. Over the years, EMM has been widely used in
manufacturing of micro nozzles, channels, various sensors and actuators in MEMS/NEMS, and micro die/mold manufacture (13). A
combination of electrochemical-based drilling, etching, polishing, etc. is also used with other microfabrication techniques to
fabricate a variety of components. Microfabrication of components in these industries employs basic integrated circuit techniques
such as film formation, lithography, and etching. EMM offers a unique application in fabrication of complex internal structures that
are usually difficult to fabricate by other micromachining processes. A variety of complex internal structures that include reverse-
tapered hole, barrel-shaped hole, spherical cavity, internal grooves in microholes, and many other complex structures can be
fabricated using EMM-based techniques. Some of the microstructures fabricated using EMM are shown in Figure 1(a)–1(g) (14–
17). Figure 1(a) shows an array of microholes drilled on stainless steel of diameter 40 mm and depth 100 mm, and a micro-
structured surface of Ti with 30 mm cavities is shown in Figure 1(b). Figure 1(c) shows a microwall of 160 mm height, 15 mm width,
and 400 mm length fabricated on stainless steel. Figure 1(d) shows microcolumns of 58 mm width and 83 mm height, fabricated
using electrochemical micromilling. Figure 1(e)–1(f) shows microgrooves of 20 mm width fabricated using wire EMM. Figure 1(g)
shows a curve generated using a 8 mm electrode, demonstrating the precision and accuracy of the process in complex-shape
generation.

11.15.2 Fundamentals of EMM

EMM involves the removal of material atom by atom. The fundamental governing process involves electrochemical reactions that
occur at the electrode surfaces. The material removal is the outcome of the electrochemical reactions that dissolve the workpiece
material into the electrolyte. As the anode (workpiece) dissolves, they combine with the electrolyte molecules to form hydroxides.
The dissolved material is in the form of soluble/insoluble products that get washed away along with the flowing electrolyte. This
section presents the fundamental processes that form the basis of the EMM process.

11.15.2.1 Basic Electrochemistry

When a potential difference is applied between two conducting electrodes that are placed in a conducting medium, known as an
electrolyte, an electric current passes through the medium. Unlike metallic conductors or other conducting materials like graphite,
where the passage of electric current occurs due to electron motion, the passage of electric current through an electrolyte is invariably
accompanied by the motion of ions in addition to electrons. This phenomenon is often termed as electrolysis. The electrolytes used
in EMM are molten solutions of a neutral salt, base, or acid. In a majority of applications, neutral salt solutions such as NaCl and
NaNO3 are used. However, acids such as H2SO4 and HCl and bases such as NaOH are used in specialized applications. The
electrolyte solution comprises cations and anions dispersed in an aqueous medium that are free to move. These anions and cations
are evenly distributed in the solution such that the electric field is uniform. Under the application of potential difference, these

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
376 Electrochemical Micromachining

Figure 1 Applications of EMM: (a) an array of microholes (reproduced from Kim, B. H.; Na, C. W.; Lee, Y. S.; Choi, D. K.; Chu, C. N. Micro Elec-
trochemical Machining of 3d Micro Structure Using Dilute Sulfuric Acid. CIRP Ann. – Manuf. Technol. 2005, 54 (1), 191–194), (b) Ti surface micro-
structured with 30 mm cavities (reproduced from Landolt, D.; Chauvy, P.-F.; Zinger, O. Electrochemical Micromachining, Polishing and Surface
Structuring of Metals: Fundamental Aspects and New Developments. Electrochim. Acta 2003, 48 (20–22), 31853201), (c) cross-sectional view
of microwall (reproduced from Park, B. J.; Kim, B. H.; Chu, C. N. The Effects of Tool Electrode Size on Characteristics of Micro Electrochemical
Machining. CIRP Ann. – Manuf. Technol. 2006, 55 (1)), (d) microcolumns (reproduced from Kim, B. H.; Na, C. W.; Lee, Y. S.; Choi, D. K.; Chu, C. N. Micro
Electrochemical Machining of 3d Micro Structure Using Dilute Sulfuric Acid. CIRP Ann. – Manuf. Technol. 2005, 54 (1), 191–194), (e)–(f)
microgrooves (reproduced from Kim, B. H.; Na, C. W.; Lee, Y. S.; Choi, D. K.; Chu, C. N. Micro Electrochemical Machining of 3d Micro Structure Using
Dilute Sulfuric Acid. CIRP Ann. – Manuf. Technol. 2005, 54 (1), 191–194), and (g) groove along a curve (reproduced from Yong, Liu; Di, Zhu; Yongbin,
Zeng; Shaofu, Huang; Hongbing, Yu. Experimental Investigation on Complex Structures Machining by Electrochemical Micromachining Technology.
Chin. J. Aeronaut. 2010, 23 (5), 578–584).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrochemical Micromachining 377

anions and cations tend to migrate toward the anode and cathode, respectively, as a result of the electric force. When the two
electrodes are joined by an external circuit, the continuous migration of ions results in the flow of current in the external circuit.
The flow of current at the electrode surface is induced by electrochemical reactions. The study of the material removal process in
EMM requires the basic understanding of the electrochemical reactions that occur at the electrode surfaces. The electrochemical
reactions are the driving mechanisms of material removal in electrochemical-based processes. Different types of reactions occur at
the electrode surfaces and in the bulk electrolyte.

11.15.2.1.1 Anode Reactions

In EMM, the workpiece is preferably chosen to be the anode as it gets dissolved into the solution due to the electrochemical
reactions. At the anode, the metal surface dissociates into atoms and then dissolves in the solution as cations leaving behind the
electrons (18). The reaction can be written as:
M/Mnþ þ ne [1]

where n is the valence of metal dissolution or the number of electrons removed from metal atom. This is the predominant reaction
that occurs at the anode surface. Apart from this, different reactions may also take place at the anode surface depending on the
metal–electrolyte combination, especially at high current densities. In the presence of passivating electrolytes that contain oxidizing
anions such as nitrates and chlorates, the metal reduction may also be accompanied by oxygen evolution and formation of
hydrogen ions as a result of electrolysis of water (18):
2H2 O/O2 þ 4Hþ þ 4e [2]

while, in the presence of non-passivating electrolytes that contain anions such as fluorides, chlorides, bromides, and iodides, the
liberation of hydrogen gas rather than oxygen may occur more:
2Cl /Cl2 þ 2e : [3]

These reactions occur simultaneously along with the anode dissolution and thereby lower the current efficiency for anode
dissolution. Therefore, not all the current is utilized in the anodic dissolution process. The current efficiency is generally observed to
be marginally lower by about 5–10% only.

11.15.2.1.2 Cathode Reactions

At the cathode, different reactions occur based on the type of electrolyte used. In the case of neutral or basic electrolytes, the main
reaction is the electrolysis of water that causes the liberation of hydrogen gas and a local increase in alkalinity of the electrolyte
solution due to the formation of hydroxyl ions (3):
2H2 O þ 2e /H2 [ þ 2OH [4]

whereas, in the presence of acid electrolytes, neutralizing the charge on hydrogen ions is the main reaction, which also results in the
liberation of hydrogen.
2Hþ þ 2e/H2 [ [5]

Metal ions may also reach the cathode in acidic electrolyte and get deposited there.
Mnþ þ ne /M [6]

Due to this, acid electrolytes need frequent replenishment, or a system that periodically reverses the direction of electrolyzing
current so as to deplete the deposits that accumulate on the cathode. Even in case of neutral electrolytes, a mild deposit is seen at the
cathode. This is perhaps due to the electrophoretic phenomenon that drives the metal atoms that get dissociated from the hydroxide
state. These migrating ions tend to deposit at the cathode.

11.15.2.1.3 Reactions within the Bulk Electrolyte

Depending on the material and electrolyte combination, metal ions may combine with the hydroxyl ions to form a hydroxide that
generally appears as a precipitate in the electrolyte solution (3).
Mnþ þ nOH /MðOHÞn ðsÞ [7]

The hydroxide precipitate remains mostly suspended, while a part of it can stick to the cathode surface and the electrolyte can be
filtered for it to be reused. The filtering of the electrolyte can be done by centrifugal separation or gravitational settling or can also be
done by simply using a filter paper or a mesh.

11.15.2.2 Material Removal

The material removal in an electrochemical process occurs due to the anodic dissolution. The rate of dissolution of the anode is
governed by Faraday’s law of electrolysis. According to this law, the rate of anodic dissolution is directly proportional to the amount
of current passed through the electrolyte. It is expressed as (19)
AI
_ ¼ hI
m [8]
nF

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
378 Electrochemical Micromachining

Table 2 Metal dissolution valence in different metal–electrolyte combinations

Metal Electrolyte Dissolution valence

Ni NaCl 2
Fe NaCl 2 and 3
Ni NaNO3 2
Fe NaNO3 2
Ni NaClO3 2
Fe NaClO3 2
Cr NaCl 6
Cr NaNO3 6
Cu KCl 1 and 2
Cu KNO3 2 and 1
Cu K2SO4 2 and 1
Ti NaCl 4
Ti NaBr 4
Mo KOH 6
Mo K2CO3 6
Reproduced from McGeough, Joseph. Micromachining of Engineering Materials; Marcel Dekker, Inc,
2002.

where I is the current (A), hI is the current efficiency, A is the molecular weight of the anode (g mol1), n is the valence of anode, and
F is the Faraday constant, which is equal to 96 485  C. 1 F is also defined as the amount of charge required to dissolve 1 g equivalent
of the anode.
As mentioned earlier in the section, not all the current goes into dissolution of the anode. The current efficiency is defined as the
ratio of observed amount of metal dissolved to the theoretical amount predicted from Faraday’s law for the same specific condi-
tions. It can be obtained from weight loss experiments.
ðDmÞnF
hI ¼ [9]
AIt

where Dm is the mass of anode dissolved and t is the machining time. It is sometimes more convenient to use volumetric removal
rate, which can be written as: AI
y_ ¼ hI [10]
hra F

The valence n purely depends on the metal–electrolyte combination. Certain metals exhibit different valences in the presence of
different electrolytes. For example, Fe exhibits a valence of þ2 and þ3 in the presence of NaCl electrolyte, while only þ2 is exhibited
in the presence of NaNO3. In some cases, the valence of anode also depends on the voltage or current density. The experimentally
determined data of valence for different metal–electrolyte combinations are presented in Table 2 (20).

11.15.2.3 Microscopic Analysis of EMM

A typical case of potential difference applied across the two electrodes in EMM is shown in Figure 2, where a voltage of þV is
applied at the anode, while the cathode is grounded. The application of voltage causes the electrolyte ions to move toward the
oppositely charged electrodes, thus inducing a flow of current inside the electrolyte. The Faraday’s law of electrolysis suggests
that the current density at the anode surface is the governing parameter of anodic dissolution. The prediction of current
density at the anode necessarily involves understanding the flow of charges inside the electrolyte due to the application of
voltage.
To illustrate the physics of charge dynamics in EMM, the microscopic view of the interelectrode gap in Figure 2 can be
represented as shown in Figure 3. The potential difference between the electrodes creates an electric field (E), which causes the
motion of ions. Naively, one might assume a uniform electric field, E ¼ V/a to exist in the interelectrode gap. However, this
assumption would be perfectly valid at macroscales, where the net charge inside the interelectrode gap would be nearly zero. In
the case of EMM, the small interelectrode gap would create high electric fields and cause charge segregation. Due to this, the
assumption of a uniform electric field will be valid only for specific conditions. Therefore, a more general approach to study the
charge dynamics in the interelectrode gap should necessarily be considered for a better understanding of the process at
microscales.
In Figure 3, the ions in the electrolyte move due to the concentration and the potential gradients. These gradients tend to
accelerate the ions. But they collide with the solvent molecules that offer resistance to the flow. The velocity of these molecules has
been experimentally found to be proportional to the gradient of their electrochemical potential, which is the sum of concentration
and the electric potential (21,22), and is given by
yi ¼ ki ðkb TVlnðni Þ þ Zi eVjÞ [11]

where, j is the electric potential, kb is the Boltzmann constant, e is the charge of electron, yi is the velocity, Zi is the valance, ki is the
proportionality constant, ni is the number density, and i refers to the type of ions (cation or anion).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrochemical Micromachining 379

Figure 2 Schematic of the EMM process. Reproduced from Marla, Deepak; Joshi, Suhas S.; Mitra, Sushanta K. Modeling of Electrochemical
Micromachining: Comparison to Experiments. J. Micro/Nanolithog., MEMS, MOEMS 2008, 7 (3), 033015.

Figure 3 Schematic representation of charged ions within the electrode gap.

The flux ji of ions is defined as the number of ions that flow across a plane of unit area placed at right angles to the direction of
flow per unit time (22). The flux can also be written as
ji ¼ ni yi [12]
Substituting eqn [11] in eqn [12] and solving gives
ji ¼ ki kb TVni þ ni ki Zi eVj [13]

The coefficient of the concentration gradient is also called as diffusion coefficient Di ¼ ki kb T. Equation [13] can be written in
terms of Di as
ni Di Zi e
ji ¼ Di Vni þ Vj [14]
kb T

In a typical one-dimensional (1D) case,

vni ni Di Zi e vj
ji ¼ Di þ [15]
vx kb T vx

The current density can be obtained by multiplying the flux of the ions by the charge Zie.
vni Di 2 2 vj
Ji ¼ Zi eji ¼ Di Zi F þ ni Z e [16]
vx kb T i vx

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
380 Electrochemical Micromachining

Figure 4 Control volume used for the EMM process.

To establish a relation between the flux and the rate of change of ions, consider the control volume analysis as shown in Figure 4.
Here, an element of length ‘dx,’ unit cross-sectional area is considered. A change in number of ions in the control volume is equal to
the difference in the flux entering and flux leaving the control volume. This can be mathematically written as (23)
vni
ji xþdx  ji x ¼ dx [17]
vt

This gives
vji vni
¼ [18]
vx vt
Further, differentiating the eqn [15], we get
 
vji vni v2 ni Di Zi e v vj
¼ ¼ Di 2 þ ni [19]
vx vt vx kb T vx vx

The above equation (eqn [19]) can be generalized for a 3D EMM process, and can be written as
vni Zi eDi
¼ Di V2 ni þ V$ðVjni Þ [20]
vt kb T

Equation [20] is also called as the Smoluchowski equation (23). This is the basic equation that can be applied to any system of
particles, governed by diffusion and electrostatic force. This is a nonlinear partial differential equation, which until now does not
have closed-form solution. Van Kampen (1993) (24) has presented an approximate solution to the Smoluchowski equation. The
steady-state equation of the Smoluchowski equation (eqn [20]) gives the Boltzmann distribution (25).
In the case of EMM, the small interelectrode gap (wmm) causes a very high electric field of the order of 106V m1. Due to this, the
motion of the ions is purely due to the electric potential gradient. The diffusion of ions due to concentration gradient will be very
less compared to the high electrophoretic movement. Therefore, the first term in eqn [20] that represents the motion of ions due to
concentration gradient can be neglected. The equation for the 1D case can thus be written as
   
vni Di Zi e v vj D0 Zi e v2 j vj vni
¼ ni ¼ ni þ [21]
vt kb T vx vx kb T vx2 vx vx

From the above equation (eqn [21]), it is observed that the rate of change of number density of the ions is a function of the
electrostatic potential and their number density. The presence of an electric field causes the cations and the anions to segregate and
move in the opposite directions. This results in a change in the charge density, which further causes the potential to vary, eventually
leading to a change in the current density. The process is time variant.
Under the conditions of electroneutrality, which is typically a case at macroscales, the electric field will be uniform and equal to
V m1. The voltage across the gap will be linear and the current density at the anode surface will be independent of time. However, at
microscales the charge segregation is significant and causes a net charge to appear in the electrolyte in the interelectrode gap, even
though the positive and negative charges add up to zero. The charge density at any given point in the interelectrode gap is given by
the difference between the positive and the negative charges per unit volume. The charge density can be written as
r ¼ Zeðnþ  n Þ [22]

Using the charge density, the electrostatic potential across the interelectrode gap can be obtained by the Poisson equation:
r
V2 j ¼  [23]
ε

The electrostatic potential across the interelectrode gap and the number densities of anions and cations are interdependent.
Using eqns [21–23], the system of governing equations for EMM can be formulated as:
 
vnþ ZeD0 v2 j vj vnþ
¼þ nþ þ [24]
vt kb T vx2 vx vx
 
vn ZeD0 v2 j vj vn
¼ n þ [25]
vt kb T vx2 vx vx

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrochemical Micromachining 381

v2 j Zi e
¼ ðn  nþ Þ [26]
vx2 ε

The above system of equations are also called as Nernst-Planck-Poisson equations (23,26). Solving the above equations requires
the initial and boundary conditions. Initially, at t ¼ 0, the number density of anions and cations is uniform throughout the
electrolyte and is equal to n0. This gives the initial conditions as
nþ ðx; 0Þ ¼ n ðx; 0Þ ¼ n0 [27]

The boundary conditions are the voltages applied across the interelectrode gap, shown in Figure 2. Neglecting the over-
potentials that exist at the electrode–electrolyte interface, the boundary conditions can be written as
jð0; tÞ ¼ 0 [28]

jða; tÞ ¼ V [29]

The above system of equations (eqns [15]–[17]) can be solved along with the initial conditions (eqn [27]) and the boundary
conditions (eqns [19] and [20]) to get the current and, hence, the MRR.
As the motion of ions due to the concentration gradient is negligible, the original eqn [16] for current density can be rewritten as
Di 2 2 vj
Ji ¼ ni Z e [30]
kb T i vx

From the governing equations (eqns [15]–[17]), the current density varies with time and the distance along the interelectrode
gap. The current can be obtained by multiplying the current density with the cross-sectional area (I ¼ JAc), where Ac is the cross-
sectional area of the electrode. The original eqn [8] can now be written as
AJAc
_ ¼
m [31]
zF

Due to the time dependence of the current, eqn [31] should be integrated with time to get the amount of material removed. The
workpiece erosion occurs when the anions impinge the surface. Therefore, the current due to anions at the surface of the workpiece
should be used in evaluating the MRR.
The amount of material removed can be written as
ZT
AAc
m¼ J dt [32]
ZF x¼a
0

Substituting eqn [30] in eqn [32], we get

ZTtot
2 e2
AAc D Z vj
m¼ n dt [33]
ZF kb T vx x¼a
0

where Ttot is the total machining time, for which the voltage has been applied.

11.15.2.4 Electric Double Layer

One of the most important aspects of the electrolyte–metal interaction is the formation of EDL. The EDL is formed when
a conducting plate is placed in an electrolyte containing mobile ions. When the potential is applied to the electrode, as is the
case in EMM, the ions of the electrolyte tend to get attracted to the corresponding oppositely charged electrodes as shown in
Figure 5. Figure 5 shows the EDL formed only at the anode surface. A similar EDL is also formed at the tool (cathode)
surface. According to the Helmholtz model of EDL, the electrolyte ions rearrange to form two different layers: compact and
diffuse layers. This is the reason for the name EDL (27). Due to the electric force, ions are pulled closer to the oppositely
electrode surface. This causes some of the counter ions to stick to the surface of the electrode. The concentration of the counter
ions decreases gradually away from the electrode surface. Close to the electrode surface is the compact layer, where the counter
ions are heavily concentrated and are immobile. From the compact layer to the uniform bulk liquid, the counter ion
concentration gradually decreases to that of the bulk electrolyte. Ions in this region are mobile. This layer is called the diffuse
layer.
The compact layer of EDL acts as a capacitor and has been exploited by using pulsed current to localize the current and thereby
achieve precision. This aspect is discussed in detail in a later part of the section.

11.15.3 Pulsed Electrochemical Micromachining

There are numerous issues that arise when ECM is applied for micromachining, such as nonlocalization of the electric field and
boiling of the electrolyte due to very small gap thickness and the removal of the debris (28). These problems have been tackled by

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
382 Electrochemical Micromachining

Figure 5 Schematic of electric double layer.

the use of pulsed voltage. In this process, pulses of voltage are applied. The machining process occurs during the pulse on-time (Ton),
which ranges from 5 to 5000 ns (9) and the dissolution products (sludge, gas bubbles, and heat) can be flushed away from the
interelectrode gap by the flowing electrolyte and tool position can be adjusted by the EMM-controller during the pulse off-time
(Toff). The pulse off-time enables uniform machining process.
The most important aspect of the pulsed EMM is the localization of the electric field. This is because of the formation of the EDL
at the surface of electrodes that acts as a capacitor. Upon application of a pulsed voltage between the tool and the workpiece
electrode, the EDL capacitances are charged and discharged. An equivalent circuit can be represented as shown in Figure 6.
Therefore, the pulse current consists of two parts: a capacitive current, IC, which charges the double layer; and the Faradaic current, IF
(29), which corresponds to the metal removal. The total pulsed current can be written as
I ¼ IC þ I F [34]
The charging or discharging of the capacitor undergoes exponentially with time. The time constant ðsÞ is the time taken for 63%
of charging or discharging of the capacitor, and is equal to the EDL capacitance multiplied by the resistance of the electrolyte along
the current path ðs ¼ RCÞ. Initially, the entire current is capacitive current (IC). The Faradaic current (IF) is zero initially, and then
gradually increases. When the double layer is completely charged (99%), the capacitor current (IC) becomes negligible and the
Faradaic current (IF) becomes equal to the pulsed current (I). The behavior of the Faradaic current (IF) with time is shown in
Figure 7. If the pulse on-time (Ton) is less than the double layer charging time ðsÞ, then IF remains smaller than I during the whole
pulse, and effective machining removal does not happen. This principle is used in the localization of the current. In Figure 7, Rsmall
and Rlarge are resistances of the larger and the smaller path between the tool and the work piece. Because of the small gap, Rsmall is
smaller than Rlarge, and hence the double layer charging time of the shorter path ðs1 Þ is smaller than the double layer charging time
of the larger path ðs2 Þ. Hence, when a voltage pulse with on-time Ton ðs < Ton << s2 Þ is applied, the electric current flows mainly
through the shorter path, thereby restricting the machining over the region close to the tool (6,30). Similarly, when the pulse is
withdrawn, the double layer discharges. Therefore, pulse off-time (Toff) should be selected such that it is more than the discharge
time; at the same time, it also should be enough for repositioning the tool and bringing in fresh electrolyte for the next pulse.
In EMM, ultra-short pulses of duration (Ton) ranging from 5 to 5000 ns are used. Under the condition of very short pulses (of
a few nanoseconds) only, the variation in the electrolyte properties like temperature and charge (ion) distribution inside the

Figure 6 Model of pulsed EMM process.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrochemical Micromachining 383

Figure 7 Variation of total current and Faradaic current with time.

electrolyte can be assumed to be constant. The concentration of the ions in the bulk of the electrolyte can be assumed to be constant,
i.e., n ¼ nþ zn0 (considering a z:z electrolyte). Therefore, the first two equations in the Nernst-Planck-Poisson equations become
zero. The system of governing equations for this case can be written as
v2 j
¼0 [35]
vx2
Equation [35] is based on the assumption that the electroneutrality exists at very short time scales. However, it is always
preferable to solve the Nernst-Planck-Poisson equation for an accurate prediction of the anodic current. Neglecting the over-
potentials, the boundary conditions can be written as
jð0; tÞ ¼ 0 [36]

jða; tÞ ¼ V [37]
Referring to eqn [30], the current density of either ions can now be written as
vj
J ¼ ke [38]
vx
where ke is the conductivity of the electrolyte, which is equal to
D0 2 2
ke ¼ n0 Z e [39]
kb T
The material removal (eqn [33]) in this case simplifies to

AAc vj
m¼ ke hd Ttot [40]
ZF vx x¼a

where Ac is the cross-sectional area of the tool, Ttot is the total time of machining, and hd is the duty factor of the pulse, which can be
defined as
Ton
hd ¼ [41]
Ton þ Toff
The current density depends not only on the voltage across the interelectrode gap but also on the gap size and the concentration
of the electrolyte. The current density as a function of applied voltage for different interelectrode gaps and electrolyte concentrations
that are solved using these equations (eqns [35]–[38]) is presented in Figures 8 and 9, respectively. It can be observed from Figure 8
that the current density decreases with an increase in the interelectrode gap. This is because with decrease in gap size, the electric field
increases and hence, current becomes high. Whereas, with an increase in electrolyte concentration, the current density increases as
seen from Figure 9. The higher the electrolyte concentrations, the higher is the number density of the cations and anions and hence
the high current density.

11.15.4 Experimental Setup

The conventional ECM machines have been effectively modified to suit the micromachining requirements. Micromanufacturing
requires the optimal control over the various process parameters that predominantly influence the machining process. The
experimental setup for EMM broadly consists of various subcomponents:
1. Mechanical machine setup
2. Electrolyte flow system
3. Power supply
4. EMM process controller
A typical laboratory setup showing the integrated view of the components is shown in Figure 10 and a schematic of the setup is
illustrated on Figure 11 (31).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
384 Electrochemical Micromachining

Figure 8 Variation of current density with voltage for an electrolyte concentration of 0.1 M NaNO3. Reproduced from Marla, Deepak; Joshi, Suhas S.;
Mitra, Sushanta K. Modeling of Electrochemical Micromachining: Comparison to Experiments. J. Micro/Nanolithog., MEMS, MOEMS 2008, 7 (3), 033015.

Figure 9 Variation of current density with voltage for varying electrolyte concentrations at a constant interelectrode gap of 10 mm. Reproduced from
Marla, Deepak; Joshi, Suhas S.; Mitra, Sushanta K. Modeling of Electrochemical Micromachining: Comparison to Experiments. J. Micro/Nanolithog.,
MEMS, MOEMS 2008, 7 (3), 033015.

Figure 10 A photographic view of the EMM setup. Reproduced from Marla, Deepak; Joshi, Suhas S.; Mitra, Sushanta K. Modeling of Electrochemical
Micromachining: Comparison to Experiments. J. Micro/Nanolithog., MEMS, MOEMS 2008, 7 (3), 033015.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrochemical Micromachining 385

Figure 11 Schematic of the EMM setup. Reproduiced from Yong, Li; Yunfei, Zheng; Guang, Yang; Liangqiang, Peng. Localized Electrochemical
Micromachining with Gap Control. Sens. Actuators, A 2003, 108 (1–3), 144–148.

11.15.4.1 Mechanical Machine Setup

The mechanical machine setup is the main part of the EMM setup where the machining process is carried out. This consists of the
tool holder, XYZ stage, and a machining chamber. A pictographic view of the machine setup is shown in Figure 12. Typically, the
tool is held by an attachment on Z-axis slide that can move vertically and is integrated with a stepper motor. The workpiece is
mounted on the machining chamber which is filled with an electrolyte. The machining chamber rests on the work mounting device
and is integrated with x-axis and y-axis stepper motors, which enable it to move along the xy-plane. All three stepper motors, i.e.,
x-axis motor, y-axis motor, and z-axis motor are interfaced with the CNC programmable controller. The axis-travels along the three
axes is typically around 100–200 mm. The motion of tool and machining chamber can be controlled by using a numerical control
(NC) programing. As the requirement for interelectrode gap is of the order of several micrometers, the resolution for the microfeed
is in the range of 0.1–1 mm per step.

11.15.4.2 Electrolyte Flow System

The electrolyte flow system provides fresh electrolyte to the machine chamber. The electrolyte is flushed in the interelectrode gap.
The flowing electrolyte carries away the dissolved material, which is later recycled. The machining process can also be carried out in
a still-electrolyte in the machining chamber. However, this has certain disadvantages. The continuous production of debris as
a result of machining makes the electrolyte impure, which becomes necessary to be filtered. Also, the presence of debris in the
interelectrode gap can lead to clogging and result in short-circuit.

Figure 12 A photographic view of mechanical machine setup.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
386 Electrochemical Micromachining

A schematic of the electrolyte flow system is shown in Figure 13. The machining chamber consists of an inlet and an outlet for
the electrolyte to flow in and out of the chamber. The flow of the electrolyte is directed at the machining zone with a moderate
velocity, ensuring that the microtool does not vibrate or bend due to the hydraulic force. The removed material during the
machining dissolves in the electrolyte. The used electrolyte flows out of the chamber into the reservoir, which is then filtered after
removing debris particles. The electrolyte is then pumped using a centrifugal pump. This ensures continuous recirculation of the
electrolyte. The internal piping and the pump are usually coated (e.g., polypropylene coating) to prevent corrosion. A flowmeter is
used to measure the flow of electrolyte and a valve is used to control the flow of electrolyte.

11.15.4.3 Power Supply

The main line AC power supply of 230 V is converted into a DC power supply that is required for EMM operation using a step-down
transformer and a silicon controlled rectifier unit. The applied voltage varies depending on the workpiece, tool, and electrolyte
combination. Typically, low voltage in the range of 2–14 V is used in EMM, which provides better machining conditions for
precision manufacturing. The continuous DC voltage is converted into pulsed voltage using a pulse generating module. Ultrashort
pulses ranging from microseconds to nanoseconds are used in EMM. This requires a very high-speed switching that can be achieved
using a metal oxide semiconductor field effective transistor. The pulse generator controls both voltage and the pulse parameters (i.e.,
pulse on-time and off-time). An oscilloscope is used to digitally store and monitor the pulse power parameters. The current in the
interelectrode gap is of 1–10 A and is measured using a multimeter.

11.15.4.4 EMM Process Controller

The process control system consists of eight main subsystems: a supervisory controller (SC), a tool position control system (TPCS),
a security control system (SCS), a flushing control system (FCS), a gap estimation system (GES), a man–machine interface (MMI),
a graphical user interface (GUI), and a pulse parameter control system (PPCS). The electrolyte flow rate can be controlled by FCS.
FCS can also be used to determine the electrolyte concentration and its temperature. Security issues such as smoke detection, status
of the emergency stop switches, and generation of security control signals such as emergency alarms, emergency lamps, and
emergency machine stop are controlled by SCS. The MMI consists of indicators that show the machining status such as voltage,
current, interelectrode gap. It also consists of manual-control switches that allows the user to manually control the functions of the
machine. The GUI is a software interface on the computer that allows the user to control the machining parameters. PPCS generates
the pulse parameters: pulse on time, pulse off time, and voltage and peak currents, according to the supervisory control signal. The
position and motion of the tool are controlled using TPCS. The main components of the TPCS are the DC servo motor/stepper
motor and a linear encoder. The linear encoder is used to obtain the feedback of the tool holder. The interelectrode gap, which plays
a crucial role in the machining process, is controlled using GES. EMM requires a very narrow interelectrode gap (wmicrometers)
that cannot be easily set manually. The interelectrode gap can be set using the GUI. Initially, the TPCS enables the tool holder down
until the tool touches the workpiece. Upon touching the workpiece, the short circuit is sensed by the current sensor and sends it to
the supervisory controller and marks the reference. From this position, the tool moves upward equal to the distance specified by the
user in the GUI. During the EMM operation, the continuous erosion of the workpiece leads to an increase in the interelectrode gap.

Figure 13 Schematic of the electrolyte flow system. Reproduced from Marla, Deepak; Joshi, Suhas S.; Mitra, Sushanta K. Modeling of Electrochemical
Micromachining: Comparison to Experiments. J. Micro/Nanolithog., MEMS, MOEMS 2008, 7 (3), 033015.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrochemical Micromachining 387

Maintaining a constant interelectrode gap throughout the machining operation is an important requirement to achieve high
accuracy. The constant interelectrode is achieved by the tool feed system. The current sensor senses the initial current when the
machining initiates. This value is maintained constant by the SC throughout the machining operation by adjusting the tool position
through TPCS. This process enables a constant interelectrode gap throughout the machining process. The motion of the tool along
the workpiece can be controlled in a similar way as done in a computer numerical control (CNC) machine using GUI. Appropriate
programs can be written using g and m codes to generate the necessary profiles on the surface.

11.15.5 Characteristics of EMM

The anodic dissolution (i.e., material removal) and the current localization are the two most important characteristics of the EMM
process. These two characteristics depend on various operating conditions like voltage, interelectrode gap, electrolyte concentration,
and pulsed on-time and off-time. While MRR is one the important characteristic that governs the speed of machining process, the
localization of current is equally important, especially in micromachining operation, which governs the precision of the machining
process. Therefore, it is very important to optimize these parameters to achieve high machining rates without compromising the
precision of the machining process. This section presents a study of the effect of various process parameters on the two most
important characteristics of EMM process.

11.15.5.1 Voltage
In ECM, two types of power supply are generally used: DC (full wave rectified) and pulsed DC. However, in the case of EMM, it is
largely beneficial to use pulsed DC. The use of pulsed DC in EMM has a lot of advantages. It was described in the earlier section
(Section 11.15.3) that the use of short pulses lead to localization of electric current on the anode surface due to the formation of
double layer capacitances. The voltage applied in EMM is generally in the range of 3–10 V depending on the machining require-
ments. The current density is directly proportional to the applied voltage. High voltage results in higher current density, which
increases the MRR. Because of the narrow interelectrode gap, the high debris concentration results in clogging of the gap. High
voltage not only contaminates the electrolyte in the interelectrode gap but also increases its temperature. The use of high voltage can
also result in boiling of the electrolyte, thereby affecting the process performance. Due to these reasons, it is advisable to use voltage
as low as possible.
From Faraday’s law, it is clear that the MRR is proportional to the current density and one would expect a liner curve for MRR
versus voltage. However, the MRR with the voltage shows a nonlinear trend, see Figure 14 (32). The nonlinear trend is because of
the dissolution efficiency that is generally low at low voltage. With an increase in the applied voltage, the dissolution efficiency
increases, due to which the current density increases.
Figure 15 shows the effect of applied voltage on overcut. Even here, the trend is nonlinear. It can be observed that the increase in
voltage also increases the stray current and reduces the localization. At high voltage, not only the MRR is high but also the area of
dissolution, resulting in poor accuracy of the process. For a precise machining operation, the use of low voltage is generally
advisable although the material removal is low.

11.15.5.2 Pulse On-Time and Off-Time

The usage of pulsed current in EMM is precisely for precision operation. The anodic dissolution occurs during the pulse on-time. The
pulse on-time provides a direct handle on the spatial accuracy of the machining process. The phenomenon is due to the formation
of the EDL, which has been explained in detail in the earlier section. The application between the tool and the workpiece results in
periodic charging and discharging of the EDL capacitances. The time constant of the charging of the double layer is locally varying
and depends on the distance between the tool and the workpiece. Therefore, by choosing a proper pulse on-time such that the
double layers are strongly charged in the region close to the electrode, the machining process can be localized. To illustrate this with

Figure 14 Plot of MRR vs voltage. Reproduced from Bhattacharyya, B.; Munda, J. Experimental Investigation into Electrochemical Micromachining
(EMM) Process. J. Mater. Process. Technol. 2003, 140 (1–3), 287–291.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
388 Electrochemical Micromachining

Figure 15 Plot of overcut vs voltage. Reproduced from Bhattacharyya, B.; Munda, J. Experimental Investigation into Electrochemical Micromachining
(EMM) Process. J. Mater. Process. Technol. 2003, 140 (1–3), 287–291.

an example, consider a 0.1 M HClO4 electrolyte with typical values of Cdl ¼ 10 mF cm–2 and r ¼ 30 U. For this case, pulse duration
equal to charging time constant of 30 ns yields a maximum separation of d ¼ 1 mm. At distances more than 1 mm from the tip of the
tool, the double layers are charged weakly and a very small (or negligible) material removal occurs in this region.
Figure 16 shows the effect of pulse on-time on the localization of the dissolution process. The six different troughs were
machined on a Ni sheet with a tungsten rod of 2 mm for different pulse on-times under identical conditions of 2.2 V. A plot of hole
diameter for different pulse on-times is shown in Figure 17 (33), which gives a quantitative estimation of the variation of hole
diameter with pulse on-time. Both figures clearly indicate the increase in spatial machining area with an increase in pulse on-time.
Clearly, the pulse on-time plays a crucial role in the spatial localization of the anodic dissolution and improves the accuracy of the
process.
The pulse off-time also plays an important role in the EMM process. During the pulse off-time, the dissolved products, i.e., the
sludge, gas bubbles, and so on, can be removed from the interelectrode gap by the flowing electrolyte. Most important, the off-time
restricts the electrolyte from reaching high temperatures. Using a sufficiently large pulse off-time, the heat in the interelectrode gap
can be dissipated. Therefore, unlike the pulse on-time, the off-time is preferred to be large, sufficient enough that a fresh electrolyte
enters the interelectrode gap and removes the debris. Figure 18 shows the effect of pulse off-time on hole diameter. The holes were

Figure 16 Troughs etched on a Ni plate at different pulse on-times and unchanged conditions of 2.2 V and 0.1 M HCl. Reproduced from Kock, M.;
Kirchner, V.; Schuster, R. Electrochemical Micromachining with Ultrashort Voltage Pulses - A Versatile Method with Lithographical Precision. Elec-
trochim. Acta 2003, 48 (20–22), 3213–3219.

Figure 17 Plot of hole diameter vs pulse on-time. Reproduced from Ahn, S. Electro-chemical Micro Drilling Using Ultra Short Pulses. Precis. Eng. 2004,
28 (2), 129–134.
www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrochemical Micromachining 389

Figure 18 Plot of hole diameter vs pulse off-time. Reproduced from Ahn, S. Electro-chemical Micro Drilling Using Ultra Short Pulses. Precis. Eng. 2004,
28 (2), 129–134.

drilled with a 30 mm tool at a voltage of 6 V and 40 ns on-time. The hole diameter is minimum for large off-time, whereas it increases
with a decrease in off-time. The hole diameters for off-time of 1 ms, 2 ms, and 500 ns are very insignificant, indicating that the spatial
accuracy of the process does not vary much after a certain length of the pulse off-time. The use of very large pulse off-time will
unnecessarily reduce the speed of machining process. In comparison, the spatial accuracy is largely dependent on the pulse on-time
than on the pulse off-time. It can be concluded that for a precise machining operation, the choice of pulse on-time and off-time is
very crucial. It is always preferable to use a very small pulse on-time and sufficiently large pulse off-time.

11.15.5.3 Electrolyte
The purpose of electrolyte in EMM is to provide free ions that can carry the electric current in the interelectrode gap and initiate the
anodic dissolution. The ions of the electrolyte take part in the electrochemical reactions that result in anodic dissolution. The
passage of electric current usually results in heating of the electrolyte due to the ohmic resistance. The electrolyte should, therefore,
preferably have a high boiling point, high heat capacity, and high thermal conductivity so as to reduce the tendency to boil. The
other most important characteristic desired for an electrolyte is low viscosity, so that the electrolyte can flush the interelectrode gap
very easily and remove the reaction products.
The most common electrolytes used are generally aqueous solutions of salts, acids, or bases. The electrolytes are classified into
two types as described in the Section 11.15.2 as passivating and non-passivating. The passivating electrolytes contain oxidizing
anions such as NaNO3 and NaClO3. The non-passivating electrolytes contain aggressive anions such as NaCl. Usually, passivating
electrolytes are preferred over non-passivating electrolytes as they are known to give better machining accuracy. This is because the
electrochemical reactions in the presence of passivating electrolytes result in the evolution of oxygen gas and formation of oxide
films in the stray current region, thereby reducing the anodic dissolution in the stray regions and improves the accuracy.
One of the most commonly used electrolyte is NaNO3 (pH 7). Although corrosive in nature, the electrolyte is capable of
producing high current efficiency without affecting the tool significantly. In the recent years, acidic electrolytes are also used mainly
because of their ability to produce soluble products during the electrochemical reactions. This becomes advantageous especially in
the fabrication of high-aspect-ratio structures, where the formation of insoluble products hampers the performance characteristics of
the process. However, acidic electrolytes are generally avoided as they are hazardous and also difficult to filter the dissolved metal.
Electrolyte concentration mainly influences the material removal in EMM. Figure 19 shows the influence of electrolyte
concentration on the MRR. Increase in the electrolyte concentration causes an increase in the current and also the dissolution
efficiency and hence, the MRR increases. However, due to an increase in dissolution efficiency with an increase in electrolyte
concentration, the material removal also occurs in the stray region and prevents the localization of the material removal. This leads
to an increase in the overcut as shown in Figure 20. Therefore, low or moderate electrolyte concentration is desired in EMM for
localizing the material removal.

11.15.5.4 Interelectrode Gap

The interelectrode gap plays an important role in accurate shape generation in EMM. The small interelectrode gap localizes the
electric current at the anode surface, thereby reducing the stray removal. The interelectrode gap in EMM is generally in the range of
10–50 mm. A small interelectrode gap produces a high electric field inside it, resulting in high current and therefore, high material
removal. The other advantage of a small interelectrode gap is that it lowers the required machining voltage and electrolyte
concentration by a considerable amount as compared to ECM, where the interelectrode gap is of several millimeters. However,
a very small interelectrode gap is also disadvantageous for several reasons. First, it makes flushing away the reaction products
increasingly difficult, and second, there is a very high possibility of the thin layer of electrolyte in the interelectrode gap reaching
high temperatures that may even lead to boiling, which severely affects the accuracy of the process.

11.15.5.5 Tool Electrode

Unlike in conventional machining process, the tool in EMM does not need to necessarily be harder. However, due to the very small
cross section of the tool in EMM, it is preferable to use a tool with high stiffness that can withstand electrolytic flow without
www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
390 Electrochemical Micromachining

Figure 19 Plot of MRR vs electrolyte concentration. Reproduced from Bhattacharyya, B.; Munda, J. Experimental Investigation into Electrochemical
Micromachining (EMM) Process. J. Mater. Process. Technol. 2003, 140 (1–3), 287–291.

Figure 20 Plot of overcut vs electrolyte concentration. Reproduced from Bhattacharyya, B.; Munda, J. Experimental Investigation into Electrochemical
Micromachining (EMM) Process. J. Mater. Process. Technol. 2003, 140 (1–3), 287–291.

vibrations. The tool materials are selected mainly based on their electrochemical and mechanical properties. The tool material should
possess high electrical and thermal conductivity. Due to the corrosive nature of the electrolytes, corrosion-resistant tool materials are
preferred. Some of the most commonly used tool materials are platinum, tungsten, titanium, stainless steel, and molybdenum. The
shape of the tool depends on the desired shape to be produced on the workpiece. Tools in EMM are often coated with insulating
materials such as SiC and Si3N4 on the side walls to reduce the stray current effect, so as to enhance the accuracy of the process.
Microtools for EMM that are less than 50 mm in diameter are often difficult to fabricate using conventional machining methods
as then can be easily bent by lateral forces. Fabrication of microtools for EMM is usually done using WEDG (wire electro-discharge
grinding), EDM, chemical etching, or even ECM-based processes (34). Out of all the above processes, WEDG provides high
precision, although it is time consuming.
While flat end tools are most commonly used, the recent advances have shown that the use of different geometries at the tool end
improves the precision of the process. Apart from flat end, tapered/conical end and round/spherical end shapes are some of the
most common tool-end shapes used (34,35). These tool-end geometries have certain advantages over that of a flat-end tool.
Figure 21 shows the central part of the machined work surface for flat-end and round-end tools. The round-end tool shows good
machinability as the tool-end profile eases the flow of electrolyte, whereas the insufficient supply of electrolyte to the central part of
the workpiece in the case of flat-end tool results in poor machining. Electrodes with a flat end can be used to fabricate complex
structures with planes due to its well-distributed current density. Tools with a conical end shape can be used to fabricate very small
microholes due to their very small diameter and concentrated current density. On the other hand, the round/spherical end shape
can be used to fabricate microholes or structures with minor taper.

Figure 21 Schematic of machined surface for flat-end and round-end tools. Reproduced from Ahn, S. Electro-chemical Micro Drilling Using Ultra Short
Pulses. Precis. Eng. 2004, 28 (2), 129–134.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrochemical Micromachining 391

11.15.6 EMM-Based Processes

The ECM process is widely used in the fabrication of a variety of microcomponents. Some of the components can be easily
fabricated using the basic EMM process by localizing the current. Mask-based techniques are also employed, which allow
complex structures to be fabricated with much ease and good accuracy. The mask-based technique involves the use of a photo
resist pattern that allows selective anodic dissolution along the desired portion of the workpiece (10,15). Apart from the mask/
maskless EMM, a number of processes that are based on EMM have been developed over the years, and they find various
applications in the field of microfabrication. Most of the processes are directly related to the EMM process, while a few other that
significantly differ are also based on the method of anodic dissolution. EMM is also combined with other micromachining
techniques in the hybrid micromachining operations. This section presents a brief overview of all the processes that are based on
the technique of EMM.

11.15.6.1 Drilling Using EMM

Hole drilling using EMM is one of the most commonly used applications. The basic concept of EMM has been modified for
applications related to achieve high accuracy and high-aspect ratio in microhole drilling. There exists a number of hole drilling
techniques (see Figure 22) that exploit the concept of EMM: (1) electrochemical drilling (ECD), (2) shaped tube electrolytic drilling
(STED), (3) capillary drilling (CD), (4) electrostream drilling (ESD), and (5) electrolytic jet drilling.

11.15.6.1.1 Electrochemical Drilling for Microholes

ECD uses cylindrical rods with tapered or round ends as tools. Especially in the case of deep hole drilling, a tubular-shaped
tool is used, into which a high pressure electrolyte is pumped that reaches the interelectrode gap and comes out from the
narrow gap outside of the coated tube to the workpiece surface carrying the sludge (see Figure 22). The tool is preferably made
of a high-strength material, so that it can withstand high electrolytic pressures without bending. The application of ECD to
microhole fabrication has faced several challenges such as nonlocalization of current, taper generation, and passive layer
formation, particularly in steel alloys. Current localization has been achieved using pulsed current and a small interelectrode

Figure 22 Electrochemical hole drilling processes. Reproduced from Sen, Mohan; Shan, H. S. A Review of Electrochemical Macro- to Micro-hole
Drilling Processes. Int. J. Mach. Tools Manuf. 2005, 45 (2), 137–152.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
392 Electrochemical Micromachining

Figure 23 Electrochemical drilling with balance electrode. Reproduced from Ahn, S. Electro-chemical Micro Drilling Using Ultra Short Pulses. Precis.
Eng. 2004, 28 (2), 129–134.

Figure 24 Through hole on 20-mm-thick 304 stainless steel plate: (a) entrance diameter of 8 mm and (b) exit diameter of 7.3 mm. 1 taper angle.
Reproduced from Ahn, S. Electro-chemical Micro Drilling Using Ultra Short Pulses. Precis. Eng. 2004, 28 (2), 129–134.

gap. The microtools are insulated on the sides to reduce taper. The use of a balance electrode (33) (see Figure 23) helps
prevent the formation of a passive oxide layer. The balance electrode is made of platinum, whose area is almost half the area
of workpiece which is used to compensate for the difference of voltage drops between electrolyte and the two electrodes. In the
absence of a balance electrode, an oxide layer (especially on stainless steel workpieces) is formed due to the relatively low
potential in the passive range that exists between the tool-electrode and the workpiece. The formation of passive layer hinders
the anodic dissolution process. Using this technique, Ahn et al. (33) produced a 8-mm-diameter hole with a taper angle of 1 ,
see Figure 24.
The use of salt solutions such as NaCl and NaNO3 as electrolytes causes clogging due to the formation of insoluble sludge. The
removal of sludge becomes extremely difficult in the case of deep hole drilling, which makes the technique difficult for generation
of microholes with high-aspect ratios. Therefore, the use of acid electrolytes proves effective, as they do not produce insoluble
sludge.

11.15.6.1.2 Shaped Tube Electrochemical Drilling

STED was mainly developed for drilling of microholes with very high-aspect ratios. The high-aspect ratio holes are difficult to
generate using ECD due to the formation of sludge that is difficult to remove. Not only this, other techniques such as LBM, electron
beam machining (EBM), and conventional drilling cannot be used to produce high-aspect ratio microholes. STED is a modified
form of ECD. In this technique, acid-based electrolytes are employed that prevent the formation of insoluble sludge. When acid-
based electrolytes are used, the dissolved anode material goes into the solution without forming a precipitate, and the problem of
sludge removal does not arise. Acid electrolytes such as sulfuric acid and hydrochloric acid at a concentration of 10–25% are used in
STED (12). In some cases, a combination of both, i.e., neutral salt electrolytes and a small amount of acid electrolytes, is used to
minimize the sludge formation.
The working principle of STED is similar to that of ECD, where the electrolyte at high pressure is fed through the tubular tool
electrode (see Figure 22) and comes out through the sides. This technique is used in small holes for cooling in turbine blades
and nozzle guide vanes, which require holes as small as 800 mm with an aspect ratio as high as 250 (36). STED is also applied
for multiple hole drilling at both parallel and compound angles to each other (12). However, the process also has certain
limitations. The use of acid electrolyte increases the possibility of acid reaction with the tool electrode, thus enforcing the use of
only acid-resistant tool materials. The use of acids produces soluble sludge, which is not only hazardous but also difficult to
recycle.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrochemical Micromachining 393

11.15.6.1.3 Capillary Drilling

CD uses a capillary glass tube into which a platinum wire is inserted, as shown in Figure 22. The electrolyte flows into the capillary
tube at a very high pressure of about 3–20 bar. This process is used to drill holes that are too deep to be drilled by EDM and too
small to be drilled by STED. The technique is used for an aspect ratio around 1:100. Holes of diameter 200–500 mm for a depth of 8–
16 mm have been successfully drilled using this technique in high-pressure gas turbine blades (12). Similar to the STED process,
acid electrolytes are used in the CD process to prevent the sludge formation. The choice of electrolytes depends on the workpiece
material. For example, hydrochloric solution is used for aluminum and its related alloys, while sulfuric acid solution is used for steel
alloy, Inconel, Rene alloys, and stainless steels. However, the process requires a very high voltage of 70–150 V (37), which is about
10 times more than that used in EMM. This is due to high resistance for current as a result of longer electrolyte flow path. The process
requires careful monitoring of acid temperature, pressure, concentration, and flow rate to achieve good machining results.

11.15.6.1.4 Electrostream Drilling

ESD process is also called EJD, which is similar in approach to CD but uses a finely drawn glass tube nozzle with a platinum wire
inserted into it as shown in Figure 22. In this technique, the acid electrolyte is negatively charged and is impinged onto the
workpiece through the nozzle with an electrolyte pressure of 3–10 bar. Typically, acid electrolytes with concentration of 10–25% are
used in ESD. Here, the negatively charged electrolytic jet acts as a cathode when a platinum wire is inserted into the nozzle, while the
workpiece acts as anode. The anode dissolution initiates when the electrolyte strikes the workpiece, which comes out from the sides
of the nozzle with the dissolved metal atoms. Much like CD, ESD also employs a very high voltage.

11.15.6.1.5 Electrolytic Jet Drilling

EJD is mainly used for drilling fine holes for aspect ratios around 1:12 (37), which is quite lower than that of STED, CD, and ESD
processes. In this process, the mechanism of anodic dissolution is similar to that of CD and ESD processes. Using a jet of electrolyte
at a very high pressure of 10–60 bar impinges on the workpiece to achieve anodic dissolution and comes out in the form of a spray.
Unlike ESD, the nozzle here does not require the entry of nozzle into the workpiece. The electrolyte jet emerges from the nozzle,
which also has a platinum rod protruding out of it, as shown in Figure 22. This process uses a very high operating voltage of 400–
800 V, and electrolytes of usually high conductivity are used to achieve high current density for good machining results (12).

11.15.6.1.6 Comparison of Drilling Processes

Apart from ECD, the other four processes use acid electrolytes to tackle the issue of sludge removal. Among all the five processes,
ECD with tapered or round tool heads can be used to achieve very small hole with good precision. However, the technique does not
allow fabrication of high-aspect-ratio holes. The other four processes are mainly developed for microdrilling of deep holes. The
techniques are used depending on the diameter-to-hole ratios. STED is typically used for very high-aspect ratios of 1:250, while CD,
ESD, and EJD are used for aspect ratios of 1:100, 1:50, and 1:12, respectively. A comparative study of these four acid-based deep
drilling processes in presented in Table 3, and Table 4 lists their limitations.
Among the four acid-based electrolyte drilling processes, STEM has the least overcut compared to CD, ESD, and EJD. It means
this process (STEM) is the most precise technique as well as an appropriate method for fabrication of holes with high-aspect ratios.
The overcut among the other three drilling processes is high because of the longer electrolyte flow path. CD has a very high overcut,

Table 3 Characteristics of acid-based electrochemical drilling processes

Characteristic STEM CD ESD EJD

Type of electrolyte HNO3, H2SO4 HNO3, H2SO4, HCl HNO3, H2SO4, HCl HNO3, H2SO4
Electrolyte pressure (bar) 3–10 3–20 3–10 10–60
Tool Titanium tube Glass capillary with gold, Glass tube with capillary end with gold, Platinum
platinum or titanium wire platinum, or titanium wire
Tool feed (mm min1) 1–3.5 1–4 1–3.5 0
Applied voltage (V) 5–15 100–200 150–850 400–800
Reproduced from Sen, Mohan; Shan, H. S. A Review of Electrochemical Macro- to Micro-hole Drilling Processes. Int. J. Mach. Tools Manuf. 2005, 45 (2), 137–152.

Table 4 Limitations of acid-based electrochemical drilling processes

Limitations STEM CD ESD EJD

Slow for single hole U U U U

Machining of only conducting materials U U U U
Complex machining and tooling U U U U
Hazardous handling and disposal of acid electrolytes U U U U
High voltage DC supply Ⅹ Ⅹ U U
Tool breakage Ⅹ U U Ⅹ

(U) and (Ⅹ) indicate limitation and no-limitation, respectively.

Reproduced from Sen, Mohan; Shan, H. S. A Review of Electrochemical Macro- to Micro-hole Drilling Processes. Int. J. Mach. Tools Manuf. 2005, 45 (2), 137–152.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
394 Electrochemical Micromachining

as it has the longest electrolyte flow path. Minimizing the overcut in CD is often difficult to achieve as it would restrict the entry of
glass capillary. In the cases of ESD and EJD, the overcut is also high. Also, the shapes of holes produced by ESD and EJD are bell
mouthed at entry and exit. This is primarily due to the electrolyte flow pattern.

11.15.6.2 Wire-EMM
Use of microwire as a tool can be employed in EMM for cutting metal sheets into desired shapes and fabrication of microgrooves.
Unlike in wire-EDM, the wire (tool) in wire-ECM does not get worn out as the tool does not get affected by the machining process.
This enables the use of a very thin microwire. The same wire can be used continuously and does not require movement of the wire, as
in the case of wire-EDM. A schematic of the wire-EMM process is shown in Figure 25. Very thin wires of diameter 5–10 mm are used in
the wire-EMM process (38). The wire used in this technique is preferably of high strength and chemically inert. Platinum, tungsten,
and copper are some of the common types of wires used in wire-EMM (39). The acid-based electrolytes with low concentrations are
used to achieve a small side gap, as an increase in electrolyte concentration causes an increase in the side gap. The most commonly
used electrolyte is H2SO4 at 0.1 M concentration (14,38,39) as the optimal concentration, below which the machining rate becomes
low, and above which the side gap increases. For better accuracy, nanosecond pulses are used in wire-ECM. A short pulse with long
pulse period gives optimum results. Typically, 60–150 ns pulses are used within a pulse period of 6–8 ms (39).
Wire-EMM is mainly used in the fabrication of microgrooves and for generating complex shapes. Figure 26 shows a microslit and
Figure 27 shows a microstructure fabricated by wire-ECM.

Figure 25 A schematic of the wire-ECM process. Reproduced from Zhu, D.; Wang, K.; Qu, N. S. Micro Wire Electrochemical Cutting by Using In Situ
Fabricated Wire Electrode. CIRP Ann. – Manuf. Technol. 2007, 56 (1), 241–244.

Figure 26 Microslits by wire-electrochemical cutting. Reproduced from Zhu, D.; Wang, K.; Qu, N. S. Micro Wire Electrochemical Cutting by Using In
Situ Fabricated Wire Electrode. CIRP Ann. – Manuf. Technol. 2007, 56 (1), 241–244.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrochemical Micromachining 395

Figure 27 Microstructures by microwire-electrochemical cutting. Reproduced from Zhu, D.; Wang, K.; Qu, N. S. Micro Wire Electrochemical Cutting by
Using In Situ Fabricated Wire Electrode. CIRP Ann. – Manuf. Technol. 2007, 56 (1), 241–244.

11.15.6.2.1 Microelectrochemical Milling

ECM when applied in the fabrication of 2D and 3D microstructures that operates similar to the conventional milling process is
referred to as microelectrochemical milling. A schematic of the process is shown in Figure 28. In this technique, the electrode is first
fed downward, into the workpiece to the desired depth as a starting point. The initial step is essentially a drilling process. This is
followed by the movement of the tool along a scheduled path in the XY plane to accomplish a layer of processing. A combination of
downward movement of the tool and its lateral movement in the XY plane is carried out to achieve layer-by-layer removal of the
workpiece material to produce the desired shapes and structures.
The dimensions of the structure that can be fabricated using this technique largely depend on the diameter of the tool electrode.
Tools with a small diameter give very small structures. The only drawback of ultrathin tools lies mainly in their manufacturing. Also,
such tools have very low strength and may find it difficult to withstand viscous forces in the electrolyte during its motion along the
XY plane. Wire diameters as small as 3 mm have been used in this technique (34). Similar to most of the drilling processes, acid-
based electrolytes are used in this process to reduce the sludge formation. The precision and accuracy of the process mainly lie
in the electrolyte concentration, applied voltage, and pulse duration. Low concentration of electrolyte is preferred to achieve
machining with minimum side-gap. A 0.1 M H2SO4 is the most commonly used electrolyte in the microelectrochemical milling

Pulsed power supply

Tool
Feed (into the w/p)

Feed (along the w/p)

Figure 28 A schematic of the microelectrochemical milling process.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
396 Electrochemical Micromachining

Figure 29 Microhemisphere with a diameter of 60 m on top of a microcylinder of diameter 100 mm. Reproduced from Kim, B. H.; Na, C. W.; Lee, Y. S.;
Choi, D. K.; Chu, C. N. Micro Electrochemical Machining of 3d Micro Structure Using Dilute Sulfuric Acid. CIRP Ann. – Manuf. Technol. 2005,
54 (1), 191–194.

process (40). This electrolyte not only gives accurate results with a very good surface finish (Ra value of 0.28 mm) but also is less
hazardous (14). Pulse voltage in the range of 3–7 V is used with very short pulse on-time (several nanoseconds) and a long pulse
duration (several microseconds) to achieve good precision. Microelectrochemical milling process is a suitable technique to be
applied in the fabrication of complex shapes and structures. Figure 29 shows a microhemisphere of diameter 60 mm on top of
a cylinder that was machined in three steps. In the initial step, a cylinder of diameter 100 mm was machined, followed by
a hemisphere of equal diameter on top of it. In the finishing cut, the diameter of the hemisphere was reduced to 60 mm. This shows
the capability of microelectrochemical milling as a promising technique for the fabrication of complex 3D microstructures that find
various applications in the micro-electromechanical systems (MEMS) industry.

11.15.6.3 Electrochemical Polishing

ECM involves dissolution of an anode by atom-by-atom removal, due to which the surface finish obtained is extremely good. This
makes the process a suitable technique for application in surface finishing operations. The technique of polishing by electro-
chemical dissolution, ECP is used for improving microsmoothness, microtopology, and material brightness.
The polishing occurs due to the difference in current densities at the peaks and valleys. The peaks of the rough surface (anode) are
closer to the cathode than the valleys. Due to this, the current density at the peaks is comparatively greater than that in the valley.
This causes higher dissolution rate at the peaks, leading to a decrease in surface roughness. A surface finish obtained using this
technique is a few micrometers. A surface finish as good as 0.1 mm has been reported using ECP.
Unlike EMM involving drilling, grooving, etc., polishing requires very small MRRs. The surface roughness decreases with
a decrease in the interelectrode gap. However, at narrow interelectrode gaps, pits are formed on the surface of the anode (41), and
sometimes it may even lead to sparking. Hence, an appropriate interelectrode gap is used. An interelectrode gap of around 1 mm is
used in ECP, which is very high compared to that used in other EMM techniques (wmm). A voltage of 2–10 V DC is used in the
process, resulting in current densities of 5–25 A cm2 (41), as high current densities give good rates of polishing. The operation time
in ECP is usually in the range of 2–20 min (42).
Acid-based electrolytes are employed in the process. Sulfuric acid, phosphoric acids, and a mixture of both are some of the most
commonly used electrolytes (43,44). High-temperature electrolytes are used, which causes an increase in the current efficiency due
to the low viscosity and increased specific conductivity. Better results are obtained when electrolyte temperature is in the range of
60–90  C (41).
ECP has several advantages compared to other polishing methods. It involves no mechanical, thermal, or chemical impact on
the workpiece surface. The process does not induce any stresses in the surface layer or cause any metallurgical properties to
deteriorate, which is generally observed in mechanical polishing. This makes it a promising technique for polishing of small and
fragile parts. ECP is very good technique especially for inner surface polishing of parts and also for polishing of complex shapes
that are difficult to polish by mechanical polishing methods. In the case of stainless steels, ECP increases corrosion resistance by
about 15–20 times. This is because of the formation of chromium oxide layer on the workpiece surface in the presence of
electrolyte.
However, the process also has certain drawbacks. The process gives good polishing rates only at high current density
(>8 A cm2), while at low current density the polishing rate is nearly zero. This is because of the formation of a thin-film passive
layer on the workpiece, which hinders the polishing rate, especially at low current density. This issue has also been tackled by using
abrasive particles in the electrolyte that remove the passive layer by abrasive action and increase the polishing rate. This technique is
also called electrochemical mechanical polishing (45). Also, acid electrolytes need to be carefully handled and are difficult to recycle.
Another technique similar to ECP called electrochemical buffing uses salt solutions as electrolytes; these are both easy to handle and

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrochemical Micromachining 397

environmental friendly. This technique uses a rotating disk with very fine abrasive as cathode and electrolyte solution of NaCl or
NaNO3 is passed into the narrow gap between the cathode and the anode (workpiece).

11.15.6.4 Solid Electrochemical Micromachining

One of the recent advances in the EMM is the replacement of liquid electrolyte by a solid electrolyte. The technique is also referred to
as solid-state EMM. This technique uses a metal ion conductor that acts similar to that of an electrolyte under the application of
voltage. The material removal occurs as a result of anodic electrochemical reaction at the microcontact between the ion conducting
microelectrode and the metal substrate. The metal substrate is locally oxidized and is later incorporated into the ion conductor in
the form of metal ions via the microcontact, as shown in Figure 30. The ion conductor, b00 -Al2O3 (typically Na-b00 -Al2O3), is the
common solid electrolyte used that has the ability to conduct a variety of metal ions. Under the application of a DC voltage, the
region of the anode at the microcontact with the tool is electrochemically oxidized to Mnþ. This ion is later absorbed into the ion-
conducting layer Na-b00 -Al2O3 through the microcontact. This results in the migration of Naþ ion through the b00 -Al2O3 and it is then
deposited as Na metal at the cathode and solid electrolyte interface. Na then readily reacts with O2 and CO2 present in air to form
Na2CO3. The process essentially involves the electrochemical substitution of Naþ ion by anode metal ion (Mnþ) in the b00 -Al2O3
layer (46). The material removal occurs as a result of the dissolution of anode metal in the form of Mnþ into the ion conducting
layer. Some of the commonly used metal targets are Ag, Cu, Zn, and Pb (47).
The process is also used for microstructuring of tough materials such as glassy carbon, SiC, W, and Mo (48) using an oxide ion
conductor. The machining involves production of gaseous or volatile substances by oxidation as a result of electrochemical reac-
tions. The cathode consists of an oxide ion conductor such as Y2O3 doped ZrO2 (yttria stabilized zirconia:YSZ) and the anode
consists of ceramic material or hard materials such as W and Mo. When a DC voltage is applied, the electrochemical reaction occurs
at the microcontact between the tool (cathode) and the workpiece (anode). The process involves oxidation of anode after reaction
with O 2 or O2 gas that leads to the formation of volatile or gaseous products, as shown in Figure 31.
The process has several advantages over the EMM with liquid electrolytes: (1) unlike a liquid electrolyte, the solid electrolyte is
very easy to handle, (2) the machining is highly localized along with a rapid attainment of steady state, and (3) any complex shape
can be transferred from the tool to the workpiece with a very high accuracy (47).

Figure 30 A schematic of the solid-state electrochemical micromachining using a metal ion conductor. Reproduced from Kamada, Kai;
Tokutomi, Masaaki; Enomoto, Naoya; Hojo, Junichi. Electrochemical Micromachining Using a Solid Electrochemical Reaction at the Metal/5-Al2O3
Microcontact. Electrochim. Acta 2007, 52 (11), 3739–3745.

Figure 31 A schematic of the solid-state electrochemical micromachining for hard materials using an oxide ion conductor. Reproduced from
Kamada, Kai; Hirata, Shodai; Enomoto, Naoya; Hojo, Junichi. Surface Micromachining via Solid Electrochemical Reaction on Oxide Ion Conductors. Solid
State Ionics 2009, 180 (20–22), 1226–1230.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
398 Electrochemical Micromachining

Figure 32 (a) A spiral trough with a depth of 5 mm and (b) a triangular trough of depth 1 mm on a Ni sheet using a tungsten STM tip. Reproduced
from Kock, M.; Kirchner, V.; Schuster, R. Electrochemical Micromachining with Ultrashort Voltage Pulses - A Versatile Method with
Lithographical Precision. Electrochim. Acta 2003, 48 (20–22), 3213–3219.

11.15.6.5 STM-Based Electrochemical Nanofabrication

The formation of an EDL was exploited by applying ultrashort pulses to localize the EMM process and make it suitable for
micromachining operations. Similarly, the recent use of an STM in EMM has made the process feasible for nanofabrication (49). In
this process, the STM tip is used as a tool. Due to the very small interelectrode gap of a few nanometers, the application of voltage
pulse to the STM tip results in a considerable depletion of electrolyte ions in the interelectrode gap. This is because of the fact that the
application of voltage leads to the charging of EDLs that consume ions from the electrolyte. For a 1 M electrolytes and an applied
voltage of 1 V, the depletion of ions will occur at an interelectrode gap of 10 nm (50). The ion depletion occurs instantaneously for
polarizing the double layer, while refilling of ions by lateral diffusion into the interelectrode gap from the bulk of the electrolyte
solution is relatively very slow. Under the application of ultrashort pulses (nanoseconds or picoseconds), the refilling of the ions
into the interelectrode gap is almost negligible, thereby causing the electrochemical reaction to be confined very close (wnm) from
the apex of the tip. Essentially, the process involves the confinement of electrochemical reaction to the tunneling region due to the
depletion of electrolyte ions in the gap between STM tip and the workpiece.
Different types of electrolytes can be employed in this process, like H2SO4, CuSO4, HCl, and KCl. High concentrations of
electrolyte of about 1 M are used to achieve anodic dissolution. The pulse duration plays a crucial role in the confinement of the
electrochemical reaction. Typically, nanosecond or picosecond pulses are used to achieve sub-micrometer or nanometric accuracy.
Using this technique, grooves of submicrometer width have been fabricated with machining precision of 100 nm by Kock et al.
(30) (see Figure 32). Figure 32(a) shows a spiral trough of nanometric wall thickness and 5 mm depth, and Figure 32 shows
a triangular trough of nanometric size and 1 mm depth, fabricated on a Ni plate using a tungsten STM tip. Trimmer et al. (51) used
STM-based ECM for nanostructuring of electrode surface by application of nanosecond pulses. A complex shaped tool, fabricated
using ion beam machining, was used to fabricate a nanostructure of size 400 nm in just 105 s. The precision of the process and also
its speed makes the process a prominent technique for fabrication of nanostructures with high accuracy and at relatively high speed
compared to other techniques.

11.15.6.6 EMM-Based Hybrid Processes

11.15.6.6.1 Laser-Assisted Jet Electrochemical Machining
Laser-assisted jet electrochemical machining (LAJECM) is a hybrid process that combines electrochemical dissolution with laser
heating of an anode. The main purpose of the laser jet is to localize the dissolution process. The laser beam thermally heats the
anode and directs the anodic dissolution to specifically targeted areas. The use of laser is only to assist and localize the dissolution
process by thermal activation of the anode surface, and the machining occurs by anodic dissolution. The use of laser also aids in
breaking down the oxide layer that is generally formed with certain metal–electrolyte combinations and speeds up the dissolution
process.
A relatively low-intensity laser is used that does not cause vaporization of the anode. The laser beam is aligned coaxially with the
electrolytic jet that is pumped into the nozzle and impinges on the workpiece, as shown in Figure 33, giving a combined noncontact
tool electrode. The process has two main advantages as compared to that of normal jet ECM. First, the use of a laser improves the
localization and enhances accuracy of the process by reducing the stray removal. Second, the use of a laser increases the rate of
machining. This is because the anode heating due to the laser lowers the activation energy for the electrochemical reaction, thereby
readily initiating the anodic dissolution (52,53). Also, the heating of the electrolyte increases its temperature, which results in a high
current density, according to Arrhenius law (54). The removal of the debris by transportation is also improved due to the electrolyte
heating process.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrochemical Micromachining 399

Figure 33 A schematic of the laser-assisted jet electrochemical machining. Reproduced from Hua, Zhang; Jiawen, Xu. Modeling and Experimental
Investigation of Laser Drilling with Jet Electrochemical Machining. Zhang Hua and Xu Jiawen 2010, 23, 454–460.

The process finds its application mainly in deep drilling of microholes. The depth of hole obtained using LAJECM is much higher
than that obtained with the assistance of laser. For example, under certain experimental conditions, the depth of hole achieved using
LAJECM is 55 mm, whereas the depth achieved using jet electrochemical drilling (JECD) was only 11 mm (12). Experiments by Pajak
et al. (52) also revealed that LAJECM has 55% higher volumetric removal rate and 40% reduction in taper as compared to that of JECD.

11.15.6.6.2 Oxide Film Laser Lithography Using EMM

Oxide film laser lithography (OFLL) using EMM is based on selective laser irradiation of an anodically formed oxide film. The
technique is widely used for Ti, particularly in the fabrication of multilevel structures and etching of 3D objects (56). The process
involves a series of steps as shown in Figure 34. Initially, the anode is oxidized using a suitable anodizing electrolyte such as sulfuric
acid. This leads to the formation of an oxide layer on the anode surface. A film thickness of 200–300 nm is usually desired for good
results. This is followed by irradiation of the oxide film with an excimer laser in air. The laser is used to form a desired pattern using
an XY stage to move the sample. Typically, the laser spot diameter is a few micrometers for micropattern generation. In the next step,
electrochemical dissolution of the metal from the irradiated areas of the patterned sample is carried out using an electropolishing

Figure 34 A schematic of the various steps (from top to bottom) involved in OFLL. Reproduced from Chauvya, P.-F.; Hoffmann, P.; Landolt, D.
Applications of Laser Lithography on Oxide Film to Titanium Micromachining. Appl. Surf. Sci. 2003, 208–209, 165–170.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
400 Electrochemical Micromachining

Figure 35 SEM images of (a) microstructure and (b) microstructured cylindrical surface with cavities fabricated using OFLL. Reproduced from
Chauvya, P.-F.; Hoffmann, P.; Landolt, D. Applications of Laser Lithography on Oxide Film to Titanium Micromachining. Appl. Surf. Sci. 2003, 208–209,
165–170.

electrolyte such as sulfuric acid–methanol. The anodic dissolution occurs only at the laser-irradiated areas as the remainder of the
oxide film acts as a mask. The anodic dissolution results in the undercutting of oxide films. Finally, ultrasonic cleaning is used to
remove the protruding oxide film resulting from undercutting.
The important characteristic of the process is its feasibility to be applied for nonplanar surfaces (15). The standard micro-
fabrication processes based on photolithographic techniques are limited to only flat surfaces due to the difficulty of applying
a photoresist on nonflat surfaces (57). This makes the process a potentially feasible method for fabrication of multilevel structures
and surface structuring of 3D objects of complex shapes. Figure 35 shows scanning electron microscope (SEM) images of some of
the microstructures fabricated using this technique.

11.15.6.6.3 Microelectrochemical Discharge Machining

Microelectrochemical discharge machining (micro-ECDM) is based on the principle of material removal by electrochemical
discharges (8). In this process, the workpiece, tool electrode, and auxiliary (or counter) electrode are immersed in an electrolyte as
shown in Figure 36. NaOH and KOH are some of the most commonly used electrolytes. The tool and the auxiliary electrode are
connected to a DC supply. The tool electrode behaves as a cathode, while the auxiliary electrode acts as the anode. The application of
DC voltage causes electrochemical reactions leading to production of H2 gas bubbles at both electrodes (58). When the voltage is
beyond a certain critical voltage, the rate of generation of H2 bubbles will be much higher than the bubbles floating at the electrolyte
surface. The bubbles become dense and coalesce into a gas film that isolates the tool from the electrolyte. The bubble film blan-
keting effect causes current density at the tool electrode to exceed a certain critical value, which leads to a discharge between the tool
and the surrounding electrolyte. The material removal at the workpiece occurs as a result of this discharge, which is induced by
electrochemical reactions. A portion of the workpiece erosion also occurs due to chemical etching.
Unlike micro-ECM and micro-EDM, which are applicable only to conducting materials, the micro-ECDM process can be applied
for nonconducting materials such as glass, quartz, and ceramics. Micromachining of materials such as glass, quartz, and ceramics
has wide application in MEMS industry (60). Compared to other machining processes such as LBM and other thermal erosion based
process, micro-ECDM is an effective technique for machining of these materials. Its advantages include good machining efficiency,
low heat affected zone, and the ability to fabricate 3D complex structures. Using this technique, microgrooves and 3D structures less
than 100 mm were fabricated on Pyrex glass with very high precision (61). The capability of the micro-ECDM process has been
improved by combining it with other processes such as ultrasonic vibration for deep hole drilling (62) and polycrystalline diamond
(PCD) grinding for surface roughness (63). Similar to microwire-ECM and microwire-EDM, microwire-ECDM has also been
developed for extending process capability (64).

Figure 36 Schematic of the ECDM process. Reproduced from Wthrich, R.; Fascio, V. Machining of Non-conducting Materials Using Electrochemical
Discharge Phenomenon; An Overview. Int. J. Mach. Tools Manuf. 2005, 45 (9), 1095–1108.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrochemical Micromachining 401

11.15.6.6.4 Ultrasonic-Assisted EMM

The efficiency of the anodic dissolution process in EMM is enhanced by using ultrasonic vibrations. The ultrasonic vibrations are
generally applied to the tool or the workpiece or sometimes even to the electrolyte (7,65). This method is called as ultrasonic-
assisted electrochemical micromachining. Ultrasonic vibrations with a frequency of several kilohertz and an amplitude of several
micrometers are employed in this technique.
The application of ultrasonic vibrations in EMM generates cavitation microbubbles at the tool and workpiece surfaces. The
dissolved gas in the electrolyte and increased electrolyte temperature create excellent conditions for cavitation bubbles to grow in
the interelectrode gap. The cavitation microbubbles grow extensively and then undergo an energetic collapse leading to generation
of extreme conditions of temperature and pressure. A temperature of about 5000 K and a pressure of approximately 1700 bar are
created (65). The entire process usually occurs in less a microsecond duration (66). The collapse of microbubbles at the workpiece
surface increases the mass and charge transport, leading to an increase in the anodic dissolution.
This phenomenon is useful particularly when passive layers are formed on the workpiece surface with certain metal–electrolyte
combinations. The ultrasonic vibrations can be used to destroy the passive layer and increase the anodic dissolution. This also
reduces the time required for machining. Ultrasonic vibrations also improve the flow of electrolyte and helps in preventing clogging
in the interelectrode gap due to the accumulation of debris.
Ultrasonic vibrations in the case of high-aspect-ratio drilling have been proved to be very useful. It has also been observed that
the use of ultrasonic vibrations creates a very good surface finish compared to the usual EMM process (67). All these aspects make
EMM with ultrasonic vibrations improve the capability of the EMM process for effective machining operation.

11.15.6.6.5 Microelectrochemical Grinding

Electrochemical grinding is also referred to as electrolytic grinding. The process is almost similar to the ECM, but uses a grinding
wheel as the cathode that is embedded with abrasive particles, while the workpiece acts as the anode. When applied for micro-
manufacturing, the process is also referred to as microelectrochemical grinding (micro-ECG) (68). The grinding wheel rotates and
a very small gap is maintained between the grinding wheel and the workpiece, such that the abrasive particles on the grinding wheel
touch the workpiece. An electrolytic jet is directed in between the rotating grinding wheel and the workpiece. The process is a hybrid
process of EMM and mechanical grinding, which involves the removal of material due to electrochemical dissolution of the anode
(workpiece) and also due to mechanical abrasive action. The abrasive action accounts for only about 5–10% of the total material
removal (68,69). The abrasive action is introduced only to remove the passive layer on the workpiece and improve the dissolution
of the workpiece due to electrochemical reaction, so that acid-based harmful electrolytes can be avoided. Also, the continuous
rotation of the grinding tool enhances the fresh flow of electrolyte in the interelectrode gap, improving the performance of the
process.
Micro-ECG employs a much smaller grinding wheel of about 10–15 mm diameter, while ECG typically employs grinding wheels
of diameter 150–200 mm (69). Copper, brass, and nickel are some of the most commonly used materials for grinding wheels.
A thin layer of diamond particles is used when grinding carbides or hard steels. Aluminum oxide is generally used as the abrasive
when grinding steel or other soft materials. The size of abrasive particles is very small, in the range of 150–200 mm (70). NaCl and
NaNO3 are the most common electrolytes used in this process.
Compared to the mechanical grinding process, this process provides scratch-free surfaces, eliminates grinding burns, causes low
tool wear, and creates no surface distortions or hardened layer on the work surface due to low heat generation. This makes it a very
good alternative for grinding of microcomponents.

11.15.7 Conclusions and Future Scope

ECM has proved to be one of the well-established nonconventional manufacturing processes for difficult-to-cut materials and in the
manufacturing of complex shapes. The microscale version of this process, EMM and its variants, is emerging as a truly competitive
alternative to the other nonconventional manufacturing processes like micro-EDM, LBM, and ion beam machining as it enjoys the
advantage of producing complex shapes with no thermal damage to the work surface besides imparting an excellent surface finish.
The process has found various applications in MEMS, electronic, and aerospace industries, for its ability to produce high-aspect-ratio
microholes, microstructures, microgrooves, etc. The process requires a complex experimental setup that contains various compo-
nents for tool holding and its movement, electrolyte flow, pulse generation, and various controllers. Though standard machines are
available for EMM, most researchers have developed their own EMM setups.
Although based on similar principles, the process differs from its macroscale version (ECM) in many ways. The formation of an
EDL, and the use of ultrashort pulses and a very narrow interelectrode gap, makes EMM quite distinct from ECM. The small
interelectrode gap (wmm) leads to a very high electric potential (w106V m1), due to which the electroneutrality of the electrolyte
becomes unrealistic, as in the case of ECM. The charge dynamics in the interelectrode gap and the evolution of current density at the
anode are more accurately governed by the Nernst-Planck-Poisson equation than the conventional ohmic equation assuming the
electroneutrality of the electrolyte in the interelectrode gap. In addition, the influence of the EDL at the surface of the electrodes
makes the physics of the EMM process significantly different from ECM.
The EDL acts a capacitor at the electrode surfaces, and the concept has been exploited by the use of ultrashort pulses to localize
the anodic dissolution process. This has indeed made EMM an accurate process to be applied for micromachining. The precision of
the process largely depends on the pulse parameters. High accuracy is achieved with very small pulse on-time to localize the anodic
dissolution and an adequate pulse-off time that enables the electrolyte to reach its initial temperature. The applied voltage is another

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
402 Electrochemical Micromachining

factor that largely contributes to the MRR. Low or moderate voltage is usually preferred in EMM, as the use of high voltage also
increases the electrolyte temperature, thus affecting the performance characteristics of the process. EMM employs both salt-based
and acid-based aqueous solutions as electrolytes based on the application. Salt-based electrolytes form insoluble sludge during
the process, while acid-based electrolytes form soluble compounds. Due to this, acid-based electrolytes are used in the fabrication of
high-aspect-ratio holes or structures, where the removal of sludge from the interelectrode gap becomes difficult. However, salt-based
electrolytes are the commonly preferred electrolytes as they are known to give better machining results, are nonhazardous, and are
easy to filter.
Over the years, a number of processes based on EMM have been developed to meet desired applications. The process is also
combined with other micromachining operations, resulting in a number of hybrid micromachining operations to fabricate
components that are difficult to be produced using a single technique. All these aspects have enabled EMM as a promising technique
for fabricating a variety of microcomponents that can also involve complex-shape generation. Some of the modern processes like
solid-EMM, micro-ECDM, and STM-based EMM hardly resemble the original EMM process, but all of them use the concept of
anodic dissolution. This indicates that research in EMM is still wide open. Further research on EMM will open up many interesting
possibilities, especially about its use in hybrid micromachining processes and innovations in its variants like solid-EMM and
STM-based EMM. As ECM enters the nanomachining operation, there exists challenging issues like precise control over the process
parameters and fabrication of submicrometer size tool. Apart from this, modeling based on the complex phenomenon of double
layer formation, ion kinetics, and electrolyte flow needs a fair amount of research to accurately estimate material removal and
overcut, in order to adequately control the process. A comprehensive theoretical models for most of the EMM variants are yet to be
developed. Mathematical modeling of these processes based on the fundamentals involved is undoubtedly a challenging task, but is
definitely required to improve the process control.

References

1. Bhattacharyya, B.; Mitra, S.; Boro, A. K. Electrochemical Machining: New Possibilities for Micromachining. Robot. Comput. Integr. Manuf. 2002, 18 (34), 283–289.
2. Bhattacharyya, B.; Munda, J.; Malapati, M. Advancement in Electrochemical Micro-machining. Int. J. Mach. Tools Manuf. 2004, 44, 1577–1589.
3. Wilson, J. F. Practice and Theory of Electrochemical Machining; R.E. Krieger Pub. Co, 1982.
4. Gusseff, W. British Patent. 335 003, 1994.
5. De Barr, A. E.; Oliver, D. A. Electrochemical Machining; Macdonald and Co. Ltd., 1968.
6. Schuster, R.; Kirchner, V.; Allongue, P.; Ertl, G. Electrochemical Micromachining. Science 2000, 289 (5476), 98–101.
7. Yang, I.; Park, M. S.; Chu, C. N. Micro ECM with Ultrasonic Vibrations Using a Semi-cylindrical Tool. Int. J. Precis. Eng. Manuf. 2009, 10 (2), 5–10. http://dx.doi.org/10.1007/
s12541-009-0020-5.
8. Kwan Chung, Do; Shik Shin, Hong; Soo Park, Min; Hyun Kim, Bo; Chu, Chong Nam Recent Researches in Micro Electrical Machining. Int. J. Precis. Eng. Manuf. 2011, 12 (2),
371–380.
9. Kozak, J.; Rajurkar, K. P.; Makkar, Y. Study of Pulse Electrochemical Micromachining. J. Manuf. Process. 2004, 6 (1), 7–14.
10. Datta, M.; Landolt, D. Fundamental Aspects and Applications of Electrochemical Microfabrication. Electrochim. Acta 2000, 45 (1516), 2535–2558.
11. De Silva, A. K. M.; Altena, H. S. J.; McGeough, J. A. Precision ECM by Process Characteristic Modelling. CIRP Ann. – Manuf. Technol. 2000, 49 (1), 151–155.
12. Sen, Mohan; Shan, H. S. A Review of Electrochemical Macro- to Micro-hole Drilling Processes. Int. J. Mach. Tools Manuf. 2005, 45 (2), 137–152.
13. Choi, S. H.; Ryu, S. H.; Choi, D. K.; Chu, C. N. Fabrication of WC Micro-shaft by Using Electrochemical Etching. Int. J. Adv. Manuf. Technol. 2007, 31 (7–8), 682–687.
14. Kim, B. H.; Na, C. W.; Lee, Y. S.; Choi, D. K.; Chu, C. N. Micro Electrochemical Machining of 3d Micro Structure Using Dilute Sulfuric Acid. CIRP Ann. – Manuf. Technol. 2005,
54 (1), 191–194.
15. Landolt, D.; Chauvy, P.-F.; Zinger, O. Electrochemical Micromachining, Polishing and Surface Structuring of Metals: Fundamental Aspects and New Developments. Electrochim.
Acta 2003, 48 (20–22), 31853201.
16. Park, B. J.; Kim, B. H.; Chu, C. N. The Effects of Tool Electrode Size on Characteristics of Micro Electrochemical Machining. CIRP Ann. – Manuf. Technol. 2006, 55 (1).
17. Yong, Liu; Di, Zhu; Yongbin, Zeng; Shaofu, Huang; Hongbing, Yu Experimental Investigation on Complex Structures Machining by Electrochemical Micromachining Technology.
Chin. J. Aeronaut. 2010, 23 (5), 578–584.
18. Pandey, P. C. Modern Machining; Tata McGraw-Hill Publishing Company Limited, 1980.
19. McGeough, J. A. Principles of Electrochemical Machining; Chapman and Hall, 1974.
20. McGeough, Joseph. Micromachining of Engineering Materials; Marcel Dekker, Inc, 2002.
21. Shaw, D. J. Electrophoresis; Academic Press: London, 1969.
22. Robinson, R. A.; Stokes, R. H. Electrolyte Solutions; Butterworth: London, 1970.
23. Marla, Deepak; Joshi, Suhas S.; Mitra, Sushanta K. Modeling of Electrochemical Micromachining: Comparison to Experiments. J. Micro/Nanolithog., MEMS, MOEMS 2008,
7 (3), 033015.
24. Van Kampen, N. G. Short First-Passage Times. J. Stat. Phys. 1993, 70 (1), 15–23.
25. Ermak, D. A Computer Simulation of Charged Particles in Solution. I. Technique and Equilibrium Properties. J. Chem. Phys. 1975, 62 (10), 4189–4196.
26. Bazant, M. Z.; Thornton, K.; Ajdari, A. Diffuse-Charge Dynamics in Electrochemical Systems. Phys. Rev. E 2004, 70 (2), 021506.
27. Albery, W. J. Electrode Kinetics; Oxford Univ. Press: London, 1981.
28. Kozak, Jerzy; Rajurkar, Kamlakar P.; Makkar, Yogesh Selected Problems of Micro-electrochemical Machining. J. Mater. Process. Technol. 2004, 149 (13), 426–431.
29. Puippe, J. Cl.; Ibl, N. Influence of Charge and Discharge of Electric Double Layer in Pulse Plating. J. Appl. Electrochem. 1980, 10, 775–784.
30. Kock, M.; Kirchner, V.; Schuster, R. Electrochemical Micromachining with Ultrashort Voltage Pulses – A Versatile Method with Lithographical Precision. Electrochim. Acta 2003,
48 (20–22), 3213–3219.
31. Yong, Li; Yunfei, Zheng; Guang, Yang; Liangqiang, Peng Localized Electrochemical Micromachining with Gap Control. Sens. Actuators, A 2003, 108 (1–3), 144–148.
32. Bhattacharyya, B.; Munda, J. Experimental Investigation into Electrochemical Micromachining (EMM) Process. J. Mater. Process. Technol. 2003, 140 (1–3), 287–291.
33. Ahn, S. Electro-chemical Micro Drilling Using Ultra Short Pulses. Precis. Eng. 2004, 28 (2), 129–134.
34. Liu, Y.; Zhu, D.; Zeng, Y.; Yu, H. Development of Microelectrodes for Electrochemical Micromachining. Int. J. Adv. Manuf. Technol. 2011, 55, 195–203.
35. Yang, Ye; Natsu, Wataru; Zhao, Wansheng Realization of Eco-friendly Electrochemical Micromachining Using Mineral Water as an Electrolyte. Precis. Eng. 2011, 35 (2),
204–213.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrochemical Micromachining 403

36. Ali, S.; Hinduja, S.; Atkinson, J.; Pandya, M. Shaped Tube Electrochemical Drilling of Good Quality Holes. CIRP Ann. – Manuf. Technol. 2009, 58 (1), 185–188.
37. El-Hofy, Hassan Abdel-Gawad; El-Hofy, Hassan Advanced Machining Processes: Nontraditional and Hybrid Machining Processes; McGraw-Hill, 2005.
38. Zhu, D.; Wang, K.; Qu, N. S. Micro Wire Electrochemical Cutting by Using In Situ Fabricated Wire Electrode. CIRP Ann. – Manuf. Technol. 2007, 56 (1), 241–244.
39. Shin, H. S.; Kim, B. H.; Chu, C. N. Analysis of the Side Gap Resulting from Micro Electrochemical Machining with a Tungsten Wire and Ultra-short Voltage Pulses. J. Micromech.
Microeng. 2008, 18 (075009).
40. Kim, B. H.; Na, C. W.; Lee, Y. S.; Choi, D. K.; Chu, C. N. Micro Electrochemical Milling. J. Micromech. Microeng. 2005, 15, 124–129.
41. Ramasawmy, H.; Blunt, L. Investigation of the Effect of Electrochemical Polishing on EDM Surfaces. Int. J. Adv. Manuf. Technol. 2007, 31 (11–12), 1135–1147.
42. Kosmac, Alenka Electropolishing Stainless Steels. Mater. Appl. Ser. 2010, 11.
43. Lee, E. S. Machining Characteristics of the Electropolishing of Stainless Steel (st316l). Int. J. Adv. Manuf. Technol. 2000, 16 (8), 591–599.
44. Jeyashree, G.; Subramanian, A.; Vasudevan, T.; Mohan, S.; Venkatachalam, R. Electropolishing of Stainless Steel. Bull. Electrochem. 2000, 16 (9), 388–391.
45. Chen, Yuan-Long; Zhu, Shu-Min; Lee, Shuo-Jen; Wang, Jong C. The Technology Combined Electrochemical Mechanical Polishing. J. Mater. Process. Technol. 2003, 140 (1–3),
203–205.
46. Kamada, Kai; Izawa, Kazuyoshi; Tsutsumi, Yuko; Yamashita, Shuichi; Enomoto, Naoya; Hojo, Junichi; Matsumoto, Yasumichi Solid-State Electrochemical Micromachining.
Chem. Mater. 2005, 17, 1930–1932.
47. Kamada, Kai; Tokutomi, Masaaki; Enomoto, Naoya; Hojo, Junichi Electrochemical Micromachining Using a Solid Electrochemical Reaction at the Metal/5-Al2O3 Microcontact.
Electrochim. Acta 2007, 52 (11), 3739–3745.
48. Kamada, Kai; Hirata, Shodai; Enomoto, Naoya; Hojo, Junichi Surface Micromachining via Solid Electrochemical Reaction on Oxide Ion Conductors. Solid State Ionics 2009,
180 (20–22), 1226–1230.
49. Kolb, D. M.; Simeone, F. C. Electrochemical Nanostructuring with an STM: A Status Report. Electrochim. Acta 2005, 50 (15), 2989–2996.
50. Kirchner, V.; Xia, X.; Schuster, R. Electrochemical Nanostructuring with Ultrashort Voltage Pulses. Acc. Chem. Res. 2001, 34 (5), 371–377.
51. Trimmer, L.; Hudson, J. L.; Kock, M.; Schuster, R. Single-step Electrochemical Machining of Complex Nanostructures with Ultrashort Voltage Pulses. Appl. Phys. Lett. 2003,
82 (19), 3327.
52. Pajak, P. T.; Desilva, A. K. M.; Harrison, D. K.; Mcgeough, J. A. Modelling the Aspects of Precision and Efficiency in Laser-assisted Jet Electrochemical Machining (LAJECM).
J. Mater. Process. Technol. 2004, 149 (1–3), 512–518.
53. Pajak, P. T.; Desilva, A. K. M.; Harrison, D. K.; Mcgeough, J. A. Precision and Efficiency of Laser Assisted Jet Electrochemical Machining. Precis. Eng. 2006, 30 (3), 288–298.
54. De Silva, A. K. M.; Pajak, P. T.; McGeogh, J. A.; Harrison, D. K. Modelling and Experimental Investigation of Laser Assisted Jet Electrochemical Machining. CIRP Ann. – Manuf.
Technol. 2004, 53 (1), 179–182.
55. Hua, Zhang; Jiawen, Xu Modeling and Experimental Investigation of Laser Drilling with Jet Electrochemical Machining. Chinese J. Aeronaut. 2010, 23 (4), 454–460.
56. Chauvy, P.-F.; Hoffmann, P.; Landolt, D. Electrochemical Micro-machining of Titanium Using Laser Oxide Film Lithography: Excimer Laser Irradiation of Anodic Oxide. Appl. Surf.
Sci. 2003, 211 (1–4), 113–127.
57. Chauvya, P.-F.; Hoffmann, P.; Landolt, D. Applications of Laser Lithography on Oxide Film to Titanium Micromachining. Appl. Surf. Sci. 2003, 208–209, 165–170.
58. Bhattacharyya, B.; Doloi, B. N.; Sorkhel, S. K. Experimental Investigations into Electrochemical Discharge Machining (ECDM) of Non-conductive Ceramic Materials. J. Mater.
Process. Technol. 1999, 95 (13), 145–154.
59. Wthrich, R.; Fascio, V. Machining of Non-conducting Materials Using Electrochemical Discharge Phenomenon; An Overview. Int. J. Mach. Tools Manuf. 2005, 45 (9),
1095–1108.
60. Zheng, Z. P.; Su, H. C.; Huang, F. Y.; Yan, B. H. The Tool Geometrical Shape and Pulse-off Time of Pulse Voltage Effects in a Pyrex Glass Electrochemical Discharge
Microdrilling Process. J. Micromech. Microeng. 2007, 17 (2), 265–272.
61. Cao, X. D.; Kim, B. H.; Chu, C. N. Micro-structuring of Glass with Features Less than 100 mm by Electrochemical Discharge Machining. Precis. Eng. 2009, 33 (4), 459–465.
62. Han, M. S.; Min, B. K.; Lee, S. J. Geometric Improvement of Electrochemical Discharge Micro-drilling Using an Ultrasonic Vibrated Electrolyte. J. Micromech. Microeng. 2009,
19 (6), 065004.
63. Cao, X. D.; Kim, B. H.; Oh, Y. T.; Chung, D. K.; Chu, C. N. Micromachining of Glass by Electrochemical Discharge Machining and Grinding. In Proceedings of the 16th
International Symposium on Electromachining (ISEM XVI); 2010; pp 657–666.
64. Yang, T.; Song, S. L.; Yan, B. H.; Huang, F. Y. Improving Machining Performance of Wire Electrochemical Discharge Machining by Adding SiC Abrasive to Electrolyte. Int. J.
Mach. Tools Manuf. 2006, 46 (15), 2044–2050.
65. Skoczypiec, S. Research on Ultrasonically Assisted Electrochemical Machining Process. Int. J. Adv. Manuf. Technol. 2011, 52 (5), 565–574.
66. Suslick, Kenneth S.; Didenko, Yuri; Fang, Ming M.; Hyeon, Taeghwan; Kolbeck, Kenneth J.; McNamara, William B.; Mdleleni, Millan M.; Wong, Mike Acoustic Cavitation and Its
Chemical Consequences. Philos. Trans. Roy. Soc. 1999, 357, 335–353.
67. Ruszaj, A.; Zybura, M.; Zurek, R.; Skrabalak, G. Some Aspects of the Electrochemical Machining Process Supported by Electrode Ultrasonic Vibrations Optimization. Proc. Inst.
Mech. Eng., Part B 2003, 217, 1365–1371.
68. Gaikwad, K. S.; Joshi, S. S. Modeling of Material Removal Rate in Micro-ECG Process. J. Manuf. Sci. Eng. 2008, 130, 034502.
69. Curtis, D. T.; Soo, S. L.; Aspinwall, D. K.; Sage, C. Electrochemical Superabrasive Machining of a Nickel-based Aeroengine Alloy Using Mounted Grinding Points. CIRP Ann. –
Manuf. Technol. 2009, 58 (1), 173–176.
70. Lupak, Mariusz; Zaborski, Stanislaw Simulation of Energy Consumption in Electrochemical Grinding of Hard-to-machine Materials. J. Appl. Electrochem. 2009, 39 (1),
101–106.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
This page intentionally left blank

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 11.16 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical
Vibration Cutting
E Shamoto and N Suzuki, Nagoya University, Nagoya, Japan
 2014 Elsevier Ltd. All rights reserved.

11.16.1 Introduction 405

11.16.2 Ultrasonic Vibration Diamond Cutting 406
11.16.2.1 Linear Vibration Cutting Process 406
11.16.2.2 Device for Ultrasonic Vibration Diamond Cutting 406
11.16.2.3 Ultrasonic Vibration Diamond Cutting of Hard Materials 407
11.16.2.3.1 Ultrasonic Vibration Diamond Cutting of Die Steel 407
11.16.2.3.2 Ultrasonic Vibration Diamond Cutting of Glasses 408
11.16.3 Ultrasonic Elliptical Vibration Cutting 411
11.16.3.1 Mechanics of Elliptical Vibration Cutting 412
11.16.3.1.1 Elliptical Vibration Cutting Process 412
11.16.3.1.2 Effects of Elliptical Vibration Conditions on Cutting Performance 412
11.16.3.1.3 Analysis of Elliptical Vibration Cutting Process 415
11.16.3.2 Ultrasonic Elliptical Vibration Cutting of Hard Materials 421
11.16.3.2.1 Devices for Ultrasonic Elliptical Vibration Cutting 421
11.16.3.2.2 Ultrasonic Elliptical Vibration Cutting of Die Steel 429
11.16.3.2.3 Ultrasonic Elliptical Vibration Cutting of Brittle Materials 435
11.16.3.3 Free-Form Surface Machining by Ultrasonic Elliptical Vibration Cutting 443
11.16.3.3.1 Machining Method and Devices for Free-Form Machining 443
11.16.3.3.2 Precision Free-Form Machining of Die Steel 446
11.16.3.4 Micro/Nanotexturing by Ultrasonic Elliptical Vibration Cutting 447
11.16.3.4.1 Machining Method and System for Micro/Nanotexturing 447
11.16.3.4.2 Ultraprecision Micro/Nanotexturing of Die Steel 450
11.16.4 Summary 453
References 453

11.16.1 Introduction

Ultraprecision/micro diamond cutting is widely utilized to produce many electronic and optical parts used for information
communication equipment and devices, such as front and back light panels for LCD, holographic optical elements, and optical
wave-guides. The demands for ultraprecision dies and molds made of hardened steel are increasing for mass production of those
devices. Machining, i.e., cutting and grinding, and chemical etching have been applied to micromachining. However, conventional
ultraprecision diamond cutting cannot be applied to machining of die steel because of excessive tool wear due to high chemical
activity with iron (1,2), while ultraprecision grinding and chemical etching are not suitable to machine such ultraprecision
microstructures especially with sharp edges.
In order to solve the problem of rapid tool wear of diamond tools, Casstevens (3) tried turning of high carbon steel in a carbon-
saturated atmosphere. Masuda et al. (4) employed single crystal cubic boron nitride tools, which have low chemical activity with
iron. Evans (5) developed a cryogenic machining system for decreasing the rate of chemical reactions at elevated temperature, and
stainless steel with a diameter of 35 mm was diamond turned with surface roughness Ra better than 25 nm. Brinksmeier et al. (6)
also proposed an approach for avoiding chemical reactions between the iron in the workpiece and the carbon in the diamond tool
by modifying the chemical composition of the workpiece subsurface; as a result, diamond tool wear was reduced by more than three
orders of magnitude. However, their attempts were not necessarily successful to realize the ultraprecision diamond cutting of
various steels in practice.
Kumabe (7) reported that the steel can be machined with a diamond tool by applying an ultrasonic vibration with optical
surface quality. The authors (8,9) developed the ultrasonic elliptical vibration cutting method, which significantly improves dia-
mond tool life and does not require the practically difficult adjustment between the vibrating and cutting directions. Recently, this
elliptical vibration cutting technology has been utilized in industry for ultraprecision/micromachining of dies and molds in wide
fields such as optics, electronics, and automobiles. Furthermore, some research works have been conducted to advance this tech-
nology, for example, application to ductile mode cutting of brittle materials such as glass, single crystal materials, and sintered
materials (10), mirror surface cutting of dies and molds with free-form surfaces (11), and high-efficient micromachining by
controlling the amplitude of the ultrasonic elliptical vibration (12). In this chapter, the above-mentioned ultrasonic vibration
cutting and elliptical vibration cutting processes are reviewed as well as their devices and applications not only to steel but also to
other difficult-to-cut materials.

Comprehensive Materials Processing, Volume 11 http://dx.doi.org/10.1016/B978-0-08-096532-1.01111-0 405

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
406 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

11.16.2 Ultrasonic Vibration Diamond Cutting

Kumabe (7) reported that the steel can be machined with a diamond tool by applying the ultrasonic vibration with optical surface
quality. Later, the authors confirmed this finding on an ultraprecision level (13). They applied a linear ultrasonic vibration to the
diamond tool basically in the cutting direction. This vibration cutting process, the vibration device, and some applications are
introduced in this section.

11.16.2.1 Linear Vibration Cutting Process

There are basically three possible directions in which the ultrasonic vibration is applied to the cutting edge in the vibration cutting,
which are the principal or cutting direction (Figure 1), the thrust direction, and the feed direction. It is reported (7,13) that only the
cutting direction is practical to achieve the precision diamond cutting of steel. The vibrating and cutting conditions are so deter-
mined that the rake face of the tool is separated from the chip in each cycle of the vibration, as shown in Figure 1. It is considered
that this high-frequency, intermittent cutting leads to long life of the diamond tool, suppressing the chemical reactions between the
steel and the diamond. The vibration in the thrust direction causes a rough jagged surface finish, while the vibration in the feed
direction causes excessive friction and tool wear in metal cutting. Therefore, the cutting process assisted with ultrasonic linear
vibration in the cutting direction is described in the following sections.

11.16.2.2 Device for Ultrasonic Vibration Diamond Cutting

Figure 2 shows an example of ultrasonic vibration tools (13). As shown, a bolted Langevin-type transducer (BLT) is often employed
to generate the ultrasonic vibration in the axial direction. In this case, the vibration amplitude is magnified by 3 times as the
vibration is transmitted through the stepped horn. The horn is supported by pointed bolts at the nodes of vibration of the horn to
minimize the interference with the resonance. The horn is supported at the two nodes of the horn so as to increase the stiffness of
support. A small diamond tool tip is bolted to the end of the horn. Since it is expected that the acceleration of tens of thousands of
gravity is applied to the tool tip, it is especially designed to minimize the effect of the weight. The device shown in Figure 2 is
designed to resonate at an ultrasonic frequency of 40 kHz, and the vibration amplitude is set to be 3 mm in machining, which
corresponds to the maximum vibration speed of 45 m min1.

Thrust direction
n
Fe tio
ed ec
dir l di r
ec t ipa
ion nc
Pri
Tool
n
cti o
g di re
tt in
Cu

e
iec
ork p n
W cti o
di re
ra tion
Vi b

Figure 1 Linear vibration cutting process.

Figure 2 An example of linear ultrasonic vibration tools.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 407

Chip Tool

Workpiece

Interference between relief face

and cut surface

Figure 3 Chipping problem in conventional linear vibration cutting.

The tool is vibrated mainly in the cutting direction. It is, however, expected that the small lateral vibration is superimposed due
to the unbalanced weight of the tool tip. It should be noted that this actual vibration at the tool tip needs to be measured and
adjusted precisely so that the cutting edge does not interfere with the cut surface when it is drawn back (see Figure 3). Otherwise,
the sharp and brittle cutting edge is easily chipped due to tensile stress on the relief face. On the other hand, the vibration cutting
must produce jagged marks (vibration marks) on the finished surface in the cutting direction, because the tool edge is not vibrated
exactly in the cutting direction to avoid the chipping. This roughness can be large on an ultraprecision level, and hence the
vibration direction needs to be aligned precisely with the cutting direction to reduce the jagged marks. This requirement conflicts
with the above-mentioned requirement to avoid the chipping, and a compromise adjustment between these two requirements is
necessary, which becomes a crucial problem in practical applications. For the same reason, the cutting direction cannot be varied
in the linear vibration cutting; for example, it cannot be applied to noncircular turning and free-form machining where the cutting
direction changes.

11.16.2.3 Ultrasonic Vibration Diamond Cutting of Hard Materials

11.16.2.3.1 Ultrasonic Vibration Diamond Cutting of Die Steel (13)
An ultraprecision facing machine equipped with the above vibration tool is shown in Figure 4. The lathe is equipped with
a hydrostatic oil spindle and a hydrostatic oil table. The vibration tool is installed with the diamond tip down, and the cutting
direction is down in this experiment. An ultraprecision single crystal diamond tool with a nose radius of 0.8 mm, a rake angle of 0 ,
and a relief angle of 7 is used to machine the free machining austenitic stainless steel SUS303Se (JIS, HV: 2.2 GPa). The depth of cut
is 2 mm, and the feed rate is set to be 5 mm per rev. The outer diameter of the workpieces is f15 mm. Nominal cutting speed is 1.4–
4.2 m min1.
Results are shown in Figure 5. As shown in Figure 5(a), the steel can be finished successfully by the ultrasonic vibration diamond
cutting with surface roughness of about 0.03 mm Rmax. The photograph shows workpiece surfaces obtained at different cutting
distances of 200 and 1600 m. It shows that mirror surfaces with roughness of less than 0.1 mm Rmax can be obtained stably up to
a cutting distance of 1600 m.
Similar cutting experiments were conducted for hardened chrome stainless steel (HV: 6.8 GPa). Workpieces with an outer
diameter of 15 mm were face-turned at cutting speed of 0.9–4.2 m min1, depth of cut of 3 mm, and feed rate of 5 mm per rev.
Figure 5(b) shows surface profiles of the finished surface. It was confirmed that surface quality and tool life for the hardened steel

Figure 4 Face-turning experiment with ultraprecision lathe and ultrasonic vibration tool.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
408 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

Figure 5 Mirror surfaces of stainless steels finished by ultrasonic linear vibration cutting. (a) Finished surfaces of nonhardened stainless steel. Left:
cutting distance of 200 m, right: 1600 m. (b) Surface profiles of hardened die steel measured in feed and cutting directions.

are similar to those for the nonhardened stainless steel, and that surface roughness of less than 0.07 mm Rmax can be obtained stably
up to a cutting distance of 1600 m. This may be because the diamond tool wear is not caused by the workpiece hardness but caused
mainly by the chemical reactions.
These experimental studies showed that steels can be finished precisely by the ultrasonic linear vibration cutting, once the
ultrasonic vibration direction is precisely adjusted to avoid both tool chipping and jagged rough surface.

11.16.2.3.2 Ultrasonic Vibration Diamond Cutting of Glasses (14)

Ductile machining of brittle materials such as ceramics and glasses is now recognized as an emerging technology with important
applications (15–17). It is normally realized by keeping the depth of cut and the feed rate below certain critical values (16–20),
which govern the transition from brittle fracture to plastic flow along the tool edge. It is, however, difficult to keep the depth of cut
and the feed rate at extremely low level over a wide range of cutting region even with ultraprecision machine tools. Therefore, it is
required to increase the critical values for ductile machining of brittle materials.
The authors applied the ultrasonic vibration cutting to ultraprecision machining of brittle materials. Before that, the ultrasonic
vibration cutting has been regarded inadequate for ductile machining, because the vibration has been normally applied to initiate
and facilitate brittle fracture by the impact power of vibration (21–24). However, the impact force does not necessarily cause the
brittle fracture. The authors expected that the ultrasonic vibration of tool enhances the lubrication or reduces the friction on the rake
face, which results in easier shear deformation in the cutting zone from experiences in research on brittle materials (25). The soda-
lime glass was selected as the workpiece material, because it is normally difficult to realize its ductile cutting by the conventional
cutting method due to the extremely small critical depth of cut and feed rate (20). Schinker et al. (17,18) tried to turn a small area of
optical glass with an experimental machine tool by the conventional cutting method, and Brehm et al. (26) turned the optical glass
by heating the work material with a point burner so as to increase the ductility of the glass material. However, ductile cutting of glass
has not been realized by other methods from the practical viewpoint.
The ductile ultrasonic vibration cutting of glasses was conducted by the same experimental setup shown in Figure 4. In this
experiment, a different ultraprecision lathe with a spindle and a table supported by air bearings was employed with the vibration
tool shown in Figure 2. The spindle is rotated by an air turbine, and the table is driven by a DC servo motor via a steel wire. The face
plate, on which the glass workpiece is glued, is fixed to the spindle via a steel ball as shown in Figures 4 and 6, so that the face of the
workpiece can be fixed at an adjustable inclination angle. The tool tips are made of single crystal diamond, and they have a nose
radius of 0.8 mm, a rake angle of 0 , and a relief angle of 7 .
Grooving experiments were carried out in order to clarify the effect of tool vibration on ductile machining. As shown in
Figure 6, workpieces of square plates are glued to the periphery of the face plate which is set to the spindle with slight inclination.
The depth of cut is therefore increased gradually as the spindle is rotated and the workpiece is cut. The feed rate is chosen to be
large enough so that the grooves are formed without interference. The maximum vibrating speed was changed in a range from 0 to

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 409

Figure 6 Illustration of grooving experiment by ultrasonic linear vibration cutting.

44 m min1 in this case to investigate the effect of the tool vibration speed. The average or nominal cutting speed changed from 0.5
to 1.5 m min1.
A photograph of a groove taken with a Nomarski-type differential interference microscope is shown in Figure 7. The groove
was formed on the glass surface by the ultrasonic vibration cutting. It is understood that the formation of the groove changes
from the ductile mode to the brittle mode as the depth of cut exceeds a certain critical value. Figure 8 shows microphoto-
graphs and profiles of grooves near the transition boundaries. The groove shown in Figure 8(a) was formed by the ultrasonic
linear vibration cutting at the maximum vibrating speed of 44 m min1, while that in (b), by the conventional cutting. The
groove shown in (a) has a smoother surface and deeper critical depth as compared to that in (b), although both were cut with
the same tool edge at the same conditions except the ultrasonic vibration. The critical depth of cut is determined here from the
profile of the groove near the transition boundary. Figure 9 summarizes the effect of the maximum vibrating speed of the tool
on the critical depth of cut. As the maximum vibrating speed exceeds about 10 times of the cutting speed, the critical depth of
cut is increased to about 7 times of that obtained with the conventional stationary tool. The increase in the critical depth of cut
is quite advantageous in ductile machining of brittle materials from the viewpoints of efficiency and ease of operation. The
cross-sectional profiles of the grooves formed by the conventional cutting and by the ultrasonic vibration cutting are measured
at the points where the grooves have the same width. Figure 10 shows the profiles of the grooves superimposed to that of the
tool edge. It shows that the profile of the tool edge is perfectly transferred to the groove surface formed by the ultrasonic
vibration cutting, while the groove formed by the conventional cutting has only a half depth as compared to the profile of the
tool edge. This result shows that the elastic recovery of the workpiece material is very small in the case of the ultrasonic
vibration cutting. The reason may be reduction in the thrust component of the cutting force due to the reduced friction and
also elastic recovery of the workpiece material in the instant when the tool separates periodically from the workpiece.
Ductile face turning of the soda-lime glass and optical glass BK7 was carried out by applying the ultrasonic linear vibration
cutting. The workpiece was fixed to the face plate so that the face of the workpiece was kept perpendicular to the spindle axis prior to
the face turning. The nominal depth of cut was selected to be 2 mm, the feed rate 5 mm per rev, and the rotational speed 90 min1,
which gives a cutting speed of 1.4–4.0 m min1. Figure 11(a) shows a photograph of the machined workpieces. Transparent
surfaces of glass with a roughness of about 0.03 mm in Rmax can be obtained successfully as shown in the photograph. Figure 11(b)
shows the profiles of the soda-lime glass measured in the feed and cutting directions. The regular feed and vibration marks are
visible respectively in Figure 11(a) and 11(b). Chips generated in this machining are continuous as shown in Figure 11(c). Ductile
face turning was also tried but could not succeed by the conventional cutting employing the same machine tool. This indicates that
the ultrasonic vibration cutting is advantageous for ductile machining of brittle materials for its larger critical depth of cut and better
transferability of cutting edge profile.

Figure 7 Illustration of grooving experiment by ultrasonic linear vibration cutting.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
410 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

Figure 8 Photographs and profiles of grooves near transition boundaries. (a) Ultrasonic vibration cutting; (b) Conventional cutting.

Figure 9 Effect of tool vibration speed on critical depth of cut.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 411

Figure 10 Profiles of cutting edge and machined grooves.

Figure 11 Results of ductile face turning of glasses by ultrasonic linear vibration cutting. (a) Glass surfaces finished by ultrasonic linear vibration
cutting. Left: soda-lime glass, right: BK7. (b) Surface profiles of soda-lime glass measured in feed (left) and cutting (right) directions. (c) SEM photograph
of chip generated in ductile cutting of soda-lime glass.

11.16.3 Ultrasonic Elliptical Vibration Cutting

The ultrasonic vibration cutting has been successfully applied to ultraprecision diamond cutting of difficult-to-cut materials,
including steels (7,13) and glass materials (14) as described in the previous sections. However, the method has the practical difficulty
in adjusting the vibrating and cutting directions. It is also disadvantageous that the cutting direction cannot be changed in machining.
On the other hand, the authors have proposed a new vibration cutting method called elliptical vibration cutting (8).
Synchronized two-directional vibration is applied to the cutting edge in such a way that the cutting edge forms an elliptical locus in
each cycle of the vibration. This new vibration cutting method significantly improves the cutting performance, including the dia-
mond tool life (9). It does not require the practically difficult adjustment between the vibrating and cutting directions, and it allows
the change of cutting direction in machining. This section provides a detailed description of the elliptical vibration cutting,

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
412 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

including the mechanics, the elliptical vibration system, and its applications to various hard and brittle materials (10) as well as
recent developments such as mirror surface cutting of dies and molds with free-form surfaces (11) and high-efficient
micromachining by controlling the amplitude of the ultrasonic elliptical vibration (12).

11.16.3.1 Mechanics of Elliptical Vibration Cutting

11.16.3.1.1 Elliptical Vibration Cutting Process (8)
Figure 12 shows a schematic illustration of the elliptical vibration cutting process. In this process, the tool is fed at a nominal cutting
speed and vibrated elliptically at the same time. As the maximum vibration speed is set to be higher than the cutting speed, the tool
is separated from the chip in each cycle of the vibration. The cutting is taken place after re-entering of the cutting edge into the
workpiece, and the chip is mainly pulled up and formed while the tool moves up in the chip flow direction. The tool moves down
without cutting while it is separated from the chip. The detail is shown as follows.
The cutting starts at time t1, and the tiny part left in the previous cycle is cut at small depth of cut until time t2, when the rake face
contacts with the chip formed in the previous cycle. In the actual process, this tiny part may be burnished and pushed into the
workpiece and the chip due to cutting edge roundness. After the time t2, the workpiece material is sheared and removed as the chip
at large depth of cut. As the tangent of cutting path exceeds the shear direction at time t6, the frictional direction is reversed. In the
actual process, the tool may move together with the chip in the shear direction near t6 without the sudden reversal of friction
because of elastic deformation. Then, the tool is separated from the chip at time t5, when the tangent is parallel to the rake face. This
reduced or reversed friction leads to significant reduction of cutting force, energy, and heat generation.

11.16.3.1.2 Effects of Elliptical Vibration Conditions on Cutting Performance (8,27)

Orthogonal cutting experiments of copper are carried out within a scanning electron microscope (SEM) in order to observe the
cutting process directly and to measure the dynamic cutting force in each cycle of the vibration. The basic effects of the elliptical
vibration conditions on cutting performance are also clarified by this series of experiments.
Figure 13 shows a sketch of an orthogonal cutting device equipped with an elliptically vibrated tool developed. The cutting
device is installed within a SEM in order to observe the cutting process. The tool is vibrated by two piezoelectric actuators (PZTs)

Chip Tool

Locus of
cutting edge

Workpiece

Figure 12 Elliptical vibration cutting process.

25 65

PZT

Tool tip

Feed
direction

Dynamometer Workpiece

Figure 13 Elliptical vibration cutting device installed within SEM.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 413

arranged at a right angle, and the cutting force is measured with a piezoelectric-type dynamometer. The tool is made of high-speed
steel. Its rake and clearance angles are 0 and 30 , respectively, and the measured radius of the cutting edge roundness is
approximately 1 mm. The workpiece material is oxygen-free copper (OFC) with a thickness of 0.25 mm and a length of 5 mm in
the cutting direction. The depth of cut is 10 mm and the nominal cutting speed, which is equal to the feeding speed of workpiece
in the cutting direction, is 260 mm min1. These conditions are kept constant, and vibration conditions are changed as follows.
The vibration frequency is varied in a range from 0 to 6 Hz. The horizontal vibration amplitude (cutting direction) is always set to
be 5 mm0p in the conventional/elliptical vibration cutting, and the vertical vibration amplitude (thrust direction) is changed
from 2.5 to 7.5 mm0p in the elliptical vibration cutting. The phase between the two directional vibrations is varied in a range of
45 –135 . The workpiece surface is finished prior to each experiment under the same conditions as the corresponding cutting
experiment.
First, the three cutting processes, i.e., the ordinary cutting, the conventional linear vibration cutting, and the elliptical
vibration cutting, were compared at the same conditions except the vibration. In the conventional vibration cutting, the tool was
vibrated mainly in the cutting direction, but the vibrating direction was slightly inclined to the thrust direction so that the tool
was separated from the finished surface during moving backward. The inclination angle was about 9.5 . In the case of elliptical
vibration cutting, the vertical vibration amplitude and the phase were set to be 5 mm0p and 90 , respectively, which corresponds
to a circular vibration with a radius of 5 mm. The vibration frequency was 1.2 Hz in the both vibration cutting methods.
Figure 14 shows SEM photographs of the chips obtained in the three different cutting processes, and their corresponding
principal and thrust components of cutting force measured. As shown in Figure 14(a), the chip is relatively thick and the cutting
forces are high and almost constant in the ordinary cutting. Figure 14(b) shows results obtained by the conventional vibration
cutting. The average cutting force is reduced as compared to the ordinary cutting due to the intermittent cutting. However, the
force measured during cutting and the chip thickness are as large as those obtained by the ordinary cutting. On the other hand,
the chip thickness becomes extremely small in the case of the elliptical vibration cutting, as shown in Figure 14(c). The reason

(a)

(b)

(c)

Figure 14 Comparison of chip formation and cutting force among three cutting processes. (a) Ordinary cutting. (b) Conventional linear vibration cutting.
(c) Elliptical vibration cutting.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
414 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

for this is considered to be as follows. There exists a period when the chip is pulled up by the tool moving upward, although the
chip is pushed down at the beginning of cutting in each cycle. The negative thrust force shown in Figure 14(c) proves the
existence of such phenomenon, and this negative thrust force causes reduction in the chip thickness and hence reduction in
the principal force.
Next, elliptical vibration parameters were varied to clarify their effects on the process. As a basic condition, the circular vibration
with a radius of 5 mm0p and a vibration frequency of 1.2 Hz was selected, and each parameter was varied from this basic
condition. Figure 15 shows results obtained at varied vibration frequency. Analytical results are also shown in the figures,
although analysis of elliptical vibration cutting process is described in the next section in detail. Figure 15(a) shows that the shear
angle increases and the cutting force decreases as the vibration frequency or the speed ratio increases. This is because the above-
mentioned negative friction period increases relatively and the chip is pulled up at high speed ratio. Figure 15(b) shows the
corresponding surface roughness measured in the cutting direction. The roughness is also improved as the speed ratio increases,
although the cutting speed or cutting efficiency decreases in practice because of limitation of vibration speed. As shown in
Figure 16, the shear angle increases and the force decreases as the vertical vibration amplitude increases. This is because the vertical
vibration causes the negative friction. However, this effect becomes small at large vertical amplitude of greater than 5 mm. The
roughness increases at large vertical amplitude. Figure 17 shows effects of the phase and its corresponding vibration locus. It shows
that the shear angle increases and the force decreases with an increase of the phase. This is also caused by longer negative friction
period. This effect is saturated at large phase of greater than 90 , and the surface roughness becomes the minimum at 90 . From
these observations and simulations, the circular vibration locus can be recommended to reduce the cutting force and the surface
roughness at the same time.

(a) Shear angle calculated

Shear angle measured
Maximum resultant force measured
6

Maximum resultant force N

60 5
°

4
Shear angle

40
3

2
20
1

0 0
0 10 20 30 40 50
Maximum vibration speed / cutting speed

0 2 4 6 8
Frequency Hz

(b) Theoretical roughness

Surface roughness measured
101
m

100
Surface roughness

10–1

10–2

10–3
1 5 10 50
Maximum vibration speed / cutting speed

0.1 0.5 1 5
Frequency Hz

Figure 15 Effects of vibration frequency or speed ratio. (a) Change in shear angle and maximum resultant cutting force. (b) Change in surface
roughness.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 415

Shear angle calculated

Shear angle measured
Maximum resultant force measured
60 2

Maximum resultant force N

50

°
40

Shear angle
30 1

20

10

0
0 2 4 6 8 10
(a) Vertical amplitude m

Theoretical roughness
Surface roughness measured
0.6
m

0.5
Surface roughness

0.4

0.3

0.2

0.1

0 2 4 6 8 10
(b) Vertical amplitude m

Figure 16 Effects of vertical vibration amplitude. (a) Change in shear angle and maximum resultant cutting force. (b) Change in surface roughness.

11.16.3.1.3 Analysis of Elliptical Vibration Cutting Process

11.16.3.1.3.1 Quick analysis with thin shear plane model (28)
An analytical model of orthogonal elliptical vibration cutting process, in which all velocities, forces, and vibrations are perpen-
dicular to the cutting edge, was developed and verified experimentally to predict theoretical roughness, critical relief angle, shear
angle, etc., and to clarify effects of various vibration conditions (27). The shear angle and roughness predicted with this model are
plotted in Figures 15–17. However, the practical cutting processes are not generally the orthogonal type. The elliptical vibration
locus is inclined around the cutting direction in most of the industrial applications. Therefore, simple analytical models of the three-
dimensional (3D) elliptical vibration cutting processes were developed in order to simulate the processes quickly and to understand
their basic mechanics. This 3D elliptical vibration cutting model and its verification are presented in this section.
Figure 18 shows typical chips observed in the practical ultraprecision elliptical vibration cutting with a single crystal diamond
tool. The chips are usually continuous and the chip thickness is almost constant as shown in the figure, although the instantaneous
cutting direction is varied. This suggests that the shear deformation may be caused by average or total force applied to the workpiece
material in one cycle of the elliptical vibration. As mentioned in the previous section, see Figure 12, the elliptical vibration cutting
process can be divided into the two processes, i.e., the microcutting or burnishing process from t1 to t2 and the cutting process with
the reduced or reversed friction from t2 to t5. When the depth of cut is large enough relatively to the vertical amplitude of vibration,
the former micromachining period is negligible. On the other hand, when the depth is extremely small and the cutting edge is sharp
enough, the latter cutting period vanishes, and the process may be similar to micromilling except that the rake face is not inclined
during cutting and that the tool path is elliptical. Judging from the observations of process and chips (8,27,29), it is considered that
the constant continuous chip formation with the unique frictional phenomenon is the main process in many of the practical
elliptical vibration cuttings. Whereas the mechanism of the former micromachining can be understood as the ordinary micro-
machining process, the latter chip formation process with the reduced or reversed friction is unique and academically interesting.
Thus, the present section is focused on the latter process. For simplicity, the cutting edge is assumed sharp enough, and the elastic
deformation is neglected. The instantaneous depth of cut is decreased gradually during the latter cutting period, but this effect is also
neglected; i.e., the depth is assumed constant. The simple thin shear plane model is employed, and the maximum shear stress and
minimum energy principles are separately applied to simple simulations and better comprehension of the continuous chip
formation process.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
416 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

Shear angle calculated Cutting direction

Shear angle measured
Maximum resultant force measured 45°
6

Maximum resultant force N

60
67.5°

°
4
40

Shear angle
90°

2
20
112.5°

0 0
40 60 80 100 120 140 135°
(a) Phase difference °

Theoretical roughness
Surface roughness measured
0.6
m

0.5
Surface roughness

0.4

0.3

0.2

0.1

0
40 60 80 100 120 140
(b) Phase difference °

Figure 17 Effects of phase between vertical and horizontal vibrations. (a) Change in shear angle and maximum resultant cutting force. (b) Change in
surface roughness.

30 m

1 m 50 m

(a) (b) (c)

Figure 18 Chips formed in practical ultraprecision elliptical vibration cutting. (a) Curled chips, (b) rake face side, and (c) free side. (Conditions)
workpiece: hardened steel (JIS: SUS420J2), diamond tool: R 1 mm, rake angle: 0 , cutting speed: 1 m min1, depth of cut: 10 mm, feed: 20 mm per rev,
vibration: circular, amplitude: 2 mm0p, frequency: 39 kHz, mist: kerosene.

11.16.3.1.3.1.1 Geometry of 3D elliptical vibration cutting process

Figure 19 shows various types of elliptical vibration cutting. Figure 19(a) is the orthogonal type. Figure 19(b) is an oblique type,
where the cutting edge is vibrated elliptically in the finished surface. This type is considered to be intermittent cutting with sidewise
motion of the cutting edge, which is equivalent to intermittent oblique cutting (29,30). In this case, the cutting force is reduced,
because the chip is pulled sidewise, and this reduces the thrust component of friction that prevents the chip from flowing out, as
shown in Figure 19(b). Figure 19(c) shows an intermediate type between the above two types, which has both effects; i.e., the
friction is reduced or reversed and its thrust component is further reduced. This intermediate type is utilized widely in practice.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 417

Oblique component
of friction
Elliptical
vibration
Top
view

Chip Average cutting direction

Tool

Side
view

Vertical component
Thrust component
Workpiece of friction
(a) (b)

y
x

z
i

z x

(c) (d)

Figure 19 Various types of elliptical vibration cutting. (a) Orthogonal type; (b) oblique type; (c) intermediate type; and (d) with inclination.

Furthermore, the cutting edge is not always perpendicular to the cutting direction, as shown in Figure 19(d), where the obliquity of
cutting direction is denoted by an inclination angle i. In this case, the inclination of cutting direction enhances obliquity of the
inclined elliptical vibration. The tool is vibrated at an angular frequency u and fed at a nominal cutting speed vc, and the workpiece
is stationary. Then, the tool path with the vibration can be expressed as follows:
xðtÞ ¼ a cosðutÞ  vc t cos i
yðtÞ ¼ b cosðut þ 4y Þ [1]
zðtÞ ¼ c cosðut þ 4z Þ þ vc t sin i
where x, y, and z are Cartesian coordinates of the relative position between the tool and the workpiece in the direction normal to the
cutting edge, the thrust direction, and the cutting edge direction, respectively, see Figures 12 and 19(d). The nominal cutting
direction lies on the x–z plane, i.e., finished surface. a, b, and c are vibration amplitudes in the x, y, and z directions, respectively. t is
time and 4y and 4z are phase shifts of the vibrations in the y and z directions, respectively.
When the minimum relative speed in the direction normal to the rake face is negative, the cutting process becomes
intermittent.
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
vc cos icos g  u ða cos g  b sin g cos 4y Þ2 þ ðb sin g sin 4y Þ2 < 0 [2]

where g is the normal rake angle, which is the rake angle in the plane normal to the cutting edge.
Under this intermittent condition, the time t5 can be obtained by solving eqn [3], since the relative speed in the direction normal
to the rake face becomes zero at t5.
fau sinðut5 Þ þ vc cos igcos g  b u sinðut5 þ 4y Þsin g ¼ 0 [3]

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
418 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

The rake face is separated from the chip at t5, and then it comes back to the same plane at t2 þ T, where T is a vibration period.
Thus, t2 can be solved numerically by the following equation.
fxðt2 þ TÞ  xðt5 Þgcos g þ fyðt5 Þ  yðt2 þ TÞgsin g ¼ 0
n o
r afcosðut2 Þ  cosðut5 Þgcos g þ b cosðut5 þ 4y Þ  cosðut2 þ 4y Þ sin g ¼ vc ðT  t5 þ t2 Þcos gcos i [4]

The times t1 and t4 can be solved by equating their distance x(t1)  x(t4) to the pitch 2pvc cos i =u and also equating their vertical
positions y(t1) and y(t4). Then, the critical relief angle q1 and the roughness Rth can be obtained by utilizing these times (27). At the
time t1, the relative speeds in the x and y directions are au sinðut1 Þ  nc cos i and bu sinðut1 þ 4y Þ respectively, and thus the
critical relief angle q1, which is measured in the xy plane, is given by
bu sinðut1 þ 4y Þ
q1 ¼ tan1 [5]
au sinðut1 Þ þ nc cos i
The roughness Rth can be calculated as difference between the vertical positions y(t1) and y(t3).
Rth ¼ b cosðut1 þ 4y Þ  b cosðut3 þ 4y Þ [6]

11.16.3.1.3.1.2 Prediction of shear direction in 3D elliptical vibration cutting process

The shear direction in the 3D vibration cutting process is assumed to be constant, judging from the above observations. The shear
direction is predicted by extending the 3D thin shear plane model (30) as follows. As the shear direction ! v s (unit vector) is constant,
the chip moves in a constant direction independently of the instantaneous tool velocity ! v t , i.e., instantaneous cutting velocity.
Thus, the instantaneous velocity of the chip material ! v m , see Figure 20(a), is derived by:
!v t $!
n
!v m ¼ j! v m j!v s ¼ ! !! vs [7]
v s$ n
where ! n is a unit vector normal to the rake face and given by ðcos g; sin g; 0Þ, and ! v is given by the time derivative of the tool
t
position.
!
v t ¼ ð  au sinðutÞ  vc cos i; bu sinðut þ 4y Þ; cu sinðut þ 4z Þ þ vc sin iÞ [8]
!
The friction f (unit vector) acts in the direction opposite to the relative motion between the tool and the chip.
!
f ¼ ð!
vt! v m Þ=j!
vt ! v mj [9]
Then, the direction of resultant cutting force ! r (unit vector) can be obtained by the following equation, see Figure 4(b),
assuming a constant frictional angle b.
! !
r ¼! n cos b þ f sin b [10]
! !
The resultant force R needs to produce the shear force F s in the shear plane, and hence it can be calculated by:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 ffi
!
! F s
! swd ! v s $! e y þ ð! v s $! e x Þ2 !
R ¼! ! r ¼ ! !  ! ! r [11]
r$v s v $ e ð r $ v Þcos i
s y s

where s is shear stress in the shear plane, w and d are the width and depth of cut, see Figure 19(d), and !
e x and !
e y are unit vectors in
the x and y directions, respectively.
The instantaneous cutting power U is a product of the cutting velocity and the resultant force.
!
U ¼ R $! vt [12]

(a) (b) (c)

Figure 20 Basic three relations in thin shear plane model. (a) Velocity relation, (b) force relation, and (c) MSS principle.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 419

The shear direction is predicted with the two fundamental principles separately. One is the maximum shear stress (MSS)
principle (31,32). It is assumed that the shear occurs in the maximum shear stress direction, where the resultant force makes 45
with both of the shear direction and the shear plane (30). In the present 3D vibration cutting process, the resultant force is varied.
Therefore, it is assumed that the shear direction is determined by summation of the instantaneous resultant force during the cutting
period from t2 to t5:
! X !
R sum ¼ j!
v m j R Dt [13]

where j!v m j is multiplied as a weight function, because the influence on the shear deformation may be increased as the removed
!
material is increased. As a result, the shear direction !v s and the summed force R sum need to satisfy the following equation.
! !
v s cos 45 þ !
R sum = R sum ¼ ! n s sin 45 [14]

where ! n s is a unit vector normal to the shear plane and given by:
!  

ns ¼ ! vs! e y ! v s! e y [15]
!
Therefore, the correct shear direction ! v s can be searched as follows. First, R sum is calculated by eqns [7]–[13], and then !
v s is
checked by eqns [14] and [15]. While the error between the left and right sides of eqn [12] is not small enough, the shear direction
!v s is updated iteratively.
The other fundamental shear angle prediction is based on the minimum energy principle (33). The shear direction is determined
here so that the total cutting energy E becomes minimum, where E is given by integrating instantaneous cutting power numerically
from t2 to t5.
X
E¼ UDt [16]

Thus, solution of the shear direction !v s can be found by searching !

v s which gives the minimum energy.
The present analytical models yield to the oblique cutting models (30) when a ¼ b ¼ c ¼ 0, and they are also valid for other 3D
vibration cutting processes with different vibration loci.

11.16.3.1.3.1.3 Simulation results and discussions

The developed analytical models are applied to the orthogonal, oblique, and intermediate types of elliptical vibration cutting, and
simulated results are compared with some experimental data. Figure 21 shows the shear angles predicted and measured at various
speed ratios in the orthogonal type (27), see Figure 15, where the speed ratio is defined as the maximum vibration speed in the
nominal cutting direction divided by the nominal cutting speed. The friction angle b utilized in the simulation is identified to be
23.5 by the forces measured in the ordinary cutting (27). Figure 22 shows the chip flow angles predicted and measured at various
speed ratios in the oblique type of elliptical vibration cutting (29). b is 41.3 (29) in the simulation.
Considering the simplicity of the models, the characteristic angles predicted with either principle agree with the experimental
data. It shows that the two types of elliptical vibration cutting processes can be predicted quickly with the developed models. The
agreements suggest that the orthogonal and oblique types of elliptical vibration cutting can be understood as the unique chip
formation processes with the reduced friction and the reduced thrust component of friction, respectively.
Transient cutting forces in the intermediate type are also predicted and compared with experimental data (29) in Figure 23. b is
41.3 and s is 94.3 MPa (29). The results are in good agreement during the cutting period. Disagreement before and after the cutting
is considered to be caused by the elastic deformation, which is neglected here. The thrust force is slightly reversed in the cutting
period by the orthogonal component of the elliptical vibration. The force in the cutting edge direction, which is shown by ‘Normal’

90
Maximum shear stress principle
Minimum energy principle
Experiment
Shear angle deg

60

30

0
0 10 20 30 40 50
Speed ratio

Figure 21 Shear angles at various speed ratios in orthogonal type of elliptical vibration cutting. (Conditions) workpiece: copper, rake angle: 0 , cutting
speed: 0.26 mm min1, depth of cut: 10 mm, vibration: circular, amplitude: 5 mm0p, frequency: changed.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
420 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

90
Maximum shear stress principle
Minimum energy principle
Experiment

Chip flow angle deg

60

30

0
0 20 40 60 80
Speed ratio

Figure 22 Chip flow angles at various speed ratios in oblique type of elliptical vibration cutting. (Conditions) workpiece: aluminum, rake angle: 0 , depth
of cut: 0.05 mm, vibration: circular, amplitude: 0.5 mm0p, frequency: 10 Hz.

400 Maximum shear stress principle

Minimum energy principle
Experiment

200
Cutting force N

Principal

Thrust

Normal

–200
0.00 0.05 0.10 0.15 0.20
Time s

Figure 23 Transient cutting forces in intermediate type of elliptical vibration cutting. (Conditions) workpiece: aluminum, rake angle: 0 , depth of cut:
0.3 mm, width of cut: 2 mm, cutting speed: 94.3 mm min1, vibration: circular, amplitude: 0.5 mm0p, frequency: 10 Hz, inclination of elliptical vibration:
80 from orthogonal type, i.e., 10 from finished surface.

in the figure, is caused by the oblique component of the elliptical vibration, and it also reduces the cutting force by reducing the
thrust component of friction. The measured peak principal force was 140 N, while average principal force measured in the ordinary
cutting was 480 N (29). It is considered that this significant reduction is caused by combination of the above effects of the
orthogonal and oblique components.
In summary, the simulated results agree roughly with the experimental data, and the models represent the unique cutting
mechanics of two types of elliptical vibration cutting. They suggest that the continuous chip formation process in the orthogonal
type of elliptical vibration cutting is the unique cutting process with the reduced or reversed friction, and the oblique type of
elliptical vibration cutting is equivalent on average to the intermittent oblique cutting with the reduced thrust component of
friction. The practical elliptical vibration cutting can be understood as the intermediate process between them.

11.16.3.1.3.2 Detailed analysis with FEM model (34)

The finite element (FE) model of the workpiece and the rigid tool is illustrated in Figure 24. A FEM software, Msc-Marc, was
employed in order to analyze the processes where the tool was vibrated linearly in conventional vibration cutting and elliptically in
elliptical vibration cutting. The corner lines in Figure 24 show the boundary conditions; i.e., the left boundary was fixed in the
cutting direction and the lower one was fixed in the cutting and thrust directions. An element type of Quad4 was utilized in the
analysis. The workpiece model initially contained 611 elements and 648 nodes, while size of the workpiece was assumed to be
0.2 mm long and 0.05 mm high. It should be mentioned that the numbers of elements and nodes changed due to automatic
remeshing during the analysis. The automatic remeshing is needed for the cutting simulation, since the deformation is large and the
material is separated at the tool tip. It was assumed that the material separation occurs when the stress in the material near the tool

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 421

Figure 24 FE model of orthogonal cutting process.

tip reaches yield stress of the material. By applying the above conditions to the tool and the workpiece during the vibration cutting
processes, cutting forces and stress distribution were obtained.
The cutting simulation was conducted under the same conditions as the above-mentioned experiments within the SEM, see
Figure 14 (8). Properties of the workpiece material are given as follows: elastic modulus E ¼ 125 GPa, yield stress sy ¼ 290 MPa,
and Poisson’s ratio n ¼ 0.34. Perfect plasticity was assumed; i.e., work-hardening effect was neglected for simplicity. Friction
coefficient between the chip and the rake face was assumed to be 0.43, which was measured experimentally in ordinary cutting (8).
The chip formation during the vibration cutting processes and the calculated equivalent stress distribution are shown in Figures
25 and 26. It should be noted that their scales are different, and chip thickness is much smaller in the elliptical vibration cutting than
in the linear vibration cutting. This indicates that the shear deformation zone is significantly small in elliptical vibration cutting. This
tendency agrees with the experimental results shown in Figure 14 (8).
As shown in Figure 25, one cycle of the conventional linear vibration cutting process can be divided into five stages. During the
first stage shown in Figure 25(a), the tool approaches the chip, and in the second stage, see Figure 25(b), the tool contacts the chip
and starts penetrating into the workpiece causing the chip formation. During the third stage shown in Figure 25(c), the workpiece
material is deformed and flows out as a chip on the rake face with severe friction. Thus, the stress becomes the maximum at this
stage. The following stage is unloading as shown in Figure 25(d). The tool moves backward but remains in contact with the chip
even after the vibration speed exceeds the cutting speed due to the elastic recovery. During this stage, relaxation of the elastic stresses
takes place. The last stage is withdrawal of the tool from the chip, as shown in Figure 25(e). The tool is fully separated from the chip,
but some stresses still remain as residual stresses.
Figure 26 shows one cycle of the elliptical vibration cutting process simulated by the developed FE model. During the first stage
shown in Figure 26(a), the tool approaches the previously finished surface. In the second stage shown in Figure 26(b), the tool
contacts the surface and starts penetrating into the workpiece. During the third stage shown in Figure 26(c), the tool generates the
finished surface, where the actual depth of cut is extremely small. Note that this small part is not formed into a chip but pushed into
the workpiece in this surface generation process. It is also interesting that the equivalent von Mises stresses become pretty high
already at this stage and that the deformation zone is split into two regions. It is considered that this splitting is caused by the small
part of material pushed into the workpiece under the chip. Then, the material is mainly removed during the fourth stage shown in
Figure 26(d), where the tool moves upward roughly in the shear direction. Thus, the friction between the chip and the tool becomes
very small. The friction or contact area is also very small in comparison with the conventional vibration cutting process shown in
Figure 25. The area is less than the uncut chip area or comparable with that, which never happens in the conventional cutting
processes. This extraordinary small contact area or length is also considered to be caused by the small part pushed into the
workpiece, judging from Figure 26(c) and 26(d). The following stages are the elastic recovery and the withdrawal of tool, as shown
in Figure 26(e) and 26(f).
The cutting forces simulated by the present FE model are shown in Figure 27, while the corresponding cutting forces measured in
the fundamental orthogonal experiments (8) are shown in Figure 14. The cutting conditions are the same in the experiments and
simulations, and the experimental and analytical results agree well as shown in the figures. The noncutting periods are shorter in the
both experimental vibration cutting processes than in the simulations, because the elasticity of the mechanical structure is neglected
in the simulations. In the elliptical vibration cutting experiment and simulation, thrust force becomes negative in the latter part of
cutting period. This indicates that the frictional direction is reversed and the frictional force becomes extremely small in average in
the elliptical vibration cutting process. This extremely small thrust force causes reduction in chip thickness and hence reduction in
the principal force. Note that the scales in Figures 14(c) and 27(b) are much smaller than those in Figures 14(a), 14(b), and 27(a).
In summary, the FEM simulation result indicates the importance of the small part of workpiece material in the elliptical
vibration cutting process, which is pushed into the workpiece before the tool contacts with the previously formed chip. It causes the
small contact area between the tool and the chip and the unusual plastic deformation zone, which is split into two regions.

11.16.3.2 Ultrasonic Elliptical Vibration Cutting of Hard Materials

11.16.3.2.1 Devices for Ultrasonic Elliptical Vibration Cutting
In order to attain higher machining efficiency in the elliptical vibration cutting, the tool needs to be vibrated at ultrasonic
frequency by exciting resonant modes of the vibrator. A special design of the vibrator, therefore, is required. The vibration locus

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
422 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

Figure 25 Chip formation process and stress distribution simulated in one vibration cycle in conventional linear vibration cutting. Machining conditions
are same as Figure 14(b). (a) Before engagement, (b) start of engagement, (c) material removal, (d) end of engagement, and (e) withdrawal of tool.

needs to be controlled ultraprecisely at the same time. For instance, in order to attain practical manufacturing of the ultraprecision
dies, the shape accuracy of the dies needs to be ensured as well as the surface quality. The shape accuracy obtained by the elliptical
vibration cutting technology heavily depends on stability of the locus of the ultrasonic elliptical vibration. It is understood that
deviation of the elliptical vibration locus, especially the vibration amplitude in the depth of cut direction, causes errors in the
workpiece profile. It is, therefore, necessary to control the vibration locus precisely during machining. The following sections
introduce the devices for ultrasonic elliptical vibration cutting, which generate a two-degree-of-freedom (DOF) elliptical vibration.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 423

Figure 26 Chip formation process and stress distribution simulated in one vibration cycle in elliptical vibration cutting. Machining conditions are same
as Figure 14(c). (a) Before engagement, (b) start of engagement, (c) generation of finished surface, (d) contact with previous chip and material removal,
(e) end of engagement, and (f) withdrawal of tool.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
424 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

1.5
5.5 Principal force Principal force
4.5
1
3.5 Thrust force

Force, N
Force, N 2.5 0.5 Thrust force

1.5
0.5 0

–0.5
0 2
0 2 –0.5
(a)
Phase of vibration, rad (b)
Phase of vibration, rad

Figure 27 Cutting forces simulated by developed FE model. Machining conditions are same as Figure 14. (a) Conventional vibration cutting, (b) elliptical
vibration cutting.

11.16.3.2.1.1 Two-DOF vibrator utilizing two bending modes and its controller (35)
Figure 28 shows the ultrasonic elliptical vibrator and its third resonant mode of bending, whose resonant frequency is 20 kHz. It
was developed by the authors in previous research (35,36). The vibrator has four large piezoelectric plates as actuators and two small
plates as sensors. It is vibrated in the cutting direction in the third resonant mode by applying sinusoidal voltage to the upper and
lower actuators with a phase shift of 180 . It is also vibrated in the normal direction to the cutting direction and the vibrator axis by
exciting the front and back actuators. The diamond tool tip is set at the end of the vibrator and vibrated elliptically by resonating the
vibrator in the both directions with some phase shift. The two directional vibrations are detected by the two sensors, respectively.
The vibration amplitude is magnified by the step horns. The vibrator is supported rigidly at the two nodal points in order not to
disturb the vibration.
It is important for the vibration control that the two directional vibrations are independent of each other. However, the two
vibrations interfere considerably in practice as shown in Figure 29. It shows frequency responses in the two directions when the
vibrator is excited in the cutting direction. The vibration in the normal direction is due to cross talk and undesirable. Assuming that
the cross talk is caused by the cross stiffness k12 and k21 shown in Figure 30, a cross-coupled control method was applied to remove
the cross talks. The vibration signals detected by the piezoelectric sensors are input to the cross-talk remover as feedback signals,
and the exciting voltages are compensated by the remover so as to cancel the cross stiffness k12 and k21. Then, the compensated
voltages are applied to the piezoelectric actuators via the power amplifiers.
The frequency responses obtained with use of the developed remover are shown in Figure 31. The cross talk to the normal
direction is removed almost completely by the cross-coupled control method. Another cross talk to the cutting direction is also
removed by the same method.
Figure 32 shows the vibration control system, which was developed by utilizing the cross-talk remover. The vibration amplitudes
and their phase shift are kept to desired values in the system. The resonant frequencies are slightly different in the two directions
even after adjustment, and they change due to cutting force and temperature change. Thus, their average resonant frequency is
chased automatically by the control system.
The developed vibration control system was applied to ultraprecision turning of hardened die steel in order to improve the
machining accuracy, and its validity was verified experimentally. The experimental setup is shown in Figure 33. The elliptical vibration

Cutting direction Upper PZT actuator

PZT sensors
Diamond tool Front PZT actuator

Lower PZT
Normal direction actuator
Ultrasonic elliptical vibrator

Supporting points

Third resonant mode of bending

Figure 28 Ultrasonic elliptical vibrator. Reproduced from Shamoto, E.; Suzuki, N.; Moriwaki, T.; Naoi, Y. Development of Ultrasonic Elliptical Vibration
Controller for Elliptical Vibration Cutting. Ann. CIRP 2002, 51 (1), 327–330; Suzuki, N.; Nakamura, A.; Shamoto, E.; Harada, K.; Matsuo, M.; Osada, M.
Ultraprecision Micromachining of Hardened Steel by Applying Ultrasonic Elliptical Vibration Cutting. In Proc. 2003 Int. Symp. MHS, 2003; pp 221–226.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 425

4
Cutting direction

Amplitude m
3 Normal direction

0
100

Phase lag deg

0

100

200

300
20.05 20.1 20.15 20.2 20.25 20.3

Frequency kHz

Figure 29 Frequency response of elliptical vibrator when excited in cutting direction.

Crosstalk remover
Remover Sensitivity of PZT sensor
gain 1

Elliptical vibrator

Exciting Power Displace-

PZT's
voltages amp. ments

Remover
gain 2

Figure 30 Method to remove cross talks (subscripts 1 and 2 indicate cutting and normal directions, respectively).

tool with the diamond tool tip is mounted on the ultraprecision lathe, and semicircular workpieces are face turned. The tool tip is
made of single crystal diamond, whose nose radius is 1 mm and rake and relief angles are 0 and 15 , respectively. The workpiece is
hardened stainless steel, JIS: SUS420J2, and its hardness is HRC40. Its outer and inner diameters are 27 and 6 mm, respectively. Its
surface has a step of 20 mm as shown in the figure, so that the depth of cut is increased suddenly from 5 to 25 mm. Kerosene is supplied
as cutting mist during early stage of machining, and then it is changed to manual supply of mineral oil during latter stage of machining.
The feed rate is 16.7 mm per rev, and the spindle rotation is 60 min1. The initial vibration locus is set to be circular with a radius of
4 mm, and its frequency is 20 kHz. The angle of elliptical vibration shown in the figure is chosen to be 60 in the present experiment.
The experimental results are shown in Figures 34–36. Figure 34 shows changes in the controlled parameters, i.e., the vibration
amplitudes, the phase difference, and the average phase lag. Figure 34(a) shows the result without the vibration control, while (b),
with the vibration control. As shown in Figure 34, the controlled parameters vary considerably due to the disturbances, such as cutting
force changes at the beginning of cutting and the step difference in the depth of cut and changes in the lubricant supply, when the
vibration control is turned off. On the other hand, all parameters are successfully kept constant by the developed control system.
Measured surface profiles are shown in Figure 35. As shown in Figure 35(a), the shape accuracy is about 2 mm when the
vibration control system is not activated. This shape error corresponds to the change in the vibration amplitude in the normal

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
426 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

Cutting direction
3

Amplitude m
Normal direction

0
200

Phase lag deg

100

100

200
20.05 20.1 20.15 20.2 20.25 20.3
Frequency kHz

Figure 31 Frequency responses with cross-talk remover when excited in cutting direction.

Ultrasonic elliptical vibration controller

Phase command Controller

Amplitude
commands Crosstalk remover
Amplifier
PZT actuators

PZT
sensors
Amplifer

Supports
Ultrasonic elliptical vibrator
Ultrasonic elliptical vibrator

Figure 32 Developed vibration control system.

Manual
Mist
Cutting fluid
25 m 5 m
Depth of cut

Spindle

Workpiece
Spindle rotation

Angle of elliptical vibration

Feed direction Elliptical Elliptical vibration

vibrator
Diamond tool tip

Figure 33 Setup for face-turning experiments.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 427

Beginning Change in depth End Beginning Change in depth End

of cutting of cut of cutting of cutting of cut of cutting

Mist Manual Mist Manual

6 Normal direction
Amplitude m

Normal direction
4
Cutting direction
Cutting direction
2

0
150
Phase difference deg

120

90

0
Average phase lag deg

–80

–100

–120

–140

–160
0 5 10 15 20 25 0 5 10 15 20 25
Time min Time min
(a) (b)

Figure 34 Changes in controlled parameters during machining. (a) Without vibration control and (b) with vibration control.

Beginning Change in End of

of cutting depth of cut Mist Manual cutting
1 m

0.5 m

(a)

Beginning Change in End of

of cutting depth of cut Mist Manual cutting
1 m

0.5 m

(b)

Figure 35 Improvement of shape accuracy by developed vibration control system. (a) Without vibration control and (b) with vibration control.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
428 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

Figure 36 Schematic illustration of longitudinal-bending vibrator.

direction shown in Figure 34(a). On the other hand, the shape error is successfully reduced to less than 0.2 mm while the mist of
lubricant is applied, when the elliptical vibration is stabilized by the present system. After stopping mist, the cooling effect of
adiabatically expanded air stops and thus the depth of cut increases due to thermal growth of the tool and the workpiece.

11.16.3.2.1.2 Two-DOF vibrator utilizing longitudinal and bending modes and its controller (10)
Figure 36 shows a schematic illustration of an ultrasonic elliptical vibrator, whose resonant frequency is approximately 39 kHz.
Some PZT actuators are sandwiched with the metal blocks, and it is resonated in the second resonant mode of longitudinal
vibration and the fifth resonant mode of bending vibration by applying exciting voltages to the actuators. A diamond tool tip is set
at the end of the vibrator and vibrated elliptically by combining the two resonant vibrations with some phase shift.
Automatic tracking of the resonant frequency is also implemented by applying feedback control in this system. The vibration
locus is, therefore, controlled in an ultraprecision manner. Figure 37 shows the vibrator and its control system developed in
a cooperative research between Taga Electric Co., Ltd. and the authors. Figure 38 shows the example of the vibration locus at the
cutting edge measured by laser Doppler vibrometers.
The vibrator design of the bending–bending vibrator is not so difficult because of its axial symmetry. On the contrary, a design of
the longitudinal-bending vibrator, which can generate longitudinal mode and the bending mode at the same resonant frequency, is
relatively difficult. In order to attain large elliptical vibration, the vibrator geometry needs to be designed to make a good agreement
not only in resonant frequencies but also in node positions for mechanical supports.
However, the bending–bending vibrator can generate the elliptical vibration in the plane, which is almost perpendicular to the
axial direction of the vibrator. In fact, this vibration plane inclines slightly because the cutting edge is not on the vibrator axis and the
vibrator geometry is not completely symmetry. Because of this inclined vibration plane, machinable workpiece geometry is

Figure 37 Two-DOF elliptical vibration control system utilizing longitudinal and bending modes.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 429

In thrust directin m
1

0 Vibrating direction

–1

–2

–3
–3 –2 –1 0 1 2 3
In cutting direction m

Figure 38 Measured vibration locus.

extremely restricted due to interference between the vibrator and the workpiece (see Figures 33, 39, and 45). On the other hand, the
longitudinal-bending vibrator can generate the elliptical vibration in the plane, which is parallel to the vibrator axis. This is
considerably advantageous to expand the extent of the machinable workpiece geometry.

11.16.3.2.2 Ultrasonic Elliptical Vibration Cutting of Die Steel

11.16.3.2.2.1 Turning applications (9)
Figure 39 shows a setup for the turning experiments. The elliptical vibration tool is fixed on the x-axis feed table of the ultraprecision
lathe, ULC100A made by Toshiba Machine Co., Ltd. During the cutting experiments, the cutting force is measured with the
dynamometer installed between the tool and the table, and the vibration is monitored with the optical sensors.
First, surface roughness in the elliptical vibration cutting of hardened steel is investigated by turning cylindrical surfaces.
Figure 40 summarizes the surface roughness obtained at various feed rates. The workpiece is hardened die steel (JIS: SUS420J2),
whose hardness is HRC39. As the feed rate is reduced, the measured maximum roughness Rz deviates from the theoretical one and
converges to about 0.025 mm. Therefore, the feed rate of 10 mm per rev is preferable for good surface quality and high productivity
under the present conditions.
The most crucial problem in diamond cutting of steel is wear of the diamond tool. Thus, the tool life in elliptical vibration
cutting is examined in comparison with conventional vibration cutting, where linear vibration is applied mainly in the nominal
cutting direction (13). Figure 41 summarizes the cutting forces and the surface roughness measured at various cutting distances.
The cutting forces and the surface roughness increase as the cutting distance increases. The cutting forces, especially the thrust force,
measured in the elliptical vibration cutting are much lower than those in the conventional vibration cutting throughout the cutting
experiments. Optical quality surfaces with maximum roughness of less than 0.05 mm Rz can be obtained up to a cutting distance of
2250 m in the elliptical vibration cutting, while the surface roughness obtained by the conventional vibration cutting increases
rapidly at a cutting distance of about 300 m. The reason is considered to be chipping of the diamond cutting edge. Figure 42 shows
the cutting edges after the above experiments. The tool used in the elliptical vibration cutting is worn without chipping, while the

Optical sensors
Elliptical vibration tool
Spindle
Cutting edge

Z-axis feed table

Workpiece

X-axis feed table

Dynamometer Rotary table

Figure 39 Ultraprecision cutting machine equipped with bending–bending vibrator.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
430 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

Surface roughness m Ry
0.1
Measured
Theoretical

0.05

0 10 20 30
Feed rate m per rev

Figure 40 Surface roughness obtained at various feed rates by elliptical vibration cutting. (Conditions) workpiece: hardened die steel (JIS: SUS420J2,
HRC39), cut area: 100 mm2, cutting speed: 3.4 m min1, diamond tool: R 1 mm, rake angle: 0 , relief angle: 15 , vibration: circular, amplitude:
4.25 mm0p, frequency: 21.5 kHz, vibration speed: 34.4 m min1.

3 0.15

Principal

Surface roughnes μm Ry
Thrust Conventional
Cutting forces N

Feed
2 0.1 Conventional vibration cutting
Principal Elliptical vibration cutting
Thrust Elliptical
Feed
1 0.05

0 1000 2000 3000 0 1000 2000 3000

(a) Cutting distance m (b) Cutting distance m

Figure 41 Life of diamond tools. (a) Cutting forces vs. cutting distances (b) surface roughness vs. cutting distances. (Conditions) workpiece: hardened
die steel (JIS: SUS420J2, HRC39), feed rate: 10 mm per rev, depth of cut: 10 mm, cutting speed: 2.5 m min1, diamond tool: R 1 mm, rake angle: 0 , relief
angle: 15 , vibration: circular or linear, amplitude: 3.5 mm0p, frequency: 20.4 kHz.

Figure 42 SEM photographs of cutting edges of worn-out tools. (a) After elliptical vibration cutting of 2800 m; (b) after conventional vibration cutting of
1000 m.

one used in the conventional vibration cutting is chipped. The conventional vibration cutting requires careful adjustment of the
vibration locus; i.e., the linear vibration should be inclined slightly from the nominal cutting direction toward the depth of cut
direction so that the clearance face does not interfere with the cut surface during the return motion of the tool. Moreover, if the
vibration locus is not exactly linear, i.e., slightly elliptical, the vibration should be clockwise in the case shown in Figure 12. In the
present experiment, the vibration was set with a right inclination, but it might be slightly elliptical and counterclockwise. This is
considered to cause the chipping shown in Figure 42(b). Although the conventional vibration cutting can also realize ultrapre-
cision turning of hardened steel (13), the same or better surface finish can be obtained by the elliptical vibration cutting without
such careful adjustment.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 431

Figure 43 Spherical mirror surface turned by elliptical vibration cutting. (a) Photograph of spherical mirror surface, (b) microphotograph of finished
surface taken with stereoscopic SEM at a radius of 23 mm. (Conditions) workpiece: hardened die steel (JIS: SUS440C, HRC61, f60 mm, R 50 mm),
diamond tool: R 1 mm, rake angle: 0 , relief angle: 15 , reed rate: 13.3 mm per rev, depth of cut: 1–10 mm (estimated), rotational speed: 15 min1, cutting
distance: 320 m, vibration: circular, amplitude: 3.5 mm0p, frequency: 20.4 kHz.

Figure 44 Microcylindrical part turned by elliptical vibration cutting. (Conditions) workpiece: hardened die steel (JIS: SUS420J2, HRC44, f0.22 
4.0 mm), diamond tool: R 50 mm, rake angle: 0 , relief angle: 15 , feed rate: 3 mm per rev, depth of cut: 3 mm, rotational speed: 380 min1, vibration:
circular, amplitude: 2.5 mm0p, frequency: 21.8 kHz.

Based on the successful result, the elliptical vibration cutting is applied to ultraprecision turning of a spherical die. Figure 43
shows the finished surface of the hardened die steel, JIS: SUS440C, with a hardness of HRC61. Figure 43(b) shows a micropho-
tograph of the mirror surface taken with a stereoscopic SEM, which emphasizes the roughness. There are clear vibration marks due to
the elliptical vibration, whose pitch and peak to valley amplitude are 1.75 mm and 0.02 mm, respectively. Feed marks are also visible
whose pitch corresponds to the feed rate of 13.3 mm.
In the elliptical vibration cutting, the cutting force can be reduced and generation of burr can also be restrained significantly.
Both of these effects are expected to be advantageous for precision machining of microparts, because the microstructures are
deformed easily by the cutting force and the relative size of the burr becomes large in the micromachining. An example of the
ultraprecision microcutting is shown in Figure 44. In the turning of the microcylindrical part, a high aspect ratio of about 18 is
realized because of the extremely low cutting force.

11.16.3.2.2.2 Planing applications (36)

The ultraprecision diamond cutting experiments were carried out with the developed elliptical vibration cutting system. The
developed elliptical vibration tool is mounted on an ultraprecision planing machine, NIC-300 made by Nagase Integrex Co., Ltd.,
which consists of three feed tables with double hydrostatic oil guideways in XYZ axes, two rotary index tables in BC axes, and a five-
axis control system. Figure 45 shows a setup for the ultraprecision elliptical vibration planing experiments, where the X-axis is
utilized for cutting feed. The elliptical vibration tool is attached to the upper index table and the workpiece is fixed to the lower index
table of the ultraprecision planing machine.
For comparison, the ordinary cutting without vibration was applied first to planing of hardened steel, JIS: SUS420J2, with
hardness of HRC42. The cutting speed was set to 1 m min1, the feed, 5 mm, and the depth of cut, 3.5 mm.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
432 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

Tool attachment
Mist nozzle Cutting feed direction
(X-axis)

Ultrasonic elliptical vibration tool

Workpiece

Cutting edge

C-axis index table

Figure 45 Setup for ultraprecision planing experiment with elliptical vibration cutting system.

Figure 46 shows the surface finished by the ordinary cutting. Figure 47(a) shows the surface profile measured in the feed
direction, which indicates that the diamond cutting edge was worn out after cutting of 18 m though the total cutting distance was
70 m; i.e., the workpiece material was removed only at the beginning and rubbed during the rest of cutting period. Figure 47(b)
shows the surface profiles at a cutting distance of 1 m in the cutting direction and the feed direction. They show that the roughness
is more than 0.52 mm Rz even at the beginning of cutting. It is confirmed that the hardened die steel cannot be finished by the
ordinary diamond cutting in practice due to the excessive tool wear (1–6). SEM photographs of the cutting edge before and after
the cutting experiment are shown in Figure 48. The cutting edge was worn out completely by 3.5 mm, which corresponds to the
depth of cut.
The elliptical vibration cutting system using the bending–bending vibrator was applied to ultraprecision planing of a relatively
large flat surface of hardened die steel, whose hardness was HRC53. The cutting speed was set to be 0.7 m min1, the feed, 10 mm,
and the depth of cut, 5 mm, in this experiment.
An optical quality mirror surface finish was attained over the whole area of the large finished surface with the developed elliptical
vibration cutting system, as shown in Figure 49. The measured surface roughness is less than 0.04 mm Rz even at a cutting distance of
1065 m, as shown in Figure 50. The optical quality surface was obtained up to the long cutting distance, even though the diamond
cutting edge was worn to some extent as shown in Figure 51. The chippings observed in the photograph occurred in another

Feed direction

Cutting direction

Cut area
20mm

Figure 46 Photograph of surface finished by ordinary cutting. (Conditions) workpiece: hardened die steel (JIS: SUS420J2, HRC42), cutting speed:
1.0 m min1, depth of cut: 3.5 mm (set value), pick feed: 5 mm, diamond tool: R 1 mm, rake angle: 0 .

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 433

(a)

End of cutting Begining of cutting

0.4 m
Cutting distance: 18m
0.2 m Cutting edge is worn out.

Cutting distance: 70m

Cutting distance: 1m

(b)
0.4 m

0.2 m
20 m

20 m

Figure 47 Profiles of surface finished by ordinary cutting. (a) Measured over whole cut area in feed direction. (b) Measured at cutting distance of 1 m in
cutting direction (left) in feed direction (right).

Figure 48 SEM photographs of cutting edge before and after ordinary cutting experiment. (a) Before cutting and (b) after cutting.

122mm

91mm

Figure 49 Photograph of large hardened-steel mirror surface finished by elliptical vibration cutting system. (Conditions) workpiece: hardened die steel
(JIS: SUS420J2, HRC53), cutting speed: 0.7 m min1, depth of cut: 5 mm, pick feed: 10 mm, diamond tool: R 1 mm, rake angle: 0 , vibration: circular,
amplitude: 3 mm0p, frequency: 19.6 kHz.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
434 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

Figure 50 Profiles of surface finished by elliptical vibration cutting at cutting distance of 1065 m (left: in cutting direction, right: in pick feed direction).

Figure 51 Microphotograph of cutting edge after elliptical vibration cutting of 1110 m.

experiment, and they are not related with the present cutting experiment. It is clarified that the tool wear is restrained drastically by
the elliptical vibration cutting, and that the diamond cutting edge can be used over the long cutting distance of 1110 m, which
corresponds to the large area of 122  91 mm2 in this case. The practical ultraprecision machining of a large area of hardened steel
with a single crystal diamond tool, which is practically impossible for the ordinary cutting, can be attained by applying the
ultrasonic elliptical vibration cutting system.
Ultraprecision micromachining of hardened dies for a front light panel of LCD was also conducted with the developed elliptical
vibration cutting system. Figure 52 shows the finished mirror surface with fine microgrooves. It is made of hardened die steel, whose
hardness is HRC53. Figure 53 shows a microphotograph of the microgrooves formed over the large workpiece surface. Figure 54
shows the profiles of the finished surface measured in the feed direction and in the cutting direction. As shown in the figures, the
surface roughness is less than 0.04 mm Rz, and the ultraprecision microgrooving of hardened die steel was successfully attained by
the developed ultrasonic elliptical vibration cutting system.
An example of the ultraprecision microgrooving by using the longitudinal-bending vibrator is shown in Figure 55. The accurate
trapezoidal groove with a width of 20 mm, whose shape is transformed from that of the cutting edge, is obtained without any shape
deteriorations.

Figure 52 Photograph of large hardened-steel mirror surface with microgrooves, machined as die for front light panel of LCD.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 435

Figure 53 SEM photograph of microgrooves.

0.1 m
2 m

10 m
200 m

Figure 54 Profiles of surface grooved by the developed elliptical vibration cutting system (left: in feed direction, right: in cutting direction).

Figure 55 Microphotograph of microtrapezoidal grooves machined by elliptical vibration cutting. (Conditions) workpiece: hardened die steel (JIS:
SUS420J2, HRC52), diamond tool: trapezoidal with a width of 20 mm, rake angle: 0 , relief angle: 11 , depth of cut: 1 mm  30 times, pitch: 40 mm, cutting
speed: 0.5 m min1, vibration: circular, amplitude: 2 mm0p, frequency: 38.7 kHz.

11.16.3.2.3 Ultrasonic Elliptical Vibration Cutting of Brittle Materials

Many brittle materials such as amorphous materials, single crystal materials, and sintered alloys and ceramics are utilized for
a variety of new products because of their unique physical and mechanical properties. Ultraprecision micromachining of those
brittle materials is sometimes required, which realizes sophisticated optical parts, functional components, and their dies for mass
production. Mechanical micromachining, i.e., cutting, grinding, and polishing, can be employed for micromachining as well as
electrical discharge machining and chemical etching. However, it is difficult to obtain optical quality mirror surfaces of those brittle
materials by the conventional ultraprecision diamond cutting because of brittle fractures (37). Electrical discharge machining and
chemical etching cannot be applied to machine optical quality mirror surfaces, and ultraprecision grinding and polishing are not
suitable to machine microstructures with sharp edges and complicated 3D shapes.
The authors have conducted ultraprecision machining experiments of several brittle materials by applying the elliptical vibration
cutting technology, and have clarified that ductile cutting of brittle materials, i.e., sintered tungsten carbide (38), tungsten alloy (10),
zirconia ceramics, single-crystal calcium fluoride, and glass (39), can be attained by applying this method.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
436 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

11.16.3.2.3.1 Application to sintered tungsten carbide (38)

Figure 56 shows a setup for the ultraprecision microgrooving experiments of the sintered tungsten carbide, whose grain size is
0.5 mm and cobalt content is 12.0 wt.%. The ultraprecision planing machine, NIC-300 made by Nagase Integrex Co., Ltd., is utilized.
The elliptical vibration tool is attached to the upper index table and the workpiece is fixed to the dynamometer on the lower index
table of the ultraprecision planing machine.
Ultraprecision microgrooving experiments with single-crystal diamond tools, whose cutting edges are formed to a circular
shape with a nose radius of 1 mm, were carried out to clarify the basic effects of the elliptical vibration on ductile machining of
the sintered tungsten carbide. It is known that there is a critical depth of cut at which the cutting process transits from ductile
mode to brittle mode when brittle materials are machined. Thus, the cutting feed was given by the X-axis table in synchroni-
zation with the Z-axis table, as shown in Figure 56, so that the depth of cut was gradually increased in cutting and the effect of
the depth of cut D on the ductile to brittle transition can be investigated. A negative rake angle of 20 was chosen in order to
extend a tool life in this experiment. The influence of the cutting speed Cs and the vibration amplitudes in the cutting direction
and the vertical direction Ac and Av was also investigated. The vibration frequency of the elliptical vibration cutting was set to be
19.8 kHz.
Figure 57 shows microphotographs of the grooves of the sintered tungsten carbide. Close-ups of the grooves are also shown in
the figure, which are taken at a depth of cut of 1.9 mm. The specific conditions of this experiment are shown below the figure. The
groove formed by the ordinary cutting is filled with asperities due to microfractures, while a smooth mirror surface is obtained by
the elliptical vibration cutting.
Figure 58 shows groove surfaces taken using a SEM. Many fractures appear to occur vertically to the cutting direction on the
surface finished by the ordinary cutting, while there are not such fractures on the surface obtained by the elliptical vibration cutting.
The ductile to brittle transition is not clearly observed in the both cutting methods for the sintered tungsten carbide.
Figure 59 shows the measured specific cutting forces. At a depth of cut of 1 mm, the force in the cutting direction in the elliptical
vibration cutting is 17 times lower than that in the ordinary cutting, while in the thrust direction it is 14 times lower. This reduction
of the specific cutting forces and the improvement of the surface quality can be explained as follows. Thickness of the workpiece
material cut in each cycle of the elliptical vibration is significantly small as compared to the nominal depth of cut, especially when
the cutting edge passes the bottom point and the finished surface is just generated as shown in Figure 60. Therefore, ductile
machining at large nominal depth of cut, where the ordinary cutting cannot be performed in the ductile mode, can be realized by
applying the elliptical vibration cutting.
Figure 61 shows microphotographs of surfaces finished at various vibration amplitudes and cutting speeds and a constant depth
of cut of 1 mm by the elliptical vibration cutting. Figure 61(a) and 61(b) indicate that the microfractures are generated as the vertical
amplitude is large in comparison with grain size of the tungsten carbide. These microfractures are considered to be caused by the
elliptical motion of the tool, which tears a mass of grains while pulling up the chip. Figure 61(b) and 61(c) show that higher quality
surfaces are obtained at lower cutting speeds. It is considered that this improvement of the surface quality is caused by decrease in
the geometric or theoretical surface roughness, which is patterned by the elliptical motion of the tool.
Ultraprecision planing experiments were also carried out to estimate practical performance of the elliptical vibration cutting. The
cutting speed was set to be 300 mm min1, the depth of cut was changed from 0 to 2 mm, and the amplitudes in the cutting/vertical
direction were set to 2/1 mmpp in the elliptical vibration cutting. Microphotographs of the finished surfaces are shown in Figure 62.
Numerous asperities appear over the surface finished by the ordinary cutting, which are similar to those observed in the grooving
experiment, see Figure 57(a). On the other hand, the surface roughness of less than 40 nm Rz is obtained at a cutting distance of
0.05 m by the elliptical vibration cutting. However, clear feed marks along the cutting direction appear and grow gradually with the

Figure 56 Setup for ultraprecision microgrooving experiments of sintered tungsten carbide.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 437

Figure 57 Microphotographs of the grooves of sintered tungsten carbide formed by both cutting methods (cutting speed Cs: 150 mm min1, rake angle
a: 20 ). (a) Ordinary cutting and (b) elliptical vibration cutting (Ac/Av: 1/0.5 mm0p).

(a) (b)

Figure 58 SEM images of machined surfaces of sintered tungsten carbide (rake angle a: 20 ). (a) Ordinary cutting and (b) elliptical vibration cutting
(Cs: 600 mm min1, D: 1.4 mm) (Ac/Av: 0.5/0.5 mm0p, Cs: 150 mm min1, D: 1.1 mm).

increase of the cutting distance. Figure 63 shows the change in the specific cutting force with the increase of the cutting distance. At
a cutting distance of 0.02 m, the force in the thrust direction in the elliptical vibration cutting is 22 times lower than that in the
ordinary cutting, whereas it becomes 1.6 times higher at a cutting distance of 1.5 m. SEM photographs of the cutting edges after the
planing experiments are shown in Figure 64. The tool wear is larger in the elliptical vibration cutting than in the ordinary cutting. It
is confirmed that a good quality flat surface with a roughness of less than 40 nm Rz can be obtained by the elliptical vibration
planing, though the tool life is limited.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
438 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

Figure 59 Measured specific cutting forces (Cs: 150 mm min1, a: 20 ). (a) Ordinary cutting and (b) elliptical vibration cutting (Ac/Av: 1/0.5).

Figure 60 Schematic illustration of motion of cutting edge and thickness of workpiece material cut in each cycle of elliptical vibration.

Figure 61 Measured specific cutting forces (D: 1 mm, a: 20 ). (a) Ac/Av: 2/2, Cs: 600; (b) Ac/Av: 1/0.5, Cs: 600; and (c) Ac/Av: 1/0.5, Cs: 300.

Figure 62 Microphotographs of surfaces finished at cutting speed of 300 mm min1. (a) Ordinary cutting and (b) elliptical vibration cutting.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 439

Figure 63 Specific cutting force.

Figure 64 Microphotographs of cutting edges after planing experiments. (a) Ordinary cutting and (b) elliptical vibration cutting.

In summary, these experimental results indicates a feasibility of micromachining of the sintered tungsten carbide for a small area
by applying the elliptical vibration cutting. According to the author’s experience, appropriate vibration/cutting conditions signif-
icantly vary depending on the material properties, such as a grain size and a binder material. These properties significantly affect the
surface quality and the tool life.

11.16.3.2.3.2 Application to tungsten alloy (10)

Tungsten alloy is expected to be used as a novel mold material for glass parts, because it has partially better thermal properties than
the sintered tungsten carbide and the CVD-silicon carbide as shown in Table 1. Furthermore, the tungsten alloy is machinable with
ordinary carbide tools, and it can be used for the glass molding without the coatings because of its superior thermochemical
stability. However, this material is also difficult to finish by the conventional ultraprecision machining methods. For example,
ordinary ultraprecision diamond cutting is not applicable because of adhesion of the tungsten material to the tool in addition to the
rapid tool wear and the brittle fracture. Therefore, the grooving and planing experiments with single-crystal diamond tools, whose
cutting edges were formed to a circular shape with a nose radius of 1 mm, were carried out to clarify basic effects of the elliptical
vibration on machining of the tungsten alloy. Grain size of the tungsten alloy is about 7 mm. The vibration locus was set to be
circular with a radius of 2 mm for the elliptical vibration cutting.
The depth of cut was increased gradually in the grooving experiments in order to observe the ductile to brittle transition.
Figure 65 shows microphotographs of the grooves machined at cutting speeds of 1.44 and 0.36 m min1 in the ordinary cutting and
the elliptical vibration cutting, respectively. Figure 65(a) shows the groove formed at a depth of cut of 0.4 mm by the ordinary

Table 1 Mechanical properties of mold materials (RT: Room temperature)

Material Tungsten alloy Sintered WC CVD-SiC

Charpy impact value kg$m cm2 RT 0.4 0.07 –

Deflective strength MPa RT 1176 1320 605
Young’s modulus GPa RT 450 570 406
Hardness HV RT 380 1350 24000
Coefficient of thermal conductivity W m$K RT 89.2 63.8 130
973 K 92.1 – 68
Coefficient of thermal expansion ppm K1 RTw673 K 4.8 4.6 4.5

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
440 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

Figure 65 Microgrooves formed on tungsten alloy. (a) Ordinary cutting and (b) elliptical vibration cutting.

cutting. It is filled with scratches due to the adhesion and brittle fractures. By contrast, a smooth surface was obtained by elliptical
vibration cutting. Figure 65(b) shows the first small brittle fracture. The depth of cut measured at this first fracture, i.e., critical depth
of cut, is 0.8 mm.
Figure 66 shows SEM images of the cutting edges used for the planing experiments. They were taken after ordinary cutting of
1.08 m and elliptical vibration cutting of 1.35 m, respectively. Both cutting experiments were conducted at a feed rate of 10 mm and
a cutting speed of 0.36 m min1. There is considerable rake and flank wear and adhesion of the workpiece material in ordinary
cutting. On the other hand, neither tool wear nor adhesion can be observed in elliptical vibration cutting.
The result shows that the ductile cutting of the tungsten alloy is attained at a depth of cut of less than 0.8 mm, and that the rapid
tool wear and the adhesion are restrained by the elliptical vibration cutting.
Ultraprecision molds of the tungsten alloy for a prism, a small spherical lens, and an array of micro V-grooves were machined by
the elliptical vibration cutting. Figure 67 shows the molds machined by elliptical vibration cutting. All molds were machined
successfully, and desirable smooth mirror surfaces were obtained. Figure 68 shows the relationship between the cutting distance
and the surface roughness of the prism mold. The surface roughness was measured by a stylus surface profiler. As shown in the
figures, higher quality surfaces of maximum roughness of less than 50 nm Rz were attained at shorter cutting distances, though feed

Figure 66 Cutting edges used for planing experiments. (a) Ordinary cutting and (b) elliptical vibration cutting.

Figure 67 Photograph of molds machined by elliptical vibration cutting. (a) Prism, (b) spherical lens and (c) V-groove array.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 441

Figure 68 Relationship between cutting distance and surface roughness.

Figure 69 Microphotograph of spherical lens mold.

Figure 70 Mold and tool for V-groove array. (a) SEM image of machined mold surface and (b) microphotograph of cutting edge after machining.

marks gradually became greater with tool wear. The maximum surface roughness was kept less than 100 nm Rz over the whole
cutting distance of about 50 m.
Figure 69 shows a microphotograph of the spherical lens mold, which was taken around its center. Its surface roughness measured
by an optical profiling system is less than 100 nm Rz. Figure 70 shows an SEM image of the V-grooved mold surface, whose valley
angle and depth are 75 and 60 mm, respectively, and a microphotograph of the cutting edge used for this machining. Its cumulative
cutting distance is 20 m. It shows that the micro V-grooves with sharp edges can be obtained with no considerable tool wear.
The ultraprecision molds were also applied successfully to the molding of borosilicate glass BK7 without any coatings at
a molding temperature of 690  C. Molded glass parts and an SEM image of a cross-section of the grooved part are shown in
Figures 71 and 72, respectively. Figure 72 especially demonstrates a possibility that the ultraprecision glass parts with fine
microgrooves can be mass-produced efficiently by the proposed process, i.e., molding with tungsten alloy molds finished by the
ultrasonic elliptical vibration cutting.

11.16.3.2.3.3 Application to zirconia ceramics (39)

Figure 73 shows surfaces of zirconia ceramics, which is a sintered material, grooved by the ordinary cutting and the elliptical
vibration cutting. It is confirmed that a smooth surface without brittle fractures can be obtained by applying the elliptical vibration

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
442 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

Figure 71 Photograph of optical glass parts formed by molds machined by elliptical vibration cutting.

Figure 72 SEM image of optical glass part formed by V-groove array mold.

Figure 73 Microphotographs of grooves of zirconia ceramics (Cs: 450 mm min1, a: 0 ). (a) Ordinary cutting (D: 4 mm) and (b) elliptical vibration
cutting (D: 0.5 and 10 mm, Ac/Av: 3/3).

cutting. On the other hand, the surface obtained by the ordinary cutting is filled with brittle fractures, which resembles that of the
sintered tungsten carbide. The ductile to brittle transition cannot be observed for this material in the both cutting methods, either. It
is considered that these similar characteristics between the zirconia ceramics and the sintered tungsten carbide are caused by their
similar material structures, since both are sintered materials.

11.16.3.2.3.4 Application to single-crystal calcium fluoride (39)

The grooving experiment was also carried out for the calcium fluoride, which is a single-crystal material. Figure 74 shows surfaces of
the calcium fluoride formed by the both cutting methods in h1 2 1i and h1 2 1i directions on a (111) plane. The grooves formed by
the ordinary cutting are filled with cleavage fractures, whose appearance depends on the crystal orientation. The critical depth of cut

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 443

Figure 74 Microphotographs of microgrooves machined on (111) surface of calcium fluoride (Cs: 370 mm min1, D: 2.0 mm, a: 0 , Ac/Av: 2/2 mm0p).
(a) Ordinary cutting in h1 2 1i direction, (b) elliptical vibration cutting in h1 2 1i direction, (c) ordinary cutting in h1 2 1i direction, and (d) elliptical
vibration cutting in h1 2 1i direction.

for the ductile cutting is less than 0.3 mm, depending on the crystal orientations. On the other hand, smooth surfaces without
remarkable cleavage fractures are obtained by the elliptical vibration cutting, especially in the h1 2 1i direction, as shown in
Figure 74(d). It was confirmed that the critical depth of cut for the single crystal material is increased significantly by the elliptical
vibration cutting, though the ductile to brittle transition is not clear. These results indicate that smooth surfaces without fractures
can be obtained for single-crystal materials at practically large depth of cut by applying the elliptical vibration cutting, while in both
cutting methods its machinability strongly depends on the crystal orientation of the workpiece.

11.16.3.3 Free-Form Surface Machining by Ultrasonic Elliptical Vibration Cutting

11.16.3.3.1 Machining Method and Devices for Free-Form Machining
Sculptured surfaces of hardened die steel are generally machined by ball end milling as shown in Figure 75(a). The rough
surfaces are subsequently polished when mirror surfaces are required. The polishing, however, is not only a costly and time-
consuming process but also deteriorates machining accuracy in general, and it is not suitable to fabricate microstructures. On
the other hand, the elliptical vibration cutting technology can attain mirror surface machining of difficult-to-cut materials,
including hardened die steel, with single-crystal diamond tools. The ultrasonic elliptical vibration cutting is, therefore, directly
applied to sculpturing in the proposed method in order to eliminate the polishing process. Figure 75(a) shows the proposed
machining method. The tool is vibrated elliptically, unlike conventional rotating end mills, and is fed along the sculptured
surface while the rotational position of the tool is precisely controlled in accordance with the sculptured surface orientation.
Thus, the proposed method needs a precision machine tool with at least 4 axes, i.e., X, Y, Z, and C, to machine the sculptured
surfaces.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
444 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

Figure 75 Comparison between end milling and proposed machining for sculptured surfaces. (a) Ball end milling process and (b) proposed method.

Both cutting processes are intermittent, but the frequency is much higher and the radius is much smaller in the elliptical
vibration cutting, as compared in Figure 75. These differences enable the mirror surface machining with diamond tools. The
vibrated tool can also be rotated like the end mills. However, such vibration milling is considered to be redundant, since the
vibration cutting is already an intermittent process itself. The present method is more advantageous for surface roughness, tool life,
and efficiency, because the trochoidal tool trajectory in the conventional milling process increases surface roughness and unnec-
essary air cut.
In order to apply the elliptical vibration cutting to proposed free-form machining, an ultrasonic elliptical vibration tool with
a three-DOF was developed. The conventional bending–bending vibrators generate the elliptical vibration in fixed planes, which are
roughly perpendicular to the vibrator axes (9,35,40). As this vibration direction causes significant restriction in machinable
geometry, a three DOF elliptical vibration tool, which can generate an arbitrary ultrasonic elliptical vibration in the 3D space, is
desired so that it can machine a wide range of sculptured surfaces.
In order to generate an arbitrary elliptical vibration in the 3D space, a longitudinal vibration is further combined to the two-
directional bending vibrations, as illustrated in Figure 76. The designed tool utilizes the fourth or fifth resonant mode of
bending vibrations in the U and V directions and the second resonant mode of longitudinal vibration in the Z direction.

Figure 76 Designed three-DOF ultrasonic vibrator and its mode shapes.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 445

Frequencies and nodal positions of those two resonant modes are generally different, and thus the vibrator shape should be
designed so that their frequencies and nodal positions are close to each other at the same time. First, the second resonant mode of
longitudinal vibration was selected so that the vibrator has two nodes and can be supported rigidly at the two nodal positions. Next,
the fifth resonant mode of bending vibration was chosen, because two of the nodal positions are relatively close to the longitudinal
ones. The resonant frequencies and the nodal positions were adjusted by making the stepped and tapered portions, see Figure 76,
and changing their dimensions. This design was assisted by computer simulations. They were roughly adjusted by the Euler-
Bernoulli beam analysis, and then the final shape was determined by utilizing the FEM analysis as shown in Figure 77.
The bending vibrations are excited by the four PZTs shown in Figure 76. The left and right PZTs are expanded and contracted
with a phase shift of 180 to bend the vibrator in the V direction. The front and back ones are used in the same way to bend it in the
U direction. The longitudinal vibration is excited by using the other four PZTs with the same phase. These three-directional
vibrations are detected by the small PZT sensors, and those signals are utilized for cross-talk removal, feedback control of vibration
amplitudes and phase differences, and resonance chase (35). Details of the control system developed for the three-DOF vibration
tool are omitted here.
It is recommended in the ultrasonic elliptical vibration cutting to apply the elliptical vibration in the plane including the cutting
direction and roughly the chip flow direction (8,28,29). Thus, the vibration is applied here in the plane including the cutting
direction, i.e., U direction, and the direction inclined from the vibrator axis as shown by the red arrow in Figure 76, which is roughly
the average chip flow direction in the machined region of angle.
Figure 78 shows the developed three-DOF ultrasonic vibration tool with a single-crystal diamond tool tip. The resonant
frequencies were adjusted to 34.4 kHz, and the developed system can generate an arbitrary locus of elliptical vibration in an
arbitrary plane in the 3D space. The maximum amplitudes were 30 mmpp in the U and V directions and 24 mmpp in the Z
direction, which correspond to 195 and 156 m min1, respectively.

Figure 77 Vibrations simulated by FEM.

U-axis PZT actuators

Tool supports for bending vibration
Z-axis

V-axis

PZT sensors
Ultrasonic
elliptical vibrator
PZT actuators
Diamond tool tip for longitudinal vibration

Figure 78 Photograph of developed three-DOF ultrasonic vibration tool.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
446 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

Figure 79 Photograph of developed precision machine tool with three-DOF ultrasonic vibration tool.

A precision machine tool, see Figure 79, was developed for the proposed machining method based on a commercial
machining center. The developed three-DOF vibration tool was mounted in the C-axis spindle, and it was connected to the
controller and the power amplifiers. Since they were connected by electric wires, the C-axis cannot be rotated infinitely. The tool is
fed along the sculptured surface at each Z level and rotated in accordance with the cut surface orientation in the proposed method,
as shown in Figure 75(b). Therefore, the C-axis was counter-rotated to rewind the wires after each rotation at the present stage of
research.
Maximum surface roughness of less than about 100 nm is generally required for a mirror surface finish. Thus, motion errors of
the machine tool should be less than about 100 nm, including the undesirable dynamic deviations between the tool and the
workpiece due to oil/air pumps, fans, and so on. By employing an air spindle and a coreless DC motor for the C-axis, the rotational
motion error, i.e., the nonrepeatable run out, is restrained to be less than 80 nm. Precision ball screws with small torque ripple and
precision linear roller guides are also employed for the X, Y, and Z axes. It was confirmed that the smaller pitch components of the
motion errors in linear drive systems, which affect the mirror surface quality, are much less than 100 nm.
In order to attain the proposed machining, the tool path generation for the four-axis controlled sculpturing also needs to be
considered. In this experimental verification, commercial CAM software is utilized, which outputs cutter location data and normal
vectors of the cut surface for the contour machining and the flowline machining. A special postprocessor is developed here to
transform the normal vectors to the rotational angles of the C-axis, so that the rake angle of the tool is kept constant with respect to
tool path direction while the cutting direction changes along the curved path.

11.16.3.3.2 Precision Free-Form Machining of Die Steel

Machining experiments were carried out, and basic performance of the developed system was examined. A simple planing
experiment was conducted first, since the basic performance of machining system appears clearly on a flat surface. Then, a spherical
shape was machined by the proposed machining method utilizing the four-axis control. Hardened die steel with a hardness of
HRC54 or 40 is used as workpiece material, and it is cut with single-crystal diamond tools. Both experiments were conducted only
by the elliptical vibration cutting, since steel materials cannot be machined by the ordinary diamond cutting.
The result of the planing experiment is shown in Figures 80–82. The cut surface was inclined here by 38.7 around the X-axis
in this machining, so that all relative motion of X, Y, and Z axes are involved in the operation. As shown in Figure 80, a mirror
surface was obtained successfully by the present method. The feed marks were barely observed, whose pitch corresponds to the
feed rate of 15 mm. The other marks in the cutting direction appear clearly with a pitch of about 2.5 mm. The elliptical vibration
cutting process causes small geometrical waviness, called scallop, in the cutting direction as shown in Figure 75(b), but its pitch
and height should be theoretically 0.48 mm and 5.3 nm, respectively, under the present conditions. It is considered that these
marks might be caused by the first resonant mode of bending vibration of the vibrator, whose frequency, 6.5 kHz, coincides with
that of the marks. From the measurement of the surface roughness in the cutting and feed directions, the roughness due to the
feed marks and the other marks was about 50 nm and the maximum roughness of about 150 nm Rz was attained without
polishing.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 447

Figure 80 Hardened steel mirror machined by developed system. (Conditions) workpiece: hardened die steel (JIS: SUS420J2, HRC54), diamond tool: R
1 mm, rake angle: 0 , relief angle: 15 , depth of cut: 3 mm, pick feed: 15 mm, cutting speed: 1 m min1, vibration: elliptical, amplitude: 4–4–2 mm (U–V–Z),
frequency: 34.4 kHz.

Figure 81 Profiles of surface finished by developed elliptical vibration cutting system. (a) Measured in cutting direction and (b) measured in feed direction.

Figure 82 Spherical steel mirror finished by proposed machining method. (Conditions) workpiece: hardened die steel (JIS: NAK80, HRC40), diamond
tool: R 1 mm, rake angle: 0 , relief angle: 15 , depth of cut: 3 mm, pick feed: 15 mm, cutting speed: 0–1.27 m min1, vibration: elliptical, amplitude: 3–3–
1.5 mm (U–V–Z), frequency: 34.4 kHz.

Figure 82 shows the spherical steel mirror finished by the proposed method. The result indicates that curved mirror surfaces of
hardened die steel can be obtained by the proposed elliptical vibration cutting with the four-axis control. The maximum roughness
of about 280 nm Rz was also attained for the spherical surface.

11.16.3.4 Micro/Nanotexturing by Ultrasonic Elliptical Vibration Cutting

11.16.3.4.1 Machining Method and System for Micro/Nanotexturing
In order to attain ultraprecision machining by applying elliptical vibration cutting, several ultrasonic elliptical vibration tools and
their control systems were developed in the past studies (10,11,35,36). Since variation of vibration amplitudes causes deterioration

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
448 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

of machining accuracy and surface quality, most research efforts were dedicated to keeping the elliptical vibration locus ultra-
precisely constant. Otherwise, ultraprecision cutting cannot be achieved in practice. On the other hand, the authors proposed to
utilize the variation in vibration amplitude in a positive manner, in contrast with conventional studies. It is considered a unique
function, i.e., the depth of cut can be actively controlled by controlling the vibration amplitude in the depth of cut direction while
machining. By utilizing this function to serve as a sort of fast tool servo (FTS), the ultraprecision sculpturing of difficult-to-cut
materials in micro/nano scale is achieved efficiently. FTS technology has been developed to manufacture precision microstruc-
tures efficiently, and is used in a variety of applications, especially in the modern optics industry (41). Although conventional FTS
technology is incapable of ultraprecision machining of difficult-to-cut materials due to rapid tool wear and defects on the finished
surface, elliptical vibration cutting is applicable directly to ultraprecision machining of those materials. Note that it is redundant and
disadvantageous to combine the elliptical vibration tool with the conventional FTS, since both devices have actuators and the
vibration tool is too heavy to be actuated at high frequency by the FTS.
Figure 83 shows the proposed machining with depth of cut control in elliptical vibration cutting. The tool is fed at a nominal
cutting speed and vibrated elliptically at the same time. Because of this intermittent process at an ultrasonic frequency, tool wear and
adhesion are restricted, and the ultraprecision cutting of hardened steel can be attained with single-crystal diamond tools. The
vibration amplitude in the depth of cut direction is controlled simultaneously in the proposed machining process. The trajectory of
the cutting edge, then, changes dynamically, and its envelope is transferred to the finished surface. By controlling the amplitude
ultraprecisely at high speed, the ultraprecision sculpturing of the difficult-to-cut materials can be achieved efficiently without using
conventional FTS technology. In other words, the elliptical vibration cutting technology is equipped with an FTS function by itself.
Although amplitude control command is not identical with the envelope of the cutting edge trajectory, as shown in Figure 83, their
difference is not crucial to the present study. The difference is insignificant in practice when the slope is not steep. The depth of cut
can be controlled within half of the maximum amplitude in the depth of cut direction, and available frequency range of the
amplitude control is limited to that which is relatively lower than the elliptical vibration frequency. Therefore, performance in the
role as FTS strongly depends on the specifications of the vibrator.
For verification of the proposed method, a vibration control system of a two-DOF elliptical vibration tool was developed.
Figure 84 shows the developed system of the high-speed amplitude control of elliptical vibration at a frequency of about 36 kHz.

Figure 83 Machining by controlling amplitude in elliptical vibration cutting.

Figure 84 Vibration control system for elliptical vibration cutting.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 449

A two-DOF elliptical vibration tool (10), which was designed to generate arbitrary elliptical vibration, is utilized in the present
study. The vibrator is actuated by using some PZT actuators, which are sandwiched with metal cylindrical parts, namely a BLT. As the
vibrator is designed to have the same resonant frequencies in the second resonant mode of longitudinal vibration and the fifth
resonant mode of bending vibration, it can generate large longitudinal and bending vibrations simultaneously at the same ultra-
sonic frequency by applying exciting voltages to the actuators. Thus, an arbitrary two-DOF elliptical vibration can be obtained at the
diamond tool tip attached to the vibrator by combining both resonant vibration modes with some phase shift. Longitudinal
vibration direction corresponds to depth of cut direction here.
Gain of the amplifier can be controlled by external input in the developed system, and thus the exciting voltage supplied to the
actuator is changed. The amplitude is, consequently, controlled by the external input. As the maximum amplitude in the depth of
cut direction is 4 mmpp, the vibration amplitude can be controlled to change the depth of cut within 2 mm by this system. Measured
frequency response of amplitude control is shown in Figure 85. When the external input is supplied without passing through the
low pass filter (LPF), dynamic gain is enlarged at a frequency of about 1.5 kHz as compared with static gain. In order to suppress this
high dynamic gain, a second-order analog LPF is placed in the practical system, which is still able to control the vibration amplitude
with a frequency bandwidth of more than 300 Hz. This frequency bandwidth is relatively narrow as compared with that of
conventional FTS. It might not, however, be a big problem because the elliptical vibration cutting technology is available only at
relatively low cutting speed.
In order to attain arbitrary sculpturing, an ultraprecision micro/nanosculpturing system was developed by using the developed
vibration control system and an ultraprecision planing machine. Figure 86 shows the developed sculpturing system, where the
planing machine is simply controlled to machine a plane surface at constant cutting speed. The ultrasonic elliptical vibration tool is
attached on a Z-axis table of the machine tool. The vibration amplitude in the depth of cut direction along the Z-axis is controlled in
synchronization with cutting feed motion in the X-axis, and then, arbitrary micro/nanosculpturing can be attained on a flat top
surface of a steel workpiece.

Figure 85 Frequency response of amplitude control. (a) With low pass filter and (b) without low pass filter.

Figure 86 Developed micro/nanosculpturing system using elliptical vibration cutting.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
450 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

The X-axis coordinate value is constantly monitored in the developed system by directly communicating with a computerized
numerical control system through High Speed Serial Bus interface made by FANUC Ltd., or by using an external optical sensor. In
every cutting feed, an X-axis coordinate is detected after the feed speed becomes constant but before the cutting starts. The dynamic
command signal for vibration amplitude control is subsequently input to the elliptical vibration control system based on the
detected X-axis coordinate, the cutting speed, and the amplitude command table, which is calculated from CAD data of objects.
Thus, the amplitude is controlled to change the depth of cut in accordance with the CAD data in real time. Consequently,
sophisticated structures can be machined on the steel materials efficiently by merely combining simple planing operations at
constant cutting speed with high-speed depth of cut control. An industrial computer is utilized to detect the X-axis coordinate and
generate the voltage signal for controlling vibration amplitude.

11.16.3.4.2 Ultraprecision Micro/Nanotexturing of Die Steel

The developed control system was applied to grooving experiments. Figure 87 shows an experimental setup. The ultrasonic elliptical
vibration tool is mounted on an ultraprecision planing machine, NIC-300 (Nagase Integrex Co., Ltd.). The vibration tool was fed in
the X-axis direction. The vibration amplitude was controlled with a sinusoidal, ramp, or zigzag wave command at the same time,
and then microtextured grooves were formed on the surface of a hardened-steel workpiece. The vibration amplitude was controlled
to change from 2 to 4 mmpp in all wave commands. This corresponds to the depth of cut variation of 1 mm. On the other hand, the
amplitude in the cutting direction is set to be constant, 4 mmpp.
Figure 88 shows microphotographs of grooves machined at the cutting speed of 0.2 m min1 by the single-crystal diamond tool
with a nose radius of 1 mm. It was confirmed that the grooves with various ultraprecision microtextures can be machined successfully
on the hardened steel, and mirror surface quality can be obtained on all grooves. Figures 89 and 90 show surface profiles of the
grooves measured by a laser microscope, VK-9500 (Keyence Corp.), and a surface profiler, ET4000 (Kosaka Laboratory Ltd.). From

Figure 87 Experimental setup used for texturing experiments.

Figure 88 Microphotographs of grooves (cutting speed: 0.2 m min1). (a) Sinusoidal (100 Hz), (b) sinusoidal (30 Hz), (c) ramp (100 Hz), and (d) zigzag
(100 Hz).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 451

Figure 89 Surface profiles measured by laser microscope (cutting speed: 0.2 m min1, frequency: 100 Hz). (a) Sinusoidal, (b) ramp, and (c) zigzag.

Figure 90 Measured surface profiles of grooves (cutting speed: 0.2 m min1).

these results, it was also confirmed that measured surface profiles of sinusoidal grooves agreed precisely with the command waves. On
the other hand, measured corners of ramp and zigzag grooves are slightly rounded, and their step heights are relatively smaller than
the variation width of command waves of 1 mm. This is considered to be caused by cutting off high-frequency components in the
amplitude control commands by the LPF, which the ramp and zigzag waves include at their sharp corners.
The developed machining system was applied to nanosculpturing experiments of picture images. CAD data for sculpturing were
produced from gray scale images, where the gray values of 8 bits (256 gradations) in pixels were converted into the amplitude
commands. As the vibration amplitude at 36.15 kHz in the experiments was changed within a range from 2 to 4 mmpp, the depth of
cut was changed within 1 mm. Total depth of cut is, therefore, controlled within 19–20 mm, as the nominal depth and the initial
amplitude are set to be 20 mm and 4 mmpp, respectively. The resolution of the depth of cut control is about 4 nm. Hardened-steel
workpieces (64  48 mm, JIS: SUS420J2, HRC53) were machined with single-crystal diamond tools with a nose radius of 1 mm,
a rake angle of 0 , and a relief angle of 10 . The size of original images is set to 3200  2400 pixels, and thus, 1 pixel corresponds to
20  20 mm. The pick feed is 20 mm, and the cutting speed is 1 m min1.
Figure 91 shows photographs of letters sculptured on the hardened steel. The whole surface is machined in about 260 min with
the developed sculpturing system. As the image data were binarized here, all letters were machined to be concave with a depth of

Figure 91 Photographs of sculptured letters (left: overview, right: close up).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
452 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

1 mm. As shown in the figure, the letters are sculptured successfully. It was confirmed that the measured step height of sculptured
letters agreed with the desired depth of 1 mm. Any unusual tool wear or chippings were not observed after the experiment.
Figure 92 shows an example of the nanosculpturing of picture images. The depth of cut was controlled in nanoscale in
accordance with the image data. The gray scale picture image, the photograph of the machined surface, and its profile, which was
measured by optical surface profiler (ZYGO NewView 6200), correspond well to each other. The result shows that picture images
can be printed successfully on hardened steel as nanoscale sculptures.
Nanosculpturing experiments involving dimple patterns were also carried out with the developed sculpturing system. Sinusoidal
commands to control the vibration amplitude were input to the elliptical vibration control system during machining hardened steel
(JIS: SUS420J2, HRC53), and the phase of the sinusoidal commands was changed by 180 in every cutting feed, so that precisely
aligned patterns were sculptured. Figure 93 shows a microphotograph and a profile of dimples sculptured. The hexagonal dimple
patterns, whose borders are sharp, can be observed. It was also confirmed that the measured dimple depth of 0.78 mm corresponded
well with an expected dimple depth of 0.75 mm, which is calculated theoretically from the vibration conditions, the cutting
conditions and the tool geometry.
Figure 94 shows microphotographs of dimples sculptured under the different conditions. As the maximum depth of cut of
0.4 mm was considerably smaller than the amplitude variation, circular independent dimple patterns with a concave depth of about
0.4 mm were sculptured successfully on the workpiece surface. On the other hand, hexagonal dimples with smaller pitch were
obtained, where the dimples were sculptured by using a tool with smaller nose radius of 0.2 mm under different cutting and
vibration conditions, as compared with other examples.

Figure 92 Gray scale image for amplitude command, machined surface, and measured surface profile. (a) Gray scale image, (b) microphotograph, and
(c) measured profile.

Figure 93 Microphotograph and measured profile of dimples sculptured. (Conditions) diamond tool: R 1 mm, rake angle: 0 , relief angle: 10 ,
maximum depth of cut: 10 mm, pick feed: 34 mm, cutting speed: 0.8 m min1, frequency: 36.7 kHz, amplitude in depth of cut direction: 2.14–3.86 mm,
command frequency: 100 Hz.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting 453

Figure 94 Microphotographs of dimple sculpturing examples. (Conditions) diamond tool: R 1 mm (left) and R 0.2 mm (right), rake angle: 0 (left) and
34 (right), relief angle: 10 (left) and 45 (right), maximum depth of cut: 0.4 mm (left) and 10 mm (right), pick feed: 150 mm (left) and 17.24 mm (right),
cutting speed: 0.8 m min1 (left) and 0.358 m min1 (right), frequency: 36.7 kHz, amplitude in depth of cut direction: 2.1–3.9 mm, command frequency:
100 Hz.

11.16.4 Summary

The technology of ultrasonic vibration diamond cutting has been developed for many years, and it has been successfully applied to
the ultraprecision cutting of difficult-to-cut materials, including steels and glass materials as described in Section 11.16.2.
However, the linear vibration cutting has the practical difficulty in adjusting the vibrating and cutting directions. It is also
disadvantageous that the cutting direction cannot be changed in machining. Hence, the ultrasonic elliptical vibration cutting
technology has been developed to solve those practical problems and to further improve the cutting performance, as described in
Section 11.16.3. An ultraprecision, ultrasonic, elliptical vibration tool has been commercialized recently, and it is utilized in
industry to finish various precise parts, e.g., molds for optical parts. As the technology is spread to various industrial applications,
many developments and research efforts are still continuing on the elliptical vibration cutting technology, such as mirror surface
cutting of dies and molds with free-form surfaces and high-efficient micromachining by controlling amplitude of the ultrasonic
elliptical vibration. It is expected that the ultrasonic vibration cutting technology summarized in this chapter will contribute to
improvement of machining accuracy and surface quality for various difficult-to-cut materials, and realization of new products that
require difficult machining processes.

References

1. Paul, E.; Evans, C. J.; Mangamelli, A.; Mc Glauflin, M. L. Chemical Aspects of Tool Wear in Single Point Diamond Turning. Precis. Eng. 1996, 18 (1), 4–19.
2. Shimada, S., et al. Thermo-Chemical Wear Mechanism of Diamond Tool in Machining of Ferrous Metals. CIRP Ann. Manuf. Technol. 2004, 53 (1), 57–60.
3. Casstevens, J. M. Diamond Turning of Steel in Carbon-Saturated Atmospheres. Precis. Eng. 1983, 5 (1), 9–15.
4. Masuda, M., et al. Ultra-Precision Cutting of Steel with CBN Tools (in Japanese). J. JSPE 1986, 54 (2), 384–389.
5. Evans, C. Cryogenic Diamond Turning of Stainless Steel. Ann. CIRP 1991, 40 (1), 571–575.
6. Brinksmeier, E.; Gläbe, R.; Osmer, J. Ultra-Precision Diamond Cutting of Steel Molds. Ann. CIRP 2006, 55 (1), 551–554.
7. Kumabe, J. Vibration Cutting (in Japanese); Jikkyou Publishing Co.: Tokyo, 1979.
8. Shamoto, E.; Moriwaki, T. Study on Elliptical Vibration Cutting. Ann. CIRP 1994, 43 (1), 35–38.
9. Shamoto, E.; Moriwaki, T. Ultraprecision Diamond Cutting of Hardened Steel by Applying Elliptical Vibration Cutting. Ann. CIRP 1999, 48 (1), 441–444.
10. Suzuki, N.; Haritani, M.; Yang, J.; Hino, R.; Shamoto, E. Elliptical Vibration Cutting of Tungsten Alloy Molds for Optical Glass Parts. Ann. CIRP 2007, 56 (1), 127–130.
11. Shamoto, E.; Suzuki, N.; Tsuchiya, E.; Hori, Y.; Inagaki, H.; Yoshino, K. Development of 3 DOF Ultrasonic Vibration Tool for Elliptical Vibration Cutting of Sculptured Surfaces.
Ann. CIRP 2005, 54 (1), 321–324.
12. Suzuki, N.; Yokoi, H.; Shamoto, E. Micro/Nano Sculpturing of Hardened Steel by Controlling Vibration Amplitude in Elliptical Vibration Cutting. Precis. Eng. 2011, 35 (1), 44–50.
13. Moriwaki, T.; Shamoto, E. Ultraprecision Diamond Turning of Stainless Steel by Applying Ultrasonic Vibration. Ann. CIRP 1991, 40 (1), 559–562.
14. Moriwaki, T.; Shamoto, E.; Inoue, K. Ultraprecision Ductile Cutting of Glass by Applying Ultrasonic Vibration. Ann. CIRP 1992, 41 (1), 141–144.
15. Krauskopf, B. Diamond Turning: Reflecting Demands for Precision. Manuf. Eng 1984, 92 (5), 90–100.
16. Blake, P. N.; Scattergood, R. O. Ductile-Regime Machining of Germanium and Silicon. J. Am. Ceram. Soc. 1990, 73 (4), 949–957.
17. Schinker, M. G. Subsurface Damage Mechanisms at High-Speed Ductile Machining of Optical Glasses. Precis. Eng. 1991, 13 (3), 208–218.
18. Schinker, M. G.; Doll, W. Turning of Optical Glasses at Room Temperature. Proc. SPIE 1987, 802, 70–80.
19. Nakasuji, T., et al. Diamond Turning of Brittle Materials for Optical Components. Ann. CIRP 1990, 39 (1), 89–92.
20. Puttick, K. E., et al. Single-Point Diamond Machining of Glasses. Proc. R. Soc. Lond. A 1989, 426, 19–30.
21. Kumabe, J., et al. Ultrasonic Super-Position Vibration Cutting of Ceramics (in Japanese). J. JSPE 1986, 52 (11), 1851–1857.
22. Mizutani, K., et al. Material Removal in Orthogonal Cutting of Ceramics by Superimposing Ultrasonic Vibration on the Tool (in Japanese). J. Soc. Mater. Sci. 1984, 33 (369),
652–658.
23. Takeyama, H., et al. Machinability of Glass Fiber Reinforced Plastics and Ultrasonic Machining. Ann. CIRP 1988, 37 (1), 83–96.
24. Weber, H., et al. Turning of Machinable Glass Ceramics with an Ultrasonically Vibrated Tool. Ann. CIRP 1984, 33 (1), 85–87.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
454 Ultrasonic Vibration Diamond Cutting and Ultrasonic Elliptical Vibration Cutting

25. Yamaguchi, K.; Shamoto, E.; Sato, M. Study of Press-Processing Methods for Brittle Sheets. JSME Ser. III 1990, 33 (4), 583–589.
26. Brehm, R., et al. Transparent Single-Point Turning of Optical Glass. Precis. Eng. 1979, 1 (4), 207–213.
27. Shamoto, E.; Morimoto, Y.; Moriwaki, T. Elliptical Vibration Cutting (2nd Report, Study on Effects of Vibration Conditions) (in Japanese). J. JSPE 1999, 65 (3), 411–417.
28. Shamoto, E.; Suzuki, N.; Hino, R. Analysis of 3D Elliptical Vibration Cutting with Thin Shear Plane Model. CIRP Ann. Manuf. Technol. 2008, 57 (1), 57–60.
29. Shamoto, E.; Song, Y. C.; Sassa, K.; Yoshida, H.; Hino, R.; Moriwaki, T. Proposal of Oblique Type of Elliptical Vibration Cutting and Its Basic Performance (in Japanese). J. JSPE
2003, 69 (7), 970–975.
30. Shamoto, E.; Altintas, Y. Prediction of Shear Angle in Oblique Cutting with Maximum Shear Stress and Minimum Energy Principles. Trans. ASME J. Manuf. Sci. Eng. 1999, 121,
399–407.
31. Krystof, J. Berichte uber Betriebswissen-schaftliche Arbeiten, Bd., 12; VDI Verlag: Berlin, 1939.
32. Lee, E. H.; Shaffer, B. W. The Theory of Plasticity Applied to a Problem of Machining. J. Appl. Mech. 1951, 18, 405–413.
33. Merchant, M. E. Mechanics of the Metal Cutting Process, II. Plasticity Conditions in Orthogonal Cutting. J. Appl. Phys. 1945, 16, 318–324.
34. Amini, S.; Shamoto, E.; Suzuki, N.; Nategh, M. J. FE Analysis of One-Directional and Elliptical Vibration Cutting Processes. Special issue on “Modeling and Simulation of Cutting
Process”. Int. J. Automat. Technol. 2010, 4 (3), 252–258.
35. Shamoto, E.; Suzuki, N.; Moriwaki, T.; Naoi, Y. Development of Ultrasonic Elliptical Vibration Controller for Elliptical Vibration Cutting. Ann. CIRP 2002, 51 (1), 327–330.
36. Suzuki, N.; Nakamura, A.; Shamoto, E.; Harada, K.; Matsuo, M.; Osada, M. Ultraprecision Micromachining of Hardened Steel by Applying Ultrasonic Elliptical Vibration Cutting. In
Proc. 2003 Int. Symp. MHS, 2003, pp 221–226.
37. Yan, J.; Syoji, K.; Tamaki, J. Some Observations on the Wear of Diamond Tools in Ultra-Precision Cutting of Single-Crystal Silicon. Wear 2003, 255, 1380–1387.
38. Suzuki, N.; Masuda, S.; Shamoto, E. Ultraprecision Machining of Sintered Tungsten Carbide by Applying Ultrasonic Elliptical Vibration Cutting. In Proceedings of the 4th
International Conference of the EUSPEN, 2004; pp 187–188.
39. Suzuki, N.; Masuda, S.; Haritani, M.; Shamoto, E. Ductile Machining of Brittle Materials by Applying Elliptical Vibration Cutting. In Proc. CIRP 2nd Int. Conf. HPC, 2006,
CD-ROM.
40. Moriwaki, T.; Shamoto, E. Ultrasonic Elliptical Vibration Cutting. Ann. CIRP 1995, 44 (1), 31–34.
41. Thomas, A. D.; Michele, H. M.; Peter, J. F. Application of a Fast Tool Servo for Diamond Turning of Nonrotationally Symmetric Surfaces. Precis. Eng. 1991, 3 (4), 43–250.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 11.17 Ultrasonic Vibration Cutting
Part II: Ductile Cutting and Analytical Force Models for the Elliptical Vibration
Cutting Process
C Nath, National University of Singapore, Singapore and University of Illinois at Urbana-Champaign, Urbana, IL, USA
X Zhang, National University of Singapore, Singapore and Machining Technology Group, Singapore Institute of Manufacturing
Technology, Singapore
A Senthil Kumar and M Rahman, National University of Singapore, Singapore
 2014 Elsevier Ltd. All rights reserved.

11.17.1 Introduction 455

11.17.2 Ductile Machining Model for the EVC Process 456
11.17.2.1 The EVC Process Theory 456
11.17.2.2 Effect of TOC on Finished Surface 457
11.17.2.3 Condition to Obtain a Smaller TOCm 458
11.17.2.4 TOCm at an Rs within the Rscr 458
11.17.2.5 TOCm at Rscr (TOCcm) 459
11.17.2.6 TOCm at a Rs beyond the Rscr 459
11.17.2.7 Effect of Rs on Cycle-Overlap and TOCm 459
11.17.2.8 Determination of the Critical Speed Ratio Rscr 460
11.17.2.9 Effects of Machining Parameters 461
11.17.2.10 Selection of Parameters for Ductile Mode Machining 461
11.17.3 Ductile Machining Model Validation and Analysis 462
11.17.3.1 Model Validation with Machining of WC 462
11.17.3.2 Mirror Surface from Sintered WC 465
11.17.3.3 Application of the Model for Machining Hardened Steel 466
11.17.3.3.1 Model Evaluation 466
11.17.3.3.2 Ultraprecision Machining of Hardened Steel Using PCD Tools 468
11.17.4 Analytical Force Model for the Orthogonal EVC Process 470
11.17.4.1 Development of the Force Model 470
11.17.4.1.1 Transient Thickness of Cut (TOCt) 470
11.17.4.1.2 Transient Shear Angle and Transition Characteristic of Friction Reversal 472
11.17.4.1.3 Transient Cutting Force Components 474
11.17.4.2 Experimental Setup 476
11.17.4.3 Validation of the Developed Force Model 478
11.17.4.3.1 Calibration of the Parameters 478
11.17.4.3.2 Model Validation 478
11.17.5 Conclusion 480
References 481

11.17.1 Introduction

Hard and brittle materials such as sintered tungsten carbide (WC), tungsten alloy, optical glass, ceramics, and hardened steels are
applied in many critical areas, for example, optical parts, functional components, dies and molds, etc. because of their physical and
mechanical properties including hardness, high-temperature strength, elastic modulus, wear and corrosion resistance, etc. (1–10).
However, when applying the conventional cutting (CC) process, particularly, turning, grooving, and planning to any of these
materials, the finished surface appears with brittle fractures if the nominal depth of cut (DOC) is set higher than its critical depth of
cut (DOCcr) (5–9). As the value of DOCcr of these materials is extremely low (typically 1 mm or less), the material removal rate is
highly limited with the CC process. In addition, continuous interaction between the tool and the workpiece in this process causes
accelerated tool wear (1–10).
The elliptical vibration cutting (EVC) process has been successfully applied in past decade for producing mirror surface from
these hard and brittle materials. In this process, the tool tip follows overlapping cycles in elliptical locus, as shown in Figure 1
(1–6,8–10). Due to this, for a given nominal DOC, the thickness of cut (TOC) of the work material on the tool rake face varies
continuously from zero to some value within the DOC value. The TOC value becomes significantly small, especially when the
cutting edge passes the bottom point of the cutting cycle. On the other hand, the TOC remains equal to the nominal DOC in the

Comprehensive Materials Processing, Volume 11 http://dx.doi.org/10.1016/B978-0-08-096532-1.01328-5 455

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
456 Ultrasonic Vibration Cutting

Chip flow
y Tool vibration locus
Tool rake
b
a x

Flank angle
ap Finished
surface
Workpiece vc

Figure 1 Illustration of the orthogonal EVC process (phase shift between two vibration modes: 90 ).

case of the CC process. Therefore, the nominal DOC can be set larger than the value of the DOCcr during ductile machining of
these materials with the EVC process (5,6,8). Along with the variation of the TOC value of the work material, the tool velocity also
continuously varies in each elliptical vibration cycle (11–14). These two inherent phenomena, in fact, introduce transient char-
acteristics into their cutting mechanics that are considered to be different from the one applied for the CC process. In this chapter,
a geometrical ductile machining model (5) and an analytical transient force model (11) for the orthogonal EVC process are
presented. Section 11.17.2 presents the ductile machining model. In Section 11.17.3, the model is applied to ductile/ultrapre-
cision machining of a superhard and brittle material for the validation study and to ultraprecision machining of hardened steel for
justifying its versatility in machining different hard and/or brittle materials. Section 11.17.4 describes the analytical force model
for the orthogonal EVC process and its validation with machining experiments. The chapter concludes with Section 11.17.5.

11.17.2 Ductile Machining Model for the EVC Process

A geometrical model for ductile machining with the EVC process is presented in this section (5). As the TOC of the work material
against the tool rake continuously varies from zero to some value within the given nominal DOC in each cutting cycle (see
Figure 1), there is a maximum TOC (hereafter, TOCm) of the workpiece material for a set of relevant machining parameters affecting
the quality of the surface (e.g., brittle or ductile) being machined. The following model describes which parameters directly
influence the TOCm value in each EVC cycle and how one can select the relevant process parameters during ductile machining of any
hard and brittle material with the EVC process.

11.17.2.1 The EVC Process Theory

Figure 1 illustrates the principle of the ultrasonic elliptical vibration cutting (UEVC) process (1–6,8–18). The elliptical motion at the
cutting edge is generated by applying exciting voltage on some piezoelectric transducers that are stacked orthogonally into a metal
block. The tool is vibrated elliptically at a frequency, f, with very small amplitudes of a in the cutting direction (i.e., x-axis) and b in
the thrust direction (i.e., y-axis) while combining two resonant vibrations in x- and y-axes with some phase shift (8,15). If cutting
speed, vc, is introduced to the workpiece in the tangential or cutting direction (i.e., x-axis), then the motion of the tool with respect to
the workpiece in the EVC process is as follows (14):
x ¼ a cosðutÞ  vc t [1]

y ¼ b cosðut þ 4Þ [2]
where u, t, and 4 are angular velocity of the tool tip, time, and phase difference between the two axes, respectively. When vc ¼ 0, the
maximum tool vibration speed, (vt)max in the cutting direction is obtained by differentiating eqn [1]:
ðvt Þmax ¼ ua ¼ 2pfa ðwhen sinðutÞ ¼ 1Þ [3]

The speed ratio, Rs, of the cutting speed to the maximum tool vibration speed in the cutting direction is expressed as
Rs ¼ vc =ðvt Þmax ¼ vc =2pfa [4]

For separating- or intermittent-type vibration cutting, the cutting condition is chosen so that Rs < 1.
As shown in Figure 1, the tool edge gets separated from the workpiece in each vibration cycle. Due to elliptical locus of the tool
(given by a and b in x- and y-axes, respectively) and cutting speed, vc, of the work material, overlapping between two or more
consecutive cycles must occur. During any overlap cycle, the cutting edge gets engaged with the workpiece (i.e., cutting starts in the

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Cutting 457

Figure 2 Effect of the nominal DOC on the finished surface generated in the UEVC technique when (x1  x3) > 0.

current cycle) that was left by the preceding machining cycle. In this elliptical locus, wherever engaged with the workpiece, the tool
initially starts cutting with the TOC value of zero. As the cutting in this cycle progresses, there is a continuous but nonuniform
variation of the TOC value until the tool edge gets separated from the workpiece. This phenomenon has been geometrically pre-
sented in detail in Figure 2.

11.17.2.2 Effect of TOC on Finished Surface

Suppose the nominal DOC, ap, is set to be larger than the thrust directional vibration amplitude, b. Figure 2 shows any three
consecutive cycles of an elliptically vibrated cutting edge. Let A be the cycle start point, B be the cross-point between any two
consecutive cycles, D be the point on the x-axis where the cutting edge reaches at half-vibration period (T/2), E be the leftmost point
of any cycle, and F be the rightmost point of any upper-half cycle. Subscripts 1, 2, and 3 denote the first, second, and third cycles,
respectively. Thus, the first cycle starts at point A1 when following the path A1–B01–C1–B12–H–D1–E1–F1–A2, where H is an arbitrary
point between the points B12 and D1. The first cycle ends at the point A2, from where the second cycle starts. Similarly, the third cycle
starts at the point A3 by following A3–B23–D3–E3–F3–A4, where A4 is the starting point of the fourth cycle. It can be seen from
Figure 2 that the finished surface progresses through points B01–C1–B12–G–C2–B23–L due to the first two consecutive cycles, where
L is an arbitrary point on any cycle beyond the point C.
In the first vibration cycle, the tool reaches point D1 at half-vibration period, T/2, though the cutting ends at the leftmost point E1
by the period longer than T/2. The tool is then free from the workpiece from time tE1 to tB2 , where tB2 is the cutting-start time for the
second vibration cycle at the cross-over point B12. Similarly, the cutting-end time for the second and third cycles are at points E2 and
E3, which are beyond and leftward of the points D2 and D3, respectively. According to Shamoto et al. (15), the cutting-end time, tE1 ,
in the first vibration cycle can be obtained from the following equation:
au sinðutE1 Þcos a þ bu sinðutE1 þ 4Þsin a ¼ vc cos a [5]
where a is the tool rake angle and u ¼ 2pf. One can calculate the cutting-end time, tE1 , from the above eqn [5] for any values of a, b,
vc, a, 4, and f. However, it is usual practice to set the zero degree rake angle during micromachining. Taking this fact into account and
for easier calculation, the value of a is considered zero in this study. Thus the cutting-end time is obtained as
 
1 vc
au sinðutE1 Þ ¼ vc rtE1 ¼ sin1 [6]
u au
In Figure 2, when the cutting edge starts cutting in the second cycle at the cross-over point B12, no material is seen to be removed
and thus the TOC is zero at that instant. As the tool tip moves toward the left by following the trajectory B12–G–C2–B23. in this
cycle and it already followed the trajectory B12–H–D1–E1 in the previous or first elliptical cycle, the TOC of the workpiece material
nonuniformly increases along the tool rake. For example, the TOC at point G is seen to be GH, but it is C2E1 at C2. Moreover, when
the tool crosses the point C2 toward the left, the TOC for the cutting tool drastically increases from C2E1 to a value that is close to the
value of the given nominal DOC, ap (e.g., B23I at the cycle cross-over point B23). Thus, the value of the TOCm for the cutting
condition illustrated in Figure 2 can be found just beyond the point C2 toward left (assuming C2M). As the nominal DOC has
a strong influence on machining hard and brittle materials, the surface produced along B12–G–C2 should be better as compared to
that in C2–B23–L, where L is the arbitrary point in the locus C2–B23–D2–E2 as defined earlier. Thus, brittle fractures on the finished
surface would be generated in the UEVC technique depending on the relevant parameters set. Since the tool starts cutting in the third
cycle at the cross-over point B23 and repeats the same principle as was explained above for the second cycle, the quality of the
finished surface due to a comparatively higher TOC beyond the point B23 (i.e., B23–L–.) is not considered in this study.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
458 Ultrasonic Vibration Cutting

Figure 3 Reduction of a given nominal DOC to a smaller TOCm value in the UEVC method (a) at the critical and (b) within the critical conditions.

In order to avoid the negative effect of such comparatively higher TOC beyond a critical tool position as explained by Figure 2,
a suitable speed ratio, Rs, has to be selected in a range such that the cross-point B23 between the second and the third cycles coincides
with the point C2 as shown in Figure 3(a). This means that the critical condition occurs when the line B23I of Figure 2, which is
parallel to the y-axis, shifts rightward to coincide with the line C2M at the same x-coordinate in this case. And the rightmost point, F3,
of the upper half of the third cycle disappears and coincides with the leftmost point of the first cycle at E1. Compared to the first case
described for Figure 2, a reduced TOCm in the UEVC cycle can be obtained in this case (Figure 3(a)), and the nominal DOC may not
directly affect the final finished surface. Because the cutting region distorts due to the direct DOC or abrupt change of the TOCm
beyond the critical location, B23 will be removed by the next (i.e., third) vibration cycle. It is seen from Figure 3(a) that the TOCm on
the surface being machined at the point B23 is B23E1 at the critical condition.
However, brittle fracture may still appear within the critical condition if the TOCm exceeds the DOCcr at any location between
two consecutive cross-over points, B12 and B23, i.e., in the region B12–B23, which is to be the ultimate finished surface. In this case,
the TOCm can further be reduced (e.g., B23H as shown in Figure 3(b)) if the rightmost point, F3, of the upper half of the third cycle
crosses the first cycle above the x-axis. During cutting, as the tool engages with the workpiece when they are opposite each other in
the cutting direction (i.e., x-axis), eqns [1] and [4] suggest that the speed ratio can be controlled by any of the following three
following important parameters: (1) work cutting speed, vc; (2) tool vibration frequency, f; and (3) tool vibration amplitude in the
tangential direction, a. The speed ratio Rs for which the leftmost point of the first cycle and the rightmost point of the third cycle just
touch each other or coincide can be defined, hereafter, as critical speed ratio Rscr.

11.17.2.3 Condition to Obtain a Smaller TOCm

Let the coordinates of points E1 and F3 be (x1,y1) and (x3,y3), respectively, and the linear distance between these two points in the x-
axis be (x1  x3), which is the cycle-overlap between the first and the third cycles. It can be realized by Figures 2, 3(a), and 3(b) that
the effect of a given nominal DOC on the finished surface depends on the positive, zero, or negative values of (x1  x3). By using eqn
[1], the following equation can be derived to obtain the cycle-overlap:
ðx1  x3 Þ ¼ ½a cosðutE1 Þ  vc tE1   ½a cosðutF3 Þ  vc tF3  [7]
where tF3 is the time when the tool reaches the rightmost point, F3, above the x-axis in the third cycle. The time tF3 can be obtained
from the following two equations. As described earlier, since tE1 > T=2, the following equation can also be derived from Figures 1
and 2:
tD1 E1 ¼ tF3 A4 ¼ ðtE1  T=2Þ [8]
and with reference to Figure 2:
tF3 ¼ 3T  tD1 E1 ¼ 3T  ðtE1  T=2Þ [9]
Therefore, to obtain a smaller TOCm against a comparatively higher nominal DOC in the UEVC process as shown in Figures 2
and 3, the cycle-overlap (x1  x3) has to be negative or at most zero.
½a cosðutE1 Þ  vc tE1   ½a cosðutF3 Þ  vc tF3   0 [10]

11.17.2.4 TOCm at an Rs within the Rscr

This condition is followed according to Figure 3(b). The cutting-start times tB1 and tB2 for the first and the second cycles at the cross-
over points B01 and B12, respectively, can be obtained numerically by the following equations (13,15):
½a cosðutB1 Þ  vc tB1   ½a cosðutB2 Þ  vc tB2  ¼ vc T [11]

b cosðutB1 þ 4Þ ¼ b cosðutB2 þ 4Þ [12]

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Cutting 459

And it is also understood from Figure 3(b) that

 
b cos utB0 1 þ 4 ¼ b cosðutB2 þ 4Þ [13]

where tB0 1 is the time when the tool reaches the cross-over point B12 in the first vibration cycle and ðtB0 1 stB2 Þ. Moreover,
tB2 ¼ T þ tB1 ; tB0 2 ¼ T þ tB0 1 ; and tB3 ¼ 2T þ tB1 [14]

where tB0 2 is the time required by the tool from the beginning of the first cycle to reach the cross-over point B23 in the second cycle and
tB3 is the cutting-start time for the third cycle. Now, let the point B23 be (x23,y23), whose coordinate can be found from the following
equations by using eqns [1] and [2]:
 
x23 ¼ a cos utB0 2  vc tB0 2 [15]

   
y23 ¼ b cos utB0 2 þ 4 ¼ b cosðutB3 þ 4Þ tB0 2 stB3 [16]

Again, let the point H be ðxH1 ; yH1 Þ. It is seen in Figure 3(b) that x23 ¼ xH1 . Therefore, tH1 can be numerically solved by the
following equation:
x23 ¼ xH1 ¼ a cosðutH1 Þ  vc tH1 [17]
By substituting the value of tH1 , yH1 can be found as

yH1 ¼ b cosðutH1 þ 4Þ [18]

Therefore, the TOCm at a desired Rs within Rscr is
   
ðTOCm Þ<Rscr ¼ y23  yH1 ¼ b cos utB0 2 þ 4  b cosðutH1 þ 4Þ [19]

11.17.2.5 TOCm at Rscr (TOCcm)

At the Rscr, the coordinate HðxH1 ; yH1 Þ coincides with E1(x1,y1). Thus the TOCm at Rscr can be called as the critical TOCm(TOCcm).
Therefore, eqn [19] becomes
 
TOCcm ¼ ðy1  y23 Þ ¼ b cosðutE1 þ 4Þ  b cos utB0 2 þ 4 [20]

11.17.2.6 TOCm at a Rs beyond the Rscr

When the speed ratio, Rs, is beyond the Rscr, then the second and third cycles cross each other at B23(x23,y23) leftward to C2 ðxC2 ; yC2 Þ
as shown in Figure 2. Thus the TOC beyond the Rscr varies from the point C2 toward the arbitrary point L. However, it is maximum
at the point just beyond the coordinate C2 ðxC2 ; yC2 Þ toward the left and minimum at the point B23(x23,y23) if only the final finished
surface is considered (i.e., B12–G–C2–B23) for the second cycle. Therefore, the TOCm at an Rs beyond the Rscr is at the point just
beyond the coordinate C2 ðxC2 ; yC2 Þ:
 
ðTOCm Þ>Rscr ¼ ap  yC2  b [21]

The x-coordinates for the points E1(x1,y1) and C2 ðxC2 ; yC2 Þ are same in this case and therefore

x1 ¼ xC2 ¼ a cosðutC2 Þ  vc tC2 [22]

where tC2 is the time when the tool is at the point C2 ðxC2 ; yC2 Þ. Thus, yC2 for eqn [21] can be obtained by

yC2 ¼ b cosðutC2 þ 4Þ [23]

Additionally, the minimum TOC at a Rs beyond the Rscr, which is B23I on the finished surface, C2–B23, can be obtained as

ðTOCmin Þ>Rscr ¼ ap  ðy23  bÞ [24]

11.17.2.7 Effect of Rs on Cycle-Overlap and TOCm

As discussed so far, the cycle-overlap, (x1  x3), between the first and the third cycles (Ref. Figures 2 and 3) should be negative or at
most zero to obtain a smaller TOCm, if the nominal DOC, ap, is set larger than the thrust directional vibration amplitude, b, in the
UEVC technique. Figure 4 shows the effects of speed ratio on the cycle-overlap, (x1  x3) at various tangential directional vibration
amplitudes, a by using eqn [7]. However, if vc and f are varied in such approach, only the speed ratio, Rs, will be affected; but the
nature or slopes of the straight lines at various values of a remain the same. It is interestingly seen in Figure 4 that all the lines for

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
460 Ultrasonic Vibration Cutting

Unacceptable
1.0
p(x, y)

Cycle-overlap (μm)
0.0
Acceptable

Unacceptable
–1.0
–2.0
Acceptable
–3.0 a = 1.0 μm
–4.0
a = 1.5 μm
–5.0
0.00 0.03 0.06 0.09 0.12 0.15
Speed ratio, R s

Figure 4 Effect of speed ratio on cycle-overlap at various tangential directional vibration amplitudes, a.

various values of a cross at a specific point p(x, y). If the Rs exceeds p(x, y), then the cycle-overlap, (x1  x3), becomes positive and
thereby the TOCm becomes approximately close to the nominal DOC (Figure 2), which is not desired. This means that the leftmost
point of the first cycle and the rightmost point of the upper half of the third cycle should overlap, or at least coincide with each other
to obtain a reduced TOCm. Thus the Rscr can be found at the coordinate p(x, y).
The values of the TOCm at different values of Rs can be obtained analytically by using eqns [19] and [21] against various nominal
DOCs (e.g., 3, 4, and 5 mm) and can be plotted as shown in Figure 5. Equations [19] and [21] are used depending on the negative
and positive values of the (x1  x3), respectively. It is clearly seen in Figure 5 that the value of the TOCm abruptly jumps to a higher
value beyond a certain Rs and they happen in the similar manner as described for Figure 4, where the values of the cycle-overlap,
(x1  x3), change from negative to positive. Due to the sudden increase in the TOCm, the finished surface beyond the coordinate
C2 ðxC2 ; yC2 Þ leftward, as illustrated in Figure 2, is severely affected with brittle fractures. Along with eqn [19], Figure 5 also reveals
that the TOCm value does not depend on the nominal DOC if the value of Rs is set lower than Rscr.

11.17.2.8 Determination of the Critical Speed Ratio Rscr

It can be seen by Figures 4 and 5 that the Rscr value is between 0.125 and 0.130. In order to obtain the value of the Rscr, the iteration
process is conducted based on the negative and positive values of the cycle-overlap, (x1  x3), as the term Rs (i.e., vc/2pfa) cannot be
directly obtained from eqn [7] or [10]. Table 1 shows the last step of the iterations conducted. At first, the Rscr is determined by
varying the cutting speed vc and then that value of Rscr is checked by sequentially varying f and a. The values of the corresponding
TOCm values are also calculated. It is seen that the TOCm, as obtained from eqn [19], is significantly smaller compared to the
nominal DOC when the condition (x1  x3) < 0 is satisfied. However, this is not so for the (x1  x3) > 0 and the TOCm value in this
case remains approximately close to the given value of the nominal DOC, ap, according to eqn [21] and Figure 5.
As determined in Table 1, iteration up to five digits of the Rscr value has special significance for the UEVC technique during
machining of hard and brittle materials. It can be clearly noticed that when the Rs value is increased at the fifth digit, i.e., from
0.128 37 to 0.128 38, the cycle-overlap between the first and the third cycles, (x1  x3), becomes positive from a negative value. This
means, in such case, a region with a significant amount of brittle fractures in each machining cycle (e.g., C1–B12 and C2–B23 zones in
Figure 2) would be left due to sudden rise of the TOCm value (Ref. Table 1). Since the UEVC technique is typically applied to obtain
ductile mode surfaces from hard brittle materials, consideration at the fifth digit iteration would guide one to select any of the three
relevant parameters (i.e., vc, f, and a in the x-axis) in precise level for a desired Rs within the Rscr value. Moreover, on the other hand,
the iteration beyond the fifth digit may not be realistic, as the cycle-overlap value at this state has already reached the angstrom (A )
or atomic level.

5 (i) 5
at ap:
4 (ii) 4
(i) 5 μm
TOCm (μm)

3 (ii) 4 μm (iii) 3
(iii) 3 μm
2 2

1 1

0 0
0.00 0.03 0.06 0.09 0.12 0.15
Speed ratio, Rs

Figure 5 Effect of speed ratio on the TOCm (condition: ap > b, where b ¼ 1.5 mm).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Cutting 461

Table 1 Determination of the Rscr value using iteration method (condition: ap > b, ap ¼ 4 mm,
b ¼ 1.5 mm. a is in mm, f is in kHz, and vc is in m min1)

Parameters
(x1  x3) TOCm
Rs ¼ vc /2pfa Fixed Variable (m) (mm)

0.12837 a ¼ 1.5 vc ¼ 2.90366 1.0655e10 1.5917

0.12838 f ¼ 40 vc ¼ 2.9039 þ1.1886e10 3.901
0.12837 a ¼ 1.5 f ¼ 39.950 1.3776e10 1.5924
0.12838 vc ¼ 2.9 f ¼ 39.946 þ1.6015e10 3.901
0.12837 f ¼ 40 a ¼ 1.4981 8.8002e10 1.5917
0.12838 vc ¼ 2.9 a ¼ 1.4980 þ1.1034e10 3.901

Thus, it is found that there is cycle-overlap, (x1  x3) z 0, when Rs approaches 0.12837. Therefore, the acceptable Rscr is 0.12837.
Hence, in order to obtain a reduced TOCm and to avoid much brittle fractures or distortions on the finished surface of brittle
materials in the UEVC process, first, the following condition should be maintained:
Rs ¼ vc/2pfa  0.12837 or simply Rs  0.128.

11.17.2.9 Effects of Machining Parameters

In this study, three vital machining parameters vc, f, and a have so far been identified, which have direct influence on the TOCm value
for a given nominal DOC in each UEVC cycle. Table 2 shows that the TOCm can be controlled based on a suitable Rs and the
positive/negative nature of (x1  x3) for a given value of b. As described above, either a negative or at most a zero value of the cycle-
overlap (x1  x3) is desired to obtain the reduced TOCm under the UEVC process.
However, from eqns [19] to [21], it can be noticed that another parameter, namely thrust directional vibrational amplitude b,
also has direct influence on the TOCm value. By using these equations for the three described conditions, for a given 4 mm nominal
DOC, the values of the TOCm against the Rs at different values of b are obtained, which are plotted in Figure 6. It is seen that the
TOCm decreases with the decrease in b, and this is favorable for ductile mode cutting. However, the value of b should not be zero,
because having the thrust vibration amplitude of the tool prolongs the tool life by reducing the average cutting force and energy,
which is the main advantage of the UEVC technique over the one-dimensional or conventional ultrasonic vibration cutting tech-
nique (9,15). Thus, while selecting a suitable value of the thrust vibration amplitude b in the UEVC process, both the advantages and
the disadvantages have to be compromised.

11.17.2.10 Selection of Parameters for Ductile Mode Machining

Suppose the DOCcr with the CC grooving technique is known for a hard and brittle material. Thus, to achieve ductile mode cutting
with the UEVC process, the values of b and Rs have to be considered by following Figure 6 so that the condition TOCm < DOCcr is

Table 2 Control of the TOCm based on the Rs and (x1  x3) by controlling the parameters
(ap > b)

Speed ratio Controllable parameters (x1  x3) (mm) TOCm (mm)

<Rscr <0 B23H (Figure 3(b))

¼Rscr a, f, vc 0 B23E1 (Figure 3(a))
>Rscr >0 zap (Figure (2))

4.0 4.0
3.5 at b: 3.5
3.0 (i) 1.0 μm 3.0
TOCm (μm)

2.5 (ii) 1.5 μm 2.5

2.0 (iii) 2.0 μm 2.0
1.5 1.5
1.0 1.0
0.5 0.5
0.0 0.0
0.00 0.03 0.06 0.09 0.12 0.15
Speed ratio, R s

Figure 6 Effect of thrust vibration amplitude b of the tool on the TOCm at different Rs (condition: ap ¼ 4 mm and ap > b).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
462 Ultrasonic Vibration Cutting

satisfied. Figure 6 shows that any desired value of the TOCm depends on Rs and b, where the value of Rs can be controlled by any of
the three parameters vc, f, and a, according to eqn [4]. Moreover, when selecting a suitable value of b, both the advantages and the
disadvantages have to be considered as described in Section 11.17.2.9.

11.17.3 Ductile Machining Model Validation and Analysis

The above-described geometrical model is first validated by machining a sintered WC, which is one of the hard and brittle
materials that possesses versatile thermophysical and mechanical properties and is thus considered for optical glass molding
applications. The model is also applied to a hardened stainless steel to justify its usefulness during machining of other hard and
brittle materials. Note that hardened steel is also potentially applied for ultraprecision die and mold parts, liquid crystal display
panels, etc.

11.17.3.1 Model Validation with Machining of WC

A sintered WC workpiece with about 15 wt.% Co (hardness 88.8 HRC) is considered to be machined using polycrystalline
diamond (PCD) tools. The workpiece was held by the vacuum chuck of a 5-axis computer numerical control (CNC)-controlled
Toshiba ultraprecision machine (ULG-100H3). The DOCcr of the used sintered WC was found to be about 0.6 mm when applying
the conventional groove cutting technique using the same machine. Three grooving experiments at three different locations on
a flat WC workpiece face were conducted with increasing DOC while the beginning DOC is zero. The DOCcr was determined at
a location where the finished material starts fracturing due to exceeding a certain DOC. Similar grooving tests, as the well-
established technique for determining the DOCcr of various hard brittle materials, were also conducted by Suzuki et al. (8)
and Li et al. (19).
With the same machine, facing experiments were then conducted on the same sintered WC workpiece of 40 mm diameter for
validation of the model. The experimental setup in a schematic view and with the machine can be shown in Figure 7(a) and 7(b),
P
respectively. An elliptical vibrator (EL-50 ) from Taga Electric Co. Ltd, Japan, was attached with the vertical spindle of the machine.
P
The ‘LONGI’ and the ‘BENDI’ amplifiers are connected to the vibrator through the ‘EL-50 controller’ to vibrate the tool in the
P
thrust and the tangential directions, respectively. The ‘EL-50 controller’ controls and ensures the proper elliptical motion for
different chosen vibration amplitudes a and b of the tool. A triangular-type fresh PCD tool (grade DA150) from Sumitomo was set at
the specified tool position of the vibrator for each set of four cutting conditions shown in Table 3.
The PCD tool was of 0 rake and 11 clearance angles, 0.4 mm nose radius, and about 3 mm edge radius. The tool was set with
30 approach angle and to feed in the traverse direction from 20 mm outer radius to 17 mm inner radius of the workpiece face at
3 mm rev1 feed rate and 4 mm nominal DOC in each set of cutting conditions. The tool was set to vibrate at a frequency, f, of
38.87 kHz with tangential vibration amplitude a of 2 mm. Therefore, the maximum tool vibration speed can be estimated as
(vt)max ¼ 2paf ¼ 29.41 m min1. The cutting speed vc was varied in these tests for obtaining different Rs values.
The values of the TOCm were obtained at different values of Rs (at 18.5 mm from the workpiece center) and b by following
the established equations for different conditions (also see Figure 6), which are shown in Table 3. First, test nos. 1 and 2 were
conducted to examine the surface quality within and beyond the Rscr (¼ 0.128 37), respectively, by selecting two different

(a)
Tool
insert
EL-50
device (b)
ULG-100 spindle
Air
spindle Feed EL-50 device
Workpiece
PCD tool

Vacuum chuck

Feedback
LONGI
Amplifier WC workpiece
Power EL-50 Air coolant
source controller
BENDI
Amplifier
Feedback

Figure 7 (a) Schematic view of the UEVC experimental setup for facing experiments and (b) the setup with the Toshiba machine spindle (5).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Cutting 463

Table 3 The predicted values of the TOCm at different radii of the workpiece for different values of Rs and b (b in mm, spindle speed n in rpm, nominal
DOC, ap ¼ 4 mm, ap > b, DOCcr z 0.6 mm)

Predicted values at radius

Test no. Parameters Terms 17 mm 18.5 mm 20 mm Condition

1 n ¼ 20 rpm b ¼ 2 mm Rs 0.073 0.079 0.086 –

TOCm 0.41 0.50 0.59 TOCm < DOCcr
2 n ¼ 35 rpm Rs 0.127 0.139 0.150 –
TOCm 1.90 3.848 3.826 TOCm > DOCcr
3 b ¼ 1 mm Rs 0.091 0.100 0.107 –
n ¼ 25 rpm TOCm 0.334 0.416 0.495 TOCm < DOCcr
4 b ¼ 2 mm TOCm 0.667 0.831 0.990 TOCm > DOCcr

spindle speeds, n, at a fixed value of b. Finally, test nos. 3 and 4 were conducted to observe the effect of the thrust directional
vibration amplitudes b at a fixed spindle speed, where the values of the TOCm vary according to Figure 6. Note that the cutting
speed varies at different radii of a workpiece during face turning operation at a desired spindle speed, unlike outer diameter
turning operation. The cutting speed vc continuously decreases when the tool moves toward the center of the workpiece
according to the formula:
nc ¼ 2prn [25]
where r is the workpiece diameter at any instance of the tool movement over the machining area and n is the spindle rotational
speed. Due to this, the speed ratio also decreases and so does the instantaneous value of the TOCm. Thus, in order to obtain ductile
mode machining on the workpiece face surface under the UEVC technique, the largest value of the TOCm for a desired cutting region
must be set lower than the DOCcr of a given hard and brittle material. Taking into account the above fact, the values of n and b for
the workpiece cutting region (from 20 mm outer radius to 17 mm inner radius) were chosen such a way that only one condition will
be satisfied in each test, i.e., the TOCm is either larger or smaller than the DOCcr. The condition of the TOCm against the DOCcr in
each case is shown in Table 3.
According to the proposed theoretical relationship, ductile mode machining should only be obtained throughout the cutting
region for test nos. 1 and 3 because of the condition TOCm < DOCcr. On the other hand, for test nos. 2 and 4, brittle-fractured
surfaces are expected due to the condition TOCm > DOCcr.
Figure 8 shows the Nomarski photographs of the machined surfaces at different locations within the workpiece cutting region,
which is from 20 mm outer to 17 mm inner radius, for the conditions stated in Table 3. Average surface roughness values measured
with the Taylor-Hobson (T-H) profilometer for all the conditions are also stated. It can be observed for test nos. 1 and 3 in Figure
8(a) and 8(c) that the ductile finished surfaces were produced only when the condition TOCm < DOCcr was satisfied. As the surfaces
are smooth in these conditions, all the feed marks (cutting passes) can be clearly seen. In contrast, when this condition was not
satisfied for test nos. 2 and 4, the corresponding finished surfaces in Figure 8(b) and 8(d), respectively, seem to be full of brittle
fractures. As described in Section 11.17.2.2, since the TOCm for test no. 2 is significantly higher (above 3.8 mm), which is close to the
given nominal DOC of 4 mm, the machined surfaces were found to be extremely fractured.
By comparing test nos. 1 and 2 in Figure 8, it is clearly understood that, at a fixed value of the thrust directional vibration
amplitude b, the speed ratio in the UEVC technique plays a significant role in terms of the TOCm (see Figure 5), when producing
surfaces from a hard and brittle material like sintered WC. The TOCm value also varies with the variation of the value of b at a fixed
value of Rs, as shown in Figure 6, which also affected the finished surfaces as shown in Figure 8(c) and 8(d) for test nos. 3 and 4,
respectively.
It is interesting to observe that the produced surfaces for test nos. 2 and 4 in Figure 8(b) and 8(d), respectively, do not look
similar, although all the surfaces are distorted due to the brittle fractures. It may be due to the following two reasons. The first
reason is that since the spindle speed (and hence the speed ratio) is larger for test no. 2 (35 rpm) compared to that for test no. 4
(25 rpm), the relative movement between the tool and the workpiece becomes higher, which opens a comparatively larger
region of the final machined surface to be brittle-fractured in each vibration cycle. For example, as explained by Figures 2 and
3(a) in Section 11.17.2, in the region between the two cross-over points, B12 and B23, a comparatively larger brittle fracture
region will be produced in Figure 2 due to a higher speed ratio as compared to that in Figure 3(a). The second reason is that
the calculated TOCm values for these two conditions are incomparable, as seen in Table 3, which affected the machined
surfaces. This means that the TOCm values for test no. 2 are far away from the DOCcr of the machined sintered WC and close to
the given nominal DOC and hence produced comparatively larger amount of the brittle fractures, compared to that for test
no. 4. Moreover, the speed ratio, Rs, for test no. 4 is within the Rscr; however, the TOCm values under this condition still remain
larger than the DOCcr of the machined sintered WC and thus produced brittle fractures, although to a smaller amount as
compared to test no. 2.
Thus, it is realized that the experimental results are in good agreement with the analytical models derived in this study. The
findings confirm that ductile mode cutting can be achieved even at a higher nominal DOC when the TOCm is set lower than the
DOCcr by controlling the four relevant parameters in the UEVC technique.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
464 Ultrasonic Vibration Cutting

Figure 8 Nomarski photographs (500) of the machined surfaces for various values of Rs and b at different locations ((i): 17.5, (ii) 18.5, and
(iii) 19.5 mm) of the workpiece cutting region for the stated test conditions in Table 3. (a) Test no. 1 (average roughness, Ra: 0.0139 mm).
(i) Rs ¼ 0.075, TOCm ¼ 0.436 mm; (ii) Rs ¼ 0.08, TOCm ¼ 0.50 mm; and (iii) Rs ¼ 0.084, TOCm ¼ 0.558 mm. (b) Test no. 2 (average roughness, Ra:
0.0197 mm). (i) Rs ¼ 0.131, TOCm ¼ 3.864 mm; (ii) Rs ¼ 0.139, TOCm ¼ 3.848 mm; and (iii) Rs ¼ 0.146, TOCm ¼ 3.834 mm. (c) Test no. 3
(average roughness, Ra: 0.0107 mm). (i) Rs ¼ 0.094, TOCm ¼ 0.358 mm; (ii) Rs ¼ 0.099 TOCm ¼ 0.416 mm; and (iii) Rs ¼ 0.105,
TOCm ¼ 0.467 mm. (d) Test no. 4 (average roughness, Ra: 0.0149 mm). (i) Rs ¼ 0.094, TOCm ¼ 0.717 mm; (ii) Rs ¼ 0.099, TOCm ¼ 0.831 mm; and
(iii) Rs ¼ 0.105, TOCm ¼ 0.935 mm.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Cutting 465

11.17.3.2 Mirror Surface from Sintered WC

In the previous section, it was observed that the ductile machining of sintered WC for the machining area of about 349 mm2 is
possible using inexpensive PCD tools when applying the EVC process. In this section, an attempt is made to justify the feasibility of
PCD tools for machining comparatively a larger machining area.
A PCD tool was applied to face turning operation from 17 mm radius to the center of the same workpiece (i.e., about 908 mm2
machining area). A single crystal diamond (SCD) tool was also applied for comparing the machine surface quality. The machining
conditions for these two experiments are chosen to be similar with two exceptions. In the case of the PCD tool, a nose radius of
0.6 mm was used instead of 0.4 mm in previous experiments (Ref. Section 11.17.3.1) because the authors previously studied and
found (3) that a PCD tool of 0.6 mm nose radius performs better over other nose radii for machining WC with the EVC process. For
the case of SCD tool (source: NewD), the nose radius was about 1 mm. Another exception is that the tool vibration amplitude b in
the thrust direction is set to be 2 mm. By controlling the four relevant parameters shown in Figure 6, with 20 rpm spindle speed, the
value of the TOCm was set to be about 0.41 mm (i.e., less than the DOCcr) at 17 mm radius (Rs ¼ 0.073) while zero at the workpiece
center (Rs ¼ 0).
Figure 9(a) and 9(b) show two mirror-finished surfaces produced at 4 mm nominal DOC using the UEVC process. The cor-
responding T-H profiles for a traverse length of about 1.4 mm of both the finished surfaces are presented in Figure 10. For the whole
region of 17 mm radius, surface roughness of 7.6 nm Ra and 44.0 nm Rz using a PCD tool and 4.6 nm Ra and 28.4 nm Rz using an
SCD tool were achieved.
While machining WC of about 412 mm2 area using the same type of PCD tools for 60 min cutting time, no brittle fracture was
observed (3). The same PCD tools in test nos. 1 and 3 in Section 11.17.3.1 also performed ductile mode machining. However, for
machining about 908 mm2 workpiece area in these tests, 283.33 min cutting time was required, according to the chosen machining
conditions. Figure 11(a) and 11(b) shows that the finished surface continuously degraded, while the cutting edge moved toward
the workpiece center. It may be because PCD tool continuously gets damaged as the cutting progressed. On the other hand, as
shown by Figure 11(c) and 11(d), SCD tool produced significantly finer surface compared to the PCD tool till the cutting ends. The
Nomarski photographs of both the PCD and the SCD tools are presented in Figure 12(a) and 12(b), respectively. It can be observed
that the maximum flank wear width for the PCD tool is significantly higher as compared to an SCD tool. Thus, the PCD tool is not
strong enough to machine the very hard and brittle material sintered WC till 283.33 min. However, Figure 8(a) and 8(c) and
another study by the authors in (3) reveal that PCD tools can be successfully applied to obtain ductile mode surface from sintered
WC for at least about 400–500 mm2 cutting area by applying the EVC process, which eventually reduces machining cost as
compared to very expensive SCD tools.
In summary, the EVC process is found to be an efficient technology for ductile machining of hard and brittle materials.
Ductile machining model developed in this chapter (Section 11.17.2) shows that the maximum thickness of cut (TOCm) of the
workpiece material removed by the tool tip in each cutting cycle determines the nature of the machined surface (i.e., ductile or
brittle). The TOCm value can be controlled by four machining parameters, namely workpiece cutting speed, tool vibration
frequency, and tool vibration amplitude in the tangential, and the thrust directions. When the speed ratio that is formed by the
first three parameters is set within the critical speed ratio 0.128 37, the TOCm becomes significantly smaller compared to the
nominal DOC. The cycle-overlap between the first and the third cycles is desired to obtain a reduced TOCm under the EVC
process. It is also observed that high thrust directional vibration amplitude increases the value of TOCm. In contrast, it helps the
chips pulling out from the workpiece during thrust directional motion and thereby reduces the cutting force and tool wear in
this method. To achieve ductile finishing of brittle materials, the condition TOCm < DOCcr must be satisfied. This phenom-
enon is found to be in good agreement with the theoretical relationship when a hard and brittle material sintered WC was
machined.

Machined
40 mm diameter 40 mm
34 mm 34 mm

(a) (b)

Figure 9 Mirror surfaces generated (with ‘NUS’ (which stands for National University of Singapore) reflection) from sintered WC applying the EVC
process at 4 mm nominal DOC using: (a) PCD and (b) SCD tools (workpiece radius 20 mm and machined radius 17 mm).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
466 Ultrasonic Vibration Cutting

(a)μm
Tool feed

Ra = 7.60 nm; Rz = 44.0 nm

(b)

Ra = 4.60 nm; Rz = 28.4 nm

10.0 Traverse length (mm) 11.4

Figure 10 The T-H profiles for 1.4 mm cutting region of the machined surfaces applying the EVC process using (a) PCD and (b) SCD tools.

(a) At 0.8 mm from center (b) At 11 mm from center

(c) At 0.8 mm from center (d) At 11 mm from center

Figure 11 Nomarski photographs of the machined surfaces at different locations from the workpiece center using (a, b) PCD tool and
(c, d) SCD tool.

VB = 34 μm VB = 21 μm

Flank Flank

Rake Rake

(a) (b)

Figure 12 The Nomarski photographs of the wear lands of the flank and rake faces (corner image) of (a) PCD and (b) SCD tools.

11.17.3.3 Application of the Model for Machining Hardened Steel

The EVC process was successfully applied to hardened steel by Shamoto and Moriwaki in 1999 (10) using single crystal diamond
(SCD) tools without establishing the ductile machining concept with this process. Recently, the ductile machining model developed
in Section 11.17.2 has also been applied during machining of a hardened stainless steel, namely Stavax (hardness 49 HRC or
490 HV) (21). The chemical compositions of Stavax used in these tests are listed in Table 4. The main objectives of this application
are: (1) to justify the effectiveness of the ductile machining model developed for the EVC process and (2) to justify the feasibility of
application of PCD tools, instead of expensive SCD tools, under the EVC process to obtain mirror-quality surface from hardened
steel for die and mold applications in the near future.

11.17.3.3.1 Model Evaluation

The experimental setup and procedure described in Figure 7 for the EVC process are followed in these tests employing the elliptical
P
vibration device (EL-50 ) in the same CNC ultraprecision machine (Toshiba ULG-100H3). A triangular-type fresh PCD tool

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Cutting 467

Table 4 Chemical composition of Stavax

Composition C Si Cr Mn V

wt% 0.38 0.9 13.6 0.5 0.3

(DA150) with rake, relief, and approach angles of 0 , 11 , and 30 , respectively, was set at the specific tool position of the vibrator
for each set of cutting conditions. The nose radius of the tools was chosen as 0.6 mm because the authors previously demonstrated
that this geometry performed better as compared to a smaller and a larger nose radius while machining sintered WC using the UEVC
process (3). A Stavax workpiece of 40 mm diameter was used in the tests.
In each cutting condition of the face turning tests, the tool was set to move toward the center of the workpiece from outer
diameter (OD: 40 mm) to inner diameter (ID: 30 mm) for machining a surface area of about 550 mm2. A schematic view of the
machining area in the EVC test is illustrated in Figure 13. The EVC vibrator vibrates at a frequency f of 38.87 kHz with a phase
difference f of 90 . The peak-to-peak vibration amplitudes in the tangential and thrust directions were set to be 4 mm (i.e.,
a ¼ b ¼ 2 mm). During the face turning operation, as described in Section 11.17.3.1, the speed ratio decreases and so does the
instantaneous value of the TOCm. Considering this variation, three spindle rotations of 15, 30, and 45 rpm were chosen. The speed
ratios at the OD of 40 mm and the ID of 30 mm and the corresponding values of TOCm are shown in Table 5. The values of speed
ratio Rs at spindle speed of 15 and 45 rpm are significantly smaller and larger than the critical speed ratio Rscr of 0.128 37,
respectively, while the value of Rs at 30 rpm is close to 0.128 37. Feed rate and DOC were chosen to be 5 mm rev1 and 10 mm,
respectively.
Photographs of machined surfaces obtained using three spindle speeds are captured by the Nomarski microscope (OLYMPUS
STM6), as shown in Figure 14(a)–14(c). It can be observed that with the increase in spindle speed or nominal cutting speed, the
vibration marks along the nominal cutting direction are found to be more obvious. As the speed ratio suddenly increases from
the critical value Rscr of 0.128 37 with the increase in cutting speed, the value of TOCm also significantly increases while cutting over
the entire cutting region, based on description in Section 11.17.2.2 with Figures 2 and 3. The average surface roughness Ra of 10 nm
measured by a Stylus Profilometer (Mitutoyo CS-5000) was achieved for the lowest speed (see Figure 14(a)).
Photographs depicted in Figure 15(a)–15(c) for these three speeds, respectively, also reveal that the tool flank/nose wear at
a faster rate at a higher spindle speed or a speed ratio larger than the Rscr. As the hardened steel is not under the category of brittle
materials, brittle fractures may not be seen at a larger speed ratio other than the vibration marks. Faster tool wear can negatively
influence the surface roughness value by inducing more plowing mechanism in the cutting process. As shown in Figure
16(a)–16(c), scanning electron microscopic (SEM) photographs of chips produced under these three spindle speeds or speed ratio,
respectively, also reveal that a lower speed, where the speed ratio is about 0.128 or less, produced longer, thinner, and continuous
chips as compared to shorten, thicker, and broken chips at the speed ratio larger than the critical speed ratio. As the value of the
TOCm at a higher speed is significantly higher, the tool in this condition produced a comparatively larger chip size.

Workpiece Spindle
Front view rotation
PCD
40 mm

5 mm

Machining Elliptical
area vibration

Figure 13 Schematic view of the machining area for the EVC tests.

Table 5 The EVC test conditions used during face turning

Nominal DOC, ap (mm) 10

Feed rate (mm rev1) 5
Cutting length (m) 110
Cutting area (mm2) 550
Spindle speed, n (rpm) 15 30 45
Nominal cutting speed (m min1) (vc)max at OD 1.885 3.634 5.655
(vc)min at ID 1.414 2.827 4.241
Speed ratio at OD, (vc)max/2paf 0.0643 0.1286 0.193
Coolant type Air

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
468 Ultrasonic Vibration Cutting

Cutting direction
Feed direction
(a) (b) (c)

Smooth surface Vibration marks

Figure 14 Microscope photographs (500) of the machined surfaces at three different spindle speeds (ap ¼ 10 mm, fr ¼ 5 mm rev1):
(a) nr ¼ 15 rpm, (b) nr ¼ 30 rpm, and (c) nr ¼ 45 rpm.

Figure 15 Microscope (100 and 500) photographs of the flank wear of PCD tools at three spindle speeds: (a) 15 rpm, (b) 30 rpm, and
(c) 10 mm (Depth of cut, ap ¼ 10 mm, Feed rate ¼ 5 mm rev1).

Figure 16 SEM photographs (250) of the curled chips produced with the EVC process under three spindle speeds: (a) 15 rpm; (b) 30 rpm; and
(c) 45 rpm (DOC, ap ¼ 10 mm, feed rate ¼ 5 mm rev1).

Therefore, it can be concluded from a speed ratio lower than the critical speed ratio that an Rscr of 0.128 produces improved or
nano surface from hard materials like Stavax. Although this material is not so brittle like sintered WC or glass, the ductile machining
model developed for the EVC process can still be applied for achieving ultraprecision surface while machining such cumbersome
materials.

11.17.3.3.2 Ultraprecision Machining of Hardened Steel Using PCD Tools

Obtaining ultraprecision (i.e., mirrorlike) surface from hardened steels for die and using inexpensive PCD tools instead of SCD
tools is the ultimate goal of this study. To achieve this using the EVC process, an evaluation test was conducted on the whole
workpiece face area (40 mm diameter, 1257 mm2 area) with 10 mm nominal DOC, 2.5 mm rev1 feed rate, and 15 rpm spindle
speed. The speed ratio at the OD is calculated as 0.0643, while it is zero at the center (Ref. Table 5). Figure 17 shows the picture of
the machined workpiece surface that reflects the word ‘NUS.’ It can be observed that the inner circular zone (approximately
806 mm2) can clearly reflect ‘NUS’ and is considered to be a mirrorlike surface, while the outer annular surface is a little blurred and
cannot give a clear reflection of ‘NUS.’ It may be because of vibration marks that appear at a higher speed ratio at a larger workpiece
diameter, as described earlier.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Cutting 469

Figure 17 Mirror surface of hardened stainless steel (Stavax, 49 HRC) machined by using a PCD tool with the EVC process (10 mm DOC,
2.5 mm rev1 feed rate, and 15 rpm spindle speed).

The surface profile of the machined workpiece face is analyzed by the Zygo white light interferometer. The surface roughness is
measured at several radial locations shown in Figure 18. It can be noticed that the entire machined surface can be apparently divided
into two zones, the inner circular zone and the outer annular zone, in terms of surface roughness values. For the outer annular zone,
where the radial distance is larger than 16 mm, the mean value of measured Ra was found to be about 30 nm. On the other hand, the
Ra value within 16 mm diameter (about 804 mm2 area) was found to be about 11 nm.
The photograph of the flank face/nose shown in Figure 19 was also captured by the optical microscope to observe the wear of the
PCD tool used in this evaluation test. It can be seen that the tool flank wear width is still not significant after machining such
hardened stainless steel for a surface area of 1257 mm2 with a cutting length of 533.3 m. Such machining area might be considered
enough for die and molding applications in the manufacturing industries. Therefore, it can be said that PCD tools under the EVC

OD = 40 mm

4 mm

Measurement
window Ra = 11 nm Ra = 30 nm

Figure 18 Average surface roughness values at different locations of the machined surface.

Figure 19 Microscope photographs (100 and 500) of the worn PCD tool flank/nose after machining 1257 mm2 of hardened steel with the EVC
process.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
470 Ultrasonic Vibration Cutting

technology are capable for long-distance machining to achieve mirrorlike surface from hardened steel, if the cutting parameters are
carefully considered based on the model developed in Section 11.17.2.
In summary, it is found that the ductile machining model developed for hard and brittle materials in Section 11.17.2 can also be
applied during machining of hardened steels like Stavax (49 HRC). Commercial PCD tools, instead of expensive SCD tools, show
their potential in producing mirrorlike/ultraprecision surface at an average surface roughness of about 10 nm when machining
a larger area about 800 mm2 even with a high nominal DOC like 10 mm. These tools did not wear of more than 20 mm and thus they
can satisfy the demand of die and mold manufacturing industries. This also proves that the developed ductile machining model is
versatile.

11.17.4 Analytical Force Model for the Orthogonal EVC Process

During machining, particularly, planning and turning with the CC process, it is understood that the TOC of the work material,
which is known also as nominal uncut chip thickness or DOC, remains constant. This constant TOC and the cutting speed or
velocity that is given to either the workpiece or the tool form an invariant shear plane or shear zone in this CC process. On the other
hand, during machining with the EVC process, as described in Sections 11.17.2 and 11.17.3, the TOC of the workpiece materials
along the tool rake face nonlinearly increases from zero value in each vibration cutting cycle, and it never reaches the value of the
given nominal uncut chip thickness due to the elliptical movement of the tool. The tool velocity also becomes variable or transient
due to this elliptical motion throughout the cutting cycle (14). By analyzing the chip formation process using the finite element
simulation, researchers (22) found that such continuous variations of both the transient TOC and the transient tool velocity in each
EVC cycle leads to a different cutting mechanism from the one that is applied to the CC process.
Since 1994, very few studies were conducted to investigate the basic cutting mechanics and predict the cutting performance of the
EVC technique based on the literature reported in (6). In the previously developed analytical model, Shamoto et al. (14) proposed
calculation methods for determining time instants at various critical tool locations, average shear angle, and the transient cutting
force values for the EVC process. However, two important inherent phenomena, i.e., the transient TOC (hereafter, symbolized as
TOCt) and the transient shear angle that is induced by the transient tool velocity, were not considered in their model. Besides,
although the characteristic of friction reversal in the EVC process has been analyzed in their study, the transition characteristic of
such reversal process was overlooked. In order to fully understand the cutting mechanics of the EVC process and to more accurately
predict the transient cutting force values, an analytical force model (11) that has recently been developed by the authors for the
orthogonal EVC process is presented. In the model, three important factors (1) transient TOC, (2) transient shear angle, and (3)
transition characteristic of the reverse friction between the tool and the chip were analyzed mainly based on geometric modeling
and the Lee and Shaffer’s slip-line solution are discussed. The model was also validated with the experiments by varying different
cutting and vibration conditions.

11.17.4.1 Development of the Force Model

The orthogonal EVC process presented in Figure 2 is reconsidered with the transient TOC (TOCt), as shown schematically in
Figure 20. In the orthogonal EVC process, the vibration modes, the cutting forces, and the nominal cutting velocity are perpen-
dicular to the tool edge (14). In each cutting cycle, the tool edge starts cutting at point A on the machined surface left by the previous
cycle, reaches the bottom point B on the cycle, passes point D and the friction reversal point E, and finally ends this cutting cycle at
point F. Point D corresponds to the location where the maximum TOCt (symbolized as (TOCt)m or TOCm defined in Figure 2) is
found. At point E, the friction force between the tool rake face and the chip becomes zero, and thereafter, the friction direction starts
reversing.
Using eqns [1] and [2], the tool velocity, x0 and y0 , along the x- and y-axes, respectively, relative to the workpiece can be found as
the time derivative of the tool position:
x0 ðtÞ ¼ ua sinðutÞ  vc [26]

y0 ðtÞ ¼ ub sinðut þ fÞ [27]

The transient tool velocity angle q(t) is defined as the angle of the transient tool velocity vector !
v t relative to the negative x-axis
(see Figure 20):
y0 ðtÞ
qðtÞ ¼ tan1 [28]
x0 ðtÞ
It can be seen from Figure 20 that the tool approaches the work material with negative velocity angle from point A until it reaches
point B, where the tool velocity direction is parallel to the x-axis.

11.17.4.1.1 Transient Thickness of Cut (TOCt)

As seen in Figure 20, while the given nominal uncut chip thickness or DOC ap in the EVC process is always measured from the
bottom of the vibration cycle to the workpiece free surface, the TOCt value at an instant in the EVC process is calculated with respect

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Cutting 471

y
b x
a
Chip

Tool
TOCt

(TOCt)m
ap

y
O
F t0 x

–b E
T
D

B A
Workpiece

Figure 20 Illustration of the EVC process with the instantaneous transient thickness of cut (TOCt) on the tool rake. Reproduced from Zhang, X. Q.;
Kumar, A. S.; Rahman, M.; Nath, C.; Liu, K. An Analytical Force Model for Orthogonal Elliptical Vibration Cutting. J. Manuf. Process. 2012, 14 (3),
378–387.

to the tool vibration locus. After the instant when the tool edge is in contact with the (TOCt)m point D, TOCt is calculated along the
y-axis from the tool edge to the workpiece free surface. On the other hand, before that critical instant, TOCt has to be calculated in
a different way, which is schematically shown in Figure 21. Point T represents the instant position of the tool edge, and point P
represents the contact location of the tool rake face on the surface machined by the previous cutting cycle. Hence, the value of TOCt
before the tool edge passes point D can be calculated as the height of TP.
In Figure 21, the moments when the tool edge passes the points A, B, T, D, and F are symbolized as tA, tB, t, tD, and tF, respectively,
and the instant it passes the point P in the previous cutting cycle is designated as tP. The definitions of those time points are listed in
Table 6.
Detailed calculation procedures for solving the time instants tA, tD, and tF can be found in researchers’ previous study (14), and
the value of tP can be calculated from the following equation given a specific value of t (tA < t < tD) by applying the Newton–
Raphson method:

xðtÞ  xðtP Þ 2p
¼ tan g; 0 < ðt  tP Þ < [29]
yðtÞ  yðtP Þ u

Chip

ap

Tool
F

P
TOC t D
T
B A

Figure 21 The TOCt (height TP) at an instant before the tool edge passes the (TOCt)m point D.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
472 Ultrasonic Vibration Cutting

Table 6 Definition of the time points in EVC process (Ref. Figure 21)

tP The moment tool edge passes the transient contact point P in previous cycle
tA The moment tool edge passes the cutting-start point A
tB The moment tool edge passes the bottom point B
t The moment tool edge passes the transient tool–workpiece contact point T
tD The moment tool edge passes the (TOCt)m point D
tE The moment tool edge passes the friction reversal point E
tF The moment tool edge passes the cutting-end point F

where g is the tool rake angle. According to the geometrical relationships shown in Figures 20 and 21, the TOCt value at different
locations in a cutting cycle can be calculated as
8
>
< 0; t < tA ; t  tF
TOCt ¼ yðt P Þ  yðtÞ; tA  t < tD [30]
>
:
ap  ðyðtÞ  yðtB ÞÞ; tD  t < tF

11.17.4.1.2 Transient Shear Angle and Transition Characteristic of Friction Reversal

According to the famous thin shear plane theory (23,24), the velocity relationship in metal cutting process can be expressed as
follows:
!vs ¼! vt þ! v ct [31]
where ! v s is the transient shear velocity, !
v t represents the transient tool velocity relative to the workpiece, and ! v ct is the transient
chip velocity relative to the tool.
For the CC technique, the tool velocity ! v t always stays invariant in direction and the shear velocity !v s and the chip velocity ! v ct
also remain constant in direction, as shown in Figure 22. Due to the sliding movement of the chip on the tool rake face throughout
the CC process, kinetic friction is induced and tends to push the chip material down toward the workpiece.
For the EVC process, each cutting cycle starts with the sliding movement of the chip or work material over the tool rake face,
which is similar to that in the CC technique. However, as ! v t continuously varies in its direction clockwise, the direction of ! v ct or
the relative sliding movement will get reversed in direction sooner or later. Due to such reverse-direction sliding movement, the
friction direction will be reversed subsequently, and such reverse-direction friction can eventually pull up the formed chip material
along the tool rake face.
As each EVC cycle starts with the sliding movement like the CC process and ends with the reverse-direction sliding movement, it
can be stated that two kinetic-friction zones with the sliding movement exist in each cutting cycle. The first kinetic-friction zone
(named as CC-like kinetic-friction zone) exists in the beginning portion of each cycle, which starts at tA, while the other kinetic-
friction zone (named as reverse kinetic-friction zone) exists in the ending portion, which closes at tF. Due to the direction
reversal of ! v ct and the kinetic-friction force, these two kinetic-friction zones will have different shear velocities ! v s and hence
different shear plane angles.
Lee and Shaffer (25) derived that in cutting process the shearing occurs along the maximum shear stress (MSS) direction, i.e., the
angle between the shear velocity and the direction of resultant force is 45 . Although their model is based on an assumption of rigid
perfectly plastic work material and neglects effects of built-up edge and thick shear zone on the rake face, it still gives a compre-
hensive physical understanding of metal cutting mechanics. Hence, in the present study, in order to determine the shear plane
angles and boundaries of the two kinetic-friction zones, Lee and Shaffer’s slip-line solution was extended and applied.
Based on Lee and Shaffer’s model, both the slip-line field and force relationships for CC-like kinetic-friction zone are plotted in
Figure 23(a). A triangular plastic region OMN above the shear direction with the slip lines parallel or perpendicular to OM is
assumed, and the resultant force R makes 45 with the direction of the shear force Fs. The other symbols, Fns, Fn, Fp, and Ft represent
the normal force perpendicular to the shear direction, the normal force perpendicular to the rake face, the principal force along the

γ
vct vs Chip

tc
vt
S vct
Tool
Workpiece t0 −v t ϕ
c
O

Figure 22 Velocity diagram in the CC process.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Cutting 473

Figure 23 Slip-line fields and force relationships for a single EVC cycle in (a) CC-like kinetic-friction zone and (b) reverse kinetic-friction zone.

nominal cutting direction, and the thrust force perpendicular to the nominal cutting direction, respectively. bk represents the kinetic-
friction angle between the tool rake face and the work material, and its value is assumed to remain constant throughout the cutting
process. According to the force relationships in CC process (23), bk can be derived from the following equation:
   
 
bk ¼ tan1 Ff =Fn ¼ tan1 Ft cos g þ Fp sin g Fp cos g  Ft sin g [32]

Although Lee and Shaffer’s model was developed for the CC process, the same fundamental laws of plasticity and MSS
principle can also be applied to obtain the slip-line field for the reverse kinetic-friction zone. Figure 23(b) shows the derived slip-
line field and the force relationships for reverse kinetic-friction zone. Due to the reverse-direction sliding movement mentioned
earlier, the friction force Ff in reverse kinetic-friction zone is opposite in direction to that in the CC-like kinetic-friction zone, and
accordingly, the direction of the resultant force R in Figure 23(b) should be rotated clockwise by 2bk relative to the one in
Figure 23(a).
As the kinetic-friction angle bk remains constant, the direction of R will stay invariant in each kinetic-friction zone, and hence the
shear force Fs will also stay unchanged in each zone according to the MSS principle. Hence, the shear angles (symbolized as fkc and
fkr) for CC-like and reverse kinetic-friction zones can be reasonably assumed to remain constant in each zone. Then, based on the
force relationships shown in Figure 23(a) and 23(b), the values of fkc and fkr can be determined by the tool rake angle g and the
kinetic-friction angle, bk as
4kc ¼ 45  ðbk  gÞ [33]

4kr ¼ 45 þ ðbk þ gÞ [34]

Given the determined shear angles, the velocity diagrams based on eqn [31] for the two kinetic-friction zones can be plotted in
Figure 24(a) and 24(c), where the two dashed lines OS and OS’ represent the two constant directions of ! v s . In Figure 24(a), !
v ct
will get close to zero when ! v t is approaching !v s in direction. Since kinetic friction only exists when the sliding movement occurs
(i.e., !
v ct is not zero), in CC-like kinetic-friction zone, the direction of ! v t should not rotate over OS. Otherwise, the relative
sliding movement and the chip velocity ! v ct may change in direction or in nature, and such variation could lead to the occurrence
of a reverse friction direction or another friction mode. Similar analysis can also be applied to reverse kinetic-friction zone, where
!v t should not stay below OS’ in order to allow the existence of the reverse-direction sliding movement, as shown in Figure 24(c).
Thus, the two shear directions OS and OS’ can be treated as the upper or lower boundaries of the two kinetic-friction zones,
respectively.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
474 Ultrasonic Vibration Cutting

Figure 24 Velocity diagrams for a single EVC cycle in (a) CC-like kinetic-friction zone, (b) static-friction zone, and (c) reverse kinetic-friction zone.

In mechanics, friction behavior can be divided into two main regimes: gross sliding and presliding, where kinetic friction and
static friction dominate the process, respectively. It is known that kinetic friction does not immediately reverse in direction from one
gross sliding movement to the reverse-direction gross sliding movement and presliding regime should exist as a transition zone
during the reversal process. In the EVC process, such transition zone (named as static-friction zone) acts as an intermediate zone
connecting the two kinetic-friction zones to prevent the chip from slipping on the tool rake face. Figure 24(b) shows the velocity
diagram for this transition zone, where its lower and the upper boundaries are assumed to be determined by the obtained two shear
directions in the two kinetic-friction zones, OS and OS’. Once ! v t passes over OS’, the static friction will get converted back to kinetic
friction due to the nonzero ! v ct and the induced gross sliding movement. Since no gross sliding movement occurs in this presliding
transition regime (i.e., !
v ct ¼ 0), according to the velocity relationship described in eqn [31], the shear velocity ! v s will be identical
!
to v t in both direction and magnitude. Based on such equivalence relation, it can be derived that the transient shear angle fs for this
transition zone is equal to the transient tool velocity angle:
!v ¼! v ; 04 ¼ qðtÞ [35]
s t s

Overall, based on the above analysis, the transient shear plane angle (symbolized as ft) in the three consecutive zones can be
described with specific zone boundaries as
8
< 4kc ; qðtA Þ  qðtÞ < 4kc
>
4t ¼ 4s ; 4kc  qðtÞ  4kr [36]
>
:
4kr ; 4kr < qðtÞ  qðtF Þ
Figure 25(a) and 25(b) show a schematic sketch for the three friction zones versus the tool velocity direction and the tool
location in an EVC cycle. It can be seen that the three consecutive zones, namely CC-like kinetic-friction zone, static-friction zone,
and reverse kinetic-friction zone, follow a kinetic-static-kinetic friction transformation procedure in each cutting cycle. In
Figure 25(a), the two dashed lines, OA and OF, represent the lower and upper boundaries of a complete cutting cycle, i.e., the
cutting-start tool velocity direction and the cutting-end tool velocity direction.
As an example, Figure 26 shows plots of simulated results of the transient shear angle ft versus the phase of vibration for an EVC
cycle for two given kinetic-friction angles bk using eqn [36]. In every experimental cutting test, bk depends on the specific cutting
conditions, such as material types of tool–workpiece combination and cooling condition. The two friction angles (15 and 30 ) in
this figure are identified from the friction coefficients of WC versus WC (0.25) and WC versus steel (0.6). The points A, F, S, and S0
represent the cutting-start point, the cutting-end point, and the lower and upper limits of static-friction zone, respectively. From this
figure, it can be remarked that static-friction zone plays a significant role in the whole cutting cycle. Moreover, it can also be followed
that a smaller kinetic-friction angle can lead to comparatively (1) a larger conventional shear angle fkc in the CC-like kinetic-friction
zone, (2) a shorter phase length in the static-friction zone, and (3) a smaller shear angle (fkr) in the reverse kinetic-friction zone.
Such characteristic of the cutting mechanics with the EVC technique looks completely different from that with the CC technique,
where the latter is characterized by an invariant shear angle through the CC process.

11.17.4.1.3 Transient Cutting Force Components

It is understood that the output cutting force during mechanical cutting process highly depends on material load on the tool rake
face. The higher the TOC, the higher is the material load. Due to the variation of TOC and the variation of shear angle in the EVC

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Cutting 475

Figure 25 Schematic sketch of the three consecutive friction zones in an EVC cycle versus (a) tool velocity direction and (b) tool location.

Time (s)
0 ...1/8f 1/4f 3/8f 1/2f 5/8f...1/f
90
o
k= 15
80 S' F
k= 30o
Transient shear angle (deg)

60

CC-like kinetic-friction zone Reverse kinetic-friction zone

40

20
Static-friction
A S
0

Phase of vibration (rad)

Figure 26 Example of transient shear angle for a cutting cycle in orthogonal EVC process at the conditions 0 tool rake angle, 0.094 mm min1
nominal cutting speed, vibration amplitude (a ¼ 20 mm, b ¼ 5 mm), 0.25 Hz frequency, and 90 phase shift.

process, the cutting force varies continuously in each cutting cycle. In fact, according to the illustration of TOCt in Figure 1, the
transient force value starts from zero at the cutting-start point A, reaches the maximum at elsewhere in the loci, and ends with zero
again at the cutting-end point F. In order to derive such transient cutting force, the famous thin shear plane model (23) is applied
and slightly modified here.
It is known that in the CC process, the shear force along the fixed shear plane is calculated as
sw
Fs ¼ t0 [37]
sin 4c

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
476 Ultrasonic Vibration Cutting

where t0 is the fixed uncut chip thickness, 4c is the constant shear angle, s is the shear stress, and w is the width of cut. For the EVC
process, if t0 and 4c in eqn [37] are replaced with TOCt and 4t, the transient shear force Fs(t) can be obtained as
sw
Fs ðtÞ ¼ TOCt [38]
sin 4t
The value of Fs(t) can be calculated with given values of TOCt and 4t, which are determined by eqns [30] and [36], respectively.
According to the Lee and Shaffer’s MSS principle (25), the angle between the shear direction and the resultant force direction is
45 . Therefore, the transient resultant force R(t) during cutting in each cycle can be obtained as
Fs ðtÞ
RðtÞ ¼ [39]
cos 45
Then, according to the force relationships shown in Figure 19, the transient principal force Fp(t) and the transient thrust force
Ft(t) along the x- and y-axes, can be derived as follows:
Fp ðtÞ ¼ RðtÞcosð45  4t Þ [40]

Ft ðtÞ ¼ RðtÞsinð45  4t Þ [41]

In order to calculate the values of these two transient cutting force components using eqns [40] and [41], the values of bk and s
need to be determined from the experimental results of the orthogonal CC test. In this study, bk and s in the EVC process are
assumed to be constant and equal to those found with the CC technique. Under the CC condition, the kinetic-friction angle bk can
be calculated from eqn [32] with given values of Fp and Ft, and the shear stress s can be derived according to the conventional thin
shear plane model (23):
   qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
R cos 4c þ tan1 Ft =Fp
s¼ ; where R ¼ Ft 2 þ Fp 2 [42]
wðt0 =sin 4c Þ
where 4c can be derived from the following equation, based on the study in (23):
t0 cos g
4c ¼ [43]
tc  t0 sin g
where tc is the measured chip thickness.
With the values of bk and s calibrated through CC tests, the transient cutting force components for the EVC process can be
subsequently calculated. Figure 27 shows a flowchart of the whole calculation procedures of the transient force components based
on the proposed model. In the following sections, experiments are conducted to substantiate the force model.

11.17.4.2 Experimental Setup

First, an orthogonal CC test (i.e., without vibration of the tool) was conducted to calibrate the two important parameters (s and bk).
Then, an orthogonal EVC test was conducted with the EVC cycles generated through G-code programming and axis motion control
of an ultraprecision machine tool.

Figure 27 Flowchart of the calculation procedures for the analytical EVC force model.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Cutting 477

The workpiece used in the orthogonal CC and EVC tests was prepared to have a 400 mm width rectangular strip with the base
material of 6061 aluminum alloy. A customized flat-nose SCD tool was used, which has a 0.76 mm nose width, 0 rake angle, and
7 clearance angle, as shown in Figure 28. Its tool edge radius was measured to be about 0.15 mm by applying an indentation
method proposed by Li et al. (26).
The elliptical vibration motion of the cutting tool relative to the workpiece is generated by a combined movement of the x- and
y-axes of a CNC ultraprecision machine tool (Toshiba ULG-100H3). The machine tool has a 10 nm linear resolution along the x-, y-,
and z-axes, which makes it reasonable to accurately drive the low-frequency EVC motion. In order to generate the elliptical vibration
motion of tool tip, first, some ideal elliptical vibration loci are created by employing eqns [1] and [2] with preset and suitable cutting
and vibration conditions. Then, the elliptical loci are fitted by sampling 360 points for each cutting cycle (i.e., every 2p phase), and
G-codes are generated based on coordinates of the derived points.
Table 7 shows detailed conditions for the orthogonal CC test for calibration and the orthogonal EVC tests for model verification.
According to researchers’ studies (5,6,14), the speed ratio expressed in eqn [4] is considered as one of the most essential parameters
in the EVC process. In the authors’ recent study (21) on machining hardened steel using PCD tools with the ultrasonic EVC
technique, it was found that among all the machining parameters, the speed ratio has the most significant effects on cutting
performances (e.g., time-averaged cutting force, tool wear, and surface roughness). Therefore, in the orthogonal EVC tests, different
speed ratios are employed to study the applicability of the proposed model.
In order to measure the cutting force components during cutting tests, a three-component minidynamometer (Type 9256A1)
that has a 1 mN resolution was attached to the tool post. The generated amplified force signal was captured using a real-time
recorder (DEWE 2010). A schematic illustration of the experimental setup for the cutting tests is shown in Figure 29.

Figure 28 Photograph of the flat-nose diamond tool used in the cutting tests.

Table 7 Cutting and vibration conditions for the orthogonal CC and EVC tests

Orthogonal CC test Uncut chip thickness, t0 (mm) 15

Width of cut, w (mm) 400
Cutting speed, (mm min1) 200
Orthogonal EVC tests Nominal uncut chip thickness, ap (mm) 15
Width of cut, w (mm) 400
Vibration frequency, f (Hz) 0.25
Phase shift, 4 ( ) 90
Tangential vibration amplitude, a (mm) 20
Thrust vibration amplitude, b (mm) 5
Speed ratio, Rs (nominal cutting speed, vc (mm min1)) 0.025 (0.047) 0.05 (0.094) 0.075 (0.141) 0.1 (0.188)

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
478 Ultrasonic Vibration Cutting

Vacuum Flat-nose
chuck diamond tool

Fixture Force
dynamometer

Workpiece

Figure 29 Experimental setup for the orthogonal CC and EVC tests.

11.17.4.3 Validation of the Developed Force Model

In this section, the proposed analytical force model for the orthogonal EVC process was experimentally validated. Using the cali-
brated parameters, the proposed force model was employed to predict values of the transient cutting force components, and these
values were compared with the experimental force values obtained from the EVC test.

11.17.4.3.1 Calibration of the Parameters

For the orthogonal CC test, Figure 30(a) shows the recorded values of the cutting force components and Figure 30(b) shows
photograph of the formed chip in the CC process. Based on the experimental results, the average thrust force Ft and the average
principal force Fp were calculated to be 0.4 N and 6.6 N, respectively. The average thickness of the generated chip tc was measured to
be about 41 mm. Therefore, by substituting the values of Ft, Fp, tc, w, and t0 into eqns [32], [42] and [43], the constant shear angle fc
is calculated to be 20 and the values of bk and s for the CC condition are found to be about 3.46 and 347 MPa, respectively.

11.17.4.3.2 Model Validation

By following the calculation method shown in Figure 27, the transient cutting forces are calculated given the cutting and vibration
conditions listed in Table 7 and the two calibrated parameters bk and s. When the speed ratio of EVC varies, the transient cutting
force will change in various terms, e.g., maximum transient resultant force (symbolized as Rmax), friction reversal time, and so on. In
order to get a simplified and clear comparison between the predicted and experimental cutting force, Figure 31 shows the variation
of Rmax at the four different speed ratios. Each value of Rmax is calculated by averaging its values of five consecutive EVC cycles, and
the transient cutting force shows good repetitiveness. It can be seen that both the predicted and experimental values of Rmax increase
with the increment of speed ratio, which is considered to be caused by the reduced TOCt and the increased transient shear angle.
Results in Figure 31 show that the predicted values of Rmax correspond well with the experimental values, indicating that the effect
of different speed ratios on transient cutting force can be effectively predicted by the proposed force model.
Figure 32 shows the predicted transient principal and thrust force (i.e., Fp(t) and Ft(t)) with the four speed ratios (0.025, 0.05,
0.075, and 0.1), and the retrieved experimental force values are superimposed in the same figure for comparison. It can be seen from
Figure 32 that the transient principle force first increases due to the increasing TOCt (see Figure 20), then decreases slightly due to
both the increasing ft (see Figure 26) and the decreasing TOCt, and finally drops to zero after the disengagement of tool and chip.
For the predicted principal force, it can be noticed that its variation trend and zero-to-peak value correspond well with those of the
experimental one, although the experimental principal force is slightly larger than the predicted one at some stage of the cutting
phase. It can also be noticed from Figure 32 that the transient thrust force first starts increasing from zero to a maximum positive
value, but then gradually decreases to negative values, indicating that the tool is pulling the chip or work material upward from the
workpiece. Such unique phenomenon, which is completely different from the CC process, is caused by the friction reversal in each
cutting cycle of the EVC process.
The predicted thrust force corresponds very well with the experimental one in the beginning portion of the cutting cycle.
However, the friction reversal point (i.e., the point where the thrust force is zero) for the predicted and experimental thrust force
differs from each other significantly, and the experimental thrust force starts reversing prior to the predicted moment. The reason for
such outcome may lie in the imperfection of Lee and Shaffer’s slip-line solution, which might not be applicable to accurately predict
the shear angle, especially at a low cutting speed with strong work hardening effect (27). Due to the comparatively smaller value of
true fkc, the actual relative chip velocity ! v ct may change into zero at an earlier point, and the true friction starts reversing in
direction prior to the predicted one, which subsequently results in the earlier reversal of transient thrust force shown in Figure 32.
From Figure 32 it can also be observed that the predicted cutting-end point (the point where the principal or the thrust force
becomes zero) does not agree with the experimental one, indicating that the tool still stays engaged with the work or chip material
after passing over the theoretical cutting-end point. This kind of disaccord is considered to be caused by elastic recovery of either the

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Cutting 479

6
Principal force
5
Thrust force

Cutting force (N)

4

0
Cutting starts
–1
0 0.5 1 1.5 2 2.5 3
Time (s)
(a)

(b)

Figure 30 Experimental results for the CC process: (a) cutting force components and (b) microscope photograph (X450) of the formed chip.

Figure 31 Experimental and predicted maximum transient resultant cutting force with different speed ratios.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
480 Ultrasonic Vibration Cutting

(a) (b)
5 5

Pred. force Pred. force

4 Principal force
4 Expt. force
Principal force Expt. force

3 3
Force (N)

Force (N)
2 2

1 1

0 0
Thrust force
Thrust force
–1 –1
0 /2 3 /2 2 0 /2 3 /2 2
Phase of vibration (rad) Phase of vibration (rad)

(c) (d)
5 5
Principal force
Principal force Pred. force
4 Pred. force 4
Expt. force
Expt. force

3 3
Force (N)
Force (N)

2 2

1 1

0 0
Thrust force Thrust force

–1 –1
0 /2 3 /2 2 0 /2 3 /2 2
Phase of vibration (rad) Phase of vibration (rad)

Figure 32 Experimental and predicted transient cutting force components for an EVC cycle for (a) Rs ¼ 0.025; (b) Rs ¼ 0.05; (c) Rs ¼ 0.075; and
(d) Rs ¼ 0.1.

work or the chip material at this stage. That means when the tool edge passes the theoretical cutting-end point F (see Figure 20), the
work or chip material will gradually rebound elastically with the receding movement of the tool, until its elastic energy is completely
released and it is truly disengaged from the tool rake face.

11.17.5 Conclusion

In this chapter, the ductile machining model and the analytical force model for the orthogonal EVC process recently developed by
the authors are presented. The models are also validated by machining experiments of different materials. The conclusions are as
follows.
The ductile machining model developed in this chapter (Section 11.17.2) shows that the TOCm of the workpiece material in each
cutting cycle determines the nature of the machined surface (i.e., ductile or brittle). The TOCm value can be controlled by four
machining parameters, namely workpiece cutting speed, tool vibration frequency, and the tool vibration amplitudes in the
tangential and thrust directions. The TOCm value becomes significantly smaller against the nominal DOC when the speed ratio that
is formulated by the first three parameters is set within the critical speed ratio 0.128 37. The cycle-overlap between the first and the
third cycles is desired to obtain a reduced TOCm under the EVC process. It is also observed that high thrust directional vibration

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Ultrasonic Vibration Cutting 481

amplitude increases the value of TOCm. To achieve ductile finishing of brittle materials, the condition TOCm < DOCcr must be
satisfied. This phenomenon is found to be in good agreement with the theoretical relationship when a hard and brittle material
sintered WC was machined. The findings confirm that ductile mode cutting can be achieved even at a higher nominal DOC when
the TOCm is set lower than the DOCcr by controlling these four relevant parameters in the EVC process.
The ductile machining model developed for hard and brittle materials is also applied to a hardened steel, namely Stavax
(49 HRC), for justifying its versatility for machining different hard materials and for studying the capability of application of PCD
tools, instead of expensive SCD tools, for reducing machining cost where achieving ultraprecision surface is the ultimate goal.
Experiments show that commercial PCD tools have potential in producing mirrorlike/ultraprecision surface at an average surface
roughness of about 10 nm when machining a larger area about 800 mm2 even with a high nominal DOC like 10 mm. These tools
did not wear off more than 20 mm and thus they can satisfy the demand of die and mold manufacturing industries. This proves that
the developed ductile machining model is versatile. The established relationships will be good guides for researchers and manu-
facturers for predicting a desired TOCm value with respect to the relevant machining parameters in the EVC process and to apply in
ultraprecision/ductile machining of various hard and brittle materials.
In the analytical force model for orthogonal EVC process, an in-depth understanding of the transient cutting mechanics is
presented for accurate prediction of the transient cutting force components. A calculation method for the value of TOCt is developed
based on the geometrical relationship in the EVC process. Based on variation of the friction modes, each cutting cycle is divided into
three consecutive zones: CC-like kinetic-friction zone, static-friction zone, and reverse kinetic-friction zone, and a calculation
method of the transient shear angles for the three zones is derived by investigating the relationship of transient velocities and
employing the Lee and Shaffer’s slip-line solution. Finally, the model is validated by comparing the experimental transient cutting
force values with the predicted ones, and they are found to be in good agreement. The model can finely express the EVC mechanism
and assist to predict accurate cutting force values. Therefore, this study is supposed to be helpful for researchers in this EVC
technology.

References

1. Nath, C.; Rahman, M.; Neo, K. S. A Study on Ultrasonic Elliptical Vibration Cutting of Tungsten Carbide. J. Mater. Process. Technol. 2009, 209 (9), 4459–4464.
2. Nath, C.; Rahman, M.; Neo, K. S. Machinability Study of Tungsten Carbide with PCD Tools Applying Ultrasonic Elliptical Vibration Cutting Technique. Int. J. Mach. Tool Manuf.
2009, 49 (14), 1089–1095.
3. Nath, C.; Rahman, M.; Neo, K. S. A Study on the Effect of Tool Nose Radius in Ultrasonic Elliptical Vibration Cutting of Tungsten Carbide. J. Mater. Process. Technol. 2009,
209 (17), 5830–5836.
4. Nath, C.; Rahman, M.; Neo, K. S. Enhancing the Performance of Polycrystalline Diamond Tools for Machining WC by Ultrasonic Elliptical Vibration Cutting Method. J. Vac. Sci.
Technol. – B 2009, 27 (3), 1241–1246.
5. Nath, C.; Rahman, M.; Neo, K. S. Modeling of the Effect of Machining Parameters on Maximum Thickness of Cut in Ultrasonic Elliptical Vibration Cutting. J. Manuf. Sci. Eng. –
Trans. ASME 2011, 133 (1), 011007–011008.
6. Brehl, D. E.; Dow, T. A. Review of Vibration-Assisted Machining. Precis. Eng. 2008, 32 (3), 153–172.
7. Liu, K.; Li, X. P.; Rahman, M. Characteristics of High Speed Micro-Cutting of Tungsten Carbide. J. Mater. Process. Technol. 2003, 140, 352–357.
8. Suzuki, N.; Masuda, S.; Haritani, M.; Shamoto, E. Ultraprecision Micromachining of Brittle Materials by Applying Ultrasonic Elliptical Vibration Cutting. In Proceedings of the
International Symposium on MHS, Japan; 2004; pp 133–138.
9. Suzuki, N.; Haritani, M.; Yang, J.; Hino, R.; Shamoto, E. Elliptical Vibration Cutting of Tungsten Alloy Molds for Optical Glass Parts. CIRP Ann. – Manuf. Technol. 2007, 56 (1),
127–130.
10. Shamoto, E.; Moriwaki, T. Ultraprecision Diamond Cutting of Hardened Steel by Applying Elliptical Vibration Cutting. CIRP Ann. – Manuf. Technol. 1999, 48 (1), 441–444.
11. Zhang, X. Q.; Kumar, A. S.; Rahman, M.; Nath, C.; Liu, K. An Analytical Force Model for Orthogonal Elliptical Vibration Cutting. J. Manuf. Process. 2012, 14 (3), 378–387.
12. Ma, C.; Shamoto, E.; Moriwaki, T.; Wang, L. Study of Machining Accuracy in Ultrasonic Elliptical Vibration Cutting. Int. J. Mach. Tool Manuf. 2004, 44 (12–13), 1305–1310.
13. Shamoto, E.; Morimoto, Y.; Moriwaki, T. Elliptical Vibration Cutting (2nd Report): Study on Effects of Vibration Conditions. J. Jpn. Soc. Precis. Eng. (JSPE) 1999, 65 (3), 411–416.
14. Shamoto, E.; Suzuki, N.; Hino, R. Analysis of 3D Elliptical Vibration Cutting with Thin Shear Plane Model. CIRP Ann. – Manuf. Technol. 2008, 57 (1), 57–60.
15. Shamoto, E.; Moriwaki, T. Study on Elliptical Vibration Cutting. CIRP Ann. – Manuf. Technol. 1994, 43 (1), 35–38.
16. Ma, C.; Shamoto, E.; Moriwaki, T.; Zhang, Y.; Wang, L. Suppression of Burrs in Turning with Ultrasonic Elliptical Vibration Cutting. Int. J. Mach. Tool. Manuf. 2005, 45 (11), 1295–1300.
17. Li, X.; Zhang, D. Ultrasonic Elliptical Vibration Transducer Driven by Single Actuator and Its Application in Precision Cutting. J. Mater. Process. Technol. 2006, 180 (1–3), 91–95.
18. Kim, G. D.; Loh, B. G. An Ultrasonic Elliptical Vibration Cutting Device for Micro V-Groove Machining: Kinematical Analysis and Micro V-Groove Machining Characteristics.
J. Mater. Process. Technol. 2007, 190 (1–3), 181–188.
19. Li, X. P.; Cai, M. B.; Liu, K.; Rahman, M. Characteristics of Ductile Mode Chip Formation in Nanoscale Cutting of Brittle Materials. Int. J. Abrasive Technol. 2007, 1 (1), 37–58.
20. Nath, C.; Rahman, M.; Neo, K. S. Ductile Mode Machining of Tungsten Carbide Using Ultrasonic Elliptical Vibration Cutting Technique. In Proceedings of the 3rd Asian
Symposium for Precision Engineering and Nanotechnology (ASPEN), 11–13 November, Kitakyushu, Japan; 2009.
21. Zhang, X. Q.; Kumar, A. S.; Rahman, M.; Nath, C.; Liu, K. Experimental Study on Ultrasonic Elliptical Vibration Cutting of Hardened Steel Using PCD Tools. J. Mater. Process.
Technol. 2011, 211 (11), 1701–1709.
22. Amini, S.; Shamoto, E.; Suzuki, N.; Nategh, M. J. FE Analysis of One-Directional and Elliptical Vibration Cutting Processes. Int. J. Autom. Technol. 2010, 4, 252–258.
23. Merchant, M. E. Mechanics of the Metal Cutting Process. I. Orthogonal Cutting and a Type 2 Chip. J. Appl. Phys. 1945, 16, 267–275.
24. Oxley, P. L. B. The Mechanics of Machining: An Analytical Approach to Assessing Machinability; E. Horwood: Chichester, 1989.
25. Lee, E. H.; Shaffer, B. W. The Theory of Plasticity Applied to a Problem of Machining. J. Appl. Phys. 1951, 18, 405–413.
26. Li, X. P.; Rahman, M.; Liu, K.; Neo, K. S.; Chan, C. C. Nano-Precision Measurement of Diamond Tool Edge Radius for Wafer Fabrication. J. Mater. Process. Technol. 2003, 140,
358–362.
27. Oxley, P. Shear Angle Solutions in Orthogonal Machining. Int. J. Mach. Tool Des. Res. 1962, 2, 219–229.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
This page intentionally left blank

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 11.18 Electrolytic In-Process Dressing (ELID) Grinding
for Nano-Surface Generation
T Saleh, International Islamic University Malaysia, Kuala Lumpur, Malaysia
M Rahman, National University of Singapore, Singapore
 2014 Elsevier Ltd. All rights reserved.

11.18.1 Introduction 483

11.18.2 Nano-Surface Generation by Grinding 483
11.18.3 In-Process Dressing of Super Abrasive Wheel 484
11.18.4 History of ELID Grinding 486
11.18.5 Classifications of ELID Grinding 488
11.18.5.1 Electrolytic In-Process Dressing 488
11.18.5.2 Electrolytic Interval Dressing 489
11.18.5.3 Electrodeless ELID 491
11.18.5.4 Electrodeless ELID 491
11.18.5.5 Ion Shot ELID 491
11.18.6 Fundamental Study on ELID Grinding 492
11.18.7 Study Related to Control of ELID Grinding Process 499
11.18.8 Study Related to ELID Grinding Machine Development 504
11.18.9 Study Related to Application of ELID Grinding 508
11.18.9.1 ELID on Ceramics 508
11.18.9.2 ELID on Coated Film 510
11.18.9.3 ELID on Metal 511
11.18.9.4 ELID on Optical Glasses 514
11.18.9.5 ELID on Silicon Wafer 517
11.18.10 Conclusions 520
References 520

11.18.1 Introduction

Grinding is a machining process which uses hard abrasive particles for cutting, where the surface speed of the abrasive tool (namely
the grinding wheel) is relatively faster compared to other machining processes such as turning and milling. Technically, grinding is
a subset of cutting where each grain of abrasive functions as a microscopic single-point cutting edge with a high negative rake angle.
It is a very old process and has been in existence since the Stone Age when man rubbed stones against each other to produce sharp
weapons. This process is conventionally regarded as a preliminary finishing process to obtain a high degree of dimensional accuracy
and surface finish on a part.

11.18.2 Nano-Surface Generation by Grinding

Rapid progress in electronic and optical devices (such as IC chips, MEMS, smart sensors, and lenses) with the increasing requirement
of their performance, demand of an ultraprecision surface finish of the specialized glass, silicon wafer in nano-scale is growing.
Conventional grinding cannot produce such surfaces with precise form accuracy; therefore, conventional grinding is still considered
to be a preliminary finishing process. Lapping and polishing are typical ways for achieving final finishing of different materials.
However, these methods also have disadvantages such as poor grindability, waste water problem, mechanical damages, etc. (1).
Chemical-mechanical polishing (also known as CMP) is another final finishing process for silicon wafer preparation in the IC
industries. Some of the disadvantages associated with the CMP process (2) are:
1. Low efficiency due to low removal rates,
2. Nonuniform wafer surface due to the variation in relative cutting speed across the wafer surface, and
3. Relatively high cost involved in this process.
In order to overcome all of these drawbacks associated with the conventional finishing processes, demand for nano-surface
generation by grinding (using super abrasives) to replace the polishing, lapping, etc. have been growing stronger and stronger in the
manufacture of optical and electrical parts. With the aim of meeting such demand, efforts are actively being made to develop high
precision grinding machines, new grinding wheels, and to review the applications of these new techniques in the field of
manufacturing.

Comprehensive Materials Processing, Volume 11 http://dx.doi.org/10.1016/B978-0-08-096532-1.01112-2 483

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
484 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation

11.18.3 In-Process Dressing of Super Abrasive Wheel

Grinding with super abrasive wheel is an efficient method of achieving nano-surface finish on hard and brittle materials; however,
the method has some difficulties associated with it. One of the major problems is the preparation of the bonding matrix for the
super abrasives. The abrasives also known as grits need to be firmly held by the bond material. Therefore, the bond material has to
be hard in nature which reduces grinding wheels’ self-sharpening and self-truing ability. Therefore, an additional dressing mech-
anism is needed to ensure good protrusion of the sharp cutting grits. When we talk about attaining precision surface finish and form
accuracy using grinding, it is an obvious requirement to have the grinding wheel dressed in-process. There are several methods of in-
process dressing of the grinding wheel, though all of them can be subdivided into four main groups as listed below.
1. Mechanical contact method.
2. Electro thermal method.
3. Laser technology.
4. Electrochemical method.
In-process dressing with mechanical contact uses a separate dressing tool that comes in physical contact, with the grinding wheel.
The dressing tool may be mounted on a rotating spindle or it can also be held stationary (3,4).
Dressing of the grinding wheel by electro thermal method employs electro-discharge machining technique. Basically, there are
two available arrangements for this method of wheel dressing. One is the block electro-discharge method where a stationary solid
block is used as the dressing electrode and another one is the wire electro-discharge method where moving wire instead of a block is
used for dressing the wheel (5,6). Electro thermal method of dressing is mainly used for fine grit wheels.
The laser machining method is another technique to achieve the in-process dressing of the wheel. It is mainly used for the
vitrified bond grinding wheel. As the laser beam is applied, bond material becomes soft and is removed during grinding and ensures
protrusion of new sharp grits (7). A laser can also be used as a boosting method for mechanical contact dressing. In this case, the
laser softens the bond material and the following mechanical dresser removes it with ease (8).
Grinding wheel dressing by electrochemical method actually employs the electrochemical machining principle to dress the
wheel where the wheel is set as an anode and another metal electrode is set as a cathode (9). Electrolytic In-Process Dressing (ELID)
grinding is one of the latest and most appropriate techniques to dress the wheel in-process by the electrochemical method (10).
Electrolytic dressing of the wheel applies the basic mechanism of electrolysis. This process converts electrical energy to chemical
energy in an electrolytic solution in the presence of two electrodes (anode and cathode) as shown in Figure 1. An electrical power
supply is used to positively energize the anode and negatively energize the cathode. As the electric energy is applied, ionic disso-
lutions occurs in the electrolytic solution and the negative and positive ions in the solution move to the positive and negative
electrodes, respectively. This will cause anodic metal dissolution and formation of anodic oxide.
The mathematical form of the metallic dissolution rate is governed by Faraday’s law of electrolysis, which concludes that
dissolution from anode material is directly proportional to the integration of current supplied to the electrolytic cell. The mathe-
matical form of anodic dissolution can be written as follows:
RT
idt
0 M
m¼ $
F z
where
m is the mass of the substance altered from anode,
i is the electric current passed through the cell,
F is Faraday constant,
M is the molar mass of the substance, and
z is the valence number of ions of the substance (electrons transferred per ion).

Electrolyte

Battery

(-) (+)

Anode Cathode

Figure 1 Basic principle of electrolysis.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation 485

Carbon DC power
brush supply

(+) (-)

Nozzle

Wheel Electrode

Figure 2 Basic elements of ELID grinding.

Basic elements of ELID grinding are shown in Figure 2. ELID cell comprises of a metal-bonded grinding wheel, a cathode
electrode, a pulsed DC power supply, and an electrolyte. The wheel is connected to the positive terminal of the DC power supply
through a carbon brush, whereas the electrode is connected to the negative pole of the power supply. Typically, alkaline liquids are
used as both electrolytes and coolant for grinding. A nozzle is used to inject the electrolyte into the gap between wheel and electrode.
The gap is usually maintained to be approximately 0.1–0.3 mm. An anodic oxide layer is formed on the circumference of the
grinding wheel as a result of electrochemical reaction. The formation of this oxide layer is extremely crucial for the success of ELID
grinding. It is soft and brittle in nature as compared to the original metal bond and easily gets worn off because of the excessive
grinding force. The basic mechanism of ELID grinding has been explained in Figure 3. After truing (Figure 3(a)), the grains and
bonding material (metal) of the wheel surface are flattened. It is necessary for the trued wheel to be electrically predressed to
protrude the grains on the wheel surface and the dressing continues during the grinding operation as well. When predressing starts
(Figure 3(b)), the bonding material flows out from the grinding wheel and an insulating layer composed of the oxidized bonding
material is formed on the wheel surface (Figure 3(c)), as explained in Figure 1. This insulating layer reduces the electrical
conductivity of the wheel surface and prevents excessive flow-out of the bonding material from the wheel. At this moment, the grits
are held by both the bonding material and the oxide layer. It is important to note that the insulating oxide layer is soft and brittle in
nature and easily wears off as it comes in contact with the workpiece during grinding operation (Figure 3(d)). As grinding
continues, diamond grains wear out and cutting force increases. This excessive cutting force will cause falling off of the blunt grits
which is now mostly held by the brittle insulating material (Figure 3(e)). As a result of breakage of insulating layer, the electrical
conductivity of the wheel surface increases and the electrolytic dressing restarts with the flow-out of bonding material from grinding
wheel. The protrusion of new diamond grains from the grinding wheel therefore remains constant. This cycle is repeated during the

Figure 3 Principle of ELID grinding.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
486 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation

Table 1 Comparative analysis of different in-process dressing method

Mechanical contact dressing Electro thermal dressing Laser machining dressing Electrochemical dressing

System complexity Bulky system Die sinking method is simpler Bulky and complex system Fairly simple design
than WEDM method
System noise level Noisy Quiet Quiet Quiet
Any damage to May cause mechanical damage May cause thermal damage May cause thermal damage No thermal or mechanical
the wheel from over feeding etc. damage to the wheel
Dressing tool wear Tool wear exists Tool wear exists No tool wear No tool wear

grinding process to achieve stable grinding. There are few terminologies widely used for ELID grinding, which are also used in this
book chapter, and are as follows:
Peak voltage (Vp): This is the maximum voltage applied to the ELID cell by pulsed voltage supply.
Peak current (Ip): Maximum current drawn by the ELID cell when the peak voltage is applied.
TON: This is the pulse ON time and TOFF is the pulse OFF time.
Duty ratio (Rc): Duty ratio is the % of ON time for one cycle of operation and can be calculated as follows:

TON
Rc ¼  100%:
TOFF þ TON
Various technologies for in-process dressing of the grinding wheel (described in this section) have their own advantages and
disadvantages. The comparative analysis of all these processes is explained in Table 1 given below.
It is understandable from the above discussion that ELID has some notable advantages over the other in-process dressing
techniques. A considerable number of researches have been carried out and still continue on various aspects of ELID grinding. This
chapter will discuss in detail about different researches on ELID grinding in the subsequent sections.

11.18.4 History of ELID Grinding

The basic principle of ELID grinding is explained in the previous section with Figure 3. This technology was first proposed by Murata
et al. (10) back in 1985. They applied the technology for abrasive cut-off of structural ceramics which are difficult to grind because of
their hard and brittle nature. Conventionally, softer grade wheels are used for grinding hard and brittle material. However, using
softer grinding wheels cannot maintain high accuracy in the finished product due to large wheel wear during grinding. Therefore,
stronger bonds with harder abrasives are necessary for grinding hard and brittle materials. Murata et al. (10), performed the
experiments with metal-bonded grinding wheels (not specified) of grit size #80, #100, #150, and #400. Constant protrusion of
sharp grits was realized during the whole process of grinding because of the principle explained in Figure 3. The experimental setup
used by Murata et al. (10) is shown in Figure 4, which explains that ELID is an in-process dressing where the wheel is continuously
dressed at one side and carrying out grinding operation on the other side. It was observed by Murata et al. (10) (Figure 5) that ELID

Figure 4 ELID grinding setup proposed by Murata, R.; Okano, K.; Tsutsumi, C. Grinding of Structural Ceramics. (Some Application of Electrolytic
In-process Dressing to Abrasive Cut-off Operation). In Milton C Shaw Grinding Symposium PED, 1985; Vol. 16, pp 261–272.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation 487

Figure 5 Significant reduction of grinding force when in-process dressing (ELID) is used. Reproduced from Murata, R.; Okano, K.; Tsutsumi, C.
Grinding of Structural Ceramics. (Some Application of Electrolytic In-process Dressing to Abrasive Cut-off Operation). In Milton C Shaw Grinding
Symposium PED, 1985; Vol. 16, pp 261–272.

grinding produces significantly lower grinding force compared to conventional grinding. Further, the surface roughness of the
finished product was also found to be much improved when ELID grinding was used. The two findings above by Murata et al. (10) is
obvious because ELID grinding ensures constant protrusion of the sharp grits throughout the grinding cycle.
ELID grinding process was further developed and established by Ohmori in 1990 (11). In this chapter, Ohmori et al. achieved
mirror surface finish by grinding on Silicon wafer. The experimental setup of the process is shown in Figure 6. Researchers used
a Cast Iron Fiber Bonded (CIFB) wheel as the anode and copper/graphite as the cathode in this work. Pulsed DC voltage was used as
the power source for electrolytic dressing of the wheel. The grinding fluid itself served as electrolytic medium in this case. Necessary
isolation between the grinding wheel and machine structure was arranged during the ELID grinding experiment. For the first time,
Ohmori et al. (11) also explained in this paper the dressing current characteristics during Electrolytic Dressing operation. Initially,
the wheel is very conductive and a large current flows through the ELID cell. Eventually, anodic dissolution takes place on the
grinding wheel and an insulating layer is formed on the wheel surface. This causes the current to drop significantly and working
voltage reaches to open circuit voltage gradually. The I–V characteristic during the ELID dressing operation is shown in Figure 7.
Main characteristics of the ELID grinding pointed by Ohmori et al. (11) are as follows:
1. ELID grinding does not require any special equipment. Conventional grinding machine can be used. Further, die sinking electro-
discharge machining (EDM) power supply can be used as an ELID power supply as well.
2. ELID produces highly stable and lower grinding force compared to conventional grinding.
3. Mirror surface can be realized by ELID grinding without further processing method such as polishing and/or lapping.
4. ELID grinding can be applied for a variety of machining systems; i.e., vertical/horizontal grinding machine, plane surface or
cylindrical surface grinding, etc.

Figure 6 Experimental setup for ELID grinding of silicon wafer, proposed by Ohmori, H.; Nakagawa, T. Mirror Surface Grinding of Silicon Wafers
with Electrolytic In-process Dressing. Ann. CIRP 1990, 39 (1), 329–332.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
488 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation

Figure 7 I–V characteristic for dressing of the wheel by ELID shows reduction of dressing current due to the formation of insulating layer. Repro-
duced from Ohmori, H.; Nakagawa, T. Mirror Surface Grinding of Silicon Wafers with Electrolytic In-process Dressing. Ann. CIRP 1990, 39 (1), 329–332.

11.18.5 Classifications of ELID Grinding

Researchers proposed three major classifications of ELID grinding, based on the materials to be ground and application method.
However, the basic principle of dressing of the wheel remains the same. The various methods of ELID grinding are as follows:
1. Electrolytic In-Process Dressing (ELID I)
2. Electrolytic Interval Dressing (ELID II)
3. Electrodeless ELID (ELID III)
4. Electrodeless ELID (ELID-IIIA)
5. Ion shot ELID (ELID IV)

11.18.5.1 Electrolytic In-Process Dressing

This is the mostly used method of ELID grinding. In this case, the wheel is continuously dressed in-process. Figure 8 explains the
basic mechanism of the ELID I method for peripheral grinding. A cathode electrode covers usually 1/4th to 1/6th of the cutting
edges of the wheel, whereas the other exposed portion of the wheel takes part in the grinding action. The gap between the electrode
and the wheel is maintained to be 0.1–0.3 mm.
ELID I principle can be applied for different configurations of grinding. Few examples are given below. Figure 9 shows
a peripheral grinding wheel used to grind Silicon Nitride (SiN) workpiece which is horizontally mounted (12).
The ELID I grinding method can also be employed for in-feed grinding orientation. Figure 10 is showing one example of in-
feed ELID I, where the workpiece (Aluminium Nitride AlN) is rotating and grinding wheel gives in-feed as shown by the arrow
direction (13).
Figure 11 shows ELID I grinding for cylindrical grinding for bearing steel (14). Figure 12 explains the principle of ELID I curve
generator (CG) grinding for micro-spherical lens fabrication where the workpiece is mounted on a tilted rotary table (15).
Feed Motion
e
iec
kp

Anode
or
W

Optical
In-Fe Sensor
ed

Cathode

RS232
Inductive Communication
Sensor Power
Supply

Figure 8 ELID I grinding principle.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation 489

Figure 9 ELID I grinding using peripheral grinding wheel on a SiN workpiece mounted horizontally. Reproduced from Bandyopadhyay, B. P.;
Ohmori, H. The Effect of ELID Grinding on the Flexural Strength of Silicon Nitride. Int. J. Mach. Tools Manuf. 1999, 39, 839–853.

Figure 10 ELID I grinding for in-feed grinding of AlN. Reproduced from Katahira, K.; Ohmori, H.; Uehara, Y.; Azuma, M. ELID Grinding Characteris-
tics and Surface Modifying Effects of Aluminum Nitride (AlN) Ceramics. Int. J. Mach. Tools Manuf. 2005, 45, 891–896.

Figure 11 Cylindrical ELID I grinding for bearing steel. Reproduced from Qian, Jun; Li, Wei; Ohmori, Hitoshi. Cylindrical Grinding of Bearing Steel
with Electrolytic In-process Dressing. Precis. Eng. 2000, 24, 153–159.

A schematic illustration of 3 D controlled ELID I arc-enveloping grinding is shown in Figure 13 (16). The workpiece surface is
divided into many small squares, a diamond grinding wheel is then controlled to move in XZ or YZ plane and these series motions
are employed to generate a desired surface. The above references (12–16) shall be discussed in further detail in the subsequent
section.

11.18.5.2 Electrolytic Interval Dressing

In order to carry out precision grinding on the small holes on ceramic materials, Zhang et al. (17) introduces the ELID II method.
In this method, the dressing of the grinding wheel is intermittently carried out in contrast to the ELID II method. The interval time

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
490 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation

Figure 12 ELID I (CG) grinding for micro-spherical lens fabrication. Reproduced from Ohmori, Hitoshi; Lin, Weimin; Moriyasu, Sei; Yamagata,
Yutaka. Microspherical Lens Fabrication by Cup Grinding Wheels Applying ELID Grinding, RIKEN Review No. 34 (April, 2001): Focused on Advances
on Micro-mechanical Fabrication Techniques, pp 3–5.

Figure 13 ELID I grinding for 3 D arc-enveloping grinding. Reproduced from Yin, Shaohui; Morita, S; Ohmori, H; Uehara, Y; Lin, W; Liu, Q;
Maihara, T; Iwamuro, F; Mochida, D. ELID Precision Grinding of Large Special Schmidt Plate for Fibre Multi-object Spectrograph for 8.2 M Subaru
Telescope. Int. J. Mach. Tools Manuf. 2005, 45, 1598–1604.

Figure 14 ELID II grinding principle. Reproduced from Zhang, Chunhe; Ohmori, Hitoshi; Li, Wei. Small-hole Machining of Ceramic Material with
Electrolytic Interval-dressing (ELID-II) Grinding. J. Mater. Process. Technol. 2000, 105, 284–293.

between the two successive dressings depends on the predecided grinding force limit and other grinding conditions. Figure 14
shows the schematic illustration of the interval ELID dressing and grinding.
Qian et al. (18) also investigated the ELID II method to grind the internal surface of cylindrical objects. They found that the pipe
electrode performs better than other shapes of electrodes for dressing the wheel in the case of ELID II. The concept of pipe electrode
is explained in Figure 15. They also achieved mirror internal surface on an ordinary grinding machine by applying ELID II grinding.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation 491

Figure 15 ELID II grinding with pipe electrode. Reproduced from Qian, Jun; Li, Wei; Ohmori, Hitoshi. Precision Internal Grinding with a Metal-
bonded Diamond Grinding Wheel. J. Mater. Process. Technol. 2000, 105, 80–86.

11.18.5.3 Electrodeless ELID

The ELID II system has an inherent drawback of extra dressing time, for which the grinding efficiency has to be sacrificed. In order to
overcome this problem, Qian et al. (19) introduced the ELID III system. In the ELID III system, the cathode electrode is removed by
the workpiece itself as shown in Figure 16. However, this technique cannot be used for nonconductive material such as ceramics,
glass, silicon, etc. In this grinding method, authors used a meta-resin bonded wheel. The metal is dissolved because of the elec-
trolysis, which causes the grits to be held by soft resin only. This makes the grit and workpiece contact more elastic and a better
surface can be achieved as described by the authors (19). However, one problem associated with this process is an unwanted spark
between the workpiece and grinding wheel may lead to cavities on the wheel.

11.18.5.4 Electrodeless ELID

The problem associated with ELID III grinding can be eliminated by changing the power supply to the alternating current power
supply. This will cause the workpiece to become anode during half cycle of the current and cathode during other half cycle of the
current. This method is known as ELID-IIIA and was proposed by Lim (20). The principle of this type of ELID grinding is explained
in Figure 17. This version of ELID will cause to create a thick oxide layer (passivation film) on the workpiece as shown in
Figure 17(c) to prevent direct contact between the wheel and the workpiece, hence the electrical discharges.

11.18.5.5 Ion Shot ELID

Recently, Ohmori et al. (21) developed a new ELID technique where the nozzles have two plates connected to the ELID power
supply as shown in Figure 18. When the electrolyte hits the wheels, it is already ionized and oxidation of the metal bond (of the
grinding wheel) can take place. However, it is also true that an electrolyte will lose some ionization as it leaves the electrode gap;

Figure 16 Electrodeless ELID grinding (ELID III). Reproduced from Qian, Jun; Ohmori, Hitoshi; Lin, Weimin. Internal Mirror Grinding with a
Metal/metal–resin Bonded Abrasive Wheel. Int. J. Mach. Tools Manuf. 2001, 41, 193–208.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
492 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation

Figure 17 Schematic diagram of ELID-IIIA (Reproduced from Lim, H. S.; Ohmori, H.; Lin, W.; Qian, J. High Productivity and High Accuracy
Electrode-less ELID Grinding on Die Material. RIKEN Rev. 2000, 24, 136–137 (in Japanese).). (a) ELID-IIIA machining system with alternating current.
(b) ELID without electrode. (c) Electrolysis of workpiece.

Figure 18 Principle of ELID IV grinding. Reproduced from Ohmori, H.; Katahira, K.; Naruse, T.; Uehara, Y.; Nakao, A.; Mizutani, M. Ann. CIRP 2007,
56, 569–572.

therefore, this technique is useful for fine grinding techniques. In spite of the presence of different types of ELID grinding, ELID I is
the most adapted technique by the researchers. Most of the research works carried out on ELID grinding used ELID I.

11.18.6 Fundamental Study on ELID Grinding

Ohmori et al. (11,22) investigated the nonlinear behavior of electrolytic in-process dressing. In their study, it was found that the
dressing current reduces over the time during predressing because of the formation of the insulating oxide layer as shown in
Figure 7. This formation of the layer varies with the bond material as described in their study (22). It was also observed that the
wheel producing thinner layer experiences lower cutting force because of the higher dressing rate. The effect of the types of power
sources on the dressing current variation was also studied. Dressing current reduces at the highest rate if pure DC power supply is
used. AC power source has the minimum rate of reduction in the dressing current and for the case of pulsed DC power supply the
dressing current reduction rate is in between the above two.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation 493

Bifano et al. (23) developed a model of effective film growth rate by assuming the film removal rate to be constant with constant
load. According to the theory, the rate of growth of the film is derived from the following equation:
xg ¼ a log t [1]

where
xg is the film thickness at time t,
a is the electrolytic constant, and
t is time.
Therefore, the film growth rate is,
dxg a
¼ [2]
dt t
After rearranging eqn [2], growth rate of the film can be expressed as a function of the film thickness itself as follows:
dxg
¼ aex=a [3]
dt
If the film wear rate is assumed to be equal to a constant b; the eqn [3] can be modified to redefine effective film growth rate as
follows:
dx
¼ aex=a  b [4]
dt
where x is the film thickness and a,b are constant related to electrochemistry and tribological parameters of the system, respectively.
At time t ¼ 0, the following conditions are true.
 
dx
At t ¼ 0; x ¼ 0 and ¼ ða  bÞ
dt t¼0

This gives the analytical solution of eqn [4], which is as follows

!
x axx=a  b b
þ log þ t¼0 [5]
a ab a

The simulation result of eqn [5] is shown in Figure 19. The result suggests for any value of a and b the film thickness attains
a steady state value. However, film thickness is positive when a > b. The time taken to reach the steady state value depends on
relative magnitude of a and b. If a < b, the wear rate is higher than the instantaneous growth rate of the oxide film and no oxide film
is thereby formed. The erosion rate of the bond material is same as the effective growth rate of the film. In their study, the authors
suggest that the bond erosion rate should be set in such a way that the time taken for the bond to erode through a distance equal to
the diameter of one diamond should be less than the time taken for the diamond to become unacceptably dull. However, authors
did not propose any quantitative model to realize the wear rate of the cutting grits. Therefore, a power supply parameter cannot be
adjusted based on the film formation model only.
Lim et al. (24) studied the fundamental mechanism of ELID grinding, where they used BK7 glass as the workpiece material. They
observed nonstability in the grinding force because of the formation and breakage of the insulating ELID layer which is evident
throughout grinding. The behavior of the insulating layer was studied by comparing the phase relations between the grinding forces
and current which is shown in Figure 20. Figure 20(a) shows that normal grinding force gradually increases and then decreases
again. This follows a periodic pattern. The tangential grinding force suddenly increases at a certain point to indicate the breakage of
the insulating oxide layer, this causes the dressing current to increase to 1.62 A from a minimum of 1.24 A. The overall observation
shows as the dressing current increases, the tangential force increases and the normal force decreases. It was concluded from the

Figure 19 Simulation result of effective film growth rate for ELID grinding. Reproduced from Ohmori, H.; Nakagawa, T. Utilization of Nonlinear Condi-
tions in Precision Grinding with ELID (Electrolytic In-process Dressing) for Fabrication of Hard Material Components. Ann. CIRP 1997, 46 (1), 261–264.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
494 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation

Figure 20 Variation of normal force, tangential force, and dressing current during ELID grinding. Reproduced from Lim, H. S.; Fathima, K.; Senthil
Kumar, A.; Rahman, M. A Fundamental Study on the Mechanism of Electrolytic In-Process Dressing (ELID) Grinding. Int. J. Mach. Tools Manuf.
2002, 42, 935–943.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation 495

above observation that the oxide layer is not fully worn out but suddenly breaks after reaching a certain condition. The condition of
breakage of the insulating layer was not explained in the study; however, the factors influencing layer breakage were identified and
they are the thickness of the insulating layer, the contact area between the workpiece and the wheel, and the feed rate. In their study,
it was also found that a higher current duty ratio produces lesser grinding force and better surface finish because of higher wheel
dressing; however, the tool wear ratio is more for high duty ratio dressing. They also studied the effect of feed speed for ELID
grinding. At very high feed rate (>400 mm min1), grinding burn occurs on the workpiece surface. This is because the formation of
the oxide layer is less than its wear rate at high feed speed, which was also, described by Bifano et al. (23), an abnormal
phenomenon. This causes the dull grit to rub against the workpiece surface, hence excessive heat generation and grinding burn.
An example of clean and burned ground surface is shown in Figure 21.
A study on the grinding wheel wear in ELID grinding was carried out by Fathima et al. (25). They proposed a phenomenon that
the active grits of grinding during ELID are held by the metal oxide matrix which is softer than the actual metal bond. They explained
three scenarios as shown in Figure 22. During grinding, two major opposing forces are acting on individual diamond grit, namely
grit holding force fh and grinding force per grit fg. Figure 22(a) shows very low holding force cannot perform grinding operation,
hence dislodged. Figure 22(b) shows a sharp grit firmly held by the metal oxide matrix can perform the grinding operation. As the
grit becomes dull, excessive grinding force cracks the metal oxide layer and the grit falls out (Figure 22(c)). This crack formation
suddenly increases the dressing current which is also in line with the finding of Lim et al. (24). Figure 23 compares the wheel profile
before oxide layer breakage and during layer breakage. It is clear from the wheel profile that the breakage of the oxide layer is not
uniform throughout the wheel circumference. Fathima et al. (25) also observed the effect of Ton time on radial wheel wear, which is
shown in Figure 24. An increasing trend of wheel wear is observed with the increase of Ton time. This finding is obvious as this will
cause the wheel to form a thicker oxide layer and dislodging of the sharp grit will occur without sufficient grinding operation. Based
on their study, authors concluded that shorter Ton time gives better performance for rough grinding with coarser wheels, whereas
longer Ton time gives better surface quality for fine grinding with finer grit wheels.

Figure 21 Microscopic view of ground surface, (a) clean surface with low feed speed, and (b) grinding burn during high feed speed. Reproduced
from Lim, H. S.; Fathima, K.; Senthil Kumar, A.; Rahman, M. A Fundamental Study on the Mechanism of Electrolytic In-process Dressing (ELID)
Grinding. Int. J. Mach. Tools Manuf. 2002, 42, 935–943.

Figure 22 Diamond grit condition during ELID grinding. Reproduced from Fathima, K.; Senthil Kumar, A.; Rahman, M.; Lim, H. S. A Study on Wear
Mechanism and Wear Reduction Strategies in Grinding Wheels Used for ELID Grinding. Wear 2003, 254, 1247–1255.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
496 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation

Figure 23 Grinding wheel profile before and during oxide layer breakage. Reproduced from Fathima, K.; Senthil Kumar, A.; Rahman, M.; Lim, H. S.
A Study on Wear Mechanism and Wear Reduction Strategies in Grinding Wheels Used for ELID Grinding. Wear 2003, 254, 1247–1255.

Figure 24 Relationship of radial wheel wear with Ton time. Reproduced from Fathima, K.; Senthil Kumar, A.; Rahman, M.; Lim, H. S. A Study on
Wear Mechanism and Wear Reduction Strategies in Grinding Wheels Used for ELID Grinding. Wear 2003, 254, 1247–1255.

A grinding force model for ELID grinding was proposed by Fathima et al. (26). They experimentally verified the model by
carrying on ELID grinding on BK7 glass using cast iron bonded (CIB) diamond wheel. According to the authors, normal grinding
force (FN) and tangential grinding force (FT) can be expressed as follows:
FN ¼ fg Ng Ar [6]

FT ¼ mðELIDÞ FN [7]

where fg ¼ normal force per grit, Ng ¼ number of grits per unit area, Ar ¼ contact area, m(ELID) ¼ frictional coefficient depending on
the work and bond material, the dressing current density, and the ELID electrolyte; and FN. Schematic illustration of single grit is
explained in Figure 25.
fg of eqn [7] can be calculated as follows:

fg ¼ Hw rint V ratio [8]

where Hw ¼ hardness of the work material.

And
vw vw
V ratio ¼ ¼
vs pDw N
vw ¼ workpiece velocity, Dw ¼ Grinding wheel diameter, N ¼ Wheel speed in RPM.
rint can be calculated as follows for spherical grit shape,
rffiffiffiffiffiffiffiffiffiffi
2W
rint ¼ [9]
pHw
where W is the load applied by the grit as shown in Figure 25.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation 497

Figure 25 Single grit action during grinding. Reproduced from Fathima, K.; Senthil Kumar, A.; Rahman, M.; Lim, H. S. Modeling of Ultra-precision
ELID Grinding. Trans. ASME, J. Manuf. Sci. Eng. 2007, 129, 296–302.

Contact area Ar can be estimated for vertical groove grinding from Figure 26.
The grinding force model described by eqns [6] and [7] was simulated by considering two conditions stated below:
1. The grinding wheel was predressed and then used without the application of the ELID (can be considered as without ELID
condition); and
2. The grinding wheel is predressed and then used with the application of the ELID.
The experimental and simulated grinding force results are shown in Figure 27. Figure 27(a) initially shows that the experimental
grinding force is lower than the simulated grinding force. This is because the simulation bonding material was considered to be cast
iron, however in actuality, at the beginning it was iron oxide. Therefore, the grinding force was small at the beginning of the
machining and when the iron oxide was completely worn off, the force was equivalent to that of the simulated result. In the case of
Figure 27(b), continuous ELID was taking place during the grinding operation, therefore, the simulated grinding force and the
experimental grinding force were found to be in very good agreement in this case. This result established the fact that ELID grinding
produces much lower grinding force because of the insulating oxide layer produced during in-process dressing.
An important modelling on the electrolysis of ELID was proposed by Hong Chen et al. (27) where authors considered the
existence of nonconductive diamond particle on a conductive metal matrix. In their study, it was found that the current concen-
tration becomes maximum at metal diamond interface; hence, the metal dissolution rate. Their study also showed that the metal
dissolution rate increases with higher diamond concentration as shown in Figure 28 for fixed applied electric field strength. Finally,
the authors concluded from the above finding that to maintain the same metal dissolution rate, the applied electric field should be
lower for the higher diamond concentration tool and higher for the lower diamond concentration tool. Further, to achieve better
performance in ELID, the rate of exposing new diamond particles must match the rate of diamond wear; however, no model was
proposed for quantifying this phenomenon.
Hong Chen et al. (28) further enhanced their model by considering the three-dimensional (3D) orientation of the diamond
particle on a metal matrix. The same phenomenon of increasing electric field concentration at the metal diamond boundary was

Figure 26 Illustration of wheel and workpiece contact for vertical groove grinding. Reproduced from Fathima, K.; Senthil Kumar, A.; Rahman, M.;
Lim, H. S. Modeling of Ultra-precision ELID Grinding. Trans. ASME, J. Manuf. Sci. Eng. 2007, 129, 296–302.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
498 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation

Figure 27 Comparison between the simulated and the experimental grinding forces. Reproduced from Fathima, K.; Senthil Kumar, A.; Rahman, M.;
Lim, H. S. Modeling of Ultra-precision ELID Grinding. Trans. ASME, J. Manuf. Sci. Eng. 2007, 129, 296–302.

Figure 28 Change in average current density with diamond concentration of the grinding wheel for H/A ¼ 100 (circular shape) H/A ¼ 1000 (square
shape), where H is the distance between anode and cathode and A is the size factor of the particle. Reproduced from Chen, Hong; Li, James C. M.
Anodic Metal Matrix Removal Rate in Electrolytic In-process Dressing I: Two-dimensional Modeling. J. Appl. Phys. 2000, 87 (6), 3151–3158.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation 499

Figure 29 ELID grinding system. (a) rigid electrode and (b) flexible foil electrode. Reproduced from Zhu, Zhenqi; Wang, Xiaohua; Thangam, Siva.
Simulation and Analysis of Rigid/Foil Electrolytic In-process Dressing (ELID) Systems for Grinding. J. Manuf. Sci. Eng. August 2004, 126, 565–570.

observed for the 3 D case. Furthermore, they investigated the protrusion effect of the diamond particle on the electric field
concentration. It was found that the field concentration effect at the metal diamond boundary rapidly decreases as the protrusion of
the grits increases and the electric field becomes more uniform. However, authors did not consider the effect of the ELID layer on the
protrusion of the diamond grits which is the actual case.
An interesting study carried out by Zhu et al. (29) analyzed and compared the fluid flow system for the solid electrode-rotating
wheel interface and foil electrode-rotating wheel interface. The pictures of the two systems are shown in Figure 29. In order to
achieve good ELID condition, it is always necessary to maintain sufficient flow of electrolytes between the gap of the electrode and
wheel annular surface. Their electrolyte flow simulation results showed a good agreement with the experimental findings. The
analysis revealed that at high wheel speed, the inlet velocity of the electrolyte into the gap should also have to be very high to
maintain good electrolytic condition. However, for low inlet velocity of the electrolyte in high-speed ELID grinding, the authors
suggested the use of an elastic foil electrode. It was shown that the foil ELID system is more effective for high-speed grinding with
low electrolyte supply rates.

11.18.7 Study Related to Control of ELID Grinding Process

Rich Boland et al. (30) developed a current feedback system to control the feed override during grinding. In their study, it was
observed that the dressing current during ELID grinding follows a nonuniform pattern. At the initial stage of the grinding, the
current is very high because of the absence of oxide layer and at the final stage of grinding the current becomes very low as the oxide
layer sufficiently grows. The authors studied the frequency spectrum of the current to identify which frequency change is dominant
during ELID grinding. In their study, it was observed that at 49 KHz there is a definite change in the frequency spectrum of the
dressing current because of the above-mentioned phenomenon. Based on previously machined parts, a threshold was set for the
amplitude of the frequency spectrum at 49 KHz. A control loop was then developed to override the feed speed during grinding;
which means the feed speed slowed down if the spectrum was above threshold value and it increased if the spectrum was below
threshold value. By applying this control technique, authors achieved much uniform current profile for the whole grinding cycle;
however, the effect of uniform dressing current on the final quality of the ground workpieces was not investigated in the chapter.
Eun-Sang Lee (31,32) studied ELID grinding of die steel using optimum dressing control. In their study, it was shown that
insulating layer thickness decreases as the gap between the electrodes increases. In conventional ELID grinding, the layer formation
is nonuniform, which is not favorable for constant condition grinding. However, it was experimentally observed from the authors’
study that the layer thickness can be controlled by varying the dressing current. The relationship between the layer thickness and
dressing current is expressed as follows:
to ¼ 2:86  0:71I [10]

where
to is the layer thickness
and I is the dressing current.
After applying the dressing current control, the layer thickness was found to be uniform throughout the grinding process.
The basic algorithm proposed by the authors for dressing current control is as follows:
First, the regular current limit, initial peak dressing current, Ton, and Toff are set. As the grinding starts, the real current is
continuously monitored and if there is any deviation from the regular current setting, the peak dressing current is adjusted
accordingly. The flowchart of the algorithm is shown in Figure 30. The experimental results show that this control algorithm reduces

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
500 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation

Figure 30 Flow chart for optimum dressing control proposed by Lee, Eun-Sang. A Study on the Mirror-like Grinding of Die Steel with Optimum
In-process Electrolytic Dressing. J. Mater. Process. Technol. 2000, 100, 200–208.

the grinding force and significantly improves the surface roughness as compared to conventional ELID grinding. The results are
shown in Figures 31 (for normal grinding force variation) and 32 (for surface roughness variation).
The name ELID grinding implies that ELID is meant for the dressing of the grinding wheel. It is not meant for the truing of the
grinding wheel. The difference between dressing and truing of the grinding wheel is as follows: dressing ensures protrusion of the
sharp grits, whereas truing improves the nonuniformity in wheel revolution. Previous researchers used different techniques to carry
out truing of the grinding wheel for ELID grinding. Tanveer et al. (33) proposed a new pulse width control of the dressing current for
the truing of metal-bonded diamond wheel. The ELID layer which formed due to the anodic dissolution metal bond is very soft and
brittle in nature and easily breaks off as it comes in contact with the workpiece. The wheel truing method proposed by Tanveer et al.
(33) made use of this phenomenon. An example of a nonuniform wheel in Figure 33(a) is used to give a brief introduction to this
unique concept. However, the idea is equally applicable in the case of wheel clamping error or spindle run out problem.
Figure 33(a) shows a high zone in the metal bond of the grinding wheel. In order to achieve a uniform metal bond, a controlled
dressing current is applied during the predressing of the wheel. This will cause higher metal dissolution from the peak region of the
wheel. Eventually after few cycles, the grinding wheel shall become such as Figure 33(b) which shows that a thicker ELID layer is
formed at the high portion of the wheel because of the controlled predressing. However, this excess layer shall quickly wear off as
grinding starts and finally the wheel workpiece contact shall be maintained uniform throughout the grinding cycle as shown in
Figure 33(c). In order to implement the truing technique, authors proposed a dressing power control method. The dressing peak
current Ipeak is kept constant in the power supply and the only parameter that can be controlled to achieve the uniformity of the
metal bond is duty ratio r. The following control unit illustrated in Figure 34 is designed for this purpose. In the figure, x(n) is the

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation 501

Figure 31 Variation of normal grinding force for controlled dressing ELID grinding and conventional ELID grinding. Reproduced from Lee, Eun-Sang.
A Study on the Mirror-like Grinding of Die Steel with Optimum In-process Electrolytic Dressing. J. Mater. Process. Technol. 2000, 100, 200–208.

Figure 32 Variation of surface roughness of the workpiece for controlled dressing ELID grinding and conventional ELID grinding. Reproduced from Lee,
Eun-Sang. A Study on the Mirror-like Grinding of Die Steel with Optimum In-process Electrolytic Dressing. J. Mater. Process. Technol. 2000, 100, 200–208.

ELID Layer

Metal Bond

(a) Nonuniform metal bond (b) Uniform metal bond but nonuniform (c) Uniform metal bond and uniform
and ELID layer (initial case) ELID layer (after controlled dressing) ELID layer (after grinding)

Figure 33 Schematic explanation of in-process truing for ELID grinding. (a) A typical example of nonuniform grinding wheel for ELID grinding. (b)
Expected grinding wheel profile after applying proposed controlled predressing. (c) Expected wheel profile after the grinding operation. Reproduced
from Tanveer, Saleh; Rahman, M. In-process Truing for ELID (Electrolytic In-process Dressing) Grinding by Pulse Width Control. IEEE Trans. Automat.
Sci. Eng. April 2011, 8 (2).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
502 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation

rmin

Zero order
+ hold (T/n)

Figure 34 Control unit for ELID truing. Reproduced from Tanveer, Saleh; Rahman, M. In-process Truing for ELID (Electrolytic In-process Dressing)
Grinding by Pulse Width Control. IEEE Trans. Automat. Sci. Eng. April 2011, 8 (2).

array of distance between the wheel and electrode for different segments on the wheel after proper signal processing. R is the
reference input to the controller which is the maximum of x(n). K1 is the proportional gain and K2/s is the system model in
s domain which can be derived from Faraday’s electrolysis model. The zero-order hold block is necessary because the control signal
needs to be held constant for T/n time. In this particular study, n was equal to 40. The control signal r(n) for nth segment is defined
by the following equation:
ðR  xðnÞÞ
rðnÞ ¼ ðrmax  rmin Þ þ rmin [11]
R  minðxðnÞÞ
where
rmax is the maximum duty ratio allowed by the controller which is 0.99,
rmin is the minimum duty ratio allowed by the controller which is 0.01.
ðrmax  rmin Þ
K1 can be deduced from eqn [11] which is equal to
R  minðxðnÞÞ
and
  
Ipeak M
K2 ¼ [12]
rAF z
The value of K2 can be easily found from the Faraday’s basic law of electrochemistry, where
Ipeak is the peak current,
r is the density of anode,
A is the peripheral area of the wheel covered by the electrode,
F is the Faraday’s constant,
M is the molar mass of anode, and
Z is the valence number of the anode.
By implementing pulse width controlled dressing for ELID grinding, the wheel profile was significantly improved, which can be
seen from Figure 35. Furthermore, in order to confirm that the proposed wheel truing method improves the uniformity of the wheel

0.35
Distance from the sensor head in mm

0.3

0.25

0.2

0.15

0.1
Initial Profile
Profile after 10min
0.05 Profile after 20min
profile after 30min
0
1 28 55 82 109 136 163 190 217 244 271 298 325 352 379 406 433 460 487
Points on the wheel

Figure 35 Continuous improvement in the actual wheel profile (measured by inductive sensor) with time. Reproduced from Tanveer, Saleh; Rahman,
M. In-process Truing for ELID (Electrolytic In-process Dressing) Grinding by Pulse Width Control. IEEE Trans. Automat. Sci. Eng. April 2011, 8 (2).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation 503

workpiece contact, Tanveer et al. (33) carried out an experiment to measure the grinding force for one wheel revolution. Figure 36
shows the result, which confirms that the proposed wheel truing method gradually improves the wheel workpiece contact. This new
approach (33) proved that ELID is not only meant for wheel dressing alone, but in addition a controlled dressing which can be
applied to achieve the truing of the wheel. This will eliminate the need for any additional truing device in ELID grinding.
Conventional ELID grinding continuously dresses the wheel without monitoring the wheel wear. This may lead to overdressing
the wheel and reduce the overall wheel life. To address this issue, Tanveer et al. (34) proposed a new ‘Dressing on Demand’
technique for ELID grinding. In this study, the author continuously monitored the grinding force ratio (K) (normal force
FN/tangential force FT) to decide how much dressing power is necessary. The authors decided the threshold K value from previous
literature for the diamond grinding wheel (35). The control scheme implemented for the study is shown in Figure 37, where the
peak dressing voltage is adjusted based on the current value of K. However, the minimum and maximum peak voltage was set to be
30 V and 100 V, respectively. Implementation of dressing voltage control by monitoring the wheel condition significantly reduced
the wheel wear about 48% of the conventional ELID grinding. The tool wear was calculated by comparing the grinding wheel profile
before grinding and after grinding. Figure 38(a) shows the profile of the grinding wheel before the grinding and after grinding for
conventional ELID. Figure 38(b) shows the same for the dressing on demand ELID grinding. In both cases, the total material
removal and experimental condition were the same. Significant reduction in wheel wear by implementing this dressing power
control for ELID was achieved without compromising the surface quality of the sample as compared to conventional ELID grinding.
It can be clearly observed from Figure 39 that surface roughness (Ra) and surface uniformity (Rt) of the machined workpieces are
comparable and equivalent for both cases.
A knowledge-based feedback control system was developed by Fathima et al. (36). In their experiment, material removal rate
(MRR) was kept constant by adjusting the rotational speed of the workpiece spindle and the feed rate. A knowledge database was
designed with the optimized grinding and ELID parameters for different work materials. At first, users specify the grinding job by
selecting the work material, setting grinding wheel specifications, defining the desired shape of the lens, etc. From the user speci-
fication, the designed algorithm will select appropriate grinding and ELID parameters and the expected grinding ratio from the
database. Grinding related information will then be sent to the machine controller and optimized ELID parameters (peak current
(Ip), voltage (Vp), and current duty ratio (Rc)) are sent to the computer-controlled pulse power source. The output current (I_p) and
the voltage (V_p) are used as feedback parameters for the ELID process. The overall flow diagram of the proposed knowledge-based

Figure 36 Normal cutting force signal for one wheel revolution at different time of grinding: (a) without in-process truing and (b) with in-process
truing. Reproduced from Tanveer, Saleh; Rahman, M. In-process Truing for ELID (Electrolytic In-process Dressing) Grinding by Pulse Width Control.
IEEE Trans. Automat. Sci. Eng. April 2011, 8 (2).

VP =100V
VP Grinding K value
VP =5K+25
Process
VP =30V

Figure 37 Dressing on demand model proposed by Tanveer, Saleh; Bishwas, Indraneel; Rahman, M. Efficient Dressing of the Wheel in ELID
Grinding by Controllable Voltage with Force Feedback. Int. J. Adv. Manuf. Technol. 2009, 46 (1–4), http://dx.doi.org/10.1007/s00170-009-2076-x.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
504 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation

Figure 38 Grinding wheel profile before and after grinding: (a) conventional ELID and (b) dressing on demand ELID. Reproduced from Tanveer,
Saleh; Bishwas, Indraneel; Rahman, M. Efficient Dressing of the Wheel in ELID Grinding by Controllable Voltage with Force Feedback. Int. J. Adv.
Manuf. Technol. 2009, 46 (1–4), http://dx.doi.org/10.1007/s00170-009-2076-x.

Figure 39 Comparison of surface properties (Ra and Rt) and force ratio between constant peak voltage dressing and controlled peak voltage
dressing. Reproduced from Tanveer, Saleh; Bishwas, Indraneel; Rahman, M. Efficient Dressing of the Wheel in ELID Grinding by Controllable Voltage
with Force Feedback. Int. J. Adv. Manuf. Technol. 2009, 46 (1–4), http://dx.doi.org/10.1007/s00170-009-2076-x.

feedback controlled ELID grinding system is shown in Figure 40. The improvement in flatness and roughness on the finished
product was observed to be quite remarkable after implementing this knowledge-based feedback control mechanism for ELID
grinding (Figure 41). For quartz, the flatness error reduced from 1.25 to 0.75 mm (40% improvement), whereas for BK7 glass, it
reduced from 0.9 to 0.7 mm (22% improvement). In case of surface roughness, the improvement for quartz and BK 7 glass were 50%
and 67%, respectively.

11.18.8 Study Related to ELID Grinding Machine Development

Not much research work has been carried out to develop a dedicated machine tool for ELID grinding. Uehara et al. (37) developed
a 4-axes desktop machine equipped with the ELID grinding system (TRIDER-X). All the axes are stepper motor driven and semi-
closed loop controlled. Resolution of the linear axes of the machine is 1 mm whereas the resolution of the tilting axis was 0.002 . The
overall dimension of the machine is 580 mm3 and the weight is 95 kg. A PC-based NC system is used to operate the machine.
Figure 42 shows the image of the developed machine. ELID experiments were carried out to evaluate the performance of the
machine. The average surface roughness of the ground mirror was 20 nm and the form accuracy was 20 microns, which are
reasonably good

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation 505

Figure 40 Flow diagram for knowledge-based feedback control for ELID grinding system. Reproduced from Fathima, K.; Schinhaerl, M.; Geiss, A.;
Rascher, R.; Sperber, P. A Knowledge Based Feed-back Control System for Precision ELID Grinding. Precis. Eng. January 2010, 34 (1), 124–132,
ISSN: 0141-6359, http://dx.doi.org/10.1016/j.precisioneng.2009.05.004.

Figure 41 Flatness and roughness results without control and with feedback control. Reproduced from Fathima, K.; Schinhaerl, M.; Geiss, A.;
Rascher, R.; Sperber, P. A Knowledge Based Feed-back Control System for Precision ELID Grinding. Precis. Eng. January 2010, 34 (1), 124–132,
ISSN: 0141-6359, http://dx.doi.org/10.1016/j.precisioneng.2009.05.004.

Figure 42 Image of the 4-axes desktop ELID machine. Reproduced from Uehara, Yoshihiro; Ohmori, Hitoshi; Yamagata, Yutaka; Moriyasu, Sei.
Grinding Characteristics of Cemented Carbide Concave Mirror by Desk-top Type 4-axes Machine Trider-X with ELID System, Initiative of Precision
Engineering at the Beginning of a Millennium (Book Chapter); Springer, 2001, pp 476–480.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
506 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation

Figure 43 Electrode mounting system for cylindrical ELID grinding machine. Reproduced from Uehara, Yoshihiro; Ohmori, Hitoshi; Yamagata,
Yutaka; Moriyasu, Sei; Lin, Weimin. Development of Small Tool by Micro Fabrication System Applying ELID Grinding Technique, Initiative of Preci-
sion Engineering at the Beginning of a Millennium (Book Chapter); Springer, 2001, pp 491–495.

Figure 44 (a) ELID grinding machine (b) OMM system developed by Tanveer, Saleh. Development of an Intelligent Electrolytic In-process Dressing
(ELID) Grinding System. Diss., 2009.

Uehara et al. (38) further developed a similar (37) compact cylindrical grinding machine equipped with an ELID system to
develop micro tools. The ELID system mounting was different for this machine compared to the conventional ELID. Here, the
electrode is free to rotate. A plastic body, a bearing for the electrode to rotate, an insulating guide plate to prevent contact between
the electrode, and the grinding wheel are the main components of this new electrode mounting system as shown in Figure 43. Using
the machine, they fabricated angular micro tools with good dimensional accuracy. Furthermore, they used these micro shafts for
micro engraving.
Unlike Uehara et al. (37,38), Tanveer et al. (39–41) took a complete system development approach for developing an ELID
grinding machine. A 3-axis CNC machine was developed by Tanveer et al. to perform ELID grinding as shown in Figure 44(a).
All the three axes are actuated by AC servo motors in a closed loop with mm level resolution. The machine structure was re-
engineered from an existing wire-cut EDM machine and configured as a gantry structure and is formed of two pillars, a cross-
beam, and a base as described by the authors (39–41). A feedback controlled ELID power supply (discussed in Section
11.18.1.7) was developed and incorporated to the machine (32,33). One significant concern in ELID grinding is the wheel wear
(reduction in diameter of the wheel) because of continuous dressing of the wheel. This introduces inaccuracy in the finished
product mostly in spherical/aspheric ground surface. In order to overcome this problem, the developed machine was equipped
with necessary on machine measurement system (touch probe-based OMM) tool wear compensation method. The hardware of
the developed OMM system is shown in Figure 44(b). The tool wear compensation algorithm is shown in a flowchart in
Figure 45. In this algorithm, first, the tool path for the spherical surface is generated by taking consideration of the tool
(grinding wheel) radius. After completion of the machining, the workpiece is scanned and the radius of curvature of the
machined spherical surface is calculated. If the measured radius of curvature is different from that of the desired value due to
some machining error, the first step is repeated until the achieved radius of curvature is within tolerance. Implementation of the
above tool wear compensation significantly improved the form accuracy of the machined surface from 2.83 to 0.04%. The result
is tabulated in Table 2.
To evaluate the performance of the developed machines, few cutting experiments were carried out by Rahman et al. (42). Nano-
surface finish (<20 nm Ra) was achieved with a #4000 series diamond wheel on BK 7 glass surface. Figure 46 shows the photograph
of a machined lens fabricated on the developed machine.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation 507

START

Generate the tool path for lens grinding by

considering the grinding wheel radius.

Scan the machined sample to

Measure diameter change.
calculate the radius of the lens

No
Is the radius within
tolerance?

Yes

Remove workpiece from the

machine

STOP

Figure 45 Tool wear compensation algorithm. Reproduced from Tanveer, Saleh. Development of an Intelligent Electrolytic In-process Dressing
(ELID) Grinding System. Diss., 2009.

Table 2 Comparison of profile accuracy

Experimental condition Desired profile radius (mm) Actual profile radius (mm) Deviation (%)

With compensation 100 99.96 0.04

Without compensation 100 98.17 2.83
Reproduced from Sazedur Rahman, M.; Saleh, T.; Lim, H. S.; Son, S. M.; Rahman, M. Development of an On-machine Profile Measurement
System in ELID Grinding for Machining Aspheric Surface with Software Compensation. Int. J. Mach. Tools Manuf. June 2008, 48 (7–8), 887–
895, ISSN: 0890-6955, http://dx.doi.org/10.1016/j.ijmachtools.2007.11.005.

Figure 46 Lens (Ra ¼ 15 nm) fabricated by ELID machine developed by Sazedur Rahman, M.; Saleh, T.; Lim, H. S.; Son, S. M.; Rahman, M. Devel-
opment of an On-machine Profile Measurement System in ELID Grinding for Machining Aspheric Surface with Software Compensation. Int. J. Mach.
Tools Manuf. June 2008, 48 (7–8), 887–895, ISSN: 0890-6955, http://dx.doi.org/10.1016/j.ijmachtools.2007.11.005.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
508 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation

11.18.9 Study Related to Application of ELID Grinding

The feasibility of ELID grinding has been investigated for a wide range of hard and brittle materials including ceramics, optical glass,
hardened steel, silicon, etc. In most of the cases, the results were very much optimistic. In this section, the studies related to the
applications of ELID grinding on various materials shall be discussed.

11.18.9.1 ELID on Ceramics

Ohmori et al. (43) carried out an experimental study to investigate the performance of ELID grinding on structural ceramics,
particularly for silicon nitrides. Two different types of silicon nitrides were used in this purpose, namely a sintered reaction-bonded
silicon nitride (SRBSN), and a cast-and-sintered silicon nitride (Si3N4). The effects of cutting speed, feed rate, and grit size on the
surface finish of the product were investigated. The findings of the study are summarized in Table 3.
Figure 47 shows the effect of different mesh size of the wheel on surface roughness of Si3N4 (a) and SRBSN (b). It can be
observed from Figure 47 that both cases sub 20 nm Ra surface finish be achieved which is considered to be an optical quality surface
finish by ELID grinding without any further material processing technique such as lapping and polishing.
Bandyopadhyay et al. (44) studied the application of ELID grinding for high MRR (rough grinding) case for silicon nitride-based
workpiece and compared it with conventional grinding. It was observed for conventional grinding that the cutting force contin-
uously increases as the volume of material removed increases. Significant reduction of normal force was observed for ELID grinding
under the same cutting condition. This finding is in line with the finding by other researchers (26). The variation of normal grinding
force can be observed from Figure 48. However, in ELID grinding at the very beginning of the cutting the grinding force increases,
stable and low cutting force can only be observed after 6000 mm3 of material removal. In order to address this problem, authors
proposed two solutions; first, one is to use dummy material for the first few cutting cycles before the start of actual workpiece
machining. The second solution, which is more convincing, is to use modified ELID dressing. The modified ELID dressing was
performed in two stages before ELID grinding. In the first stage, the ELID dressing was performed with 90 V for 30 min. The oxide
layer produced during the ELID dressing was mechanically removed by an aluminium oxide stick. Then, the second dressing was
performed also with 90 V for 30 min. The ELID grinding was then performed with 60 V, by applying modified ELID dressing as
explained above, a stable and low grinding force can be achieved from the beginning of the grinding operation. The effect of
modified ELID grinding can be observed from Figure 49.

Table 3 Summarized result of the study

Variable Material SRBSN Material Si3N4

Grit size Higher mesh size wheel produces better surface and Same trend was observed as SRBSN. Surface finish achieved by rougher
#4000 no wheel can produce mirror surface. wheel (#320) on Si3N4 is worse than that was achieved on SRBSN
material, however for finer wheel equivalent surface can be produced
on both materials.
Cutting speed No significant effect on the surface finish of the material. Same as SRBSN.
Feed speed Rough wheel produces bad surface finish at higher feed Similar trend was observed as SRBSN.
rate, finer wheel has no significant effect.
Reproduced from Ohmori, H.; Takahashi, I.; Bandyopadhyay, B. P. Ultra-precision Grinding of Structural Ceramics by Electrolytic In-process Dressing (ELID) Grinding. J. Mater.
Process. Technol. 1996, 57, 272–277.

Figure 47 Effect of grit size on surface finish of Si3N4 (a) and SRBSN (b) with constant cutting speed, feed rate, and depth of cut. Reproduced
from Ohmori, H.; Takahashi, I.; Bandyopadhyay, B. P. Ultra-precision Grinding of Structural Ceramics by Electrolytic In-process Dressing (ELID)
Grinding. J. Mater. Process. Technol. 1996, 57, 272–277.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation 509

Figure 48 Variation of normal grinding force with volume of material removed (a) conventional grinding (b) for ELID grinding. Reproduced from
Bandyopadhyay, B. P.; Qhmori, H.; Takahashi, I. Efficient and Stable Grinding of Ceramics by Electrolytic In-process Dressing (ELID). J. Mater.
Process. Technol. 1997, 66, 18–24.

Figure 49 Stable and low grinding force after applying modified ELID grinding as explained in the text. Reproduced from Bandyopadhyay, B. P.;
Qhmori, H.; Takahashi, I. Efficient and Stable Grinding of Ceramics by Electrolytic In-process Dressing (ELID). J. Mater. Process. Technol. 1997, 66, 18–24.

Bandyopadhyay et al. (12) further studied the effect of ELID grinding on the strength of Si3N4 materials, and compared it with
conventional grinding. The experimental setup is shown in Figure 9. Two types of grinding modes were investigated in this study,
namely longitudinal or parallel direction grinding (PG) and traverse direction grinding (TG). PG ground specimens were found to
have a higher bending strength than TG ground specimen; however, the strength property was significantly improved for the TG
ground sample when it was annealed at 1200  C. On the contrary, if ELID grinding is used for the finishing operation of the TG
ground samples, then no annealing is required and the bending strength is significantly high as shown in Figure 50. AFM and SEM
studies of the samples suggest that ductile machining (without surface fragmentation) can be achieved by ELID grinding when
#4000 or higher mesh wheels are used.
K. Katahira et al. (13) investigated the surface modifying effect caused by ELID grinding on AlN ceramics. Figure 10 explains the
grinding setup for this experimental work as described in the earlier section. ELID grinding technology can produce an extremely
smooth surface with nanometric roughness of only 8 nm. If the mesh size of the wheel is greater than #4000, a regular ductile
surface can be achieved with ELID grinding and the transition between ductile to brittle machining occurs between mesh sizes
#2000 to #4000. Figure 51(a) shows the effect of wheel mesh size on the average surface roughness of the material and Figure 51(b)
explains the transition from brittle mode machining to ductile mode machining. Surface properties were also found to be greatly
improved by ELID grinding. It was observed that the ELID ground workpiece have better surface characteristics such as higher
hardness, less friction coefficient, etc., over the conventionally polished workpiece. The authors’ investigation suggests that oxygen is
diffused into the workpiece when ELID grinding is used for the finishing operation and they identified this diffusion phenomenon
as the cause of the improvement of the surface characteristics of the ELID ground samples.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
510 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation

Figure 50 Effect of ELID grinding on the bending strength of Si3N4. Reproduced from Bandyopadhyay, B. P.; Ohmori, H. The Effect of ELID
Grinding on the Flexural Strength of Silicon Nitride. Int. J. Mach. Tools Manuf. 1999, 39, 839–853.

Figure 51 (a) Variation of surface roughness with wheel mesh size, (b) brittle/ductile mode transition. Reproduced from Katahira, K.; Ohmori, H.;
Uehara, Y.; Azuma, M. ELID Grinding Characteristics and Surface Modifying Effects of Aluminum Nitride (AlN) Ceramics. Int. J. Mach. Tools Manuf.
2005, 45, 891–896.

11.18.9.2 ELID on Coated Film

K. Katahira et al. (45) also investigated ELID grinding on TiAlN films. In their study, authors used ELID grinding and other finishing
process for comparison. ELID grinding can achieve an average surface roughness of 2.4 nm, whereas other finishing process can
generate a surface finish of 56.2 nm. It was found that the ELID process creates an oxide layer on TiAlN film because of high
reactivity of Ti, Al. This oxide layer enhances tribological properties of the TiAlN film. Ball on disk experiments of ground surface
suggests that the friction coefficient of the ELID ground TiAlN film obtained in this study is approximately two-thirds that of the
non-ELID ground TiAlN film as shown in Figure 52. The wear rate of the ELID ground TiAlN film (disk specimen) obtained in this
study is approximately one-half of that of the non-ELID ground TiAlN film; and the wear rate against of Al2O3 ball for the ELID
ground TiAlN film is approximately one-tenth of that for non-ELID ground TiAlN film. The result is shown in Figure 53.
The rationale behind these improvements in the tribological behavior of ELID ground samples is the formation of a diffused
oxide layer on the TiAlN film which was confirmed by energy-dispersed X-ray diffraction system (EDX) study. The result suggests the
existence of an oxide layer on the ELID ground TiAlN film (Figure 54).
Zhang et al. (46,47) compared the grinding of CVD-SiC (chemical vapor deposited silicon carbide) film by two methods; one is
ELID grinding, and the other one is ordinary grinding with saw-like abrasive stone dressed CIB diamond wheel. In their study, the
authors observed that during ELID study, an oxide layer of a few microns in thickness was formed which holds the new sharp grits.
The layer is soft in nature and it changes the contact status between the wheel and workpiece, which is a unique feature of ELID
grinding. However, in the case of conventional dressing of the wheel, no such layer is formed and only the cutting edge of the
diamond grain protruded at a certain amount. The surface roughness of the ground samples was found to be better in quality in the

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation 511

Figure 52 Friction coefficient obtained by ball and disk experiment for ELID ground TiAlN and non-ELID finished TiAlN film. Reproduced from
Katahira, K.; Watanabe, Y.; Ohmori, H.; Kato, T. ELID Grinding and Tribological Characteristics of TiAlN Film. Int. J. Mach. Tools Manuf. 2002, 42,
1307–1313.

Figure 53 Wear rate of disk and ball specimens. Reproduced from Katahira, K.; Watanabe, Y.; Ohmori, H.; Kato, T. ELID Grinding and Tribological
Characteristics of TiAlN Film. Int. J. Mach. Tools Manuf. 2002, 42, 1307–1313.

Figure 54 Analytical results of intensity of mass obtained by use of EDX diffraction system. Reproduced from Katahira, K.; Watanabe, Y.; Ohmori,
H.; Kato, T. ELID Grinding and Tribological Characteristics of TiAlN Film. Int. J. Mach. Tools Manuf. 2002, 42, 1307–1313.

case of ELID ground samples. The variation of the surface roughness with grinding time was found to be more stable in the case of
ELID grinding as shown in Figure 55. Moreover, the morphological study of the samples suggests that in ELID grinding, ductile
regime machining is the predominant mode of machining (with less pits and fractures), whereas in ordinary grinding both ductile
regime and brittle facture machining occurs equally as shown in Figure 56. Therefore, in their study, the authors suggested that ELID
grinding is a better choice in ultraprecision grinding.

11.18.9.3 ELID on Metal

Qian et al. (14) studied the feasibility of ELID technology on finishing bearing steel materials by cylindrical grinding (Figure 11).
Both traverse and plunge mode grinding were used and compared. Further ELID grinding was compared with typical finishing
operation of bearing steel, such as honing and electro-finishing methods. Experimental findings showed ELID grinding produces
a better surface finish on the workpiece while comparing with other finishing processes (Table 4).
In ELID, plunge mode grinding generates a poorer surface finish than traverse grinding, especially with coarser grits. Moreover,
better and stable waviness can be achieved in traverse mode grinding with ELID. In this study, the effect of spark out process on
waviness and roundness of the workpiece were investigated. Both waviness and roundness were improved with 3–4 ‘spark out’

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
512 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation

Figure 55 Variation of surface roughness Ra over grinding time. Reproduced from Zhang, Chunhe; Ohmori, Hitoshi; Kato, Teruko; Morita, Noboru.
Evaluation of Surface Characteristics of Ground CVD-SiC Using Cast Iron Bond Diamond Wheels. J. Int. Soc. Precis. Eng. Nanotechnol. 2001, 25, 56–62.

Figure 56 Laserscopic topography of ground surface of CVD-SiC film: (a) Ordinary grinding method and (b) ELID grinding method. Reproduced
from Zhang, Chunhe; Kato, Teruko; Li, Wei; Ohmori, Hitoshi. A Comparative Study: Surface Characteristics of CVD-SiC Ground with Cast Iron Bond
Diamond Wheel. Int. J. Mach. Tools Manuf. 2000, 40, 527–537.

Table 4 Effect of finishing operation on bearing steel

Surface parameters
Surface quality
Methods Ra (mm) Rz din (mm) Skew rsk Delta q

Honing 0.06 0.5 1.1 0.95

Electric polishing 0.04 0.3 0.6 0.47
ELID grinding 0.02 0.2 0 0.2

rsk is ruled surface kernal; delta q is surface slope; and din means DIN standard.
Reproduced from Qian, Jun; Li, Wei; Ohmori, Hitoshi. Cylindrical Grinding of Bearing Steel with Electrolytic In-process
Dressing. Precis. Eng. 2000, 24, 153–159.

passes. Metrological results of the samples showed that ELID grinding produces comparable waviness (L band, M band) result on
the samples as honing and electro-finishing processes, shown in Figure 57(a). Residual stress on the ELID ground sample was also
found to be very low (Figure 58(b)), however, the cycle time for ELID ground samples was found to be high when compared to
existing honing and electro-finishing processes (Figure 58(c)). Therefore, ELID technology should be used for small batch
production situations.
Saito et al. (48) studied the effect of bond material on the performance of ELID grinding. In their study, ELID grinding was used
to produce cemented carbide alloys lens mold. Two types of metal bonded diamond wheels were used for the study, namely Cr
bond and Cu bond. The ELID grinding result was compared with conventional polishing method (termed as P-series). The finding
shows Cr bonded wheels produced best surface roughness on the workpiece (Figure 58). Authors further investigated the elec-
trolytic behavior of the Cr bonded grinding wheel. In ELID grinding, most of the metal bonded wheels produce an insulating layer
after the initial electrolytic dressing; however for the Cr bonded wheel, the phenomenon was found to be nonmatching. In Cr
bonded wheels, no oxide layer is formed after and during the electrolytic dressing, therefore the protrusion of sharp cutting grits are
more for Cr bonded wheels. Moreover, higher oxygen diffusion to the workpiece material occurs in case of Cr bonded wheel that
enhances the surface quality. This finding is similar to that of K. Katahira et al. (13).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation 513

Figure 57 Comparison of ELID performance with different finishing processes for bearing steel: (a) surface waviness; here, MAX mean maximum
allowable limit, (b) surface residual stress, and (c) cycle time of the process. Reproduced from Qian, Jun; Li, Wei; Ohmori, Hitoshi. Cylindrical
Grinding of Bearing Steel with Electrolytic In-process Dressing. Precis. Eng. 2000, 24, 153–159.

Figure 58 Comparison of surface roughness for wheels with different bond material (wheel mesh size #8000). Reproduced from Saito, Tomoyuki;
Katahira, Kazutoshi; Ohmori, Hitoshi; Komotori, Jun; Mizutani, Masayoshi; Nemoto, Akihiko. Fabrication of High-quality Surfaces on Precise Lens
Mold Materials by a New ELID Grinding Wheel, towards Synthesis of Micro-/Nano-systems. In The 11th International Conference on Precision Engi-
neering (ICPE) Tokyo, Japan, August 16–18, 2006, pp 315–318.

A previous study suggested that the formation of an oxide layer or diffusion of oxygen into the work material enhances the
surface property of the material (13,48). Ohmori et al. (49) proposed a controllable method of oxide layer formation during ELID
grinding of stainless steel workpiece by applying a weak positive potential at the workpiece as shown in Figure 59. Three different
types of abrasive particles (diamond, alumina, and silicon dioxide) were used for comparison. All three abrasives produced a similar
surface roughness of 8–10 nm. Also, ELID ground surface (by diamond particle C series) was compared with normal polished
surface (P-series). ELID ground samples exhibited superior surface properties including hardness, tribological and fatigue prop-
erties, corrosion and high temperature oxidation resistances, and adhesive strength with coating films over conventionally polished
samples. Further, the observation suggests that the ELID mirror surface grinding process produced a stable oxide layer on the
workpiece surface, and also that the abrasive elements of the grinding wheel penetrated and diffused into the workpiece. In
addition, it was found very interesting that the wettability of the processed surface varies with different methods and wheel grit
materials. Therefore, authors proposed that by varying the processing conditions and grinding wheel, one can possibly control the
surface potential characteristics and even hydrophilicity of the surfaces.
The use of ELID grinding was studied by Raffles et al. (50) for finishing operation of bearing balls using metal-resin bonded CBN
wheel. Using #12000 series wheel 2 nm Ra (average surface roughness) is achievable which is significantly better than balls
produced by polishing. Without ELID, the MRB-CBN wheel was proven to be ineffective for grinding operation as the wheel
becomes quickly glazed. Different ELID techniques were used for the finishing operation, namely ELID I, ELID II, and ELID III

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
514 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation

Figure 59 Overview of the proposed grinding system and a schematic illustration of surface modification reactions. Reproduced from Ohmori, H.;
Katahira, K.; Komotori, J.; Mizutani, M. Fictionalization of Stainless Steel Surface through Mirror-quality Finish Grinding. CIRP Ann. Manuf. Technol.
2008, 57, 545–549.

(explained in Section 11.18.1.5). The ELID III operation was found to be unstable and unpredictable, whereas grinding by ELID II
and ELID I method were successful. Sub-10 nm surface finish was consistently achievable using ELID II; however, ELID I was found
to be the most effective because of continuous wheel dressing. It was possible to achieve 2 nm Ra using the ELID I method.

11.18.9.4 ELID on Optical Glasses

Ohmori et al. (15) used ELID grinding to fabricate spherical glass lens and compared the characteristics of the machined lens with
ordinary ground samples. The experimental setup of the study is shown in Figure 12. With the help of ELID grinding, the average
surface roughness (Ra) achieved was 0.018 microns, whereas by ordinary grinding method, it was found to be 0.138 microns. The
variation of the surface roughness with different wheel mesh size is shown in Figure 60. Morphological study of the samples also
suggests that ELID grinding can produce a much smoother surface than the conventional grinding method. The shape accuracy of
the lens ground by ELID was also found to be in the less than one micron range.
Kumar et al. (51) studied the effect of ELID grinding on BK7 glasses. It was found that current duty ratio plays an important role
on the grinding performance and surface finish of the product. Both the surface finish and grinding force improves (cutting force
reduces) with the increase of current duty ratio as shown in Figure 61. This is because higher current duty ratio dresses the grinding
wheel more to ensure better protrusion of the sharp cutting grits. However, the grinding ratio also decreases with high current duty
ratio which indicates a shorter grinding wheel life. Authors also found that the #4000 series wheel with proper grinding conditions
can generate a flawless ductile surface with sub-10 nm average surface roughness on BK7 glass.
Stephenson et al. (52) investigated the AE detection method to identify wheel loading and to assess the grinding state of a wheel
for machining of BK7 and Zerodur glass. This investigation suggests that the acoustic emission energy has a potential to be used as
grinding condition indicator because of its strong dependence on wheel loading. In the study, it was found that ELID grinding with
a fine grit size cup wheel is less likely to encounter wheel loading compared to a resin bonded wheel when wheel work contact zone
is big. A high level of AE signal during grinding with resin bonded wheel indicates the occurrence of excessive friction and rubbing
between the wheel and workpiece, which implies that a resin bonded wheel cannot perform effective self-dressing when the

Figure 60 Variation of surface roughness of fabricated lens with wheel mesh number. Reproduced from Ohmori, Hitoshi; Lin, Weimin; Moriyasu,
Sei; Yamagata, Yutaka. Microspherical Lens Fabrication by Cup Grinding Wheels Applying ELID Grinding, RIKEN Review No. 34 (April, 2001):
Focused on Advances on Micro-mechanical Fabrication Techniques, pp 3–5.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation 515

Figure 61 (a) Normal grinding force variation for different dressing duty ratio. (b) Tangential grinding force variation for different dressing duty
ratio. (c) Surface roughness for different ELID duty ratio. Reproduced from Senthil Kumar, A.; Lim, H. S.; Rahman, M.; Fathima, K. A Study on the
Grinding of Glass Using Electrolytic In-process Dressing. J. Electron. Mater. 2002, 31 (10), 1039–1046.

machining arcs of the abrasive grits were long. However, resin bonded wheels tend to generate lower AE amplitude when the wheel/
workpiece contact area is small (Figure 62). Authors also found that if the dressing parameter is low in ELID grinding with finer
wheels, the AE signal level was also considerably high. Therefore, they recommend in-process dressing with more aggressive dressing
parameters for finer grit size grinding wheels to thereby minimize wheel loading and AE signal level (Figure 63). However, this
selection of dressing parameters depends on many factors such as wheel configuration, grinding parameters, and material properties
of the workpiece, etc. Thus, the authors suggested the use of the AE sensing technique for monitoring the complex ELID grinding
process to ensure optimum grinding/dressing conditions are maintained.
Yin et al. (53) studied the characteristics of ELID grinding on Zerodur glass where they achieved an average surface roughness of
5–10 nm with #4000 series wheel. The flatness and perpendicularity were also found to be very good. AFM study of the ground
surfaces also indicated that the material removal was predominantly by ductile machining as shown in Figure 64(a). The influences
of the grinding parameters on the grinding performances were also studied. It was observed that a higher feed rate causes high
grinding force (Figure 64(b)) and more thermal damages to the surface. Depth of cut was another important factor for ELID
grinding of Zerodur glass. Ideal depth of cut for better surface finish and stable grinding conditions was found to be 0.4 microns. Yin

Figure 62 AE signals generated by resin bonded wheel (no-ELID) and CIB wheel (ELID) with variable wheel/workpiece contacting area. Reproduced
from Stephenson, D. J.; Sun, X.; Zervos, C. A Study on ELID Ultra Precision Grinding of Optical Glass with Acoustic Emission. Int. J. Mach. Tools
Manuf. 2005, 46, 1–11.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
516 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation

Figure 63 (a) Influence of duty ratio on AERMS voltage of AE signals. (b) Effect of peak voltage on AERMS voltage of AE signals. Reproduced
from Stephenson, D. J.; Sun, X.; Zervos, C. A Study on ELID Ultra Precision Grinding of Optical Glass with Acoustic Emission. Int. J. Mach. Tools
Manuf. 2005, 46, 1–11.

Figure 64 (a) AFM images of surfaces ground by #4000 wheel. (b) Effects of feed rate on grinding force. Reproduced from Yin, Shaohui; Ohmori,
Hitoshi; Dai, Yutang; Uehara, Yoshihiro; Chen, Fengjun; Tang, Hengning. ELID Grinding Characteristics of Glass-ceramic Materials. Int. J. Mach. Tools
Manuf. 2005, 49, 333–338.

Figure 65 Form deviation profile before compensating form error (a) in X direction (b) in Y direction. Reproduced from Yin, Shaohui; Morita, S;
Ohmori, H; Uehara, Y; Lin, W; Liu, Q; Maihara, T; Iwamuro, F; Mochida, D. ELID Precision Grinding of Large Special Schmidt Plate for Fibre
Multi-object Spectrograph for 8.2 M Subaru Telescope. Int. J. Mach. Tools Manuf. 2005, 45, 1598–1604.

et al. (16) further developed a new grinding process and system for the fabrication of large nonaxisymmetric aspheric surfaces of
fused silica (Figure 13). In their study they employed a precision truing method for the grinding wheel as well as the form error
compensation of the machined workpiece. In order to compensate the form error, the ground surface was measured using an on
machine measurement system and the tool path was regenerated by taking into consideration the grinding wheel wear.
Improvement in form accuracy was achieved by implementing compensation scheme as shown in Figures 65 and 66.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation 517

Figure 66 Form deviation profile after compensating form error (a) in X direction (b) in Y direction. Reproduced from Yin, Shaohui; Morita, S;
Ohmori, H; Uehara, Y; Lin, W; Liu, Q; Maihara, T; Iwamuro, F; Mochida, D. ELID Precision Grinding of Large Special Schmidt Plate for Fibre Multi-
object Spectrograph for 8.2 M Subaru Telescope. Int. J. Mach. Tools Manuf. 2005, 45, 1598–1604.

11.18.9.5 ELID on Silicon Wafer

Chao et al. (54) studied the effect of grinding with a metal bonded grinding wheel and resin bonded grinding wheel on
silicon wafers. Amorphous layer, nanocrystals, polycrystals, dislocations, and micro-cracks were observed in both cases
depending on the wheel and grinding parameters. It was found that the ELID ground surfaces were very sensitive to the
ELID parameters and the optimized conditions were difficult to obtain and be maintained. However, in the case of grinding
with a resin bond wheel, the surface integrity was relatively less dependent to the machining conditions in comparison to
ELID grinding. In the case of ELID grinding, thick oxide layer and high feed rate favors the wheel to produce sharp new
protruding grains so that a stable cutting condition can be reached, but there is still chances for some protruding grains
getting excess attritious wear before it is pulled out in ELID grinding. The worn grains will dull the wheel and generate
friction heat and subsurface damage. This chance is very low for the resin bond wheel because of its self-sharpening effect.
Moreover, in the case of the resin bond wheel, except when the feed rate is very low (say 2 mm min1) or the wheel is
loaded, the amorphous layers generated were normally thinner than those generated by ELID grinding under the same
grinding conditions.
Itoh et al. (55) performed an experimental investigation on ELID lap grinding of silicon workpieces. The results of the exper-
iments showed that stable grinding can be achieved with #8000 to #3000000 series wheel and ultrafine surface with average surface
roughness of Ra ¼ 0.38 nm can be achieved using #3000000 wheel, whereas conventional lap grinding can generate surface
roughness Ra ¼ 0.47 nm. The peak to valley roughness which is another very important parameter for silicon polishing was also
significantly reduced (from 5.05 to 2.8 nm) when ELID lap grinding was used as shown in Figure 67. Moreover, a morphological
study of the ground samples with SEM and AFM also showed that produced surfaces are very smooth in the order of several
nanometers.
M.M. Islam et al. (56) studied the injection electrode (IE)-assisted ELID grinding for the thinning of the silicon wafer. The
schematic illustration of the newly developed mechanism is shown in Figure 68. The injection electrode flushing system
ensures better electrolyte filling between the electrode-grinding wheel gap than the conventional ELID flushing system. This
ensures better dressing of the wheel, hence IE-assisted ELID grinding exhibits less force than conventional ELID grinding as

Figure 67 Comparison of surface roughness for silicon workpiece. (a) Conventional lap grinding. (b) ELID lap grinding. Reproduced from Itoh,
Ohmori; Moriyasu, Kasai; Karaki-Doy, Bandyopadhyay. Finishing Characteristics of Brittle Materials by ELID Lap Grinding Using Metal-resin Bonded
Wheels. Int. J. Mach. Tools Manuf. 1998, 38, 747–760.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
518 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation

Figure 68 Injection Electrode ELID grinding system (IE) for silicon wafer thinning process. Reproduced from Islam, M. M.; Senthil Kumar, A.;
Balakumar, S.; Lim, H. S.; Rahman, M. Performance Evaluation of a Newly Developed Electrolytic System for Stable Thinning of Silicon
Wafers. Thin Solid Films 2006, 504, 15–19.

Figure 69 Comparison of grinding force and force ratio between Injection Electrode ELID grinding system (IE) and conventional ELID grinding
system. Reproduced from Islam, M. M.; Senthil Kumar, A.; Balakumar, S.; Lim, H. S.; Rahman, M. Performance Evaluation of a Newly Developed
Electrolytic System for Stable Thinning of Silicon Wafers. Thin Solid Films 2006, 504, 15–19.

shown in Figure 69. With the help of the IE-assisted ELID grinding process, authors succeeded to thin down the silicon wafer
from 750 mm to approximately 70 mm with sub-ten nano-surface finish and without any visible cracking, pilling, and
chipping.
M.M Islam et al. (57) further characterized the ELID lapping process for machining silicon wafers and following salient
conclusions were drawn in this study. The rate of the formation of the oxide layer was found to be dependent on the current duty
ratio; however, total oxide layer growth is independent of current duty ratio. Compared to other conditions, only intermediate range
of current duty ratio and grinding parameter (feed rate) showed better wheel surface condition; the results are shown in Figure 70.
Higher current duty ratio results in denser and harder oxide layer formation that makes breakage of the ELID layer along with dull

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation 519

Figure 70 Variation of ground surface roughness of silicon wafer by ELID grinding: (a) effect of current duty ratio and (b) effect of feed rate.
Reproduced from Islam, M. M.; Senthil Kumar, A.; Balakumar, S.; Lim, H. S.; Rahman, M. Characterization of ELID Grinding Process for Machining
Silicon Wafers. J. Mater. Process. Technol. 2008, 198, 281–290.

grits less probable. Hence, the surface finish becomes worse due to grinding by dull grits. In the case of very low feed speed, wheels
repeatedly grind the same area which may cause excessive attrition wear of the grits resulting in a poor surface finish. Conventional
grinding process for a silicon wafer exhibited unstable grinding forces with very high magnitude, on the contrary, the ELID grinding
especially under intermediate range of value, showed a stable grinding force with very low magnitude. The ELID grinding process
under intermediate range of experimental conditions (current duty ratio and feed rate) showed better grinding performance
(smaller percentage of wear flat and smaller value of specific grinding energy) with superior ground wafer surface quality
(Ra ¼ 4 nm). Under all experimental conditions, the ELID grinding process performs better compared to the conventional grinding
process. Moreover, the subsurface damages exhibited on the samples are significantly lower for ELID grinding process than
conventional grinding.
Researchers also studied the effect of diamond grain size on different aspects of silicon wafer machining by ELID grinding such
as surface roughness, subsurface damage, and material removal (58). The results are shown in Figure 71. It is clear from the result
that smaller grain size improves the surface roughness and reduces the depth of the subsurface cracks. However, the MRR is much
lower for superfine abrasive wheels. Therefore, it is recommended to use an ultrafine abrasive wheel for the final finishing
operation.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
520 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation

Figure 71 Effect of grain size for ELID grinding of silicon wafer (a) on surface roughness, (b) subsurface crack, and (c) material removal rate.
Reproduced from Liua, J. H.; Peia, Z. J.; Fisher, Graham R. ELID Grinding of Silicon Wafers: A Literature Review. Int. J. Mach. Tools Manuf. 2007,
47, 529–536.

11.18.10 Conclusions

In-process dressing of a grinding wheel has been under research interest for several years. In-process dressing is very important to
achieve a nano-surface finish by grinding using super-abrasive wheels. This may eliminate the need for further finishing operations
such as polishing and/or lapping. Significant studies were carried out by many researchers on this subject area. ELID is proven to be
the most effective method of in-process dressing. Various researches have been carried out by different researchers since its invention
back in the 1990s. Continuous anodic dissolution and formation of soft insulating anodic oxide layer ensures more protrusion of
sharp cutting grits and lower grinding force than conventional grinding. Sub-10 nm average surface roughness is achievable on hard
and brittle materials such as optical glasses, silicon wafer, ceramics, bearing steels, etc., by ELID grinding. However, wheel wear rate
in ELID grinding is considerably high because of continuous anodic dissolution of metallic bond of the grinding wheel. This leads to
a low grinding ratio and wheel life. Furthermore, it hampers the form accuracy of the finished products, especially for an aspheric or
curved surface. Researchers proposed a tool wear compensation scheme to overcome this problem. One very interesting thing about
ELID grinding is that it can be easily implemented to the existing grinding without much modification. However, researchers also
tried to develop a dedicated intelligent ELID grinding system. Researchers also proved that pulse width control of the dressing
current can make ELID grinding useful for in-process truing of the grinding wheel. In a nutshell, it can be concluded that ELID
grinding is a promising technology to achieve a nanometric surface finish on hard and brittle materials and has the potential to
eliminate the final finishing process such as polishing.

References

1. Liu, C.; Ohmori, H.; Lin, W. Development and Characteristics of Fixed Abrasive Polishing Utilizing Small Grinding Tools with ELID for Aspheric Optical Surfaces. ELID Grinding
Project 1999, 25, 247–253.
2. Rahman, M.; Lim, H. S.; Neo, K. S.; Senthil Kumar, A.; Wong, Y. S.; Li, X. P. Tool-based Nanofinishing and Micromachining. J. Mater. Process. Technol. 2006, 185, 2–16.
3. Shih, Albert J. An Experimental Investigation of Rotary Diamond Truing and Dressing of Vitreous Bond Wheels for Ceramic Grinding. Int. J. Mach. Tools Manuf. 2000, 40,
1755–1774.
4. Nakagawa, T.; Suzuki, K. Highly Efficient Grinding of Ceramics and Hard Metals on Grinding Center. Ann. CIRP 1986, 35 (1), 205–210.
5. Suzuki, K.; Uematsu, T.; Nakagawa, T. On-machine Trueing/dressing of Metal Bond Grinding Wheels by Electro-discharge Machining. Ann. CIRP 1987, 36 (1), 115–118.
6. Nakagawa, T.; Uematsu, T.; Kirnura, M. Development of a New Turning Center for Grinding Ceramic Materials. Ann. CIRP 1988, 37 (1), 319–322.
7. Xie, X.-Z.; Chen, G.-Y.; Li, L.-J. Dressing of Resin-bonded Super Abrasive Grinding Wheels by Means of Acousto-optic Q-switched Pulsed Nd:YAG Laser. Opt. Laser Technol.
2004, 36, 409–419.
8. Zhang, C.; Shin, Y. C. A Novel Laser-assisted Truing and Dressing Technique for Vitrified CBN Wheels. Int. J. Mach. Tools Manuf. 2002, 42, 825–835.
9. Schopf, M.; Beltrami, I.; Boccadoro, M.; Kramer, D. ECDM (Electro Chemical Discharge Machining), a New Method for Trueing and Dressing of Metal Bonded Diamond Grinding
Tools. Ann. CIRP 2001, 50 (1), 125–128.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation 521

10. Murata, R.; Okano, K.; Tsutsumi, C. Grinding of Structural Ceramics. (Some Application of Electrolytic In-process Dressing to Abrasive Cut-off Operation) In: Milton C Shaw
Grinding Symposium PED 1985, Vol. 16, 1985, 261–272.
11. Ohmori, H.; Nakagawa, T. Mirror Surface Grinding of Silicon Wafers with Electrolytic In-process Dressing. Ann. CIRP 1990, 39 (1), 329–332.
12. Bandyopadhyay, B. P.; Ohmori, H. The Effect of ELID Grinding on the Flexural Strength of Silicon Nitride. Int. J. Mach. Tools Manuf. 1999, 39, 839–853.
13. Katahira, K.; Ohmori, H.; Uehara, Y.; Azuma, M. ELID Grinding Characteristics and Surface Modifying Effects of Aluminum Nitride (AlN) Ceramics. Int. J. Mach. Tools Manuf.
2005, 45, 891–896.
14. Qian, Jun; Li, Wei; Ohmori, Hitoshi Cylindrical Grinding of Bearing Steel with Electrolytic In-process Dressing. Precis. Eng. 2000, 24, 153–159.
15. Ohmori, Hitoshi; Lin, Weimin; Moriyasu, Sei; Yamagata, Yutaka. Microspherical Lens Fabrication by Cup Grinding Wheels Applying ELID Grinding, RIKEN Review No. 34
(April, 2001): Focused on Advances on Micro-mechanical Fabrication Techniques, pp 3–5.
16. Yin, Shaohui; Morita, S.; Ohmori, H.; Uehara, Y.; Lin, W.; Liu, Q.; Maihara, T.; Iwamuro, F.; Mochida, D. ELID Precision Grinding of Large Special Schmidt Plate for Fibre Multi-
object Spectrograph for 8.2 M Subaru Telescope. Int. J. Mach. Tools Manuf. 2005, 45, 1598–1604.
17. Zhang, Chunhe; Ohmori, Hitoshi; Li, Wei. Small-hole Machining of Ceramic Material with Electrolytic Interval-dressing (ELID-II) Grinding. J. Mater. Process. Technol. 2000, 105,
284–293.
18. Qian, Jun; Li, Wei; Ohmori, Hitoshi. Precision Internal Grinding with a Metal-bonded Diamond Grinding Wheel. J. Mater. Process. Technol. 2000, 105, 80–86.
19. Qian, Jun; Ohmori, Hitoshi; Lin, Weimin. Internal Mirror Grinding with a Metal/metal–resin Bonded Abrasive Wheel. Int. J. Mach. Tools Manuf. 2001, 41, 193–208.
20. Lim, H. S.; Ohmori, H.; Lin, W.; Qian, J. High Productivity and High Accuracy Electrode-less ELID Grinding on Die Material. RIKEN Rev. 2000, 24, 136–137 (in Japanese).
21. Ohmori, H.; Katahira, K.; Naruse, T.; Uehara, Y.; Nakao, A.; Mizutani, M. Microscopic Grinding Effects on Fabrication of Ultra-fine Micro Tools. Ann. CIRP 2007, 56, 569–572.
22. Ohmori, H.; Nakagawa, T. Utilization of Nonlinear Conditions in Precision Grinding with ELID (Electrolytic In-process Dressing) for Fabrication of Hard Material Components.
Ann. CIRP 1997, 46 (1), 261–264.
23. Bifano, T.; Krishnamoorthy, R.; Fawcett, H.; Welch, E. Fixed-load Electrolytic Dressing with Bronze Bonded Grinding Wheels. Trans. ASME, J. Manuf. Sci. Eng. 1999, 121, 20–27.
24. Lim, H. S.; Fathima, K.; Senthil Kumar, A.; Rahman, M. A Fundamental Study on the Mechanism of Electrolytic In-process Dressing (ELID) Grinding. Int. J. Mach. Tools Manuf.
2002, 42, 935–943.
25. Fathima, K.; Senthil Kumar, A.; Rahman, M.; Lim, H. S. A Study on Wear Mechanism and Wear Reduction Strategies in Grinding Wheels Used for ELID Grinding. Wear 2003,
254, 1247–1255.
26. Fathima, K.; Senthil Kumar, A.; Rahman, M.; Lim, H. S. Modeling of Ultra-precision ELID Grinding. Trans. ASME, J. Manuf. Sci. Eng. 2007, 129, 296–302.
27. Chen, Hong; Li, James C. M. Anodic Metal Matrix Removal Rate in Electrolytic In-process Dressing I: Two-dimensional Modeling. J. Appl. Phys. 2000, 87 (6), 3151–3158.
28. Chen, Hong; Li, James C. M. Anodic Metal Matrix Removal Rate in Electrolytic In-process Dressing II: Protrusion Effect and Three-dimensional Modelling. J. Appl. Phys. 2000,
87 (6), 3159–3164.
29. Zhu, Zhenqi; Wang, Xiaohua; Thangam, Siva. Simulation and Analysis of Rigid/Foil Electrolytic In-process Dressing (ELID) Systems for Grinding. J. Manuf. Sci. Eng.
August 2004, 126, 565–570.
30. Boland, Rich. Computer Control and Process Monitoring of Electrolytic In-process Dressing of Metal Bond Fine Diamond Wheels for NIF Optics. In Proc. SPIE 1999, Vol. 3782;
pp 61–69.
31. Lee, Eun-Sang. A Study on the Mirror-like Grinding of Die Steel with Optimum In-process Electrolytic Dressing. J. Mater. Process. Technol. 2000, 100, 200–208.
32. Lee, Eun-Sang; Kim, J. A Study on the Analysis of Grinding Mechanism and Development of Dressing System by Using Optimum In-process Electrolytic Dressing. Int. J. Mach.
Tools Manuf. 1997, 37, 1673–1689.
33. Tanveer, Saleh; Rahman, M. In-process Truing for ELID (Electrolytic In-process Dressing) Grinding by Pulse Width Control. IEEE Trans. Automat. Sci. Eng. April, 2011, 8 (2),
338–346.
34. Tanveer, Saleh; Bishwas, Indraneel; Rahman, M. Efficient Dressing of the Wheel in ELID Grinding by Controllable Voltage with Force Feedback. Int. J. Adv. Manuf. Technol.
2009, 46 (1–4). http://dx.doi.org/10.1007/s00170-009-2076-x.
35. Kramer, D.; Rehsteiner, F. ECD (Electrochemical In-process Controlled Dressing) a New Method for Grinding of Modern High Performance Cutting Materials to Highest Quality.
Ann. CIRP 1999, 48 (1), 265–268.
36. Fathima, K.; Schinhaerl, M.; Geiss, A.; Rascher, R.; Sperber, P. A Knowledge Based Feed-back Control System for Precision ELID Grinding. Precis. Eng. January 2010, 34 (1).
ISSN: 0141-6359, 124–132. http://dx.doi.org/10.1016/j.precisioneng.2009.05.004.
37. Uehara, Yoshihiro; Ohmori, Hitoshi; Yamagata, Yutaka; Moriyasu, Sei. Grinding Characteristics of Cemented Carbide Concave Mirror by Desk-top Type 4-axes Machine Trider-X
with ELID System, Initiative of Precision Engineering at the Beginning of a Millennium (Book Chapter); Springer, 2001; pp 476–480.
38. Uehara, Yoshihiro; Ohmori, Hitoshi; Yamagata, Yutaka; Moriyasu, Sei; Lin, Weimin. Development of Small Tool by Micro Fabrication System Applying ELID Grinding Technique,
Initiative of Precision Engineering at the Beginning of a Millennium (Book Chapter); Springer, 2001; pp 491–495.
39. Tanveer, Saleh; Rahman, M. Sazedur; Lim, H. S.; Rahman, M. Development and Performance Evaluation of an Ultra-precision ELID Grinding Machine. J. Mater. Process.
Technol. October 2007, 192–193, 287–291.
40. Tanveer, Saleh; Rahman, Mustafizur. A System Development Approach for Electrolytic In-process Dressing (ELID) Grinding. Int. J. Automat. Technol. 2011, 5 (1), 21–29.
41. Tanveer, Saleh. Development of an Intelligent Electrolytic In-process Dressing (ELID) Grinding System. Diss., National University of Singapore, 2009.
42. Sazedur Rahman, M.; Saleh, T.; Lim, H. S.; Son, S. M.; Rahman, M. Development of an On-machine Profile Measurement System in ELID Grinding for Machining Aspheric
Surface with Software Compensation. Int. J. Mach. Tools Manuf. June 2008, 48 (7–8). ISSN: 0890-6955, 887–895. http://dx.doi.org/10.1016/j.ijmachtools.2007.11.005.
43. Ohmori, H.; Takahashi, I.; Bandyopadhyay, B. P. Ultra-precision Grinding of Structural Ceramics by Electrolytic In-process Dressing (ELID) Grinding. J. Mater. Process. Technol.
1996, 57, 272–277.
44. Bandyopadhyay, B. P.; Qhmori, H.; Takahashi, I. Efficient and Stable Grinding of Ceramics by Electrolytic In-process Dressing (ELID). J. Mater. Process. Technol. 1997, 66,
18–24.
45. Katahira, K.; Watanabe, Y.; Ohmori, H.; Kato, T. ELID Grinding and Tribological Characteristics of TiAlN Film. Int. J. Mach. Tools Manuf. 2002, 42, 1307–1313.
46. Zhang, Chunhe; Kato, Teruko; Li, Wei; Ohmori, Hitoshi. A Comparative Study: Surface Characteristics of CVD-SiC Ground with Cast Iron Bond Diamond Wheel. Int. J. Mach.
Tools Manuf. 2000, 40, 527–537.
47. Zhang, Chunhe; Ohmori, Hitoshi; Kato, Teruko; Morita, Noboru. Evaluation of Surface Characteristics of Ground CVD-SiC Using Cast Iron Bond Diamond Wheels. J. Int. Soc.
Precis. Eng. Nanotechnol. 2001, 25, 56–62.
48. Saito, Tomoyuki; Katahira, Kazutoshi; Ohmori, Hitoshi; Komotori, Jun; Mizutani, Masayoshi; Nemoto, Akihiko. Fabrication of High-quality Surfaces on Precise Lens Mold
Materials by a New ELID Grinding Wheel, towards Synthesis of Micro-/Nano-systems. In The 11th International Conference on Precision Engineering (ICPE) Tokyo, Japan;
August 16–18, 2006; pp 315–318.
49. Ohmori, H.; Katahira, K.; Komotori, J.; Mizutani, M. Fictionalization of Stainless Steel Surface through Mirror-quality Finish Grinding. CIRP Ann. Manuf. Technol. 2008, 57,
545–549.
50. Raffles, M. H.; Stephenson, D. J.; Shore, P.; Jin, T. Electrolytic In-process Dressing Superfinishing of Spherical Bearings Using Metal–resin Bond Ultra-fine CBN Wheels. Proc.
Inst. Mech. Eng., Part B January 1, 2011, 225 (1), 112–122.
51. Senthil Kumar, A.; Lim, H. S.; Rahman, M.; Fathima, K. A Study on the Grinding of Glass Using Electrolytic In-process Dressing. J. Electron. Mater. 2002, 31 (10), 1039–1046.
52. Stephenson, D. J.; Sun, X.; Zervos, C. A Study on ELID Ultra Precision Grinding of Optical Glass with Acoustic Emission. Int. J. Mach. Tools Manuf. 2005, 46, 1–11.
53. Yin, Shaohui; Ohmori, Hitoshi; Dai, Yutang; Uehara, Yoshihiro; Chen, Fengjun; Tang, Hengning. ELID Grinding Characteristics of Glass-ceramic Materials. Int. J. Mach. Tools
Manuf. 2005, 49, 333–338.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
522 Electrolytic In-Process Dressing (ELID) Grinding for Nano-Surface Generation

54. Chao, C. L.; Ma, K. J.; Liu, D. S.; Sheu, S. C.; Lin, Y. S.; Lin, H. Y. A Study on the Surface Integrity of Single Crystal Silicon Ground by CIFB Diamond Wheels (ELID) and Resin-
bonded Diamond Wheels, Initiative of Precision Engineering at the Beginning of a Millennium (Book Chapter); Springer, 2001; pp 376–380.
55. Itoh, Ohmori; Moriyasu, Kasai; Karaki-Doy, Bandyopadhyay. Finishing Characteristics of Brittle Materials by ELID Lap Grinding Using Metal-resin Bonded Wheels. Int. J. Mach.
Tools Manuf. 1998, 38, 747–760.
56. Islam, M. M.; Senthil Kumar, A.; Balakumar, S.; Lim, H. S.; Rahman, M. Performance Evaluation of a Newly Developed Electrolytic System for Stable Thinning of Silicon Wafers.
Thin Solid Films 2006, 504, 15–19.
57. Islam, M. M.; Senthil Kumar, A.; Balakumar, S.; Lim, H. S.; Rahman, M. Characterization of ELID Grinding Process for Machining Silicon Wafers. J. Mater. Process. Technol.
2008, 198, 281–290.
58. Liua, J. H.; Peia, Z. J.; Fisher, Graham R. ELID Grinding of Silicon Wafers: A Literature Review. Int. J. Mach. Tools Manuf. 2007, 47, 529–536.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 11.19 Microelectrochemical Deposition
MA Habib, Islamic University of Technology, Gazipur, Bangladesh
 2014 Elsevier Ltd. All rights reserved.

11.19.1 Introduction 523

11.19.2 Process Mechanism 524
11.19.2.1 Basic Concept of ECD 524
11.19.2.2 ECD Using Pulse Voltage 525
11.19.3 Micromanufacturing Using ECD 527
11.19.3.1 3D Microstructure Fabrication Using the Anode as a Counter Electrode 527
11.19.3.1.1 Principle of Operation 527
11.19.3.1.2 LECD with Conventional Analog Feedback Control 528
11.19.3.1.3 LECD with Adaptive Tip Withdrawal Control 528
11.19.3.1.4 ECD Structures Using the Anode as a Counter Electrode 529
11.19.3.2 3D Microstructure Fabrication Using Anode as Counter Electrode and Micro-EDM 531
11.19.3.2.1 Principle of Operation 531
11.19.3.2.2 Open-Loop Control for LECD 532
11.19.3.2.3 Closed-Loop Control for LECD 533
11.19.3.2.4 Effect of Different Operating LECD Parameters 533
11.19.3.2.5 LECD Structures Using the Anode as a Counter Electrode 539
11.19.3.2.6 Microholes Fabricated by LECD Electrodes 541
11.19.4 Conclusions 544
References 545

11.19.1 Introduction

Today, fabrication of products and its miniaturization with a broad range of materials enable microsystem technology to enhance
health care and quality of life, to attain new technological breakthroughs, and to cover engineering applications with envi-
ronmentally friendly and energy-saving practices. Currently, state-of-the-art fabrication techniques refer to the fabrication of
components and parts for microelectromechanical systems (MEMS), subminiature actuators and sensors, components for
biomedical devices, high-precision equipment, components for advanced communication technology, long microchannels for
lab-on-chips, shape memory alloy ‘stents,’ fluidic graphite channels for fuel cell applications, and many more (1–4). The more
recent trends have shown that the drive has gone beyond the earlier challenge of precision and minuteness in dimension to a new
level where components of the same precision and invisible dimensions are demanded to be machined on tough materials with
lower cost.
Semiconductor processing technologies such as photolithography on a silicon substrate are used for fabricating MEMS
components (5,6). The material properties of silicon often do not meet the requirement of recent applications of these
microparts because they require high-quality structure and capability to withstand high strength. Such applications are in
microsurgery, biotechnology, fluidics, and high-temperature environments (7). Moreover, the photolithography technique is
not capable of fabricating high-aspect ratio microstructures (8,9). On the other hand, the LIGA process (from the German
Lithographie Galvanformung und Abformung – a combination of lithography, electroplating, and molding) can fabricate high-
aspect ratio components with submicron structure using the synchrotron radiation process and the focused ion beam
machining process. However, LIGA requires special and extremely expensive facilities such as a synchrotron system and requires
fabrication of expensive masks, which are not economical for microparts fabrication on a laboratory scale and fabrication
industries (8,10).
Nonconventional micromachining technology such as microturning, microgrinding, micro electro-discharge machining
(micro-EDM), and microelectrochemical machining (ECM) have many advantages in productivity, efficiency, flexibility, and
cost effectiveness; consequently, these nonconventional methods have been applied to a variety of substrates and materials to
fabricate microstructures (6,11–14). Among the nonconventional micromachining techniques, micro-EDM has provided an
efficient solution for machining hard conductive materials and fabricating complex cross-sectional structures. In order to
fabricate these complex cross-sectional structures effectively, the noncircular electrode is required, which is one of the chal-
lenges in the micro-EDM area. As an alternative, localized electrochemical deposition (LECD) is a fashionable method in the
fabrication of small and shaped electrodes directly. People have been using electrochemical deposition for centuries. The term
LECD means electrochemical deposition (ECD) in a predetermined and controlled area. By using this method, three-dimen-
sional microstructures can be made easily on high-strength metals. This process has advantages over any other microfabrication
method in terms of fabrication time and cost.

Comprehensive Materials Processing, Volume 11 http://dx.doi.org/10.1016/B978-0-08-096532-1.01109-2 523

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
524 Microelectrochemical Deposition

11.19.2 Process Mechanism

11.19.2.1 Basic Concept of ECD
ECD is a process whereby the metallic ion can become solid metal and deposited on the cathode surface if sufficient amounts of
electric current go through an electrolyte or plating solution, which is known as a solution that contains charged ions. These charged
ions, especially positively charged ions, can be achieved by dissolving metallic salt into water. Although traditionally this process
has been used for depositing coatings, it can also be used to fabricate 3D microstructure as well as both metallic and semiconducting
nanowires. If a metal M of valency n, then n electrons would be required to reduce the cation into its metallic form. A more realistic
equation is
Mnþ þ ne 0M [1]
where M metal will be deposited on the cathode by receiving n number of electrons. The total cathodic charge used in the deposition
(Q) (coulomb) is the product of the number of g mol of the metal deposited (m), the number of electrons taking part in the
reduction (n), Avogadro’s number (NA) (the number of atoms in a mole), and the electrical charge per electron (Qe) (coulomb). To
reduce 1 mol of a given metal, (n) moles of electrons are therefore required.
Q0m n NA Qe [2]
The product of the last two terms in the equation above is the Faraday constant (F). Therefore, the number of moles of metal
reduced by charge (Q) can be obtained as:
Q
m¼ [3]
nF
The total charge used in the deposition can be obtained as the product of the current (I) (ampere) and the time of deposition
(t) (second) if the deposition current is held constant. If, however, the current varies during ECD, we arrive at the following
equation.
Z
Q ¼ I dt [4]

And so the number of moles deposited can be calculated as:

Z
1
m¼ I dt [5]
nF
The weight of the deposit (w) (gram) can now be obtained by multiplying the equation above by the atomic weight (WM) of the
deposited metal. To calculate the height of the deposit, we have to use the density of the metal DM (g cm3), where Vol is the volume
of the deposited metal in cm3, AD is the area of the deposit in cm2, and Ht is its height in cm.
w w
DM ¼ ¼ [6]
Vol AD Ht
Solving the height of deposition for the unit cross-sectional area and the mole volume of deposited metal (VM) (cm3g1) is the
ratio of WM and DM, we arrive at the following equation (15,16).
Z Z
w WM VM
Ht ¼ ¼ I dt ¼ I dt [7]
AD DM nFAD DM nF
Figure 1 shows a schematic of an electrolytic cell for the ECD of metal M from an aqueous solution of metal salt MA. However, in
order to localize the ECD, pulse voltage is very helpful. In the next section, the mechanism of pulse ECD is described.

e– e–

Cathode
Anode
Deposition Deposition
M metal M metal
Mask

MA MA
Cathode Solution Anode Solution

e– e–
(a) (b)

Figure 1 Schematic diagram of ECD setup: (a) anode as a counter electrode and cathode works as a metal substrate, (b) cathode as a counter
electrode. Here mask is used in order to localize the deposited metal on the cathode.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Microelectrochemical Deposition 525

11.19.2.2 ECD Using Pulse Voltage

To make a microstructure by ECD, the deposition area should be localized. Ultrashort pulses can be applied for localizing the
deposition region because the double layer has the property of the electric condenser. Capacitance value multiplied by electrolyte
resistance value equals the double-layer charging time constant in the electrochemical cell. This charging time constant varies
according to location on the electrode surface. The distance between tip end and substrate is closer than that of tip side and
substrate. Therefore, the double layer between the former one is charged first and current flows in advance. The pulse is cut off before
charging the latter double layer, so the current flow through the electrode side is negligible. This means that localization can be
achieved without insulation of the electrode. If we choose proper pulse duration considering the charging time constant of double
layer, we can deposit copper on a substrate beneath the electrode.
When a cathode and anode electrodes are immersed in a solution, an interface consists of two equal and opposite layers of charge,
one on the metal (4m) and the other in solution (4s). This pair of charged layers, called the double layer, is equivalent to a parallel-
plate capacitor (Figure 2). The variation of potential in the double layer with the distance from the electrode is linear. The capacitance
of the double layer is a function of potential. On both the electrode surfaces, the electrochemical double layer forms a capacitor. This
double layer is charged when a potential is applied between the two electrodes. The charging time constant (sc) for the double layer is
the product of resistance (R) and capacitance (cDL). The charging current has to flow through the electrolyte, whose resistance is
proportional to the length of the current path; that is, the distance between the electrodes (dgap). Therefore, resistance is the product of
the gap distance between the electrodes (dgap) and the specific electrolyte resistivity (r). Finally, the time constant:
sc ¼ R  C ¼ r$cDL $dgap [8]

In this process, pulse potential is applied to deposit the metal ions. The charging time (tc) of the double layer should be at least
four times the time constant, that is, 98% of the pulse on time. If the duration of the pulse on time (ton) is longer than the charging
time (tc), the double layer will be charged properly for metal deposition. On the other hand, if the charging time (tc) is longer than
the pulse on time (ton), the double layer will not be charged sufficiently for metal deposition. Since the chemical reaction rate is
exponentially proportional to the potential drop in the double layer, metal deposition can be controlled by controlling the pulse
duration.
The charging time of the double layer is the time before the charged pulse potential (4c) reaches the value corresponding to the
applied pulse potential (40) (Figure 3(b)). If the charging time is longer than the duration of the pulse on time (ton), the double
layer is not completely charged and 4c never reaches to 40 (Figures 3(c) and 3(d)). A similar phenomenon occurs after the end of
the pulse. The double layer must be discharged, and it takes some time before the potential drops to the value corresponding to zero
value. Therefore, it takes some time before 4c drops to zero. If this time is longer than the off time (toff), the double layer is not
completely discharged and 4c never decreases to zero. Therefore, the charged potential of a double layer at any time (t):
  t
4c ¼ 40 1  esc z40
t
[9]
sc
When an electrode is made a part of an electrochemical cell through which current is flowing, its potential will differ from the
equilibrium potential. If the equilibrium potential of the electrode is E and the potential of the same electrode as a result of external
current flowing is E(I), then its difference is known as overpotential (h).
h ¼ EðIÞ  E [10]

m d s
Helmholtz Plane, HP

– Solution Cathode
+
–
Metal (Cathode)

–
Double layer cDL
– capacitance, cDL
Hydrated
– + ions m s R
– +
– – + cDL
– +
–
– d +
– –
+ – + Anode
– – +

(a) (b) (c)

Figure 2 (a) HP model of double layer: 4m, excess charge density on metal; 4s, excess charge density in solution, (b) HP double layer: a parallel-plate
capacitor, and (c) electrochemical cell upon application of a voltage pulse. Reproduced from Habib, M. A.; Shaleh, T.; Rahman, M. Modeling for
Fabrication of Micro Electrodes by Localized Electrochemical Deposition for Micro-EDM. Proc. Inst. Mech. Eng. Part B J. Eng. Manuf. 2010, 224 (11),
1741–1755.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
526 Microelectrochemical Deposition

t period
0 0
t on t off C

0 0
time time
(a) (c)

Applied potential ( 0)
Charged potential ( C)

0 0
C
C
0 0
time time
(b) (d)

Figure 3 Applied pulse voltage in LECD and double layer (DL) time constant effect (a) tc >> ton no damping, (b) tc < ton small damping, and (c, d)
tc > ton, tc >> ton strong damping. Reproduced from Habib, M. A.; Shaleh, T.; Rahman, M. Modeling for Fabrication of Micro Electrodes by
Localized Electrochemical Deposition for Micro-EDM. Proc. Inst. Mech. Eng. Part B J. Eng. Manuf. 2010, 224 (11), 1741–1755.

The overpotential (h) is required to overcome hindrance of the overall electrode reaction, which is usually composed of the
sequence of partial reactions. There are four possible partial reactions and thus four types of rate control: charge transfer, diffusion,
chemical reaction, and crystallization. Thus, four different kinds of overpotential are distinguished, and the total overpotential (h)
can be considered to consist of four components
h ¼ hct þ hd þ hr þ hc [11]
Here, hct, hd, hr, and hc are charge transfer, diffusion, chemical reaction, and crystallization overpotential, respectively. In order to
complete the deposition model, and simplify the formulation of determining the deposition rate and height, the following
assumptions are considered (18):
l In the deposition reaction, only copper ions are deposited.
l In the electrolyte, there are no concentration gradients; therefore the solution is well stirred.
l The diffusivity of the reacting species is constant during the deposition, and the rate of change of shape of the deposit is slow
compared with the establishment of the concentration field, and
l LECD current efficiency is unity.

Since there is no electrochemical reaction or metal deposition during the pulse off time, pulse off time voltage is comparatively
less than pulse on time voltage. For this reason, charged potential (4c) can be judged as overpotential (h). From the Butler–Volmer
equation, during the pulse on time, reaction current density (i) is:
         
ð1  aÞnF anF ð1  aÞnF anF
i ¼ i0 exp h  exp h ¼ i0 exp 4c  exp 4c [12]
RT RT RT RT

For large negative values of overpotential, the Butler–Volmer equation can be simplified. As the first exponential term in the
equation (corresponding to the anodic partial current) decreases, the second exponential term (corresponding to the cathodic
partial current) increases and the second exponential term can be neglected.
   
ð1  aÞnF anF
exp 4c [exp 4c [13]
RT RT

Therefore, reaction current density (i) during the pulse on time is:
   
ð1  aÞnF ð1  aÞnF t
i ¼ i0 exp 4c zi0 exp 40 [14]
RT RT sc
Here, i0, exchange current density; a, leakage factor; F, Faraday constant; R, gas constant; T, temperature; n, the number of electrons
taking part in the reduction. Since the reaction rate is proportional to the reaction current density, that is, z(t)Ni. This can be
represented as

i
zðtÞ ¼ [15]
nF
   
i0 ð1  aÞF t i0 ð1  aÞF 40 t
zðtÞ ¼ exp 40 ¼ exp [16]
nF RT sc nF RTrcDL dgap

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Microelectrochemical Deposition 527

The electrochemical reaction or deposition occurs only during on time of pulse. For this reason, the deposition rate can be
calculated by integrating the reaction rate during pulse on time. Therefore, the LECD rate, Z, is:
Zton Zton  
  1 1 i0 ð1  aÞF 40 t
Z 40 ; ton ; tperiod ; dgap ¼ zðtÞdt ¼ exp dt
tperiod tperiod nF RTrcDL dgap
0 0 [17]
   
dgap f i0 RTrcDL ð1  aÞF 40 ton
¼ exp 1
40 tperiod ð1  aÞzF 2 RTrcDL dgap

In order to control the LECD rate and height of deposited electrodes, four experimental parameters are used. These four variable
parameters are pulse potential amplitude (40), pulse frequency ( f ), pulse duty ratio (D), and effective gap distance between two
1 ton ton D
electrodes (dgap). We know that frequency f ¼ and duty ratio D ¼ ¼ ¼ ton  f 0ton ¼
tperiod ton þ toff tperiod f
Therefore, LECD rate, Z, is:
   
  dgap f i0 RTrcDL ð1  aÞF 40 D
Z 40 ; f ; D; dgap ¼ exp  1 [18]
40 ð1  aÞzF 2 RTrcDL dgap f

So the deposition height (Ht) at time t ¼ 0, 1, 2, 3 . will be:

H0 ¼ 0
     
dgap  H0 f i0 RTrcDL ð1  aÞF 40 D
H1 ¼ H0 þ Z1 ¼ H0 þ exp    1
40 ð1  aÞzF 2 RTrcDL dgap  H0 f
     
dgap  H1 f i0 RTrcDL ð1  aÞF 40 D
H2 ¼ H1 þ Z2 ¼ H0 þ exp    1
40 ð1  aÞzF 2 RTrcDL dgap  H1 f
.. .. .. .. .. [19]
.. .. .. .. ..
     
dgap  Ht1 f i0 RTrcDL ð1  aÞF 40 D
Ht ¼ Ht1 þ Zt ¼ Ht1 þ exp    1
40 ð1  aÞzF 2 RTrcDL dgap  Ht1 f
     
dgap  Ht f i0 RTrcDL ð1  aÞF 40 D
Htþ1 ¼ Ht þ Ztþ1 ¼ Ht þ exp    1
40 ð1  aÞzF 2 RTrcDL dgap  Ht f

The deposition rate can be obtained by eqn [18] over deposition time. The change of the deposition height can be calculated
from the deposition rate. Equation [19] shows that the deposition rate is not constant for every unit time. This is why, in order to
calculate the deposition height, the deposition rate needs to be calculated in every unit time (19). As discussed earlier, to control the
deposition rate and quality, the four variable parameters need to be optimized and properly controlled.

11.19.3 Micromanufacturing Using ECD

ECD was introduced by Madden and Hunter about a decade ago as a realistic technique for the inexpensive free-form micro-
fabrication method. ECD has a huge prospective to afford solutions to a variety of challenges for the microfabrication of three-
dimensional metal structures (20,21). Jansson et al. had deposited a nickel structure from a different kind of nickel plating
solutions (22). El-Giar and Thomson and El-Giar et al. deposited long, thin micrometer-size copper columns, copper electrical
interconnects, and tips for scanning probe microscopy applications (23,24). Yeo et al. had used opened-loop (without analog
feedback) and closed loop (with analog feedback) conditions in order to investigate the deposition phenomena of the Ni
microcolumn structure in LECD (25). Afterward, they studied the effects of ultrasonic vibration on the rate of deposition,
concentration, and porosity of the nickel microcolumns and the rotation of the electrode on the growth of nickel microcolumn
structure (26,27). Park et al. fabricated microstructures such as micropatterns, microcolumns, and microsprings by applying
ultrashort pulses with LECD (28). In order to form noncircular shaped deposition, a nonconductive mask is used, which is shaped
by a micromilling operation. The substrate is fixed on the machine z-axis, which is over the anode electrode because it eases the
next operation subsequent to the deposition. The electrode clamping error can be minimized, and the production rate can be
increased with this fabrication method.

11.19.3.1 3D Microstructure Fabrication Using the Anode as a Counter Electrode

11.19.3.1.1 Principle of Operation
The schematic drawing in Figure 4 illustrates a general arrangement of a typical setup used for LECD. A microelectrode is placed very
close to a conducting substrate, while both are immersed in an ionically conducting electrolyte that contains ions of the material to
be deposited. The microelectrode is usually insulated from all sides except for an exposed tip region with micrometer-scale

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
528 Microelectrochemical Deposition

Figure 4 An illustration of a typical LECD setup arrangement, including reaction current monitoring for feedback and probe positioning purposes.
Reproduced from Said, R. A. Microfabrication by Localized Electrochemical Deposition: Experimental Investigation and Theoretical Modeling.
Nanotechnology 2003, 14, 523–531.

dimensions. An electric potential is applied between the microelectrode and the substrate, thus causing a faradic current to flow
through the electrolyte between the microelectrode and the substrate.
Since the electrolyte contains reducible metal ions (e.g., Cu2þ ions in the present work) and the substrate is connected to
a negative potential with respect to the microelectrode, then the flow of faradic current results in an oxidation process at the
microelectrode tip and a deposition of metal ions at the substrate. Unlike typical electroplating methods, where deposition occurs at
a uniform rate on all exposed regions of the substrate, the deposition process outlined in Figure 4 is much localized to the region
beneath the electrode tip. This is due to the highly localized electric field in the space between the microelectrode tip and the
substrate region directly below the tip. The result is highly localized growth with an extent of approximately the dimensions of the
microelectrode tip, as illustrated by the lower circled drawing in Figure 4. At the start of fabrication, however, the extent of deposition
is usually less confined and extends outside the region beneath the tip due to a fringing electric field emanating from the tip
boundaries and terminating on the substrate, which will be demonstrated in the next section. As the deposition end moves farther
away from the substrate, the geometry of the deposit becomes more confined to a region of the same extent as the tip boundaries.
Copper columns were formed from CuSO4$5H2O (250 g L1) and H2SO4 (75 g L1), with thiourea (0.04 g L1) onto copper or
steel substrates. Microelectrode tips were prepared by sealing Pt, Pt–Ir, or Pt–Rh wires of 25 mm diameter. When a satisfactory seal is
obtained, the disk is exposed by polishing with successive grades of 600, 1200, and 1500 silicon carbide paper. Final polishing was
with 6.0 and 1.0 mm diamond polish on nylon cloth followed by 0.05 mm alumina on another cloth to provide a smooth surface
with nanometer scale roughness. In order to increase the aspect ratio of the deposited structure, it is very important to withdraw the
tip of the microelectrode with proper speed and control. For this reason, analog feedback control and adaptive feedback control are
discussed in the next sections.

11.19.3.1.2 LECD with Conventional Analog Feedback Control

In the conventional feedback arrangement of Figure 5, the output of a buffer current amplifier is connected to the electrode contact
pin and serves as constant voltage source. As deposition starts, a difference amplifier samples the deposition current, in the form of
an electric voltage as indicated by node a in Figure 5, and compares it with a reference value, Vref, that is usually set to represent high
current amplitudes. The addition of a current limiter at the buffer output can be of an advantage to cut the deposition current to
a limited value for the cases where the contact area between the tip and deposition end is large enough to sink a current beyond the
electrode capabilities. The difference amplifier output is cascaded to an inverter that controls an analog switch used to couple a pulse
generator to the driving circuitry of a microstepping positioner holding the electrode. Every pulse passed to the positioner circuitry
triggers the micropositioner, causing an incremental step withdrawal. Thus, the repetition rate of the pulses, ftrig, determines the
withdrawal speed of the tip and is a user controllable parameter utilized by the experimenter for adjustment relative to the
deposition rate. Successful deposition of structures with desired geometry and characteristics is critically dependent on the proper
selection of this parameter (29).

11.19.3.1.3 LECD with Adaptive Tip Withdrawal Control

Figure 6 shows a block diagram of a possible arrangement of the proposed adaptive tip withdrawal control. The output of the
differential stage is connected to an integrator, thus yielding the total instantaneous change in the deposition current. The output of
the integrator is used to tune a voltage-controlled oscillator that generates pulses at a rate proportional to the tuning voltage
amplitude and is also used to trigger positioner movement at a speed corresponding to the generated pulse rate. The pulse
generation rate would thus be modulated by the total change in the deposition current resulting from either a decrease in the tip-
deposit spacing due to deposition growth, or from an increase in the tip-deposit spacing due to tip withdrawal away from the
deposit end (29).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Microelectrochemical Deposition 529

Figure 5 A block diagram of a conventional analog feedback implemented in an LECD. Reproduced from Said, R. A. Adaptive Tip-Withdrawal Control
for Reliable Microfabrication by Localized Electrodeposition. J. Microelectromech. Syst. 2004, 13 (5), 822–832.

11.19.3.1.4 ECD Structures Using the Anode as a Counter Electrode

11.19.3.1.4.1 Microcolumn
Figure 7 shows a scanning electron microscope (SEM) image of an example of a successfully deposited copper column, which has
a diameter of about 25 mm and a length of about 6.5 mm, thus giving an aspect ratio of 1:282. The deposition rate of the deposited
column is estimated to be around 8 mm s1. After deposition, the tip shape was inspected, for possible effects such as retraction due
to erosion, and was found to have been preserved. In Figure 8 the SEM images of a column microstructure were deposited at a tip
withdrawal speed almost equal to the estimated deposition growth rate, along with the deposition current signal (Va) and control
signal (Vb) monitored during the deposition session, where Va and Vb are voltage signals at the nodes indicated in Figures 5 and 6.

Figure 6 A block diagram of the adaptive tip withdrawal control feedback. Reproduced from Said, R. A. Adaptive Tip-Withdrawal Control for Reliable
Microfabrication by Localized Electrodeposition. J. Microelectromech. Syst. 2004, 13 (5), 822–832.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
530 Microelectrochemical Deposition

Figure 7 An SEM image of a copper microcolumn deposited by LECD using conventional feedback control. Reproduced from Said, R. A. Adaptive
Tip-Withdrawal Control for Reliable Microfabrication by Localized Electrodeposition. J. Microelectromech. Syst. 2004, 13 (5), 822–832.

The quality of deposition shown in Figure 8(b) in terms of geometry confinement, reduced porosity, and enhanced uniformity has
clearly improved, even when compared to the results of equal tip withdrawal speed and deposit growth rate using the conventional
feedback control illustrated in Figure 8(a) (29).

11.19.3.1.4.2 Microspring
If an electrode moves circularly on an XY plane and upward simultaneously, microspring can be fabricated by LECD. An insulated
tip was used because spring was easy to be affected by the exposed tip side. The experimental condition was fixed with applied
voltage 2.5 V and duty ratio 0.45. Motion was controlled to make a spring with a 100 mm diameter and a 350 mm pitch. The final
shape of the deposited microspring is shown in Figure 9. The diameter of the spring coil was maintained with 10–12 mm, and the
deposited shape was uniform (28).

Figure 8 SEM images of a column microstructure deposited using (a) conventional feedback control and (b) adaptive tip withdrawal control
under same operating condition. Reproduced from Said, R. A. Adaptive Tip-Withdrawal Control for Reliable Microfabrication by Localized Electrode-
position. J. Microelectromech. Syst. 2004, 13 (5), 822–832.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Microelectrochemical Deposition 531

Figure 9 Spring made with applied voltage 2.5 V and duty ratio 0.45. Reproduced from Park, J. W.; Ryu, S. H.; Chu, C. N. Pulsed Electrochemical
Deposition for 3D Micro Structuring. Int. J. Precis. Eng. Manuf. 2005, 6 (4), 49–54.

11.19.3.1.4.3 Micropatterning
LECD using ultrashort pulses can be applied for micropatterning. Micropatterns can be made by moving the electrode on an XY
plane sustaining gap of a few micrometers from the substrate. In this research, micropatterns such as alphabets and a spiral were
written as shown in Figures 10 and 11. The growth direction of structure was perpendicular to the feed direction of electrode. So,
deposition could occur discontinuously. The width of pattern line was 10–15 m (28).

11.19.3.2 3D Microstructure Fabrication Using Anode as Counter Electrode and Micro-EDM

11.19.3.2.1 Principle of Operation
The schematic diagram of the LECD experimental setup is shown in Figure 12. The acidic super sulfate is used as an electrolyte, and an
anode is immersed in this electrolyte. A cathode is placed above the anode and between the anode and cathode; a nonconductive
mask is located to create the noncircular shape of the deposition. A small constant gap is maintained between the anode and mask
during deposition time. When both of the electrodes are conducted electrically, current will pass through the plating solution. The
positively charged metal ions get (Cu2þ) deposited as solid metal on the cathode through the nonconductive mask. In this way, metal
can be deposited on the cathode surface and electrodes can be fabricated. In this process, the cross section of the electrode will be
same as the mask. Finally, the deposited electrode can be used directly in the micro-EDM process without changing its orientation.
The LECD subsetup consists of two main parts: a cathode–electrode holder and a deposition tank (Figure 13). The electrode
holder is attached to a voice coil motor, which has a resolution of 0.1 mm fixed on the z-axis of the machine. The voice coil motor is
capable of sensing the mask and maintaining a constant distance between the anode and cathode. It does so by giving the feedback
motion after measuring the current of the system through the pico ammeter. The voice coil motor and pico ammeter are connected
with PC by RS232 serial communication. The cathode electrode where deposition will take place is attached to this voice coil motor.
The deposition tank consists of the anode electrode and the mask. In the deposition tank, a micrometer screw and two wedges are

Figure 10 Image of letters made with applied voltage 2.5 V and duty ratio 0.45 by micropatterning. Reproduced from Park, J. W.; Ryu, S. H.; Chu, C. N.
Pulsed Electrochemical Deposition for 3D Micro Structuring. Int. J. Precis. Eng. Manuf. 2005, 6 (4), 49–54.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
532 Microelectrochemical Deposition

Figure 11 Image of a spiral with applied voltage 2.5 V and duty ratio 0.35 by micropatterning. Reproduced from Park, J. W.; Ryu, S. H.; Chu, C. N.
Pulsed Electrochemical Deposition for 3D Micro Structuring. Int. J. Precis. Eng. Manuf. 2005, 6 (4), 49–54.

used to adjust the gap between the mask and anode. The mask is made from a nonconductive material like PMMA (poly methyl
methacrylate) because of its advantages over other materials; it has greater transparency, ease of fabrication, excellent alkaline, and
good acidic chemical resistance. The masks are machined in different kinds of cross-sectional shapes by using the micromilling
process such as ‘X,’ ‘Y,’ and ‘NUS.’ The thickness of the mask was 250 mm. In order to increase the aspect ratio of the deposited
structure, it is very important to withdraw the cathode with proper control. For this reason, both open-loop control and closed-loop
control are discussed in the next sections.

11.19.3.2.2 Open-Loop Control for LECD

An open-loop controller, also called a nonfeedback controller, is a type of controller that computes its input into a system using only
the current state and its model of the system. It is often used in simple processes because of its simplicity and low cost, especially in
systems where feedback is not critical. A characteristic of the open-loop controller is that it does not use feedback to determine
whether its input has achieved the desired goal. This means that the system does not observe the output of the processes that it is
controlling. Consequently, a true open-loop system cannot engage in machine learning and cannot correct any errors that it could
make. It also may not compensate for disturbances in the system. In this process, two height values input is required: one is for the

Figure 12 (a) A simple illustration of a typical LECD setup arrangement and (b) concept of the LECD and EDM setup. Reproduced from Habib, M. A.
Development of Localized Electrochemical Deposition Process for the Fabrication of On-Machine Micro-EDM Electrode. Ph.D. Thesis, National
University of Singapore (NUS), 2010.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Microelectrochemical Deposition 533

Figure 13 Schematic diagram of LECD EDM combined process. Reproduced from Habib, M. A. Development of Localized Electrochemical Deposition
Process for the Fabrication of On-Machine Micro-EDM Electrode. Ph.D. Thesis, National University of Singapore (NUS), 2010.

initial growth height (Hi), and the other is for the final desired height (Ht). The controller feedback system will start when the
deposition reaches Hi height. When the controller receives the interrupt, then it will shift the motor by (Ht–Hi) distance. In this way,
it is possible to increase the aspect ratio by open-loop control. The detailed algorithm of the open-loop control system is shown in
Figure 14 (19).

11.19.3.2.3 Closed-Loop Control for LECD

This section provides the design of a controller for the LECD process in order to increase the height of the deposited structure, which
was one of the main objectives of this project. To increase the height of the deposited structure, it is necessary to lift up the cathode
from the mask. The reason is, if the cathode is kept in the same initial place, then after a certain time the deposited structure will
come out from the mask and the actual structure of the deposited electrode will be affected. This is why, to keep the deposited
structure inside the mask, it is required to lift up the cathode from its original position. However, in order to lift up the cathode, it is
very necessary to know the deposition height at the particular time. This is why two different controllers are designed: One is an
open-loop controller, and the other is a closed-loop controller in order to control the lifting mechanism of the cathode. For the two
controllers, two input heights are needed to be given for control simulation. One is the initial growth height, and the other is the
final desired height. The next subsections present the determination of the initial growth height, as well as the design of an open-
loop and closed-loop controller.
Generally, to obtain a more accurate or more adaptive control, it is necessary to feed the output of the system back to the inputs
of the controller. This type of system is called a closed-loop system. A closed-loop system utilizes feedback to measure the actual
system operating parameter being controlled such as temperature, pressure, flow, level, or speed. This feedback signal is sent back to
the controller where it is compared with the desired system set point. In order to design such a controller, a system model has been
derived from Faraday’s basic law of electrochemistry (19).
The detailed algorithm of the closed-loop control that is implemented in the process is given in Figure 15.

11.19.3.2.4 Effect of Different Operating LECD Parameters

Figures 17–24 present the simulation of deposition height and rate with respect to deposition time at different levels of voltage
amplitudes, frequencies, duty ratios, and gap distance based on eqns [18] and [19]. The details LECD parameters are tabulated in
Table 1. In order to study the effect of pulse amplitude (40)f, D and dgap are kept constant on 100 kHz, 0.33, and 350 mm. Similarly,
to study the effect of pulse frequency (f)40, D and dgap are kept constant on 1.5 V, 0.33, and 350 mm. Likewise, to study the effect of
the pulse duty ratio (D)40, f and dgap are kept constant on 1.5 V, 100 kHz, and 350 mm. Lastly, to study the effect of the electrode gap
distance (dgap)40, f and D are kept constant on 1.5 V, 100 kHz, and 0.33. In order to verify the mathematical model, experimental
results are also shown in the figures. Results show that in case of a higher deposition rate the experimental results are not properly
matched with the simulation results. The simulation results show that the deposition height and rate are higher than those of
experimental results (19).
At the beginning of the deposition process, the anode electrode touches the mask. The anode electrode surface is properly
polished, but still there is some horizontal misalignment. This is why the electrolyte leaks through the gap (Figure 16). This causes

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
534 Microelectrochemical Deposition

START

Electrode Positioning

Starting of Deposition

Calculating the cumulative initial Initial growth

growth height from equation (4.12) Height (Hi)

Total initial growth height Hi

False

True

Feeding backward of (Ht-Hi)μm

in every t sec interval according to
equation (4.12)

Final
Count of total height = required
Total backward height + Hi Height(H )

Total Height H
False

True

END

Figure 14 Algorithm for open-loop control. Reproduced from Habib, M. A. Development of Localized Electrochemical Deposition Process for the
Fabrication of On-Machine Micro-EDM Electrode. Ph.D. Thesis, National University of Singapore (NUS), 2010.

the metal deposited on the places the electrolyte reaches. For this reason, in the experimental results the initial rate is not matched
with simulation results. However, this phenomenon has also occurred in a low deposition rate, but the effect is comparatively less.
After the above leakage, the gap seals within a very short time. This is why the effect is not that great. Once the deposited metal enters
into the mask, the gap is sealed. After the above conditions, the results show that the deposition height of experimental data is
higher than simulation data. One probable occurrence may be that when the deposition starts the top surface is not perfectly flat all
the time. Some peaks and valleys can be visible on the top surface (Figure 16). For measuring the height of the deposition, only
the peaks are taken into account. This is why the total height becomes higher. However, the actual equivalent height is less than the
measurement height. The effects of different parameters are given in the following (19).

11.19.3.2.4.1 Effect of Pulse Voltage amplitude

Figures 17 and 18 indicate that the deposition height and rate increase with the increase of deposition time at all voltage ampli-
tudes. Moreover, at a certain point, the deposition height and rate increase almost suddenly and the system becomes uncontrollable.
This point of transition is different for different voltages. For lower voltage, the transition point arrives at a higher deposition time.
On the other hand, for higher voltage, the position is the reverse. For a voltage value lower than 1.5 V, the deposition rate is
comparatively lower than the higher voltages. The results indicate that below 1.5 V, the supplied energy is not sufficient to deposit
the material at a higher rate (17).
On the other hand, for a voltage value higher than 1.5 V, the deposition rate is comparatively higher than the lower
voltages. Due to the higher deposition rate after the transition point, the deposition structures become tree and powder type.
One possible explanation is that at a certain point of high-voltage deposition, the corresponding current value exceeds the
limit value of the electrolyte, which causes the deposit to become powdery. It is probable that the area surrounding the
anode becomes depleted of ions for discharging the anode. At the same time, a high volume of hydrogen gas is produced at
the anode (17).

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Microelectrochemical Deposition 535

START

Electrode Positioning

Starting of Deposition

Calculating and summing the initial Initial growth

growth height from equation (4.12) Height (Hi)

Total initial growth height Hi

False
True

Measurement of initial current

Reference
Current
(I ref)
Measure of current value (I) of
present position

I > I ref
True False

Feeding backward with the minimum Feeding forwardward with the minimum
resolution of motor (0.1μm) resolution of motor (0.1μm)

Count of total height = Count of total height =

Hi + Total backward height Hi - Total forward height

Input of Desired
Total Feedback height H
False Height(H )

True

END

Figure 15 Algorithm for closed-loop control. Reproduced from Habib, M. A. Development of Localized Electrochemical Deposition Process for the
Fabrication of On-Machine Micro-EDM Electrode. Ph.D. Thesis, National University of Singapore (NUS), 2010.

Extra
Gap Anode deposited metal
Electrode

Mask Mask
Electrolyte

Deposited
Anode electrode
Electrode Peaks and
valleys
Mask Mask
Electrolyte
(a) (b)

Figure 16 (a) Showing the gap between the electrode and mask and (b) SEM image showing the extradeposited material through the gap. Reproduced
from Habib, M. A.; Shaleh, T.; Rahman, M. Modeling for Fabrication of Micro Electrodes by Localized Electrochemical Deposition for Micro-EDM.
Proc. Inst. Mech. Eng. Part B J. Eng. Manuf. 2010, 224 (11), 1741–1755.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
536 Microelectrochemical Deposition

180

160

140

Deposition height (μm)

120

100

80

60

40

20

0
0 500 1000 1500 2000 2500
Deposition time (s)

1.2V (sim) 1.5V (sim) 1.6V (sim) 1.8V (sim) 2.0V (sim)
1.2V (exp) 1.5V (exp) 1.6V (exp) 1.8V (exp) 2.0V (exp)

Figure 17 Effect of pulse voltage amplitude on deposition height (simulation and experimental). Reproduced from Habib, M. A.; Shaleh, T.; Rahman,
M. Modeling for Fabrication of Micro Electrodes by Localized Electrochemical Deposition for Micro-EDM. Proc. Inst. Mech. Eng. Part B J. Eng. Manuf.
2010, 224 (11), 1741–1755.

11.19.3.2.4.2 Effect of Pulse Voltage Frequency

The simulation and experimental results of Figures 19 and 20 show that the deposition height and rate increase with the increase of
deposition time at any value of frequency. For a frequency value higher than 100 kHz, the deposition rate is comparatively lower
than the lower frequencies. In contrast, for frequency values lower than 100 kHz, the deposition rate is comparatively higher than
the lower frequencies (17).
It is well known that the frequency has an inverse relationship with the pulse period. This is why the pulse period or duration
increases with the decrease of the pulse frequency. With the increase of the pulse period time, the amount of energy per pulse also
increases. Owing to the increased deposition energy, the number of ions also increases for deposition, which causes the deposition
rate to increase. Alternatively, when the frequency increases, the condition will be the reverse (17).

0.3

0.25
Deposition rate (μm s–1)

0.2

0.15

0.1

0.05

0
0 500 1000 1500 2000 2500
Deposition time (s)
1.2V (sim) 1.5V (sim) 1.6V (sim) 1.8V (sim) 2.0V (sim)
1.2V (exp) 1.5V (exp) 1.6V (exp) 1.8V (exp) 2.0V (exp)

Figure 18 Effect of pulse voltage amplitude on deposition rate (simulation and experimental). Reproduced from Habib, M. A.; Shaleh, T.; Rahman, M.
Modeling for Fabrication of Micro Electrodes by Localized Electrochemical Deposition for Micro-EDM. Proc. Inst. Mech. Eng. Part B J. Eng. Manuf.
2010, 224 (11), 1741–1755.
www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Microelectrochemical Deposition 537

180

160

140

Deposition height (μm)

120

100

80

60

40

20

0
0 500 1000 1500 2000 2500
Deposition time (s)

70kHz (sim) 85kHz (sim) 100kHz (sim) 115kHz (sim) 130kHz (sim)
70kHz (exp) 85kHz (exp) 100kHz (exp) 115kHz (exp) 130kHz (exp)

Figure 19 Effect of pulse voltage frequency on deposition height (simulation and experimental). Reproduced from Habib, M. A.; Shaleh, T.; Rahman,
M. Modeling for Fabrication of Micro Electrodes by Localized Electrochemical Deposition for Micro-EDM. Proc. Inst. Mech. Eng. Part B J. Eng. Manuf.
2010, 224 (11), 1741–1755.

11.19.3.2.4.3 Effect of Pulse Voltage Duty Ratio

For any particular duty ratio, the deposition height and rate increase with the increase of deposition time, and the results are shown
in Figures 21 and 22. For the 0.2 and 0.25 duty ratio, however, the deposition rate is relatively lower than the higher duty ratio.
Conversely, for the 0.4 and 0.5 duty ratio the deposition rate is relatively higher than the lower duty ratio, and after a certain point it
increases rapidly. A reasonable explanation is that this study is conducted on a fixed pulse frequency where the pulse period is fixed
and the duty ratio is the ratio of pulse on time and pulse period. This is why, due to the decrease of duty ratio, the pulse on time also
decreases (17).
If the pulse on time is much less than the double-layer time constant, then there will be strong damping in the system. This
incidence causes the insufficient charging and discharging of the double layer. As a final point, the deposition rate decreases owing

0.45

0.4

0.35
Deposition rate (μm s–1)

0.3

0.25

0.2

0.15

0.1

0.05

0
0 500 1000 1500 2000 2500
Deposition time (s)

70kHz (sim) 85kHz (sim) 100kHz (sim) 115kHz (sim) 130kHz (sim)
70kHz (exp) 85kHz (exp) 100kHz (exp) 115kHz (exp) 130kHz (exp)

Figure 20 Effect of pulse voltage frequency on deposition rate (simulation and experimental). Reproduced from Habib, M. A.; Shaleh, T.; Rahman, M.
Modeling for Fabrication of Micro Electrodes by Localized Electrochemical Deposition for Micro-EDM. Proc. Inst. Mech. Eng. Part B J. Eng. Manuf.
2010, 224 (11), 1741–1755.
www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
538 Microelectrochemical Deposition

180

160

140

Deposition height (μm)

120

100

80

60

40

20

0
0 500 1000 1500 2000 2500
Deposition time (s)

0.20 (sim) 0.25 (sim) 0.33 (sim) 0.40 (sim) 0.50 (sim)
0.20 (exp) 0.25 (exp) 0.33 (exp) 0.40 (exp) 0.50 (exp)

Figure 21 Effect of pulse voltage duty ratio on deposition height (simulation and experimental). Reproduced from Habib, M. A.; Shaleh, T.; Rahman, M.
Modeling for Fabrication of Micro Electrodes by Localized Electrochemical Deposition for Micro-EDM. Proc. Inst. Mech. Eng. Part B J. Eng. Manuf.
2010, 224 (11), 1741–1755.

to this occurrence. Inversely, for a higher duty ratio the pulse on time is much higher than the double-layer time constant, causing an
increase in deposition energy. Due to the increase of deposition energy, the number of ions also increases for deposition, which
causes the deposition rate to increase (17).

11.19.3.2.4.4 Effect of Electrode Effective Gap Distance

As can be seen from the simulation as well as experimental results, the deposition height and rate increase with the increase of
deposition time at any electrode effective gap distance (Figures 23 and 24). In addition, with the increase of gap distance, the overall
deposition rate decreases. These results can be adequately explained by the double-layer time constant characteristics, which is

0.6

0.5
Deposition rate (μm s–1)

0.4

0.3

0.2

0.1

0
0 500 1000 1500 2000 2500
Deposition time (s)

0.20 (sim) 0.25 (sim) 0.33 (sim) 0.40 (sim) 0.50 (sim)
0.20 (exp) 0.25 (exp) 0.33 (exp) 0.40 (exp) 0.50 (exp)

Figure 22 Effect of pulse voltage duty ratio on deposition rate (simulation and experimental). Reproduced from Habib, M. A.; Shaleh, T.; Rahman, M.
Modeling for Fabrication of Micro Electrodes by Localized Electrochemical Deposition for Micro-EDM. Proc. Inst. Mech. Eng. Part B J. Eng. Manuf.
2010, 224 (11), 1741–1755.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Microelectrochemical Deposition 539

180

160

140

Deposition height (μm)

120

100

80

60

40

20

0
0 500 1000 1500 2000 2500
Deposition time (s)

350 μm (sim) 400 μm (sim) 450 μm (sim) 500 μm (sim) 600 μm (sim)
350 μm (exp) 400 μm (exp) 450 μm (exp) 500 μm (exp) 600 μm (exp)

Figure 23 Effect of gap distance on deposition height (simulation and experimental). Reproduced from Habib, M. A.; Shaleh, T.; Rahman, M. Modeling
for Fabrication of Micro Electrodes by Localized Electrochemical Deposition for Micro-EDM. Proc. Inst. Mech. Eng. Part B J. Eng. Manuf. 2010,
224 (11), 1741–1755.

a product of resistivity, capacitance, and gap distance. If the gap distance increases, then the time constant also will increase. When
the time constant increases, it takes more time to charge the double layer. This will lead to a strong damping condition. Due to this
insufficient charging and discharging of the double layer, the deposition rate decreases (17).

11.19.3.2.5 LECD Structures Using the Anode as a Counter Electrode

LECD is an extraordinary method for easily making noncircular cross-sectional electrodes. This process has advantages in terms
of fabrication time and cost compared with any other microfabrication methods. To fabricate an electrode, a plating solution of
acidic CuSO4$5H2O with a Cu2þ ion concentration of 1.0–1.25 M and some organic additives such as thiourea is used to

0.25

0.2
Deposition rate (μm s–1)

0.15

0.1

0.05

0
0 500 1000 1500 2000 2500
Deposition time (s)

350 μm (sim) 400 μm (sim) 450 μm (sim) 500 μm (sim) 600 μm (sim)
350 μm (exp) 400 μm (exp) 450 μm (exp) 500 μm (exp) 600 μm (exp)

Figure 24 Effect of gap distance on deposition rate (simulation and experimental). Reproduced from Habib, M. A.; Shaleh, T.; Rahman, M. Modeling
for Fabrication of Micro Electrodes by Localized Electrochemical Deposition for Micro-EDM. Proc. Inst. Mech. Eng. Part B J. Eng. Manuf. 2010,
224 (11), 1741–1755.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
540 Microelectrochemical Deposition

Table 1 LECD parameter for simulation and experiments

Parameters Value

Pulse amplitude, 40 1.2, 1.5, 1.6, 1.8, and 2.0 V

Pulse frequency, f 70, 85, 100, 115, and 130 kHz
Pulse duty ratio, D 0.20, 0.25, 0.33, 0.40, 0.50
Electrode gap distance, dgap 350, 400, 450, 500, and 600 mm
Exchange current density, i0 1.5 mA cm2
Specific electrolyte resistivity, r 10 U cm
Leak factor, a 0.5
Stoichiometric number, n 2
Specific capacitance, cDL 10 mF cm2
Temperature, T 298.15 K
Copper mole volume 7.11 cm3 mol1
Faraday constant, F 96485 C mol1
Gas constant, R 8.314 J mol1 K1

improve the microstructure of the copper deposit. Based on the parametric study of the previous section, an optimized elec-
trode gap of 350 mm, voltage amplitude of 1.6 V, voltage frequency of 100 kHz, and duty ratio of 0.33 is used to fabricate
a good structure of copper. These optimum conditions produce a smooth, fine-grained, and low-porosity copper electrode
suitable for use in EDM.
Figures 25(a) and 25(b) show the SEM images of deposited electrodes before and after the EDM process. Figures 25(c) and
25(d) show that the structural accuracy is better for a closed-loop than for an open-loop control. However, in the top side of the
structure the deposited structure is not uniform. This may be due to the change of copper concentration during the deposition
process. When the concentration becomes less than a certain limit, then the number of ions available for discharging is low when
the concentration is low, creating a depletion layer just beneath the electrode. There is no such effect on the deposition rate due to
this occurrence, but the deposited structures are irregular and highly porous and it will become tree type.
Figure 26 represents the EDX spectrum analysis of the deposited electrode before and after EDM. The mass percentages of copper
(Cu), carbon (C), and oxygen (O) obtained in the EDX spectrum in Figure 26(a) were 81.37%, 12.31%, and 2.73%, respectively.
On the other hand, in Figure 26(b) the relative masses of Cu, C, and O were 55.94%, 39.87%, and 4.09%, respectively. This analysis
shows that after EDM there is more carbon and oxygen content than before EDM. This excess amount of carbon came from the
dielectric oil in the form of debris and burrs that remain resolidified over the surface of the electrode. Sometimes oxidization of
debris occurred during resolidification at the edge; hence, the oxygen content (19).

Figure 25 X shape deposited electrode (a) before EDM and (b) after EDM. Deposited structure for (c) open loop control and (d) close loop control.
Reproduced from Habib, M. A. Development of Localized Electrochemical Deposition Process for the Fabrication of On-Machine Micro-EDM
Electrode. Ph.D. Thesis, National University of Singapore (NUS), 2010.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Microelectrochemical Deposition 541

Figure 26 EDX analysis of a deposited electrode (a) before EDM and (b) after EDM. Reproduced from Habib, M. A. Development of Localized
Electrochemical Deposition Process for the Fabrication of On-Machine Micro-EDM Electrode. Ph.D. Thesis, National University of Singapore (NUS),
2010.

11.19.3.2.6 Microholes Fabricated by LECD Electrodes

Figures 27(a) and 27(b) shows that, at almost all levels of discharge energy the microholes obtained were free of burrs and recast
layers, and improved circularity was achieved. By using a very low value of capacitance, the discharge energy can be minimized,
which can give good surface finish and edge linearity. Therefore, as the energy per pulse is smaller in the RC circuit, smaller craters
are generated, which means smaller amount of material is removed per cycle. For this reason, it is easy to wash away the debris from
the machining zone by the low-pressured side dielectric flushing.

Figure 27 (a) Entrance and (b) exit side SEM image of microhole with LECD electrode at different energy level of discharge energy on stainless steel.
Reproduced from Habib, M. A.; Shaleh, T.; Rahman, M. Modeling for Fabrication of Micro Electrodes by Localized Electrochemical Deposition for
Micro-EDM. Proc. Inst. Mech. Eng. Part B J. Eng. Manuf. 2010, 224 (11), 1741–1755 and Habib, M. A. Development of Localized Electrochemical
Deposition Process for the Fabrication of On-Machine Micro-EDM Electrode. Ph.D. Thesis, National University of Singapore (NUS), 2010.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
542 Microelectrochemical Deposition

11.19.3.2.6.1 Parameters Influencing the Micro-EDM of LECD Electrode

The RC-type pulse generator is mainly determined by the gap voltage and the capacitance. In RC-type pulse generator, the maximum
discharge energy per pulse that can be obtained from RC circuit is:
1
Eds ¼ CV 2 [20]
2
where, C ¼ capacitance and V ¼ gap voltage.
So in the RC-type generator, the performance of the micro-EDM process can be more precisely controlled by knowing the effect
of only the gap voltage and the capacitance. For the fine-finish micro-EDM, the discharge energy should be minimized, which can be
more easily done in the RC generator by using low values of voltage and capacitance. To determine the optimum conditions for
quality microholes in different workpiece materials with LECD electrodes, a series of experiments were conducted by varying the
major operating parameters. The machining conditions for this study are listed in Table 2. Therefore, to assess the optimum
conditions and dimensional accuracy of the microholes the material removal rate (MRR), relative wear ratio (RWR), average spark
gaps (ASG), and average taper angel (ATA) were also measured. Equations [21] to [24] were used to calculate the MRR, RWR, ASG,
and ATA, respectively. Figures 28(a) and 28(b) shows the ASG and ATA measuring procedure.
Amount of material removed from workpiece ðvolm Þ
MRR ¼ [21]
Unit time

Amount of material removed from electrode ðvolm Þ

RWR ¼ [22]
Amount of material removed from workpiece ðvolm Þ
 
1 g1 þ g2 þ g3 þ g4
ASG ¼  a [23]
2 4
 
1 dtop  dbottom
ATA; q ¼ tan [24]
2h
Here, a is actual dimension (mask), g1, g2, g3, g4 are the machined dimensions (hole), dtop and dbottom are the top and bottom
diameter of the hole, and h is the height of the workpiece.

11.19.3.2.6.2 Effect of Gap Voltage

The gap voltage plays an important role in micro-EDM applications. Figures 29(a), 29(e), and 29(g) shows that with the increase of
voltage, the MRR, ASG, and ATA increase for all values of capacitance. This is because; from eqn [20], it is clear that when the gap
voltage increases the discharge energy as well as the spark gap also increase. Taper angle depends on the spark gap and material

Table 2 Machining parameters of RC pulse generator

micro-EDM for microholes machining of LECD electrode

Parameters Values

EDM circuit R–C

Supply voltage (V) 60, 80, 100, 120, 140
Capacitance (pf) 100, 220, 470, 1000, 2200
Resistance (kU) Fixed to 1 kU
Dielectric coolant EDM oil 3

Figure 28 Measurement of (a) average spark gap and (b) taper angle q. Reproduced from Habib, M. A.; Shaleh, T.; Rahman, M. Modeling for
Fabrication of Micro Electrodes by Localized Electrochemical Deposition for Micro-EDM. Proc. Inst. Mech. Eng. Part B J. Eng. Manuf. 2010, 224 (11),
1741–1755 and Habib, M. A. Development of Localized Electrochemical Deposition Process for the Fabrication of On-Machine Micro-EDM Electrode.
Ph.D. Thesis, National University of Singapore (NUS), 2010.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Microelectrochemical Deposition 543

0.0035 0.0035
0.0030 0.0030

MRR (mm3 min–1)

MRR (mm3 min–1)

0.0025 0.0025
0.0020 0.0020
0.0015 0.0015
0.0010 0.0010
0.0005
0.0005
0.0000
0.0000
60 80 100 120 140
100 220 470 1000 2200
Gap Voltage (V) Capacitance (pf)
c = 100 pf c = 220 pf c = 470 pf c = 1000 pf c = 2200 pf 60 Volt 100 Volt 140 Volt
(a) (b)
60 60

50 50

40 40

RWR (%)
RWR (%)

30 30
20
20
10
10
0
0 100 220 470 1000 2200
60 80 100 120 140
Capacitance (pf)
Gap Voltage (V)
c = 100 pf c = 220 pf c = 470 pf c = 1000 pf c = 2200 pf 60 Volt 100 Volt 140 Volt
(c) (d)
20 20
18 18
Average spark gap (μm)

Average spark gap (μm)

16 16
14 14
12 12
10 10
8 8
6 6
4 4
2 2
0 0
60 80 100 120 140 100 220 470 1000 2200
Gap Voltage (V) Capacitance (pf)
c = 100 pf c = 220 pf c = 470 pf c = 1000 pf c = 2200 pf 60 Volt 100 Volt 140 Volt
(e) (f)
12 12

10 10
Taper angle (degree)

Taper angle (degree)

8 8

6 6

4 4

2 2

0 0
60 80 100 120 140 100 220 470 1000 2200
Gap Voltage (V) Capacitance (pf)
c = 100 pf c = 220 pf c = 470 pf c = 1000 pf c = 2200 pf 60 Volt 100 Volt 140 Volt
(g) (h)

Figure 29 Effect of gap voltage on (a) MRR, (c) RWR, (e) average spark gap, (g) taper angle; effect of capacitance on (b) MRR, (d) RWR, (f) average
spark gap and (h) taper angle. Reproduced from Habib, M. A.; Shaleh, T.; Rahman, M. Modeling for Fabrication of Micro Electrodes by Localized
Electrochemical Deposition for Micro-EDM. Proc. Inst. Mech. Eng. Part B J. Eng. Manuf. 2010, 224 (11), 1741–1755 and Habib, M. A. Development
of Localized Electrochemical Deposition Process for the Fabrication of On-Machine Micro-EDM Electrode. Ph.D. Thesis, National University of Singapore
(NUS), 2010.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
544 Microelectrochemical Deposition

Figure 30 (a) NUS shape deposited electrode; and (b) NUS shape hole was machined by NUS shape electrode with EDM discharge energy of 2.35 mJ.
Reproduced from Habib, M. A.; Shaleh, T.; Rahman, M. Modeling for Fabrication of Micro Electrodes by Localized Electrochemical Deposition for
Micro-EDM. Proc. Inst. Mech. Eng. Part B J. Eng. Manuf. 2010, 224 (11), 1741–1755; Habib, M. A. Development of Localized Electrochemical Deposition
Process for the Fabrication of On-Machine Micro-EDM Electrode. Ph.D. Thesis, National University of Singapore (NUS), 2010 and Habib, M. A.;
Rahman, M. Performance Analysis of EDM Electrode Fabricated by Localized Electrochemical Deposition for Micro-Machining of Stainless Steel. Int. J.
Adv. Manuf. Technol. 2010, 49 (9–12), 975–986.

removal rate. This is why when the spark gap increases the taper angle also increases. However, in the lower voltage, the taper angle
is higher than the higher voltage. This is due to the material removal rate is low in lower voltage that means the machining time
longer which cause the taper angle larger. In an RC-type pulse generator, relatively smaller craters are generated due to its lower
energy per pulse, and debris created by machining is flushed away from the machining zone by the dielectric. Figure 29(c) shows
that the relative wear ratio increases with the increase of gap voltage due to increased discharge energy. In the case of dimensional
accuracy, Figure 29(e) shows that the dimensional accuracy decreases as the average spark gap and taper angle increase due to the
increase of gap voltage. In order to improve the dimensional accuracy, appropriate gap voltage, and capacitance value can be used to
achieve around a 2 mm average spark gap and a 2 of taper angle (17,19, 30).

11.19.3.2.6.3 Effect of Capacitance

The capacitor serves a very important role in micro-EDM applications. In the RC-type pulse generator, the capacitor controls the
charging and discharging pulse frequency. Therefore, the nano pulse can be generated in the RC type with very short pulse
duration. For this purpose, by using a very low value of capacitor, the pulse energy minimization can be easily fulfilled. This is
why, by changing the capacitor value, good dimensional accuracy can be achieved. Figure 29(b) shows that with the increase of
capacitance the MRR increases as the discharge energy increases. Therefore, the larger capacitance results in deeper craters,
which increase the material removal. The RWR also increases with the increase of capacitance. However, Figure 29(d) shows
that at a very high value of capacitance the RWR decreases. Actually, this is not investigative of lower electrode wear. In this
study, the electrode wear was measured as a ratio of the volume of electrode material eroded to the volume of material
removed from the workpiece. In the case of lower capacitance, RWR is high for lower voltage, because in this condition MRR is
very low, which causes the electrode to erode more. For this reason, the ratio shows a decreased value, although more material
is removed from the electrode compared to that of lower capacitance. Therefore, at a very high value of capacitor, the average
spark gap increases, which causes more material to be removed with respect to electrode erosion (Figure 29(f)). Moreover, the
dimensional accuracy is also reduced as the average spark gap and taper angle (Figure 29(f) and 29(h)) increase with respect to
gap voltage (17,19, 30).
Figure 30 shows the deposited electrode with NUS shape, and the NUS shape microhole is fabricated on an austenitic stainless
steel (SUS 304) workpiece.

11.19.4 Conclusions

LECD is a remarkable method for easily making a 3D microstructure. This process has advantages in a variety of materials and cost
efficiency compared with other microfabrication methods. The LECD process is capable of fabricating on-machine noncircular
microelectrodes. These microelectrodes can be used directly in the micro-EDM process. This fabrication process will be a good
solution in MEMS and bio-MEMS industries, where concircular-shaped holes and cavities are required to fabricate. Moreover, this
fabrication process is very effective for industrial applications, where production time and cost can be minimized.

See also: Introduction to Advanced Machining Technologies; Compound and Hybrid Micromachining Processes; Compound and
Hybrid Micromachining; Micromilling; Micro-Electrical Discharge Machining (Micro-EDM): Processes, Varieties, and Applications;
Electrochemical Micromachining.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬
 Microelectrochemical Deposition 545

References

1. Corbett, J.; McKeown, P. A.; Peggs, G. N.; Whatmore, R. Nanotechnology: International Developments and Emerging Products. CIRP Ann. 2000, 49, 523–546.
2. Lang, W. Reflexions on the Future of Microsystems. Sens. Actuators 1999, 72, 1–15.
3. Madou, M. J. Fundamentals of Microfabrication; CRC Press: Boca Raton, 1997.
4. Weck, M.; Fischer, S.; Vos, M. Fabrication of Micro Components Using Ultra Precision Machine Tools. Nanotechnology 1997, 8, 145–148.
5. Meeusen, W.; Clijnen, J.; Reynaerts, D.; Van, H. Micro-Electro-Discharge Machining as Microsensor Fabrication Technology. IEEE Sens. J. 2003, 3 (5), 632–639.
6. Schoth, A.; Förster, A.; Menz, W. Micro Wire EDM for High Aspect Ratio 3D Microstructuring of Ceramics and Metals. Microsyst. Technol. 2005, 11, 250–253.
7. Kuo, C. L.; Huang, J. D.; Liang, H. Y.; Huang, J. D. Fabrication of 3D Metal Microstructures Using a Hybrid Process of Micro-EDM and Laser Assembly. Int. J. Adv. Manuf.
Technol. 2003, 21, 796–800.
8. Okuyama, H.; Takada, H. Micromachining with SR and FEL. Nucl. Instrum. Meth. Phys. Res. B 1998, 144, 58–65.
9. Rajurkar, K. P.; Yu, Z. Y. 3D Micro-EDM Using CAD/CAM. CIRP Ann. 2000, 49 (1), 127–130.
10. Ananthakrishnan, A.; Sarma, R.; Ananthasuresh, G. K. Systematic Mask Synthesis for Surface Micromachined Microelectromechanical Systems. J. Micromech. Microeng.
2003, 13, 927–941.
11. Fang, F. Z.; Liu, K.; Kurfess, T. R.; Lim, G. C. Tool-Based Micromachining and Applications in MEMS. In MEMS/NEMS Handbook Techniques and Applications, Vol. 3; Springer:
US, 2006; pp 678–740.
12. Li, T.; Gianchandani, Y. B. A Micromachining Process for Die-Scale Pattern Transfers in Ceramics and Its Application to Bulk Piezoelectric Actuators. J. Microelectromech. Syst.
2006, 15 (3), 605–612.
13. Yu, Z. Y.; Masuzawa, T.; Fujino, M. Micro-EDM for Three-Dimensional Cavities–Development of Uniform Wear Method. CIRP Ann. 1998, 47 (1), 169–172.
14. Zhao, W.; Yang, Y.; Wang, Z.; Zhang, Y. A CAD/CAM System for Micro-ED-Milling of Small 3D Freeform Cavity. J. Mater. Process. Technol. 2004, 149, 573–578.
15. Habib, M. A.; Gan, S. W.; Rahman, M. Fabrication of Complex Shape Electrodes by Localized Electrochemical Deposition. J. Mater. Process. Technol. 2009, 209 (2),
4453–4458.
16. Habib, M. A.; Gan, S. W.; Lim, H. S.; Rahman, M. Fabrication of EDM Electrodes by Localized Electrochemical Deposition. Int. J. Precis. Eng. Manuf. 2008, 9 (2), 75–80.
17. Habib, M. A.; Shaleh, T.; Rahman, M. Modeling for Fabrication of Micro Electrodes by Localized Electrochemical Deposition for Micro-EDM. Proc. Inst. Mech. Eng. Part B J. Eng.
Manuf. 2010, 224 (11), 1741–1755.
18. Said, R. A. Microfabrication by Localized Electrochemical Deposition: Experimental Investigation and Theoretical Modeling. Nanotechnology 2003, 14, 523–531.
19. Habib, M. A. Development of Localized Electrochemical Deposition Process for the Fabrication of On-Machine Micro-EDM Electrode. Ph.D. Thesis, National University of
Singapore (NUS), 2010.
20. Hunter, I. W.; Lafontaine, S. R.; Madden, J. D. Three-Dimensional Microfabrication by Localized Electrochemical Deposition and Etching. U.S. Patent Specification
5641391, 1997.
21. Madden, J. D.; Hunter, I. W. Three-Dimensional Microfabrication by Localized Electrochemical deposition. J. Microelectromech. Syst. 1996, 5 (1), 24–32.
22. Jansson, A.; Thornell, G.; Johansson, S. High Resolution 3D Microstructures Made by Localized Electrodeposition of Nickel. J. Electrochem. Soc. 2000, 147, 1810–1827.
23. El-Giar, E. M.; Cairo, U.; Thomson, J. D. Localized Electrochemical Plating of Interconnectors for Microelectronics. In Proc. IEEE Conf. Communications, Power and Computing,
WESCANEX 97, 1997, pp 327–332.
24. El-Giar, E. M.; Said, R. A.; Bridges, G. E.; Thomson, D. J. Localized Electrochemical Deposition of Copper Microstructures. J. Electrochem. Soc. 2000, 147, 586–591.
25. Yeo, S. H.; Choo, J. H.; Yip, K. S. Localized Electrochemical Deposition – The Growth Behaviour of Nickel Micro-Columns. Proc. SPIE 2000, 4174, 30–39.
26. Yeo, S. H.; Choo, J. H. Effects of Rotor Electrode in the Fabrication of High Aspect Ratio Microstructures by Localized Electrochemical Deposition. J. Micromech. Microeng.
2001, 11, 435–442.
27. Yeo, S. H.; Choo, J. H.; Sim, K. H. A. On the Effects of Ultrasonic Vibrations on Localized Electrochemical Deposition. J. Micromech. Microeng. 2002, 12, 271–279.
28. Park, J. W.; Ryu, S. H.; Chu, C. N. Pulsed Electrochemical Deposition for 3D Micro Structuring. Int. J. Precis. Eng. Manuf. 2005, 6 (4), 49–54.
29. Said, R. A. Adaptive Tip-Withdrawal Control for Reliable Microfabrication by Localized Electrodeposition. J. Microelectromech. Syst. 2004, 13 (5), 822–832.
30. Habib, M. A.; Rahman, M. Performance Analysis of EDM Electrode Fabricated by Localized Electrochemical Deposition for Micro-Machining of Stainless Steel. Int. J. Adv. Manuf.
Technol. 2010, 49 (9–12), 975–986.

www.iran-mavad.com
‫ﻣﺮﺟﻊ ﻣﻬﻨﺪﺳﯽ ﻣﻮاد و ﻣﺘﺎﻟﻮرژي‬