MATERIALS AND PROCESSES FOR NDT TECHNOLOGY ASNT jal Continuing Education Testing The American Society for Nondestructive TestingMaterials and Processes for NDT Technology was edited by: Harry D. Moore Publication and review of this text was under the direction of the Personnel Training and Certification Committee Of the American Society for Nondestractive Testing: ‘George Wheeler, Chair (1976-80) Allen Whiting (1976-1979) Frank Satter, Vice Chair (1976-79) Robert Brostrom (1978-81) Robert Anderson, Secretary (1976-79) ‘Ward Rummel (1978-81) FC. Berry (1976-78) John Weiler (1978-81) ‘Chet Robards (1976-79) Robert Baker (untenured) Carl Shaw (1976-82) EN. Moschini (untenured) Kermit Skeie (1976-79) Ed Briggs (1979-82) Robert Spinetti (1976-80) ack Spanner (1979-82) Conteibutors: Donald R. Kibby De. Robert C. MeMaster Dr. Vernoa L. Stokes ASNT Production Staff Robert Anderson, Technical Director George Pherigo, Director of Education Diana Nelson, Coordinator of Education Services ublishad by ‘The American Society for Nondestructive Testing, In, 1711 Arlingate Lane PO Box 28518 ‘Columbus, OH 43228-0518 Copyright © 1981 by The American Society for Nondestructive Testing, Inc. ASNT is not responsible for the authenticity or accuracy of information herein, Published opinions and statements do not necessarily reflect the ‘opinion of ASNT. Products or services that are advertised or mentioned do not carry the endorsement of recom- ‘mendation of ASNT, IRRSP, NDT Handbook, The NDT Technician and www asnt.org ate trademarks of The American Society for Nondestructive Testing, Inc. ACCP. ASNT, Level II] Study Guide, Materials Evaluation, Nondestructive Testing Handbook, Research in Nondestructive Evaluation and RNDE are registered trademarks of Tae American Society for Nondestructive Testing, Inc ASNT exists to create a safer world by promoting the profession and techaologies of nondestructive testing ISBN-13: 978.0-931403-06-4 ISBN-10: 0-931403-06-5 Printed in the United States of America. frst prinsing 1981 second printing 12/87 ‘ind printing 05/88 fourth printing 10788 si pining 12/91 sixth printing 01/94 Seventh printing 04/96 cighth printing 0898 ‘int pesting 1100 10th psinting 09/05 th printing 10103 1th printing 0209Table of Contents CHAPTER Page 1 RELATION OF NDT TO MANUFACTURING 1 NONDESTRUCTIVE TESTING: NDT Defini- tion. REQUIREMENTS FOR NDT SUPER. VISORY PERSONNEL. QUALIFICATION AND CERTIFICATION OF NDT LEVEL III ‘TESTING PERSONNEL: Sources of Technical Information Available to NDT Personnel. MANUFACTURING: MATERIALS AND PROCESSES: Material Failures. Purpose for Use of NDT. NDT IN FRACTURE CON- ‘TROL. INTRODUCTION TO MANUFACTURING TECHNOLOGY 9 History. INDUSTRIAL RELATIONSHIPS: ‘Competition in Industry. Personnel. Nomencla- ture, SUBJECT MATTER: Materials. Pro- cesses. Economics. Order. PROPERTIES OF MATERIALS 18 INTRODUCTION: Classes of Properties. Sig- nificance of Properties of Design. LOADING SYSTEMS AND MATERIAL FAILURE: Loading Systems. TESTING: The Tensile Test. ‘True Stress-True Strain. Compression Testing. ‘Transverse Rupture Testing. Shear Testing. Fa- tigue Testing. Creep ‘Testing. Notched Bar i ‘Testing. Fac:‘THE NATURE OF MATERIALS AND SOLID STATE CHANGES INMETALS 31 ‘The Effect of Energy on the Atom. Metallic ‘Structure. Solidification. Grain Size, SOLID STATE CHANGES IN METALS: Work Har- dening. Plastic Deformation, Cold Work. RE- CRYSTALLIZATION: Recovery. Rectystalli- zation. Grain Growth, AGE HARDENING. ALLOTROPIC CHANGES. HEAT TREAT- MENT OF STEEL: Approximate Equilibrium Heat-Treatment, Processes. Austenitization. Annealing. Normalizing. Spheroidizing. Har- ening of Steel. Tempering. CORROSION: Di rect Chemical Action, Electrolytic (Electro- chemical) Reaction. Corrosion Rate Dependent ‘on Several Factors. Types of Corrosion, Corto sion Protection. FERROUS METALS 46 Choosing Metals and Alloys. Ferrous Raw Ma- terials, CAST IRONS, STEEL: Wrought Iron, ‘Steel Making. Plain Carbon Steel. Alloy Steels. Low Alloy Structural Steels. Low Alloy AISI Steels, Stainless Steels. Tool and Die Steels. Cast Steels. MATERIAL IDENTIFICATION SYSTEMS, NONFERROUS METALS AND PLASTICS 81 ALUMINUM ALLOYS: General Properties. ‘Wrought Aluminum Alloys, Property Changes. Cast Aluminum Alloys. COPPER ALLOYS: General Properties. Brasses and Bronzes. NICKEL ALLOYS. MAGNESIUM ALLOYS. ZINC ALLOYS. SPECIAL GROUPS OF NONFERROUS ALLOYS: Heat- and Corro- sion-Resistant Alloys. Other Nonferrous Metals, NON-METALS: Plastics. Plastic Ma- terials. Types of Plastics. Characteristics of Plastics. ‘THE NATURE OF MANUFACTURING 13 MODERN MANUFACTURING: Markets. De- sign. Processing. States of Matter. Shape Changing Processes. Summary. THE CASTING PROCESS 79 The Process. SOLIDIFICATION OF METALS; Solidification. Shrinkage. POUR- ING AND FEEDING CASTINGS: Casting Design. Pouring. The Gating System. Risers. Chills. FOUNDRY TECHNOLOGY. SAND MOLDING: Green Sand, Patterns. Flasks. Sand Compaction. Cores. Green Sand Advan tages and Limitations. Dry Sand Molds. Floor and Pit Molds. Shell Molds, METAL MOLD AND SPECIAL PROCESSES: Permanent Mold Casting. Die Casting. Investment Cast- ing. Plaster Mold Casting. Centrifugal Cast- ings. Continuous Casting. MELTING EQUIP- 10 ul 32 MENT: Cupola, Crucible Furnaces. Pot Pur aces, Reverberatory Furnaces, Electric Arc Furnaces. Induction Furnaces. FOUNDRY MECHANIZATION. ‘THE WELDING PROCESS 95 BONDS: Nature of Bonding. Fusion Bonding. Presoure Bonding. Flow Bonding, Cold Bond ing. WELDING METALLURGY: Composi tion Bfects. Effects on Grain Size and Struc ture, Effects of Welding on Properties. DIS: PORTIONS AND STRESSES. WELDING PROCESSES AND DESIGN 105 HEAT FOR WELDING: Chemical Reactions. The Electric Are. Welding Equipment and Pro- cedures. Are Welding Electrodes, Modification of Are Welding for Special Purposes. Automa- tic Welding. Electric Resistance Heating. SPECIAL WELDING PROCESSES: Elec tronBeam Welding. Plasma Arc. Ultrasonic Welding. Friction Welding. Electroslag Welding. Explosion Welding. Diffusion Weld- ing. WELDING DESIGN: Joints. Design Con- siderations. Weldability. WELD DEFECTS: Fusion Welding. Dimensional Defects. Struc- tural Discontinuities, Weld Metal and Base Metal Properties. Basic Symbols for NDT PLASTIC FLOW 1 EFFECTS OF DEFORMATION: Work Har- dening and Reerystallization, Effects of Flow Rate. Direction Effects. Temperature and Load- ing Systems Effects, Grain Size, RELATIVE EFFECTS OF HOT AND COLD WORKING: Mechanical Properties. Finish and Accura Process Requirements. MILLWORK, FORGING, AND POWDER METALLURGY 127 MILLWORK: Hot Rolling. Cold Finishing. ‘Tube and Pipe Making. Extrusion. FORGING AND ALLIED OPERATIONS: NDT of Fors ings. Open Die Forging. Closed Die Forging. Forging with Progressive Application of Pres- sure. Powder Metallurgy. Pressing. Sintering. Sizing and Postsintering Treatments. Applica- tion for Powdered Metal Products. PRESSWORKING OF SHEETMETAL 141 ‘Sheoring, Bending. Drawing. New Develop- ments in Sheet Metal Forming. MACHINING FUNDAMENTALS May ‘The Machining Process. Chip Formation. Cut- ting Tool Materials. Abrasives. MACHINE ‘TOOLS: Machinability. Finish. NUMERICAL CONTROL.16 wv Index MISCELLANEOUS PROCESSES 187 PLASTIC PROCESSING: Compression Mold ing, Closed Die Molding. Casting, Extrusion. Reinforced Plastic Molding. Postforming, De sign Considerations. ADHESIVE BONDING. COMPOSITES: Laminates. Mixtures. METAL REMOVAL PROCESSES: Electrical Dis- charge Machining. Electrochemical Machining. Other Possible Material Removal Methods. DE- POSITION PROCESSES: Electroforming. GROSS SEPARATION PROCESSES: Torch Cutting, Friction Sawing, SURFACE FINISHING im CASE HARDENING OF STEELS: Carburiz- ing. Flame Hardening. CLEANING: Choice of Cleaning Method. Liquid and Vapor Baths. Blasting. ABRASIVE BARREL FINISHING: Wire Brushing. Polishing. Buffing. Electropol- ishing. COATINGS: Preparation for Coatings. Paints, Varnishes, and Enamels, Lacquers. Or ganic Coating Application. Vitreous Enamels. ‘Metallizing. Vacuum Metallizing. Hot Dip Plat- ing. Electroplating. CHEMICAL CONVER- SIONS: Anodizing. Chromate Coatings. Phos- phate Coatings. Chemical Oxide Coatings. INSPECTION 183 INSPECTION PROCEDURES: Organization of Inspection. Quantity of Inspection. Process Control Charts. PRINCIPLES OF MEA: SUREMENT: Dimensional References. Toler- ances. Sources of Measurement Variation. Basis for Measurement. INSPECTION EQUIPMENT: Micrometer Caliper. Other Ad- justable Tools. Indicating Gages and Compara- tors, Fixed Gages. Surface Finish. Surface Fin- ish Measurement. Surface Specification. 199Preface ‘This book has been compiled as a reference and source of general information concerning manufacturing for ‘use by personnel involved in designing, using, or evalu- ating nondestructive testing of products and struc: ‘tures, Tho text material has been kept as general as pos- sible to still retain technical value but broad enough to include all phases of manufacturing industry and most of the materials used. The depth of treatment has inten tionally been kept low in order that NDT personnel ‘without a great amount of formal education might gain an interest and develop understanding of the material However, the techniques of NDT are not included in the coverage. Itis intended that persons using this text be already informed regarding that subject or acquire the necessary knowledge from other more specialized sources. Wherever practical, though, throughout the ‘book mention is made of certain NDT methods that might be suitablo for the kinds of defects under consid eration, Some of the limitations of the methods are also indicated in some applications. One of the essential needs for satisfactory use of NDT ie recognition of its limitations; kmowledge of the source of defects and the materials in which they are found is an aid in determin- ing the validity of any test. and its evaluation Appreciation is hereby expressed: to Grid Incorporated for giving permission to use large portions of Manufacturing: Materials and Pro- cesses by Harry D. Moore and Donald R. Kibbey, 1975, to Vernon L. Stokes for providing many of the illustra- tions used and copy which served as a guide for some of the text material, to Robert T. Anderson, Technical Director, ASNT, and George L. Pherigo, Director of Education, ASNT, for their invaluable assistance in selecting and or- ‘ganizing the material and for supplying the majori- ty of the NDT tierins scattered throughout the text, and to Diana Nelson, Coordinator of Educationel Services, ASNT, for her aid in oditing the text material H. Don Moore 1979Relationof NDTto 1 Manufacturing NONDESTRUCTIVE TESTING Nondestructive testing is a fandamental and essen- tial tool for control of quality of engineering materials, ‘manufacturing processes, reliability of produets in ser- vices, and maintenance of systems whose premature failure could be costly or disastrous. NDT DEFINITION Like most complex procedures, NDT is not definable by a few simple words. Nondestructive testing is nor- mally interpreted to mean the use of physical methods for testing materials and products without harm to those materials and products. Many inspection proce- dures such as dimensional measurements, visual exam- ‘nation for completeness, functional tests, and others, although required in a manufacturing process, are not normally considered part of a NDT program. Nondestructive Tests Are Always Indirect. It is frequently important to know a property or character- istic of a material or product which, if tested directly, would be destructive. Therefore it becomes necessary to perform a nondestructive test on some property or characteristic which can be related to that about which knowledge is desired. The test may be very simple in. ‘some cases, but in others may be complex and difficult. However, in every case, reliable correlation must be es- tablished between the desired property and the mea- sured property (or properties). Correlation May Be Costly And Difficult. Analysis to provide accurate knowlege of the relation between a testable quality and one which cannot be tested direct- ly without destruction is likely to require a great ‘amount of knowledge, skill, and background experi- cence together with good judgement which, in a broad sense, can be described as an instinctive knowledge of the laws of statistical probability.2 Materials and Processes for NDT Technology Decisions to accept or reject following a test result must be based on a thorough knowledge of materials and the properties, processes and their effect on pro- perties, test techniques, design requirements, product applications, service conditions, and suitable life ex- pectancy. Clearly this much knowledge is seldom lo- ‘cated in a single individual, and group decisions or con- sultations may be necessary. NDT correlation may require the cooperation be- tween test supervisors, designers, metallurgists, man- ufacturing personnel, customer personnel, and test personnel REQUIREMENTS FOR NDT SUPERVISORY PERSONNEL From the above it can be seen that supervisory per- ‘sonnel in charge of nondestructive testing operations ‘must have adequate background knowledge for resolu- tion of complex problems in establishing tests and in- terpreting results, Background Knowledge. It is important that a NDT supervisor be well versed in all the available NDT methods, their applications and limitations. In addition, the reliability of the methods and their corre- lation with desired material and product characteris. tice are very important, Knowledge of the product design, purpose, and fune- tion together with process details may enhance appli- cation of test methods by supplying information re- garding the importance of the test interpretations and possible sources of discontinuities, faults, andior de- fects that could cause product failure. Familiarity with all policies, local, industry-wide, governmental, or safety and environmental agencies may affect the inspection methods chosen. Although these policies may be seemingly unimportant at times, they can have very strong implications and influences, Ability To Communicate Ie Always Important. ‘The ability to communicate in both written and oral manner ‘cannot be overemphasized in consideration of success- ful job accomplishment, particularly in supervisory positions, It is necessary for a NDT supervisor to direct, ine struct, and manage the personnel of that group. It is essential that cooperation be maintained with other manufacturing or working personnel with whom the projects are associated. ‘There is need for concise accurate reports to higher, management. These reports, by the way, particularly when written, are the principal evidonce by which repa- tation and advancement are developed. In addition, NDT supervisory personnel are often called upon to interface with customer personnel in solving problems and interpreting test results. QUALIFICATION AND CERTIFICATION OF NDT LEVEL II TESTING PERSONNEL ‘The American Society for Nondestructive Testing, in its Recommended Practice No. SNT-TC-1A entitled “Personnel Qualification and Certification in Nondes. tructive Testing” indicated the responsibilities and capabilities of Level ITI nondestructive test personnel in the following worde:* ‘An NDT Level INT individual should be capable of establishing techniques and procedures; interpreting codes, standards, specifications, and procedures: and designating the particular vest’ methods, techniques and procedures to be used. The NDT Level III should be responsible for the NDT operations for which queli fied and to which assigned, and should be capable of in- terpreting and evaluating results in terms of existing codes, standards, and specifications. The NDT Level IIT should have sufficient practical background in ap- plicable materials, fabrication, and product technology ‘oeestablish techniques and to assist in establishing ac- ceptance criteria where none are otherwise available ‘The NDT Level 111 should have general familiarity with other appropriate NDT methods, and should be qualified to train and examine NDT Level | and Level TI personnel for certification, SOURCES OF TECHNICAL INFORMATION AVAILABLE TO NDT PERSONNEL Regardless of the sources of technical knowledge by which NDT personnel obtain their positions, perpetual updating and extension of that knowledge is essential. Continual Improvement of Entire Knowledge Base Needed. Additional knowledge of NDT techniques, par- ticularly new developments, is very important, Accom panying this, however, should be broadening of know|- edge, as well as keeping up with new developments, in tthe field where the NDT techniques are applied. Greater familarity with engineering materials, manu- facturing processes, changing designs and service re quirements, management. policies, government. re- quirements, environmental impacts, personal safety and other areas can be vital to successful application of NDT. Many Sources of NDT Information Available. Per- sonal ambition and expenditure of time permit informa- tion to be gained in a variety of ways. Activity {not simply attendance) in suitable technical societies can be a good source of new methods and ap- plications of both testing and manufacturing tech- niques. ‘The statement of NDT Lavel III roguzements i taken fromm the ‘June 1980 Ecition of ASNT Recommended Practice No. SNT-TC1A, ‘ap Mustrative of guneral ents, Note that future changer may occur in SNT-TC-LA, and the current dacitment should be ferved to in all cases,Attendance at manufacturers’ training courses and short courses offered by universities often supplies ‘g00d up-to-date information. Studying current technical literature, including ad- vertising, often provides leads for new applications. Personal contact with other NDT peronnel is a vital source of information that may lead to applications of entirely new test techniques or to use of well-known techniques in entirely new ways or to new applications. Personal interchange of knowledge may also prevent costly errors and disasters that others have learned to avoid ‘This personal relationship can sometimes be estab- lished locally in large organizations, but, more often best accomplished by visits to other organizations for a specific purpose, with contacts made during attendance at local, regional, and national technical society meet- ings, and by arranged discussions with representatives of equipment manufacturers, The study and use of standard references such as ASNT Nondestructive Testing Handbook (2 volumes) published in 1959 and reprinted in 1963 and 1977, and ‘Volume 11 of the eighth edition of the Metals Hand- book on Nondestructive Testing and Quality Control, published by the American Society for Metals in 1976, should be valuable to all nondestructive testing per: sonnel, Other organizations such as the American Welding Society have prepared material on nondestructive test ing. A variety of NDT materials is available through the American Society for Testing and Materials. A compre- hensive list of material related to nondestructive tost- ing and related subject matter is available through the National Headquarters of the American Society for Nondestruetive Testing (ASNT), 3200 Riverside Drive, Columbus, Ohio 43221. ‘Many Sources Of Materials and Properties Informa- tion Available. Tt would be virtually impossible to name or list all of the important sources of information doaling with materials, their properties, and their pro- cessing. Any of the sources could be important to NDT personnel for aid in solving specific problems; aside from that, it is important the NDT personnel, particu- larly at Level IIT, have a basic, broad understanding of these subjects. ‘The purpose in use of NDTiis to locate various faults in materials and products that heve been processed to at least some degree, The flaws, defects, or imperfec- tions that may be located by inspection may bea result from the original material, caused by the processing used, created by some human error, or be a result of some combination of these. In most cases it is impor- tant for the NDT inspector to be able to locate the source of the problem when it exists, or even when & fault is not located, know that the possibility of one ex ists. An understanding of the materials, the processes, 3d the possible interactions between them is therefore a “must.” Relation of NDT to Menufacturing 3 In performing their work, NDT personnel ae of neces- sity in close contact with manufacturing personnel and, toobtain the needed cooperation, must be able to speak ‘a suitable language and understand some of the prob- lems, requirements, policies, and operations involved in the manufacturing process. NDTiis also performed in the field during or after con- struction and sometimes after failure to determine the real reason for failure. It is important therefore to also have an understanding of the effects of various environ- mental factors on material properties. To prevent recommendation of lengthy publications written with great depth of treatment or the use of mul- tiple publications, this text has been developed with the hope that a brief, concise treatment of the materials and processing subject area will provide maximum benefit to those that nocd a broad background knowledge. ‘When more detailed knowledge is needed, it may be found in many sources in whatever depth needed. MANUFACTURING: MATERIALS AND PROCESSES The text material in the following chapters presents anintroduction to the many facets of industry which in- volve needs for nondestructive evaluations of materi- als, relating to control of properties, effects of proces- sing, problems in welding and assembly, finishing and protection, and serviceability of engineering materials. ‘The response of metallic materiale to various stages of manufacture, construction, or service life can vary widely as a function of chemical composition, thermal ‘treatment, mechanical working, surface conditions, presence of discontinuities, and other material charac- teristics. Most nonmetallic materials have quite differ- ent properties and response characteristics, and so their manufacture, applications, and test requirements may differ from those of most metalic materials Material Characteristics Often Critiesl. NDT person- nel need familiarity with these many material charac- teristics in order to evaluate their suitability for service, through nondestructive testing. They also should be propared to advise management of possible methods for alleviating undesired response characteristics of mate- rials, especially during processing, manufacturing, and assembly operations. Determining the source or cause of defects is frequently necessary in order to eliminate these defects from production parts. Often, the causes of defects lie in early forms or stages of material produc- tion or processing. In some cases, these prior processes control the response of materials to later processes, dur- ing which defects or failures are induced. MATERIAL FAILURES ‘Some products are purely decorative in use or have such low strength requirements that they are inherent- ly over-designed from the strength point of view. These4 Materials and Processes for NDT Technology may require inspection to be certain that they will main tain their as-manufactured qualities such as color, polish, stability, ete. Products more likely to need careful testing and eval- uations are those used in load carrying applications where failure may involve loss of use, expensive repair, or danger to other products, structures, and even life. Although the manufactured item is a product, it is the ‘material of that product that may fail, so material fail- ure types and causes are of interest. ‘Material Failure Definition. The simplest definition of failure is that the item of interest becomes unusable, but there are several ways in which a product may be- come unusable. Itis usually important to know the type of-failure that. might be expected in order to know for what to inspect, how to inspect, how to eliminate the fault, and how to assess the risk of failure, Complex units with moveable parts may become in- operable because of failure of some minor element. An automobile, for example, may not run for lack of fuel, a tire goes flat, or the ignition is out of adjustment. ‘There are two generally accepted types of material failure: one is the easily recognized fracture ot separa- tion into two or more parts; the second is the less easily recognized permanent deformation or change of shape andior position. ‘Although complete fracture is unmistakable, an inci- pient type whieh will be discussed in connection with “fatigue failure” with suitable inspection methods can be determined before complete failure occurs. Fracture failure in some complex structures may also become progressive. An example of progressive failure would be the release of load by some weak component in a struc- tural configuration such as a large bridge. The released load must be absorbed by neighboring structural ele- rents, Unless these neighbors can spread the new load and become stabilized, they will become overloaded and, if stressed above their elastic limit, will deform, crack, or fracture, causing additional load to be passed along in a way that causes the entire structure to col- lapse almost immediately. A simple exomple of progressive failure would be the breaking of a gear tooth in a mechanical power system. ‘The following gear tooth is then subjected to shock (im- pact) loads which increase the stress levels so that fail ure is more likely. In this case also it is probable that if power is maintained, all the gear teeth will be broken off in a short period of time. ‘Material Failure Causes. Products and structures may be subject: to a number of service conditions. Im- posed loads may be static (stationary or fixed) or dyna- mic varying). The use environment may contribute cor- rosion, vibration, or temperatures and pressures highet or lower than normal. The product may also be subject to abuse. Mechanical failure is always a result of stresses, above some critical value for each material ‘that cause deformation or fracture, Such excessive stresses are set up by some combination of material de- fect, excess load, improper type load, or design error. 1. As far as failure is concerned, static loads some- times include dynamic loads that are slowly applied ‘The principal reasons for failure under static loads in- clude large discontinuities (both internal and external), poor dimensional control during manufacturing, mas- sive overloading during use, and unsatisfactory ori- ginal design or combination of these factors, 2. Dynamic loads are varying loads that can be single-directional or multi-direetional with multi-direc- tional loading being more serious as a cause for failure. When the cycles of loading become high (usually mil- lions but dependent on the material), failure can occur at stress levels far below those determined by static load tests, Although millions of cycles seems high, there are many applications, such asa rotating shaft under bend- ing load, where millions of loading cycles can be reached in a relatively short time. As pointed out, slow or low frequency dynamic toad- ing is similar to static loading except that even low fre- quency loads applied suddenly create a condition of shock which can cause failure at a level lower than nér- mally expected, 3. Service at high temperature reduces most of the desirable material properties of metals including the ability to support load. The tendency for creep also in- creases with increased temperature. The temperatures, at which property values become critical depend upon, the particular material end the previous treatment it has received. Most metallic materials also exhibit a brittle charac- teristic (much like cold glass) through @ transition tem- perature range usually at lower temperatures. Pressure creating stress above a material's elastic limit may cause material flow (plastic flow). distortion. and cross-sectional weakening, effects that would be in- tensified at elevated temperatures. Fluctuating pres- sures of high frequency create dynamic loading that may decrease safe operating levels. It was reported that several early day pressurized cabin aircraft failed by fuselage skin failure due to this cause. 5. Corrosive environments or a combination of mate- rials that cause corrosion can produce failure in two ways. The corrosion may actually reduce the amount of material available to carry load, but even more impor: tant in many cases is that the corrosion may create small discontinuities which serve as stress risers that become the nucleus for fatigue failure, 6. Many structures and systems are subjected to vi bration during service. Included are transportation equipment, machines, and devices that have moving parts. In addition, some structures may vibrate be- cause they are excited by some outside influence. ‘Stresses from vibration may be superimposed on stresses from other loading sources. The principal prob- Jem created by vibration is the introduction of cyclic loading leading toward fatigue feilure. 17, Excess loading from abuse may be accidental but nearly always has the human element as a source. Con- trol is atternpted by use of design factors of safety,usually based on yield strength and ranging from slightly more than one to five or more, Factors of safety are applied during design dividing the nominal allow- able strength of the material by the safety factor. 8, Useof equipment in improper environment may be considered a type of abuse but is sometimes unavoid- able. The main problems experienced are high tempera ture which may decrease material strength and corro- sive conditions which may initiate failure, or, as a mini- ‘mum, decrease the aesthetic properties of the product. 9. Another form of abuse is improper maintenance, including lack of suitable lubrication of moving parts, and improper cleaning and finishing which may permit corrosion to begin, 10. Some materials deteriorate with age and that de- tetioration accelerates with relatively small increases of temperature. Many plastics, most glass, and some metals can develop a brittle characteristic with natural aging and become particularly susceptible to failure under shock loading. Suitable Tests Essential. If failures are to be pre- vented by use of nondestructive tests, these testa must, be selected, applied, and interpreted with care and on the basis of valid knowledge of the failure mechanisms ‘and their causes. The purpose of the nondestructive test design and application should be effective control of materials and products, leading to satisfactory service without premature failures or objectionable damage. Nondestructive Tests Are Performed on Materials. tis rather obvious that knowledge of materials and their properties should be important to any niondestruc- tive testing person. Most test procedures are designed to allow detection of some kind of interior or exterior fault, or measure some characteristic, of a single mate- rial or group of materials, The souree of the problom may be a discontinuity, or it could be a material that is chemically incorrect, or that has been treated in such a way that its properties are not suitable. Discontinuities. ‘The term “discontinuity” is used to escribe any local variation in material continuity in- cluding change in geometry, holes, cavities, cracks, structure, composition, or properties. Some discontinu- ities such as drilled holes, or irregular surface shapes, may be intentionally designed and should have been given full consideration by the designer. These nor- mally do not require testing unless the material is being used under critical conditions or trouble has been experi- enced in service. Other discontinuities may be inherent in the material because of its chemical make-up and structure. Struc- ture refers to the three-dimensional atomic arrange ‘ment in which solid metals and other engineering mate- rials exist. This type of discontinuity ean vary widely depending on the particular material, the treatment it has received fintentional or incidental), and its environ- ‘mental exposure. Because this type discontinuity can vary so much in size, distribution, and intensity, testing to determine its effect may be in order. An exception ‘would be when the discontinuities fall well within the Relation of NDT to Manafacturing 5 limits expected by the designer and when there is little probability of their being affected by outside influences. Discontinuities therefore are not always bad or haz- ardous and may even sometimes be needed in thedesign. or may be helpful in some kinds of processing. Defecis. When any discontinuity, single or multi- ple, is of such size, shape, type, and location that it ere- atesa substantial chance of material failurein service, it is commonly called a “defect.” Finding defects is one of the most frequent objectives of NDT. It must be under- stood, however, that a fault that is a defect under one set of conditions may be only a simple discontinuity that is not harmful in a different application. For example, cast iron is a material that is “loaded” with discontinuities consisting of free graphite flakes, voids (both microscopic and macroscopic), and some- times cracks or tears where the atomic structureiscom- pletely separated. Because of this internal structure, the material is never intentionally used under more than relatively small tensile loading (then usually the result ofa bending load), but is found to be very satisfac- tory in many applications whore the loads are princi- pally compressive. Cast iron, because of the kind of use it receives, is seldom the objective of NDT, although it, possibly contains more internal flaws than any other commonly used material. Discontinuities May Grow Into Defects, In light of the above statements, it should be pointed out that under some conditions, discontinuities believed to be harmless can change into serious defects that can cause disastrous failure. This is most likely to occur under ser- vice conditions and could be because of the effects of fa- tigue or corrosion, especially when accompanied by ‘eyclic loading. A small discontinuity started by corro- sion, a slight scratch, ora discontinuity that is inherent in the material, may develop into a crack from thestress concentration that, under varying loads, propogates ‘with time until thereis no longer sufficient solid materi- al to carry the load. Sudden total failure by fracture then occurs. ‘An example of this type failure is the collapse of the Silver Bridge across the Ohio River at Point Pleesant, ‘West Virginia in 1967. Many aircraft parts require care- ful nondestructive testing and evaluation because they are designed with high stress levels and low safety fac- tors to keep down weight; very small discontinuities ‘may develop into failure defects. Processing Affects Materials Properties. To this point, it would appear that all theemphasis would be on materials, their structuro, and their propertice. Regard. ing NDT thisis true, except that it must beremembered that the processing of those materials from the state through to the completed product has a large in- fluence on the characteristics of the final material. ‘Some processes such as heat treating are for the ex- pressed purpose of affecting material properties. Other Processes such as casting, welding, forming, and ‘machining makes use of heat and/or deformation forces to perform their function and the reaction of the mate-16 Materials and Processes for NDT Technology rial is the same as though the procedure were carried on to change the material properties. "To some degree, knowledge of the processing is there- fore necessary in order to understand the effect on the material, to evaluate the material properties, and to trace down the source of problems. PURPOSE FOR USE OF NDT ‘Acritical task for persons responsible for nondestrue- tive testing operations is often the determination of the true reasons why nondestructive tests have been re- quested, specified, or needed. The interpretation of Lest indications depends critically upon the purpose of the ‘tests, and this often determines the stage of manufac- turing or assembly at which tests should be done. Ultimate Purpose—Reliability, Serviceability. In the preceding paragraphs, the common assumption has ‘been made that the tests are to be used to assure relis- bility and to prevent premature failures of materials, parts, or assemblies during their intended service. In general, this serviceability is the ultimate purpose of most nondestructive tests. However, itis often the case that the test itself does little to predict the serviceabil ity or safety of the final product, assembly, structure, or system. In this case, there is no way todetermine that a discontinuity or material condition constitutes a dan- gerous “defect” which may lead to premature failure in service. This by no means invalidates or countermands the need for the nondestructive test. Other valid rea- sons for nondestructive testing may include the follaw- ing. ‘Specific Purposes for NDT. 1. Identification or sorting of materiel. 2. Identification of material properties and the relia- bility associated with their existence, 3. Indication of proper material and suitable quality control during processing in order to prevent further costly processing. 4. Tests to assure completeness, proper dimensions and geometry, and proper relationships among as- sembled components. 5, Toots during service to discover initiation of pos- sible failure before it actually occurs, 6. Diagnostic tests after failure to determine the fail ure reason. This knowledge might be useful for produet design change, test method change, quality control re- cords, and for records to combat possible product abil. ity suits, NDT IN FRACTURE CONTROL. Fracture Control Design Philosophy. When using factors of safety in design, itis assumed that competent judgement in providing a safety factor can provide high ‘assurance of the safe life of a product or structure. At the same time, consideration must be given to conserva tion of material. Historically, many complex structures and machines have survived admirably under such de sign concepts, However, others have failed from seem. ingly inexplicable causes, some with costly and disus tous consequences. “Most design procedures still are based upon over two and a half centuries of using Hooke’s aw which relates stress and strain in elastic bodies. The safety factor is applied toa value of strongth that the material used ina design is presumed to possess. The value of strength is that which is hoped to be representative of the material ‘yoed under the assumption that the material is continu- us and bas uniform properties throughout. In this de- ‘sign process, discontinuities can be accounted for, ifitis Anown they exist, by their effect on reduction of cross- sectional area available to sustain the applied loads, ar their effect on the local volumetric strength of the mave- rial. Conventional design practices incorporating safety factors ordinarily will succeed if: 1, the material used will not be accidentally or other~ wise overloaded, 2. the variations in ordinary loads will be‘as in- tended, 3, the environmental factors are properly antici- pated, 4, unkaiown or undetected discontinuities will not grow to a critical size, Notable catastrophic failures have been analyzed and invariably found to involve an oversight in oneor acom- bination of the above factors that caused failure. Sud- den, complete, and unsuspected fracture failures occur in components and structures when a crack or other de- fect reaches some critical size and rapidly propogates. ‘Where the consequences of unexpected failure are un- usually expensive in terms of public safety) and/or money, more and more designs of such critical nature re- quire consideration of fracture-esistant qualities in roaterials selection and usage. Implicit in frature con- trol design criteria is the need, by some means, toassure that unexpected flaws of some specific critical size are not present when the product is introduced into service. Nondestructive testing and proof testing are the princi ‘pal means upon which such assurance is based. Proof testing always involves some risk that the test itself will either cause immediate failure during the test, or cause flaw growth to a point below actual failure but be- ‘yond the point where additional flaw growth can be ‘safely sustained during the service life of the compon- ent in question, Nondestructive testing is heavily relied upon as the basis of assurance against the presence of flaws large enough to either cause immediate fracture or to grow large enough to cause later premature fracture. Often, tho successful implementation of a fracture control de- sign depends solely upon the reliability of nondestruc- tive vesting to detect flaws in otherwise sound components 80 that repairs or other corrective measures can be taken. Fracture Mechanies. ‘The study and description of « ‘material property related to fracture resistance is frac- ture mechanics. Fracture mechanics analytical studies and fracture mechanics testing attempt to quantify the fracture toughness of a material. The property of frac- ‘ure toughness is expressed quantitatively in terms of the stress intensity factor, K. K is a measure of the in- tensity of the stress field surrounding the tip of an ideal sharp crack in a linoar elastie material. The stress inten: increases rapidly in the small, localized volume sur- rounding the crack tip when the crack faces are pulled apart by loads imposed normal to the crack plane. The local stress at the crack tip is greater than the stress in vonflawed areas. If a critical value of this stress is ex- ceeded due to increased load, the crack can advance by stress-rupturing the material just ahead of the crack tip. In brittle materials, this sudden, small stress- rupture may release enough energy into the new crack tip region to permit the process to continue in a rapid, uncontrolled manner. At this point, structural stability is compromised and the part fails completely and sud- Genly, often catastrophically. ‘The most common condition under which K, the stress intensity factor, is considered is that of plane strain, Plane strain is the condition where strain sur- rounding the crack tip is zero in the through-thickness direction. This condition results when the material is relatively thick compared with crack size and the con- tained crack is under severe tensile constraint. Under such conditions, the property of interest is the critical pplane stress intensity factor, K,.. Knowing the value of jefor aparticular material under particular conditions enables the designer to calculate the critical flaw size. Abthe design stress, a flaw larger than this size can trig- ger brittle fracture. rior to the time that serious consideration was given to fracture mechanics, some materials selected for their high strength capability failed under relatively low loads. Flaws initially small in size were found as the ori- «gins of failure, having propagated to critical size. In some instances, the initial flaws wre smaller than could have been reliably detected by nondestructive testing, Inspection Reliability. Fracture mechanics pre- sumes the presence of flaws in finished structural ele- ments. Furthermore, through analysis and testing, fracture mechanics predicts the size of flaw which can cause brittle failure either as an initial critically-sized flaw or as a smaller flaw that ean grow to critical size under cyclic loads. In order torealize an advantage from ‘sueh an analysis, some form of inspection or testing rust be applied to the produet. Proof testing and non- destructive testing are the most adaptable means to de- tect flaws. There are advantages and disadvantages of each form of testing ‘Tho risks in proof testing have boon previously men- tioned. in addition. proof testing often is quite expen- Relation of NDT to Manufecturing 7 sive, Nondestructive testing ean aleo be expensive and, ‘without special care, is not routinely called upon to pro- vide assurance that flaws exceeding a certain size are not present in a test object. In most cases, nondestruc- tive testing is applied as either « qualitative or semi- qualitative tool in inspection and process control. In fact, most nondestructive tests provide only indirect in- dications of actual discontinuities. Nondestructive testing personnel are truly challenged to answer the questions posed by engineers using fracture mechanics ‘concepts: 1, Willthe inspection procedure to be used guarantee that ail flaws greater than some critical size will be de- tected? 2. What is the largest flaw that can escape detection using a particular inspection procedure? With the present state-of-art of NDT and for some in- definite future time, these questions cannot be an- swered with the precision desired by design engineers wishing to use fracture mechanics concepts. However, ‘an approach presently accepted uses statistical meth: ‘odology to define inspection reliability. Probability of DetectioniConfidence Level. Given an inexhaustible continuum of flaw sizes and shapes for all the materials of interest along with substantial fi- nancial resources, each inspection procedure could be tested with flawed specimens. Straightforward statis- tical methods could be used to demonstrate the probe- bility that a particular procedure would (or would not) detect flaws of given sizes. In simple terms. if a large ‘number of specimens each contained a flaw of the same size and all were subjected to a given inspection proce- dure, the ratio of flaws detected to total number of flaws looked for would constitute the probability of detection. For example, if 100 flaws of the same size were present and 90 were detected, for the circumstances of this par- ticular experiment it could be stated that the probabil ity of detection of flaw size X ina given material is 0.9 or 90%. Based on this example, repeated experiments on addi tional flawed specimens could be conducted. Instead of 100 flawed specimens being examined, 1000 or even 10,000 could be tested. £9,000 out of 10,000 flaws were detected, the confidence would be increased that the true probability of detection is 90%. In practical situations, 100 flawed speciments would bea luxurious sample, indeed. Itis possible, however, to estimate from a limited sample size the probability of detection for a larger population. In the original exam ple, where only 100 units were tested, it would be useful to know how precisely true is the inference that the de- tection probability is 90%. In statistical analyses, the term “evel of confidence” refers to the probability that the 90% detection probability inference is truly valid. ‘The interpretation of statement that.a particular ex- periment produced « 90% probability of detection with ‘95% confidences that thereis a 5% probability that the ‘90% probability of detection is overstated.{8 Macerials end Processes for NDT Technology NDT Demonstration Programs. Pionesring offorts to include meaningful fracture control criteria into structural design first appeared in the specifications for military aircraft in the late 1960s. Since then, fracture ‘control criteria have been applied to aircraft, space- craft, nuclear components, pipelines, and pressure ves sels, ‘The most ambitious programs to date have in- cluded several military aircraft and NASA's Space Shuttle. Typically, the overall vehicle specifications hhave required the contractors involved to demonstrate inspection capabilities to detect certain size Naws in components designated fracture critical at 90% proba: bility of detection, 95% confidence level. Several approaches have been used by the contrac: tors to provide the required demonstration, Generally, some economically practical numbers of fatigue cracked specimens are preparad and intermingled with unflawed specimens. Well-iefined procedures are used. by inspection personnel under production conditions. ‘The parts used in the demonstration are usually entered into the normal sequence of inspection so that the in: spection personnel are not biased toward particular awareness that a demonstration is being conducted. in ‘other cases, only certain inspection personnel have been classified as being qualified to inspect fracture critical components, Such components are clearly marked ant channeled specifically to their selected inspectors. In ‘any event, through experiment design and statistical analysis, the outcome of NDT demonstration programs is intended to provide a workable interface between the needs of fracture mechanics and the realities of the ap: plications and limitations of NDT.Introduction to 2 Manufacturing Technology Webster defines “manufacture” as “to make by hand, by machinery, or by other agency; to produce by labor, especially now with division of labor and usually by machinery.” Such a definition is all-inclusive, It covers the making of foods, drugs, textiles, chemicals, and, in. fact, everything made usable or more usable by the conversion of shape, form, or properties of natural materials. ‘Special interests have developed in the mechanical and industrial phases of industry concerned with the making of durable goods of metals and plastics. The majority of metals and some other materials fall in a class that is often referred to as engineering materials Characteristic of this group are the properties of rela- tively high hardness, strength, toughness, and dura- bility. Glass, ceramics, wood, concrete, and textiles, although they may compete with metals in many applications, have usually been excluded from these structural materials because of a difference in the combination of properties, a difference in processing requirements, and a difference in type of goods pro- duced. The list of so-called engineering materials continues to grow with the addition of new metall ‘combinations, plastics, and even materials that have been previously excluded from the list, as they are developed with better properties or used in new appli cations. Present interpretation of the term engineering ‘materials includes most metals and those plastics that, are solids and have reasonable strength at room temp- erature. This book will be concemed with these10 Materials and Processes for NDT Technology materials and the processes that are used to shape them or change thelr properties to a more usable form HISTORY ‘The growth of industry in the United States is typical of industrial development throughout the world. Early settlers were concerned primarily with food and shelter. Most manufactured goods were imported but some ‘manufacturing was done in the family units, Eventual- ly, as conditions were stabilized, efficiency improved and excess goods were available for sale and trade, The factory form of industry finally resulted, under control of single families. Some of those still exist but most have changed to corporate enterprises under ownership of many individuals Early Manufacturing. The first manufacturing was devoted mainly to agricultural and military needs. One of the earliest industrial operations to grow to large size was the reduction of ore to metal. By its very nature, particularly for ferrous metals, this process is not adaptable to very small operations. The trend in this in- dustry to increasing size has continued to the present. ‘A few very large corporations produce nearly all of the basic metals, even though there are many small fabrica- tors, Interchangesbility. ‘The Civil War and the expand- ing frontier created much incentive for the manufacture of firearms. Many will remember that the first example of true interchangeability and the development of bet- ter transportation following the Civil Wer resulted in rapid growth of production goods. Many of the pro- ducts were considered luxuries at the time but since have become necessities to the modern life style. ‘Importance of Manufacturing. Manufactured pro- ducts are an integral part of everyone's life, but most persons do not realize the great, amount of investment and labor that makes those products possible. Realiza- tion comes with thought that almost every activity, re- gardiless of field, is in some way dependent on hardware produced by the manufacturing industry. Approxi- mately 25% of the gross national income is spent for manufactured goods and about the same proportion of ‘the United States’ working force is employed in the manufacturing industry. INDUSTRIAL RELATIONSHIPS COMPETITION IN INDUSTRY In the American way of life, the profit motive is, the root of most business, including manufacturing, ‘The system presumes direct competition, so that if a number of companies are engaged in the manufacture of similar products, the sales volume will be in pro- portion to the product quality, promotional acti ties, service policies, and price. The cost of manu- facturing therefore hecomes of prime importance, for the company that can produce at the lowest cost anc maintain quality can spend mote for sales activities can‘sell at a lower cost, or can make a larger profit per sale than competitors in a less fortunate positior For this reason industry is continually engaged in 2 battle to lower production costs and to gain this favored position, Direct. Competition Limited. Because of the complexity of the overall manufacturing operation, many decisions are, of necessity, rather arbitrary. For nearly all products, there are many altematives of design, materials, and processing that will satisfy the function the product is to have, For many products. direct sales-price comparisons are not adequate, fo: different demands for similar products rade of dil- ferent materials ox having different designs may exist ‘The purchaser is truly the final decision-maker, which makes advertising and sales promotion @ most impor- ‘tant phase of the business. Adequate time is often not available to study the effect of a design on the market or to investigate all the possible processes of manufacture, particular! for new products, Sometimes, to determine the exact ‘material that would serve best even for a fixed design is too time consuming. In any case, reasonable dex sions must be made, and when absolute knowledge is hot available, they are based on past experiences of similar nature. Because of tie interrelationships exist ing in manufacturing, accurate decisions will depenc not only on exact knowledge of a specific area ut algo on knowledge of interaction from related areas, PERSONNEL Several kinds of workers are needed in any manu: facturing operation. Some work directly with the product, and some are only indirectly connected with the product but are more concemed with the organi zation producing the goods, Those directly connected with the product include the designer, those respon sible for choosing the processes, establishing. control ‘over the operation, and supervising the manufactur ing, and the machine and equipment operators who perform the actual work of converting raw material into useful objects. Each of these, to function ef: fectively in his job, must have varying degrees of knowledge concerning the product requirements, the material properties, and the equipment limitations. Most jobs directly connected with the product call for specifi knowledge in depth concerning certain phases of the work and more general knowledge of volated areas. Products, from the simplest single part items to the ‘most complex assemblies costing millions of dollars each, go through a series of chosen steps of manufacture as they proceed from raw material to completed useful products. In order to conserve energy, material, time, effort, and to reduce cost, itis necessary at each stageofproduct development that qualified personnel examine the processed material to insure that the final product has the quality and reliability expected from the design. A large part of the manufacturing effort therefore is in addition to modifying material and adding to the pro- duct development, Essentially all products requireade- gree of inspection of the material to see that it conforms tothe requirements that providea high quality product. Although not nosmally classed as direct labor, sales personnel usually must have complete familiarity with the product and its manufacture. They are called upon to recommend, compare, troubleshoot, and even install a product. Indirect. Other personnel are only indirectly « nected with the product or the manufacturing opers- tion. These inchide most workers in administration, accounting, finance, purchasing, custodial service, and other support areas. The personnel who work in these areas may be highly skilled or trained in their own field, They do not need extensive technical knowledge of the product or its manufacture. How- ever, they may still make decisions that are far-reach- ing in effect on the products, Therefore, they do need broad understanding of the product and the manu- facturing facility. NOMENCLATURE, ‘The ability of personnel from one area of manu- facturing to discuss and understand problems with people from another area will depend directly on their knowledge of the nomenclature used in the area of concer, A designer, to discuss intelligently with a production man the effects of various design changes ‘on the method and cost of production, must be able to understand and use the language of the production ‘man, a most cases, he needs to know at least the ames of the various machines and tools that might be used and have some understanding of their capa- vilities. In the final analysis, the problems of the production of a product become the problems of the machine and equipment operators. The loyalty, cooperation, and respect for supervision of these operators, necessary for the proper solution of production problems, can be gained only when a full understanding exists between the two groups. Of necessity, this understanding must be based on suit; able language, including proper terminology, even to the point of using local terms and nicknames when appropriate, Similarly, NDT personnel must communi: ate with production and other personnel, SUBJECT MATTER, Even with the limitations that have been placed on the term manufacturing processes for use in this text, many possible variations of content and organization Introduction to Manufacturing Technology 1 of subject matter exist. The principal objective of this text will be to present a broad discussion of the materials used in manufacturing and the principal processes by which these materials are made into usable products, The subject of materials and manu: facturing processes is truly a single subject when the orientation of discussion is toward the end product that must be manufactured to fulfill some function. Although the attempt has been made in this book to show this singleness of subject matter, it is still nec- ‘essary to treat specific areas as specific topics. Simi- larly, manufacturing plants are normally divided into areas in which the equipment and personnel concen: trate on particular manufacturing operations. For example, a foundry may produce only iron castings ‘of a certain weight range because of specialized experiance and equipment. MATERIALS An understanding of materials is important to any manufacturing procedure. One or more materials are required for any product, and most can be processed in @ number of different ways. However, for many materials, the processing possibilities are very limited, and the process may be dictated by the particular material chosen, Properties. The practical differences between various materials is in their properties or combina tions of properties. Compared to many other ma terials, steel is hard and strong and may be chosen as a manufacturing material for these reasons. Steel is elastic to some extent. However, if elasticity is the important property of interest, it may be necessary to choose material like rubber for the application. An intelligent comparison of materials depends on pre- cise meanings of the terms used and an understanding of how properties are defined and measured. Some properties are defined by tests, such that the results may be used directly as design data, For example, from a standard tensile test, the modulus of elasticity of a material may be determined, and a designer can ‘use this value to predict accurately the deflection of a certainsize beam under known loads, On the other hand, many properties are defined no less specifically bout ina more arbitrary manner, which makes the use of the test results for calculation difficult or impos ible, However, the tests still provide the opportunity for accurate comparisons with data obtained from similar vests from other materials. For example, har ness measurements may give an indication of relative wear resistance for different materials, or hardness numbers may correlate with tensile strength for a given material, but the number values can seldom be Used dlrectly in computation for dosign loads. Property Variations. Bach elemental material has at least some properties different from those of all other elemental materials. Some or all of the prov-12 Materials and Processes for NDT Technology erties of an element may be changed by the addition fof even small parts of another element. In many cases the properties obtained from the combination will be better than those of either element alone. In a similar manner, the properties of elements or combinations can be varied by the type of treatment given the material, The treatments that affect propertios are often intentionally selected for this purpose. How- ever, the properties are no less affected, often in an undesirable way, by the processes being used with the objective of shaping the material. Sufficient knowl: edge of the relationship between the properties and the processing of materials may permit the improve- ment of the properties as a natural result of the processing for a different main objective. Reducing the crosssectional size during the shaping of most ‘metals results in an increase in hardness and strength that may be undesirable if the metal must. undergo further deformation processing. In many cases, this increase in hardness and strength that occurs as a result of the processing can be beneficial and part of the product design, PROCESSES Manufacturing consists of converting some raw material, which may be in rough, unrefined shape, into a usable product. The selection of the material and the processes to be used seldom can be separated Although in a few cases some unusual property requirements dictate a specific material, generally a wide choice exists in the combination of material and processing that will satisfy the product requirements, ‘The choice usually becomes one of economic com- parison. [n any case, a material is usually selected first, sometimes rather arbitrarily, and a process must. then be chosen. Processing consists of one or many separate steps producing changes in shape or proper- ties, or both. Shape Changes. Shape changing of most materials can be accomplished with the material in one of sev- eral different forms or states: liquid, solid, or plastic. Melting of a material and control of its shape while it solidifies is referred to as casting. Reshaping of the material in the plastic or semisolid form is called molding, forging, pressworking, rolling, or extrusion, Shaping by metal removal or separation in the solid state is commonly performed to produce product shapes. If the removed material is in chip form, the process is machining. ‘The joining of solid parts by welding usually involves small localized liquid areas that are allowed to solidify to produce « complete union between solid parts. Energy Form, ‘The material condition and the energy form used to effect these shape changes may vary. As noted, the material may be in a liquid, solid, or plastic form. The energy may be supplied in the form of heat, mechanical power, chemical reaction, electrical energy, oF, as in one of the newest proce dures, light, In nearly every instance, one principal objective is shape changes, but usually part of the energy is consumed in property changes, particularly in those processes involving state changes or solid deformation. Different materials react differently io the same energy system, and the same materials reac differently to different energy systems. Process Effect on Properties. Many concepts ant fundamentals in reference to materials are comma: to different kinds of processes. When studied in con- nection with the material, these concepts, then, car be applied regardless of the kind of process by whic! the materjal is treated, The metallurgical changes that take place during solidification during casting are of the same nature as those that take place in fusior welding. Auxiliary Steps. ‘The completion of a product for final use generally includes the various finishing pro cedures apart from basic shape-changing processes, ‘The dimensions and properties that are produced by any process are subject to variation, and, in practi- cally all cases, some inspection of nondestructive type is necessary for controlling the process and for assuring that the final product meets certain specifications as to size and other properties. As one of the final steps, or sometimes as an intermediate step, control of prop- erties by heat treatment or other means may be necessary. The final steps may also require surface changes for appearance, wear properties, corrosion protection, or other uses. These steps may involve only the base material or may require the addition of paints, platings, or other coatings. Few finished products are constructed of single pieces of material because of the impracticality of producing them at a reasonable cost. Also, it is fre- quently necessary that properties that can be ob- tained only from different materials be combined into a single unit. The result is that most manu- factured articles consist of assemblies of a number of separate parts. The joining of these parts can be accomplished in many ways, with the best method being dependent on all the factors of shape, size, and material properties involved in the particular design ECONOMICS ‘The private ownership systems of business and industry in the United States are profit motivated. In a competitive market, the manufacturer who makes the most profit will be the one who has the best combination of design, materials choice, and manu- facturing processes. Ultimately, most decisions become 2 compromise between the most desirable from a design, life, and function standpoint and the most practical from a production and cost. stand. point.Design. The designer must not only know the functional requirersents of the product but also have some knowledge of the probable market demands for various levels of quality and appearance. He certainly must be familiar with the mechanical properties of the various materials he might choose. Less obvious at times is tne importance of the part the designer plays in the selection of manufacturing processes. If the designer designates a shect-metal housing for a radio, obviously, the housing cannot be { plastic molded part or a die casting. If he specifies certain tolerances, these not only may dictate that a ertain dimension be achieved by machining but also may even dictate the specific type of machine to be used. Clearly then, in every case, the designer's choices of materials, shapes, finishes, tolerances, and other factors restrict the possible choices to be made in the manufacturing process. The designer may also ‘specify the NDT criteria, thus influencing the choice of NDT. Choice of Materials. Engineering materials, metals and others, have properties that vary over wide ranges with many overlaps. Costs also vary widely, but the cheapest material suitable for the product does not necessarily insure the product will have the lowest cost, For example, 2 lower cost steel substituted for another may satisfy the functional requirements of ‘the product but may lead to increased inspection costs, thus decreasing or eliminating the margin of necessary Quantity. The number of a product that is made can have more influence on the cost than the design or the type of material used. Most manufacturing processes involve both a get-ready, or setup cost, and 4 production cost. The setup cost can range from nothing to many thousands of dollars, depending on tne type af process and the amount of special tooling needed. The actual production time for each product is usually inversely related to the setup cost. Quality. Quality costs_money. Higher quality Implios longer life, better finishes, better materials, quieter operation, and more precision. These factors ali involve greater costs that may be justified by market demand. if not justified, competition wall sat- isfy the demand with lower quality at lower cost, inspection. Inspection also costs money to perform, but, in another sense, like advertising, it pays: infact, it i essential to assure better quality product output and to improve customer relations. Tntroduction to Manufacturing Technology 13, Modern technology has produced much inspection equipment nooded for nondestructive testing, How: ever, proper application of inspection methods and in- terprotation of their test indications is not possible without relying upon qualified nondestructive test per- sonnel. Capable individuals are needed to provide input, to the decision processes regarding the integrity and. serviceability of the test objects, stemming from the in- direct indications provided by nondestructive tests, ‘Such persons must have an adequate background of knowledge concerning the materials and manufactur: ing technologies involved in their specific industries, and the service conditions to which their products will, bbe subjected, in order to make valid decisions. ORDER ‘The enormous quantity of knowledge available about manufacturing processes can be discussed in varying degrees of depth and coverage. The following chapters of this book have been chosen with the hope that the order will seem logical and conducive to maximum learning. The discussion does not go into ‘great detail in the belief that for the purpose of this book broad knowledge of the overall manufacturing system is more important than the development of depth in any special but restricted area. Materials. As has been indicated in this chapter, the properties of materials are very important and cannot be divorced from the manufacturing proc- esses, The first topic of discussion will therefore be properties, with their definitions, which generally consist of a description of the test procedure used to measure the property, followed by the fundamentals of metallurgy as they apply to the commonly used manufacturing materials and processes. The proper- ties of specific materials will be discussed only as they affect the process choice and as the process affects them: Processes. The major processes of casting, defor- ‘mation shaping, welding, machining, and finishing ‘will be discussed with an emphasis in length and depth commensurate with their use and importance to NDT personnel. The experiences of many individuals fre quently leads to a belief that one area of manufacturing is more important than others, but the interrelation- ships are such that no one area can exist alone, and the importance of any process in an individual case is entire- ly dependent upon its relation tothe product with which it is associated.Properties of 3 Materials INTRODUCTION Because manufectured itemas are made from materi- als with various * roperties, responsible NDT person nel must be generally familiz with engineering materi- als and their eapabilities and limitations. Selection of ‘an engineering mzterial to implement the design of a usable part or a‘sembly requires knowledge of the ‘material’s chemical, physical, and mechanical proper- ties, Moat structs: al materials are loaded by external forces which generate high lovels of internal mechani ‘eal stress within the mat-rials. The reaction of the component to a new stress distribution caused by the development of discontinuities may be critical to its ‘continued functioning. In order to perform meaningful inspections, the “esponsible nondestructive testing personnel must be cognizant of both the normal mate- ial properties and of the effects of discontinuities upon the material serviceability in its intended applica- tions. ‘As indicated in Chapter 2, the qualities of materials that are of practical interest to manufacturing ave measured quantit‘tes called properties, as distin- guished from the physical makeup of materials called atomic structure. Science in recent years has made great strides in determining the atomic structure of materials. Figure 3-1 shows that an atom of iron con: tains twenty-six electrons ind an atom of aluminum contains thirteen electrons, arranged in ¢ efinite order. ‘The number and the arrangement of pa ticies in each atom actually determine a'l the pror-tties of any materi], and it should be theoretical y possible to predict the properties of a maierial fror. the structure of its stoms. Physicists and chemists ¢ n make some predict ons of properties, particularly chemical and electrical, based on structure but tle mechanical properties of greatest interest to a s udy of manu-Properties of 3 Materials INTRODUCTION ‘Because manufactured items are made from materi- als with various properties, responsible NDT person- nel must be generally familiar with engineering materi- als and their capabilities and limitations. Selection of ‘an engineering material to implement the design of a usable part or assembly requires knowledge of the material's chemical, physical, and mechanical proper- ties, Most structural materials are loaded by external forces which generate high levels of internal mechani- cal stress within the materials. ‘The reaction of the ‘component to a new stress distribution caused by the development of discontinuities may be critical to its continued functioning. In order to perform meaningful inspections, the responsible nondestructive testing personnel must be cognizant of both the normal mate- rial properties and of the effects of discontinuities upon the material serviceability in its intended epplica- tions. ‘As indicated in Chapter 2, the qualities of materials that aro of practical interest to manufacturing are measured quantitites called properties, as distin- ‘guished from the physical makeup of materials called atomic structure, Science in recent years has made great strides in determining the atomic structure of materials. Figure 3-1 shows that an atom of iron con- ‘tains twenty-six electrons and an atom of aluminum contains thirteen electrons, arranged in definite order. ‘The number and the arrangement of particles in each atom actually determine all the properties of any material, and it should be theoretically possible to predict the properties of a material from the structure of its atoms. Physicists and chemists ean make some predictions of properties, particularly chemical and electrical, based on structure, but the mechanical properties of greatest interest to a study of manu-16 Materials and Processes for NDT Technology facturing processes must still be defined and ‘measured by empirical test for each material. ss WON ALUM Figure 34 Atomie structure CLASSES OF PROPERTIES ‘The application to which a material is put deter- mines which of its properties are most. important, ‘Chemical Properties. ‘The chemical properties (reaction with other materials) are of interest for all material mainly because of the almost universal need for resistance to corrosion. Although aluminum is chemically more active than iron, in most. atmos- pheres the corrosion byproducts of aluminum form a denser coating, which acts as a shield to further corro- sion, than do the corrosion byproducts of iron. While the atomic and crystalline structure of all metals gives them high electrical and thermal condue- tivity compared to nonmetals, individual metals still differ considerably. Aluminum is among the best electrical conductors, while iron, although much more conductive than nonmetals, is a poor conductor compared to aluminum, On the other hand, the magnetic properties of iron make it much more desir- able for some electrical uses than aluminum. Physical Properties. Physical properties for each material are constants associated with the atomic structure. ‘These properties include density (weight per unit volume), crystalline type, atomic spacing, specific heat, cohesive strength (theoretical), and melting point. Iron has a much higher melting point and density than aluminum, Iron is allotropic, mean- ing it can exist in several different crystalline struc- tures as opposed to aluminum, which always exists in single crystalline pattern, This difference makes pos- sible, for irorrbased alloys, methods of property con- trol by heat treatment that are not possible for alurni- num. Some aluminum-based alloys may be heat treated for property control, but the reaction is entire: ly different, Mechanical Properties. Of most interest to man- ‘ufacturing are the mechanical properties of hardness, strength, and others that are of prime importance in, design considerations for determining sizes and shapes necessary for carrying loads. These qualities will also determine the work loads for any deformation type of manufacturing process. Neither iron nor alumninny in the pure state has many applications in manufac turing because their strengths are low, hut thei alloys, particularly iron alloys, are the most com ly used of all metals. Both of these materiais 22m 9 strengthened over their weakest forms by factors almost ten by suitable alloying and treatment, with alloys of iron being approximately five times as strong as those of aluminum on a volume basis: Processing Properties. As pointed out at the beginning of the chapter, the properties that have been discussed are actually dependent on the atomic structure of a material, but in practice these proper ties must be separately measured. In a similar wa different properties that are related to hardness, strength, ductility, and other physical and mechanical properties and that are frequently of even greater Importance to manufacturing must in practice be defined by separate tests. These include tests for castability, weldability, machinability, and bending that describe the ability of the material to be proc essed in definite ways. Tests of this type m developed at any time there is need for determining the ability of the material to meet eritival needs of processing, and they are usually performed under conditions very similar to those under which the process is performed. SIGNIFICANCE OF PROPERTIES TO DESIGN ‘A designer is necessarily interested in properties because he must know material strengths beiore he ccan calculate sizes and shapes required to cary loads, chemical properties to meet corrosive conditions, and other properties to satisfy other functional require- ments. Knowledge of processing properties is likely to be of more importance to manufscturing personnel than to the designer, although even he must be able to choose material that can be manufactured in a reasonably economical manner. Many manufacturing problems arise from choice of materials hased only 6n functional requirements without considering which is the most suitable for the processing requited. Similar results can occur when inspectability has not bbeen given proper consideration in design Material Choice a Compromise. Most products ‘can be manufactured from a number of different pos sible materials that will satisfy the functional requlie- ments. However, some are more desirable from the product standpoint than others, and one particular material may have the best. posible combination of properties. Likewise, all materials can be manu: factured by some means, although costs of manufae: turing will vary, and there will likely be one single material from which a usable product could be manu factured at lowest cost. Seliiom ean a material be ehosen that has optimum properties for both theproduct and the manufacturing, $0 the majority of material chaiges tiny out to be compromises, The final choice may be a result of trial and error tests among several possible best mnaterials and processes. New choices may be required with changes of design, material availability, processes, or market demand. LOADING SYSTEMS AND MATERIAL FAILURE LOADING SYSTEMS Physical loading of material is a result of applying foree under one or more simple, basic loading sys: ems. In nearly all cases, even when a piece is loaded by only a single set of outside forces, the internal loads developed are more complex than those ap- lied. However, in many testing procedures this complexity is disregarded, and the forces are treated as though they are uniform thoughout the material. Stresses, Internal forces, acting upon imaginary planes cutting the body being loaded, are called stresses. For purposes of ease in understanding and comparison, stresses are usually reduced to unit stress by assuming that the force acts uniformly over the cross-sectional area under consideration. The load: per-unit area can then be ealeulated by dividing the total load or force by the area on which it acts. The common units used for measurement and description in the United States are pounds for force and square inches for area, so unit stress becomes pounds per square inch (psi) or when dealing with large figures, ‘thousands of pounds per square inch (kips/in*), With the changeover to the international metric system of measurement, the units for stress become newtons per square meter (pascals) in which a newton is equal to approximately 0.2248 pounds of force (poundals}. Conversion of kips/in* involves multipli- cation by the factor 6.894767 to obtain megapascals (MPa), Approximate conversion can be accomplished by use of the multiplying factor 7. ‘Normal Stresses. Figure $-2 represents a bar sub- jected to a pulling foree of P. If the load is uniform ly distributed over the ends of the bar, it can be assumed that the internal loads are uniformly di tributed. Examination at any plane x-x perpen- dicular to the line of applied force will show that the erystals along one side of the plane are trying to sep- arate from the adjacent crystals along the plane. This internal force tending to separate the material is known as stress, If tne surface area cut by the imagin- ary plane x=x is A, then the unit stress (s) is P/A, oF ao a Figure 3-2 ‘Simple loading Properties of Matersle 17 vwritven as @ formula, s = P/A, Because in this case the applied force is a pulling force or tensile force, the internal loads are tensile stresses (S), and the formula may be written St = P/A Reversal of the external load P would cause the internal stress to be compressive instead of tensile ‘The unit stress on any plane x — x perpendicular to the line of force would then be calculated from the fosmula Se = PIA. Shear ‘Stresses. Tension and compression forces and their resulting stresses are always considered to act normally, or perpendicular, to a plane. A third term, shear stress, is used to describe the effect of forces that act along, or parallel to, a plane. No provi sion has been made for describing forces meeting plane at an angle. Because, however, an infinite num- ber of planes may be of interest, it becomes necessary to resolve the stresses to various angles to determine critical vatues and positions. Figure 3-3 illustrates a ber, similar to that of Figure 3-2, with tensile load theing applied to the end. As already iltustrated, ten- sile stresses, and tensile stresses only, are set up on any imaginary plane x ~ x perpendicular to the line of force. If, however, a plane not perpendicular to the line of force is examined, it can be seen that a differ- ‘ent situation exists, The imaginary plane 2~z is at any angleé. The area cut by the imaginary plane 2~2 is equal to the area of the plane x-x multiplied by the secant of the angle 9. Therefore, the unit shear stross is 8,.Bsind 2 Asec A = Gan sin @ cos @ Figure 33 Resolved loading ‘Substitution of the values for ¢ in this formula shows that for zero or 90", the shear stress is equal to zero. ‘The maximum shear stress occurs when ¢ is 45° and sine 2 9 is 1, in which case the shear stress, Ss, equals, P/2A. The maximum value is one-half the tensile stress, $1, established on 2 plane that is perpendicular to the applied force, Tf, in the preceding case, the external load were ‘compression instead of tension, shear stress would18 Materials and Processes for NDT Technology hhave been developed to the same magnitude and in the same way but opposite in direction and combined with compressive stress instead of tensile stress, Shear stress exists alone only in a bar subjected to pure torsion, that is, a bar being twisted with no tension, compression, or bending present. Shear stresses are important to our manufacturing processes because these are the forces that cause material to shift in plastic flow and permit shape changing by deforma- tion processes, Bending. Bending loads create a combination of stresses. The concave side of a bent body will be in compression and the convex side in tension with transverse shear occurring along the axis between them. The maximum unit stress will be in the outer fibers of the bent body and is represented by the formula Sp = Me/I where. M equals bending moment, ¢ equals distance from neutral axis, and J equals, moment of inertia of the body. Effects of Stresses. The principal point to be ‘made in this discussion of forces and stresses is that structural designs must be of suitable size and shape ‘and must be made of material with proper strength ‘values to withstand the loads imposed upon them. When a structural member (almost any object) is physically loaded by weight, by pressure from mechanical, hydraulic, or pneumatic sources, by thermal expansion or contraction, or by other means, internal stresses are set up in the member. The size, direction, and kind of stresses are dependent upon the loading system. ‘The magnitude of the unit stresses will be dependent not only upon the applied force but also upon the area of material resisting the stresses. As loads are increased, unit stresses will increase to the point where, in some direction, one oF more reach critical values in relation to the material. Failure by plastic flow or by fracture can then be expected, depending upon which eritieal values are reached first. In nearly all cases of fracture failure, the separation of material is preceded by at least « small amount of plastic flow. In those cases in which plastic flow occurs to a large degree, fracture failure will finally result. TESTING ‘Testing of material is essential to gain practical knowledge of how materials react under various situa- tions. The ultimate goal of any test is to enable the making of decisions that provide the best economic results. In practice, two general methods of testing are used. Direct Testing. The only test that supplies abso- lute information about a workpiece or a material is a test of the particular property of interest conducted con that part itself. In this method of direct testing, an attempt is made to use the materials under the exact, conditions of practical use, and the test may be con- cerned with a product, a process, or both. Direct test ing is usually time-consuming, and, for the results to hhave statistical significance, often requires compila- tion of data from many test samples. The procedure is necessary, however, for those cases in which simp- ler methods are not available and in which sufficient historical information has not been accumulated to permit correlation between the attribute about which information is desired and some other measurable factor. Indirect Testing. Indirect testing involves the use of such a correlation, such that accurate knowledge of the relationship between the two factors must ‘exist. The ability of grinding wheels to resist the centrifugal forces imposed in use is directly tested by rotating them at higher speeds than those of actual use. Such a test indicates that the wheel strength is sufficient for normal use with some safety margin. An indirect test that is sometimes used for the same purpose can be performed by rapping a suspended wheel to cause mechanical vibrations in the sonic range. A clear tone indicates no cracks. A danger of Indirect testing is that the conclusions depend on the assumption that the correlation betwern the meas- ured factor and the critical factor exists under all conditions. The rapping test for grinding wheels does not give any real indication of strength, unless know!- edge of the wheel’s history permits the assumption that with no cracks it has sufficient strength for use. Destructive Testing. A large number of diroct tests are destructive. These also are dangerous because ‘the assumption must be made that those materials not tested are like the ones for which test informa- tion has been obtained, A portion of weld bead may be examined for quality by sectioning it to look for voids, inclusions, penetration, bond, and metallurgical structure by visual examination. By this operation, this portion of the bead has been destroyed; regard- less of the quality that was found, the only know! edge acquired about the remaining portion of the weld comes from an assumption that it is similar to that examined because it was made under the same conditions. Nondestructive Testing. In addition to the nondes- tructive feature, these tests almost entirely are indi- rect tests that require first, correlation with the de- fects that are being sought, and second, expert evalus- tion or interpretation of the evidence that is gathered. Nondestructive tests may be for faults and discontin- tunities located on either the surface or internally and may be performed before, during, and after the ‘manufactaring process, ‘These tests are performed by (1) exposing the prod- ‘vet material to some kind of probing medium (radia- tion energy, sonic energy, magnetic and electricalenergy, and other media), (2) obtaining some kind of in- dicating signale from the probing medium, and then (3) interpreting the signals as evidence of the presence oF absence of possible defects. To function properly, a suitable probing medium must be one that can be ap plied in euch a manner that it will be affected by any defects present, and the signals obtained must be cor- related with the defects, Standardized Tests. Over the years a numoer o1 tests have been standardized for checking of material properties, Some of these provide data that are useful for design calculation, while others have the primary purpose of aiding in materia) choices by supplying ‘comparative information. Many properties are de- fined only by the test procedure that has been devel- oped for their measurement. To cover the wide range of valies occurring with different materials, shapes, and sizes, different sets of conditions have been established for some of the tests. Por any test for which this is true, it is necessary that the test con- ditions used be indicated as part of the measurement. ‘THE TENSILE TEST One of the more important tests for determination of mechanieal properties of materials is the tension est, Material specimens are fastened between a fixed ‘able and a movable table on a machine designed specifically for this purpose (Figure 3-4). A weighing scale is attached to the tables so that as they are moved apart (together for compression testing), the load imposed on the specimen can be measured. Universal testing machine ‘Setup ae chown for compression test Some machines are fitted with auxiliary equipment that takes into account the loads imposed and the resulting elongation of the specimen to actually plot « stress-strain diagram of the test. The same results can be accomplished without this special equipment by measuring the elongation as the loads are increased and plotting the individual points to develop the curve. Tensile Specimens. In order that these standard tests can be accurately reproducible and valuable for comparison with other tests, test specimens are made to one of several standard designs. Figure 3-5 shows the dimensions for a standard tension test bar with Binch gage length for rolled, flat stock. The radii from outside the gage-length portion to the increased section size at the ends are designed, in this and other test bars, to minimize stress direction effects from clamping loads on the end of the bar. Round test bars with the same S.inch gage length are standard for testing rod and bar materials, but because it is often impossible to produce test samples of this length from castings and forgings and other material sources, Z-inch gage length Is frequently used. The diameter ‘of the parallel section of round, tensile test bars is made to 0.505 inch (0.2 square inch cross-sectional ‘area) to facilitate calcalations. Adoption and use of the international metric system of measurement require that these dimensions be expressed in centimeters. Pee HE rere pa in tor, FLAT ear eanens seinen (Fil ROUND BAR ie Figure 35 Tension test bars &:inch gage length ‘Strese-Strain Diagram. An understanding of a tensile test can best be acquired from a stress-strain diagram made by plotting the unit tensile stress against the unit strain (elongation), as shown in Fig- ure 3-6. The illustration displays data from a tensile ‘test on ductile steel and is representative of this kind of material only. Curves for other materials take on slightly different shapes. Elastic Deformation and Plastic Flow. The straight line from A to B represents loads and defor-20. Materials and Processes for NDT Technology ‘mations in the elastic range, nd as long as the load at Bis not exceeded, the material will resume its original position and shape after removal of the load. B is the elastic limit for this particular material, and loads above that limit will cause permanent deformation (plastic flow) that cannot be recovered by removal of the load. At the load represented by the point at C, plastic flow is occurring at such a rate that stresses are being relieved faster than they are formed, and strain imereases with no additional, or even with a reduction of, stress, The unit stress at C is known as the yield point. UNIT STRESS, os ‘UNIT DEFORMATION Figure 36 Stress-strain diagram Plastic Now occurring at normal temperature is called cold working, regardless of the kind of loading system under which it is accomplished, As plastic flow takes place, the erystals and atoms of the materi al rearrange internally to. take. stronger positions resisting further change, ‘The material becomes, stronger and harder and is said to be work hardened! At the point D in Figure 36, the curve suddenly turns upward, indicating. that the material has. beooine stronger because of work hardening and that higher Joads are required to continue deformation. ‘The deformation rate, however, increases until at point F the ultimate strength is indicated Uitimate and Breaking Strengths. The ultimate tensile strength of a material is defined as being the highest strength in pounds per square inch, based on the original erost-sectional’ area, By this definition, ductile materials that elongate appreciably and neck down with considerable reduction of cross-sectional ‘rea, rupture at @ load lower than that passed through previous to fracture. The breaking strength, or rup- ture strongth, for this material is shown at F, consid- erably below the ultimate strength. This is typical of ductile materials, but as materials become less ductile, the ultimate strengths and the breaking strength get closer and closer together until there is no devectable difference. Yield Point and Yield Strength. Many materials do not have a well-defined or reproducible yield point, Plotting of tensile stress-strain values produ a curve of the type shown in Figure 8-7. For tess materials, an artificial value similar to the yield point called yield strength, may be calculated. The yield strength is defined as the amount of stress required to produce a predetermined amount of permanent strain. A commonly used strain or deformation is 0,002 inch per inch, or 0.2% offset, which must be necessarily indicated with the yield strength value. ‘The yield strength is the stress value indicated by the intersection point between the stress-strain curve and the offset line drawn parallel to the straight portion of the curve, Modulus of Elasticity. In the stress range below the elastic limit, the ratio of unit stress to unit devor mation, or the slope of the curve, is referred to as che UNIT STRESS, psi for aun UNIT DEFORMATION Figure 37 Yield strength modulus of elasticity, or Young's modulus, and is represented by 6. E, therefore, equals s divided by 5 Following ate listed the values of £ far some of the more common structural materials TABLE 31 10 «illion psi (6.8 x 10" Pe) 14 10 19 milion p= 12 to 19 million ps 28 0 30 milion 931 approx 60 million 2 Aluminum alloys Copper alloys Gray iron ‘Steel and high-strength irons Cemented carbides The gross values of the modulus of elastici Important to the design of members when deflectionor deformation in the elastic range must be given con: sideration, ‘The relative stiffness or rigidity of differ- fent materials can be ascertained merely by comparing their moduli, By rearrangement of the formula for E, the unit deformation becomes equal to the unit stress divided by E If a bar of steel with a cross-sectional area of 1 square inch and with a modulus of elasticity ‘of 20 million pounds per square inch is subjected to a tensile pull of 1,000 pounds, each inch of length of the bar will be stretched 1/30,000 of an inch, A 30.inch-long steel bar with this’ cross section would then be elongated 1/1,000 (0,001) of an inch overall with a 1,000-pound tensile load Ductility. ‘The tension test provides two measures of ductility, One is called percent elongation, repre sented by the formula percent elongation = (LE 1 > final gage length Ly = original gage length For ductile material the major portion of the elonga- tion will occur over a relatively small portion of the gage length after the specimen begins to neck as it approaches the breaking point. Because much of the elongation is localized, a variation of gage length would cause a difference in calculated percent elon- gation Another measure, percent reduction of area, is sleulated by comparing the original area of the spect- men to the smallest area of the neck at rupture. Resilience and Toughness. The area under a curve is influenced by both factors that are used to make that curve, In a stress-strain diagram the area under any portion of the curve represents the energy re- jquited to deform the material, Up to the elastic limit, this energy is recoverable and is called resilience. Toughness is defined as the ability of a material to absorb energy without fracture, For the tension test, the total area under the curve is @ measure of tough- ‘TRUE STRESS—TRUF STRAIN In the tensile test just described, stresses were cal- elated as though the original specimen size did not change. More precisely, the vertical axis of the dia- gram should be labeled load/original area rather than sivess. If each time a load reading were made, the smallest diameter of the specimen were found and the calculation for stress based on this actual diameter, this axis could be labeled true stress. The definition of trae strain is somewhat more complex, and, in any case, true strain does not differ greatly from olonga: von normally plotted. The greatest difference be- tween the diagram of Figure 9-6 and a true stress-flow strain diagram would be in the plastic flow region. Proportion of Materials 21 ‘True stress would continue to increase throughout the test, as shown in Figure 38, and maximum stress ‘would occur at the final break. The test of Figure 3.8 is usually called an engineer's stress-strain diagram. This curve is shown as a dotted line in Figure 3:8. Not only is it easier to proparo than a true stress-true strain diagram, but the value for ultimate strength obtained from it fe more useful for design than the maximum true stress that occurs when the specimen breaks. The true concern of a designer is the maximum load that can be supported, not the maximum stress, UNIT STRESS, pi for Pa DEFORMATION Figure 38 True stress:tue strain diagram COMPRESSION TESTING Up to the elastic limit, most metals are approxi- mately equal in properties under either tensile or ‘compressive loading. Cast iron, however, has a tensile strength of only about one-half its’ compressive strength and is therefore used mostly in applications where the principal loads are of the compressive type. ‘Many nonmetals such as timber, concrete, and other aggregates are also used almost entirely for supporting compressive or compactive loads. This is due in part to higher compressive strength, but also these materi- als have a high incidence of flaws and faults that might cause sudden failure in tension but produce relatively small effect under compressive loading, ‘The testing of materials in compression is con- ducted in much the same manner as in testing under tension. Specimens are placed between tables of a testing machine that sre brought together to subject the specimen to compressive loads. Compression specimens must be short compared to their diameter so that column affect will not cause bending with eccentric, unequal loading,22 Materials and Processes for NDT’ ‘Technology ‘TRANSVERSE RUPTURE TESTING Limitations of Tensile Tests for Brittle Materials. in a number of cases a substitute for the standard tensile test is necessary. With some materials that are ‘difficult to shape or very brittle in nature, it is im- practical to produce a specimen for tension testing. ‘This condition occurs particularly with ceramics. With most materials that are very brittle in character, even though a tensile specimen might be produced, the results from the standard tensile test would have only limited significance. It is almost impossible to insure in the tension test that the applied load will be precisely centered in the specimen and will be exactly parallel to the axis of the specimen. If this is not the case, bending moments are introduced in the speci ‘men, With a ductile material, small amounts of plastic flow take place in the specimen, particularly where the load is applied; the specimen aligns itself properly with the load; and the stresses are uniform across the tested area. With a brittle material in which this align: ment cannot take place, the bending moments result in higher stresses on one side of the specimen than on the other. The specimen fails when the highest stress reaches some critical value, but the observed stress at this time, based on the assumption of uniformity, is somewhat lower. As 2 consequence, the results from testing a number of similar brittle specimens exhibit, wide variations and are not representative of the true strength of the material. ‘The Transverse Rupture Tost. The transverse rup- ture test, while it gives less complete information ‘than the tension test, isa fast and simple test, making ‘use of more easily prepared specimens, and is especi- ally well suited to brittle materials. In many instances the specimen can be an actual workpiece. The test is particularly well suited for those materials that are to be used in beam applications. It is really the only ‘meaningful type of strength test for reinforced con- rete. ‘The test consists of loading a simple beam as illus- trated in Figure 3-9. While some standards have been jt for particular materials, there are no univeral tandards for specimen sizes and shapes as there are ‘or the tension test he modulus of rupture, or beam strength, is cal- ulated by the formula 3PL Sr" Oba Limitations of Transverse Rupture Testing. While nis formulas the formula that is used to calculate we maximum actual stress in the outer fibers in a cam, it is based on the assumption that stress re- ‘ains proportional to strain. This is not the case for vost materials when highly loaded, with the result sat the calculated “stress” is higher than the actual ress in the outer fibres at rapture, and direct com- parison cannot be made with ultimate tensile strength values taken from a tension test, nor can the values of ‘modulus of rupture be used as design tansile strength values. The values are useful for comparing matertals, and they are useful in design when the material is to be used as a beam. Figure $9 Transverse rupture test SHEAR TESTING In the section dealing with material failure, it was pointed out that when a bar is subjected to a tension load as in the tension test, the value of shear stress existing in the bar at failure can be calculated from the load and the dimensions of the bar (Figure 3-10). Figure 3-10 Hoads of a torsion testing machine. Torsion is the simplest way of obtaining pure shear stress. Results are useful for evaluating cola-warking properties of metals Th eat tom Act tem sure, thar, whic cate Ais the as sh FATIG Am peated, applied material alterat: cycles a time. 7) of servic whieh fr tural me of load must be fatigue p Fatigu Fatigue stress. cor the mem| material, voids, ere and fault‘The term shear, however, has a broader meaning than. shear stress only and is used to describe loading sys- toms that subject @ material to a shearing action Actually, the stress distribution in such loading sys tem is quite complex, but a rather simple shear strength test has been developed that simulates the conditions of actual loading and provides information that may be used in deisgn where the loading situa: tion is similar to that of the test. Such loading occurs in using bolts or rivets and in shearing operations in which material is being separated. In the test ind) cated in Figure 3-11, the bar with cross-sectional area 4 ig made to fail simultaneously in two places so that the area of failure is 2A, and shear strength js defined ‘as shoar strength = P/2A Pe 3 PATIGUE TESTING A metal may fail under sufficient cycles of re peated stress, even though the maximum stress applied is considerably less than the strength of the material determined by static test, Failure will occur at a lower stress level if the cyclic loading is reversed, alternating tension and compression, than if the icles aye repeated in the same direction time after time, The conclusion from one comprehensive study of service failures was that in 90% of such failures in which fracture occurred, fatigue was involved. Struc tural members subject to vibration, repeated variation of toad, or any cyclic disturbance causing deflection ‘must be designed to have low enough stress levels that fatigue phenomena will not cause failure. Fatigue Failure Initiation and Development. Fatigue failure normally starts at some spot where stress concentration is high because of the shape of, fhe member or some imperfection. Holes through the material, notches in the surface, internal flaws, such as flds, cracks, oF inclusions or even minor seratehes and faults caused by corrosive attack on the grain Figure 9-11 Shear strength test Proportion of Materia 20 boundaries, may be sources of fatigue failure. With repeated siressing, a crack starts at one of these fatigue nuclei ana grows until insufficient solid metal remains to carry the load. Complete failure in a sud: den, brittle manner results. As seen in Figure 3:11, the exposed surface of a fatigue failure shows part of tho surface to be smooth and polished, while the rest exhibits a welllefined grain structure, The crystal line-appearing portion was separated in the sudden, final break. The smooth part was polished and burnished by the movement of the material with repeated deflection as the crack developed and grew Fatigue failure is more frequent than commonly ‘thought. There have been estimates that with oquip- ment having moving parts or subject to vibration as ‘much as 90% failures inchude fatigue in some form. Be- cause any kind of discontinuity, particularly those at; (or near) the aurface whore tensile stresses are likely to be highest, can be the nucleus for fatigue failure, loca- tion of these spots by NDT may prevent a later cata- atrophic failure. Endurance Limits. Because a material may fai under conditions of a great many repeated Toads at a stress level far below that determined by the standard strength test, 2 designer must know how different materials stand up under these conditions. Tests have ‘been developed with special machines that bend plate-shaped test specimens or subject a rotating beam to a bending load for large numbers of cycles. From data collected from such tests, the endurance limit of a material can be determined. ‘The endurance limit is the highest completely re- versed stress whose repeated application can be endured for an indefinitely large number of cycles without failure, Figure 3-12 shows a typical S-N, or endurance limit, curve. The material represented by this curve would have an endurance limit of 42,000 pounds per square inch (200 MPa) because the curve 60) - MAX, STRESS, kipalin? $ 7 BF ee TF overs Figure 3.12 ‘Typical SN curve24. Materials and Processes for NDT Technology has flattened out, and stressing at this level could be continued indefinitely without failure, Endurance limits correlate fairly’ closely with tensile strength and for most materials are from about one-third to one- half the stress required to break a tensile specimen. Fatigue Strength. For some materials the curve does not flatten even after several hundred million cycles, When the endurance limit cannot be de- termined, of it is impractical to carry on a test long enough for this determination, it is common practice to use another value, fatigue strength, to evaluate the ability of a material to resist fatigue failure. Fatigue strength is the stress that can be applied for some arbitrary number of cycles without failure, The mum- ber of cycles for which a fatigue strength is valid must always be specified because the operating stress chosen may be at a level where the S-N curve still slopes, and indefinite cyclic operation could cause fatigue failure. CREEP TESTING ‘The term creep is used to describe the continuous deformation of a material under constant load, producing unit stresses below those of the elastic limit, At normal temperature, the effect of creep is very small and can be neglected. As operating temper- atures increase, however, this deformation by slow plastic flow becomes very important in the design and use of material. Recognition of this phenomenon is ‘most important for the higher strength materials that are to be used at elevated temperatures. Creep tests are conducted by applying a constant load to a material specimen held at the desired temp- erature and measured periodically for deformation ‘over a long period of time. The results may be plotted ‘on a graph of elongation against time, as in Figure 3-18, with an indication of the maintained tempera- ‘ture’ and stress level under which the test was con ducted. Most creep tests are carried on for periods of at least 1,000 hours, so this is a time-consuming test. The creep strength of a material is the stress required to produce some predetermined creep rate (the slope of the straight portion of a curve) for a prolonged period of time. Commonly, the stress required to produce a creep rate of 1% in 10,000 hours is used as ‘creep strength. Stress rupture strength is defined as the stress required to produce failure at prescribed values of time and temperature. NOTCHED BAR TESTING Materials are often used in situations in which dynamic loads are suddenly applied to produce shock that increases the effective load far above that which would be expected from gradual application of the same load or a similar static load. Tests designed to ‘check the ability of a material to withstand this kind ‘of loading are energy absorption tests that seldom can ‘x euoNesmion = Tae Figure 313, Creop test be used to give information that can be used directly in design, but primarily provide data for com- parison of different materials. While such tests are frequently called impact tests, the energy required to ‘cause failure does not differ greatly from that re- quired if the load were applied slowly. True impact failure, in which the energy-absorbing capacity of a ‘material is greatly reduced, occurs only at much high- er speeds. Charpy Test. The most commonly conducted ‘tests are bending impact tests, using one of two kinds of notched speciments (Figure 3-14). The Charpy specimen is supported at both ends by a standard Figure 914 Impact specimens nguces Br be tic (co ducimpact testing machine and struck on the side oppo- site that of the notch. The testing machine is con- structed with a weighted pendulum, which is lifted to start the test. Upon its release, the pendulum swings past the specimen, and breaks it. As the pendulum swings past, the remaining energy can be measured by the height of the swing and the absorbed energy de- termined, lod Test. The Izod specimen is supported in the esting machine by one end only and is loaded as a eantilever beam with a notch on the side of impact. Energy absorption is measured in the same way as with the Charpy specimen ‘Fest Specimens. Two kinds of notches are used on bending impact specimens. The Izod specimen is usually made with a 45° angular notch with a G.010-inch radius at the bottom. The specimen is, extremely sensitive to variation of notch size or change of radius, and extreme care in manufacture of the Lest specimen is necessary for reproducibility of, est results, The keyhole notch shown on the Charpy specimen can be duplicated more accurately but is, limited in the smaliness of the hole producing the notch effect by the size of the smallest drill that will not “drift” in making the hole, The notches in the lest specimens act as points of stress concentration, and the smaller the notch radius, the more severe is, ‘ue stressing at this point, These notched test speci mens actually provide only information regarding ‘material that is to be used in a similar notched condi. tion but are often practical because materials are fre- quently used with design shapes of structural imper- fections that cause a structural member to be, in effect, a notched beam. Tensile Impact ‘Test. Greater reproducibility and greater similarity between the Lest and some use con- sitions can be provided by tensile impact tests. The specimens for these tests are not notched and are supported so that uniaxial tensile impact loads may be applied. The standard impact testing machine with pendulum weight can be tooled for testing small specimens of this type. Por larger specimens a special machine with a variable-speed flywheel to store energy can be obtained, BEND TESTING Materials that are to be deformation processed by being subjected to bending loads and materials that may have been affected by localized heating, such as in Welding, are sometimes tested by bend tests to provide comparative data, Free Bend Test. Free bends are accomplished by sebending a flat specimen slightly to produce eccen- tricity and then loading the specimen in compression (column) until failure occurs or a 180° bend is pro- suiced, Normally, the loads to accomplish this are so Propertios of Materiels 25 variable that they are of little value and are not re- corded, Instead, the angle of bend at failure is com- pared with results of other tests. Guided Bend Test. In guided bend tests, the test specimen is bent about a fixed radius to 180°. The bend angle of a failure before 180° bending usually cannot be satisfactorily compared with other test results because of nonuniform plastic flow of material in the specimen caused by pressures set up by the guided bend fixture. Multipleradius guided bends may be used for rating specimens by determining the smallest radius about which a standard specimen will bend 180°, HARDNESS TESTING The most frequently used tests for determining material properties are hardness tests. With sufficient Knowledge of material composition and previous processing, hardness tests can be used as indirect measures of properties entirely different from hard: ness. For example, hardness can sometimes be used to separate raw materials of different composition, to determine whether or not satisfactory heat treating or other processing has been accomplished, or to mea: sure the strength and wear-resistant properties of a product. Hardness measurements, therefore, are fre- quently made on raw material, on parts in process, ‘and on finished goods ready for use. With some metal alloys, electrical conductivity and hardnese are related within limited ranges. Eddy cur- rent tests standardized to measure electrical conduc- tivity can thorofore be used as an indirect measure of hardness. Such tests must be applied cautiously since the ranges are restricted ovar which the relationship between hardness and conductivity are reasonably lin- ear. Aluminum alloys and other non-ferrous metals are more reliably tested by this method than are ferrous alloys. ‘Most hardiness tests result in some kind of measure of the ability of a material to resist penetration of the near surface material. Penetration of material with any kind of indentor requires the use of force and involves plastic flow of the tested material. The work-hardening qualities of a material, therefore, become part of most hardness measurements and partially explain the difficulty of converting from one type of hardness measure to another, because dif- ferent methods of measuring hardness do not measure exactly the same thing. They are, however, well enough standardized to provide useful and practical information. ‘Mohs Test. One of the first standardized systems of measuring hardness made use of the Mohs scale of hardness, which specifies ten standard minerals ar ranged in order of their increasing hardness and num- bered according to their position. Starting with26 Materials and Processes for NDT Technology number 1 as the softest, the standard Mohs scale is as. follows: 1... Tale 6 ... Orthoclase (Feldspar) 2... Gypsum 7... Quartz 3... Calcite 8... Topaz 4... Fluorite 9... Corandum 5... Apatite 10... Diamond If a material can be noticeably scratched by the mineral topaz (number 8) but cannot be scratched by ‘quartz (number 7), it would have a hardness value between 7 and 8 on the Mohs scale, The Mohs scale of hardness has little value for hardness testing of metals but is still widely used in the field of minera- logy. File Test. Another abrasion or scratch method of ‘measuring hardiness that does have some practical use in metal working is the file test. Standard test files can be used to gage quickly the approximate hardness of a material and, although not very accurate, can be used in many shop situations with satisfactory results, Experience and comparison with standard test blocks will permit a fair degree of accuracy to be attained. Brinel) Test. In 1900 Johan August Brinell, a ‘Swedish engineer, introduced a new universal system. for hardness measurement. The method involves impressing, with a definite losd, a hardened steel ball into the material to be tested and calculating a Brinell hardness number from the impression size (Figure 3-15), A wide range of hardnesses can be tested by varying the size of the ball and the loads imposed, but in the hardness range most frequently tested, a ball 10 millimeters in diameter is impressed into the ma- terial under a toad of 3,000 kilograms for 10 seconds to check steel and under a load of 500 kilograms for 80 seconds to check nonferrous materials. The weiseTR (0-VIEE Figure 315 Brinoil hardness measurement numerical value of the Brinell hardness number is obtained by dividing the load in kilograms by the area of the spherical impression in millimeters. In practice, the average diameter of the impression is usually read with a measuring microscope and the Brinell hardness number determined directly from a table. Advantage and Limitation of Brinell Tests, ‘The Brinell hardness method has the advantage, as com- pared to most other measuring methods, of determin- ing a hardness value over a relatively large area, thus reducing the inconsistencies caused by flavs, imper- fections, and nonhomogeneity in the material, likely to be introduced with small area measurement that includes only a few metallic grains. With plain carbon and low alloy steels, the relation between tensile strength and Brinell hardness is so consistent in the medium hardness range that the tensile strength of the steel can be closely approximated by multiplying ‘the Brinell hardness number (BHN) by 500. The principal disadvantages of the Brinell method are that the machine to supply the load for impressing the ball, into the material is often cumbersome and cannot always produce the impression where desired. The all cannot be impressed in very thin materials and, of course, cannot be used to examine extremely small samples, and the impression is of such size that it may harm the appearance or uso of finished surfaces, Rockwell Test. Because of its convenience and the fact that only small marks are left in the work: tested, one of the most frequently used tests is the Rockwell hardness test (Figure 3-16). This also is an impression test, but the hardness number is deter- mined by a differential depth measurement that can, be read directly on a dial indicator of the machine ‘used to impose the load (Figure 917 shows the Rockwell hardness tester.) To obtain a Rockwell Figure 316 Rockwell hardness measurement Ro. special the be materi the har tippedhardness reading, the equipment is first used to place ‘a minor load of 10 kilograms on the penetrator. This reduees the effect of dirt, oil films, scale, and other surface conditions that might affect the reading. A major load of 60, 100, or 150 kilograms, depending upon the type of penetrator and scale being used, is then imposed to force the penetrator into the work material, After the penetrator has seated to its full depth ~ the time usually being controlled by a dash pot built into the equipment — the major load is removed. The permanent differential depth between, the minor and major loads is then read directly as a Rockwell hardness number. Standard Rockwell Seales. Although provision has been made for use of a 1/8inch-diameter ball as a penetrator, almost all hardness testing with the Rockwell equipment is done with two standard pene- trators. The one used for softer materials is a 1/16- inchdiameter hardened steel ball supported in a Figue 317 Rockwell hardness tester. The tester impresses a enotrator into the work to provide @ Girect surface hardness measurement special chuck that permits easy replacement should the ball become damaged. The testing of harder materials that would cause excessive deformation of the hardened stee! ball is performed with a diamond- tipped penetrator with a 120° conical point and a Propertios of Materials 27 spherical tip of 0.200-millimeter radius. The diamond penetrator, or indentor, is known as a brale. ‘The penetrator used and the size of load impressing it into the test material are defined by a letter that, becomes part of the Rockwell reading. The accom- panying Table 3-2 shows the relationship among the scale designation, the loads, and the penetrators. TABLE 32 Load Seale kilogram Penetrator Aastra se vei aensats ot pop COLT Oro Ee 8 a siss1100 1/16" bal c fost 150 Brae D 100 Brale FB. 1/16" batt 6 1198" batt ‘The letter designating the test conditions is a very important part of a hardness notation because the number alone could represent several different hard- ness conditions. For example, a Rockwell hardness reading of B60 would represent a relatively soft material, such as a medium hard copper alloy. A Rockwell hardness reading of C60, sometimes written Re 60, on the other hand would represent a hardness such as might be used for a hardened tool steel to cut metals, Superficial Rockwell Test. Another machine, the Rockwell superficial hardness tester, is contructed and used in much the same manner as the standard machine but is a special-purpose tester designed to be used when only a very shallow impression is permis- sible or when measurement of hardness of material very close to the surface is the principal aim. The superficial hardness tester makes use of the same penetrators, except that the brale is of higher pre- ‘cision and is designated as N brale. The loads used to cause penetration are lighter: 15, 30, and 45 kilo- ‘grams. Table 3-3 shows tho testing conditions for Rockwell superficial hardness testing. TABLE 9-3 Seal IGN earner IO CEE alo 30N 48N Norale wT 1/16" bat 30T 1/16" ball a5T 4/16" bait ‘As in the previous case, the scale indication must bbe used as a prefix to the hardness number read from the dial.28 Material and Processes for NDT Technology Vickers Test. ‘The Vickers hardness tester operates ‘on the same principle as the Brinell instrument but makes use of a diamond penetratot shaped as « four- sided pyramid. The impression made by the pene- trator is accurately measured by swinging a micro- scope into position without moving the test piece in the machine. As in the Brinell method, the Vickers hardness number is the ratio of the force imposed on the indentor to the area of the pyramidal impression. In the lower range of hardness, under Brinell 300, Vickers and Brinell hardness numbers are almost iden- tical, but above this range they separate as hardness increases, primarily because of distortion of the steel ball used for Brinell testing when it is forced against the harder materials. Microhardness. It is frequently important, partic- ularly in research or development work, to test the hardness of material that is very thin or very small in area. A number of special machines have been devel- oped for determining “microhardness.” One of the more commonly used pieces of equipment of this type is the Tukon microhardness tester. Normally, the machine is fitted with an elongated diamond: shaped penetrator. Microscopic measurement of the impression provides information that can be con- verted to Knoop numbers. Knoop hardness measure ment often cannot be compared directly with Brinell or Viekers hardness measurement because the elonga- ted impression is rather strongly affected by the directional properties of the material being tested. ‘The use of a symmetrical, square-based, pyramid. shaped indentor will provide hardness data com- parable with that of the other systems, It should be self-evident that the lighter the in- dentor loads and the smaller the impressions made, the greater the care that must be used to perform & hardness test, and the better must be the quality of surface on which it is made. In Brinell testing, small surface imperfections tend to be averaged out because of the large area covered, but in microhardness checks, in which the impression may be only a few thousandths of an inch long, small scratches and sur- face imperfections may contribute large errors. Micro- hardness testing is usually performed on a highly polished surface, and in many cases, to obtain repro- aucibility, it is necessary to etch the surface to reveal the constituent structure in order to locate the im- pression properly. FACTOR OF SAFETY No property, structural or otherwise, whether cal- culated from theoretical considerations or determined by test procedures, can be safely used at or very close to its ultimate (maximum) value. Tests are neither consistent enough nor accurate enough, particularly as they are not conducted under exact use conditions, to permit strong confidence to be placed in their re sults, Also, because of the complexity of stress- analysis problems, it is almost essential that simpli Jying assumptions be made during design to prevent design costs and time from becoming prohibitive. A factor of safety is therefore used to prevent working too close to maximum values. The factor of safety is the ratio between the maximum value and the work- ing value and is determined by competent judgment, taking into consideration all conditions of use. Fac: tors of safety vary from as low as one to as high as five or more, They may be applied to any quality but ‘are most commonly used in connection with strengths. As an example of its use, if the ultimate tensile strength of a certain grade of steel is 80,000 pounds per square inch and its elastic limit, 60,000 pounds per square inch, an allowable stress, or working stress, ‘of 20,000 pounds per square inch would provide a safety factor of four, based on the ultimate strength, or of three, based on the elastic limit, ‘The closer the factor of safety approaches one, the ‘more the danger that an unforeseen fault ot condition of use may cause failure. On the other hand, the Jarger the factor of safety, the greater the volume and weight of material needed, with a corresponding increase in cost and in space-need problems. Factors of safety in the range of two to four are most com- ‘mon, but a satisfactory value depends upon a groat number of conditions, some of which are described in the following paragraphs. Allowances must be made for unexpected loads ot conditions. This is particularly true if the human ele- ment is large in the use of the equipment, since the human mind is most unpredictable. It is common to include a factor of at least two in the factor of safety when a design is based on static tensile strength values but subjected in use to varying loads. This cor- responds approximately to the ratio of static tensile strength to endurance limit. Allowances must be made for environmental and time factors. Strengths of most materials are greatly reduced by corrosion and other chemical effects. Other materials lose strength or become brittle with age. The consistency of test data should influence the factor of safety choice. Test information should be of large enough volume to be statistically significant. Larger safety factors are necessary with materials varying widely in quality than with those that are quite uniform, Whether or not the use of a material may affect human life has a large influence on the factor of safe- ty, In the designing of hoists, cranes, and other lifting equipment, factors of safety of five or more are com- monly used because failure could mean injury or loss of life, The same consideration applies, of course, to aircraft design. Here, however, space and weight are very important, and large factors of safety could easi-ly prevent a usable design; consequently, the problem is handied in a different way. Extreme care is used in selecting and testing materials. Stresses are carefully calculated and, as far as possible, the structures built so that they cannot be overloaded in use. Thus, by spending more cate, time, and money preceding and during manufacturing, it is possible to use a smaller factor of safety because of greater certainty of not exceeding the design condition. The smaller the safety factor and the more important any possible failure, the more reliable must be any nondestructive testing pro- cedure that is used. Properties of Materials 38The Nature of 4 Materials and Solid State Changes in Metals ‘The chemist ordinarily considers the smallest func- tional portion of matter to be the atom. The atom consists of a nucleus, made up of positively charged protons and uncharged neutrons, surrounded by electrons. ‘The electrons carry negative charges and move in orbit at different levels. Each level of orbit can contain only a definite number of electrons, and the number of levels or shells is determined by the atomic number of the element. All the shells will usually be full except the outer one, which is short ‘of the maximum possible number of electrons for most materials, All of the electrons are in constant motion, spinning about their own axes and traveling through thei orbits about the nucleus with speeds dependent on their energy level, which in tum is strongly affected by the pressure and the temperature conditions, The physicist’s picture of an atom depicts it as a heavy nucleus containing most of the mass, surrounded by a cloud of moving electrons. ‘THE EFFECT OF ENERGY ON THE ATOM Forees on the Atom. A number of different forces exist among the atoms making up a material, some of them attractive, some repulsive. The nature ‘of any material depends primarily on the nature of these forces, which themselves depend not only on the type of atom, but also on the energy level of the atom, At high energy levels, the repelling forces predominate, and the atoms tend to move as far from each other as possible. This condition is called the gaseous state. If the energy of the material is lowered, the forces change, and a condition of equilibrium is reached in which the atoms assume fixed average dis tances from each other, although still frae to move92 Materials and Processes for NDT Technology land not tied closely together. In this liquid state, the materials have fixed volume but assume the shape of the container in which they are placed. As the energy level is further decreased, the mobili- ty of the atoms decreases. There are at least four different mechanisms by which the atoms ean assume positions well fixed enough that for practical pur- poses the material could be called solid. Of the ‘materials of interest to manufacturing, all the metals ‘occur as erystattine solids, METALLIC STRUCTURE Definition of a Metal. Metals are usually defined ‘as materials having some degree of plasticity, relatively high hardness and strength, good electric and thermal conductivity, crystallinity when solids, and opacity, A definition based on atomic structure is more pre: cise. A metallic solid is one that has free electrons available in the structure to carry a current and that has a negative coefficient of conductivity with in creasing temperature. States of Matter. Figure 4-1 shows the relation- ship that exists among the three states of matter for crystalline material. At the intersection of tempera: ture T, and pressure P, on the curve, notice that an increase of temperature of a material for which this curve is valid would cause the material to change directly from a solid to a gas. Similarly, a reduction of pressure (a shift toward the left) would also cause the same change. Such a change of state from solid directly to gas is known as sublimation. Arsenic is the only metallic material that sublimates at atmospheric pressure. When the temperature is raised to 7; at Pressure Pz, the atoms of the material will become sufficiently active that a change is made from a solid to a liquid. A further increase in temperature at this same pressure to point. Ty will cause agecond change ‘TEMPERATURE ——> PRESSURE ———» Figure 4-1 States of mattor from a liquid to a gas. The intersecting point of curves at the temperature Ty and pressui known as the tiple point and occurs ot she tem ture and pressure conditions under whic a mate may exist as a solid, a liquid, a gas, or partially al! three at the same time. For most metals, this point ‘oceurs below normal temperatures and well holow atmospheric pressure; consequently, most metals upon being heated go through the changes from solic to liquid to gas as the temperature increases, Space Lattices. As the energy’ of a liquid metal ie reduced by taking away heat, the attraction between atoms increases until they arrange themselves in definite three-dimensional geomotrie patterns that are characteristic of the metal. Thess structures are called space lattices and consist of network groupings of identical unit cells that are aligned in jvallel ple ‘There are fourteen types of crystal iatlices. but most of the common and commercially importaat metals exist, in the solid state, in one of three struc tures, These are, as shown in Figure 4-2, body-cen- tered cubic, facecentered cubic, and hexagonal closed: packed. In tho illustrations of unit cells, the dots repre senting atoms should be considered as centers of activ. ity for the atoms and not as graphic illsutrations of the atoms themselves. wooy-cenreaco cual: LATTICE FAce- CENTERED CUBE HOXCAGONAL CLOSE-PAOKED Lar HCE Figure 4.2 Common metallic space laiticeA single unit cell does not exist alone. To attain stability, it must grow past some eritical size by being joined with other cells that share the atoms on the outer adjacent surface. For purposes of iilustration, it hhus been assumed that a unit eell can exist by itself ‘and that all its atoms belong to it alone, Body-centered Cubie Lattice. ‘The body-centered cubie cell is made up of nine atoms. Bight are located fn the corners of the cube with the nth positioned ventrally between them, The body-centered cubic is @ strong stucture, and in general, the metals that are hard and strong are in this form at normal tempera- tures. These metals include chromium, iron, molyb- denuin, tantalum, tungsten, and vanadium, Face-centered Cubic Lattice. Face-centered cubic calls consist of fourteen atoms with eight at the cor- ners end the other six centered in the cube faces, This structure is characteristic of ductile metals, which include aluminum, copper, gold, lead, nickel, plati- num, and silver. Tron, which is body-centered eubie at room temperature, is also of the face-centered struc: ture in the temperature range from about 910° C to 1400" C. This is a solid-state change that will be dis- cussed more thoroughly in the following chapter. Hexagonal Close-packed Lattice. Seventeen atoms combine to make the hexagonal close-packed unit cell, Seven atoms are located in each hexagonal face vyath one at each comer and the seventh in the center. ‘The three remaining atoms take up a triangular posi tion in the center of the cell equidistant from the two faces, The metals with this structure are quite sus. ceptible to work-hardening, which will be discussed in, the following chapter. Some of the more commonly used metals that crystallize with this structure are cadmium, cobalt, magnesium, titanium, and zinc. Tin is an exception to the other commonly used metals in that the atomic configuration is body centered tetragonal, which is similar to the body- centered cubic but has wider atomic spacing and an elongated axis between two of the opposite faces. SOLIDIFICATION Growth of a Crystal. As the temperature of the liquic metal is reduced and the atoms become less active, they are attracted to each other and take defi- nite positions to form unit cells. Because cooling can not be exactly the same for every atom, certain ones wil assume their positions ahead of others and become a nucleus for erystal formation. In the proe- ss of assuming their positions, these first atoms will give up Kinetic energy In the form of heat, which retards the slowing down of other atoms; but as heat removal is continued, other atoms will take their places along the sides of the already solidified unit cell, forming new cells that share atoms with the first and with others to come later. Orderly growth con- fanues in all directions until the crystal, or as usually ‘Materials and Solid State Changes in Metals 33 referred to for metals, the grain, runs into interfer. cence from other grains that are forming simultaneous ly about other nuclei Although with some metals and with special treat- ments it is possible to grow single crystals several inches in diameter, with most metals and at the usual cooling rates, great numbers of crystals are nucleated and growing at one time with different orientations. If two grains that have the same orientation meet, they will join to form a larger grain, but if they are forming about different axes, the last atoms to solidi- fy between the growing grains will be attracted to each and must assume compromise positions it an attempt to satisfy a double desire to join with each ‘These misplaced atoms are in layers about the grains and are known as grain boundaries, They are inter ruptions in the orderly arrangement of the space lat- tices and offer resistance to deformation of the metal A fine-grained metal with large numbers of interrup: tions, therefore, will be harder and stronger than a coarse-grained metal of the same composition and condition. Grain size, grain orientation, and the composition of grain boundaries are factors that can influence some nondestructive tests. In radiography, at certain x-ray ‘energies, diffraction effects can produce images that resemble flaws and, at best, make interpretation diff- cult. Ultrasonic testing of large grained castings and welds also may be radically influenced by excessive noise and attenuation, which may produce false indice tions or mark the presence of dangerous flaws. GRAIN SIZE ‘The grain (crystal) sizes produced during soliditica- tion are dependent both upon the rate of nucleation and upon the rate of growth of grains. For most ‘materials the rate of growth is relatively slow, and the primary influence on grain size is the rate of nuclea tion. Grain size can be used as an indication, or measure, of properties, For this reason, visual stand- ards have been set up to aid accurate comparisons, While not in routine usage, ultrasonic methods have been applied to grain size determination. If the grains are randomly oriented, at high ultrasonic frequencies, reflection from grain boundaries that would usually be considered noise can be related to grain size. Importance of Grain Size. Grain size exerts an important influence on the mechanical properties of materials and, fortunately, can be controlled by methods much more precise than manipulation of the factors that influence growth during solidification. In some processes though, particularly casting, the soli dification grain size is important, because with some materials and some shapes, grain size cannot be readi- ly changed after the first formation. In those cases in which changes can be effected, additional processing34 Materials and Processes for NDT Technology costs will be added. The methods, other than solidifi- cation, that can be used for grain-size control involve solid-state changes, ‘As has already been indicated, coarse grains in the harder materials have lower strength than fine grains Coarse-grained materials machine more easily, requir- ing less power, although the quality of surface pro- duced will not be as good as with a finer-grained ‘material. Coarse-grained ferrous material is easier to harden by heat treatment than fine-grained material of the same composition but has increased suscepti- bility to eracking under the thermal loads. Coarse- grained material will cascharden on the surface more readily than fine-grained. It is evident, then, that coarso grains may sometimes be desirable during processing, but fine grains are usually necessary in the final product to provide the best mechanical proper. ties, Some deformation processes of shaping materials can be used so as to cause grain-size reduction auto- ‘matically during the shaping process with little or no additional cost involved, SOLID STATE CHANGES IN METALS In the previous section the process of metal solidifi- cation was briefly described. The properties of a mate- rial are derived from the erystalline structure, includ- ing the atomic arrangement and the erystal sizes, and are affected by the boundary layers that join the grains together. The atomic arrangement is primarily a fune- ton of the material composition, which may consist of ‘a single material or a combination of materials that are completely soluble, pertially soluble, or totally insolu- ble in each other in the solid state, The structure and grain size also may be influenced by the operating tem- perature changes and by mechanical loads that stress the material sufficiently to cause plastic flow in com- bination with time and heat effect. Some materials, particularly those that are cast to shape, may be used with the structure in which they solidify, but some of the cast materials and nearly all metals processed by other methods are treated in some way in the solid state to obtain improved mechanical properties. ‘These treatments include work hardening, recry- stalization, age hardening, and heat treating of allo- tropic materials to cause crystal transformations. In many cases, treatment may be inherent in the process, This may be beneficial, as in many cases of deformation shaping with associated work hardening, ‘or may be detrimental, as in other cases in which cold working develops directional properties in @ material to make some kinds of further cold work difficult or ‘impossible. WORK HARDENING Erfects of Deformation. The application of loads to a solid material in processing or in service can ‘cause two kinds of deformation. If the load does not stress the material past its elastic limit, the deforma. tion is “elastic,” and the material returns to its original position upon removal of the load. If, how- ever, the elastic limit is exceeded, the material does not retum completely to its original position when the load is removed and is permanently deformed by plastic flow within its erystalline structure. When the elastic limit is passed, elastic properties are not lost, but instead are enhanced, providing the deformation is produced by cold work. The strength of metal is increased by plastic flow and the elastic limit is raised. Some of the deformation processes produce improved properties at the same time the shaping is being performed PLASTIC DEFORMATION Permanent deformation of metallic crystals occurs fn three ways: slip, twinning, and rotational deforma. tion. The degree of each is dependent largely on the characteristics of the particular metal, UNSTRANED GRAN SUP DEFORMED GRAIN Figure 43 ‘lp Slip Deformation. Slip deformation is illustrated in Figure 4-3 and occurs by translation or sliding be: tween the atomic planes within a grain. If the defor. mation causes mote than a very minor shift, a large number of atomic planes in each grain will slide over adjacent planes to occupy new locations with new neighbors. The planes through the crystal that. are Usually most subject to slip are those of the greatest atomic population and greatest distance between planes, The orientation of the planes along which slip takes place most easily will, of course, be different for different types of crystal lattices. Because of the usual random orientation of the crystals, the slip planes of many will not be in line with the direction of loading. When the best slip planes are completelyout of alignment, slip may occur along other Joss preferred planes, ‘Twinning Deformation. Figure 4-4 shows a type of grain deformation referred to as twinning, which seems to occur most easily under loads applied sud denly, rather than gradually. With twinning, the grain FANT | SAY Sos PNY Figure 44 Twinning deforms by twisting or reorienting a band of adjacent lattice forms, with each unit cell remaining in contact with the same neighbors it had before deformation took place. Rotational Deformation. A third type of shift in a grain is a kind of rotational deformation of portions fof the crystal lattice. Stresses below the elastic limit cause the crystals to be tomporarily bent and de. formed, but when the elastic limit has been exceeded ‘and slip has occurred on a number of difforent planes, sections of the lattice tend to bend and rotate to a neve, preferred orientation, After a large percentage of grains have been reoriented by action of consider able deformation work, the metal is likely to take on ‘ireetional properties called fibering, Fibering may be veneficial or harmful, depending upon the use to which the material is put. COLD WORK According to dislocation theory, as plastic flow takes place, existing dislocations (atomic discontinui- Lies) are reinforced and new dislocations are created to ‘sist further plastic movement. Regardless of what the exact mechanisms may be by which plastic flow takes place in the metal graine, itis a proven fact that when metals are cold worked to produce plastic defor- mation, they become harder and stronger. The word cold in this instance refers to different temperatures {or different metals. Cold work is work accomplished below the reerystallization temperatures for the partic: ular material. ‘The mechanical strain energy necessary to produce the plastic deformations described above is converted io other forms of energy within the material. In most metals, dislocation processes are accompanied by dis- crete releases of mechanical energy, sometimes called strese waves, These stress waves produce acoustical Mateiale and Solid Steto Changes in Metala 86 vibrations that travel at high velocity through the ma- terial and can be detected by sensitive microphones or transducers coupled to the metal surface. Analysis of, the intensity, duration, and rate of such acoustic emis- sions can provide information about the formation and ‘growth of microcracks that result from continued load: ing. Acoustic emission monitoring is a relatively new ‘means of nondestructive testing that has the capabil- ity of signaling etates of over-stress and producing ear- ly warning of impending failure RECRYSTALLIZATION Metals that are cold worked are left with their ‘grains in a strained and unstable condition. The grains have a tendency to return to the equilibrium of a lower energy state by equalization of internal cry- stalline stress ot by changing to new, unstrained grains, The greater the deformation strain, the greater the instability and the easier it is for the change to take place. Time and temperature also have strong influence. Two kinds of change, recovery and recry- stallization, take place upon the heating of a cold- worked metal. RROOVERY) FRectt Recovery, sometimes referred to as stress relief, involves rearrangement of some of the ‘more strenuous dislocations or imperfections with lite or no effect on the external form of the crystals or grains. Although the changes that take place during recovery are rather minor in respect to the crystal. they have a marked effect, on some properties. Elec trical properties and corrosion resistance are im: proved and residual stresses are reduced. AfRECHUA by ‘Time, Temperature, Cold Works Re: covery occurs completely for some metals at room temperature, For some others it occurs partially over a long period of time without increase of temp- erature. For most it is necessary to heat treat to a specific temperature that will depend upon the degree of recovery desired, The temperature chosen will, of course, be dependent upon the metal and to soine extent on the amount of cold work that has been performed previously. The objective of recovery is usually to regain electrical and chemical properties without sacrifice of mechanical properties. If the temperature is raised too high or maintained for too Jong a time, hardness and strength of the metal will, decrease appreciably, but high temperature treatment, is sometimes nevessary to remove residual stresses in forgings and steel weldments, RBCRYSPALLIZATION Furteer-Trestment for Maximum Duetility. though some of the major distortions are eliminatedby treatment for recovery, most of the distorted crystalline lattice remains as it was produced by cold work. The elastic limit for the material has been raised close to the ultimate strength, and further deformation will cause fracture failure. Recovery of ductility to permit further change of shape by defor- mation can be obiained only by elimination of the deformed grains, and this can be accomplished by recrystallization. By this heat-treating process, new, smaller, unstrained grains with fully recovered capac ity for plastic flow can be formed by solid-state change in the metal. It is important to note that in the absence of allotropic changes, which will be dis- cussed later, no grain-size changes by heating metal to any temperature below the melting point can be accomplished unless the strained condition of cold- worked metal is present. Recrystallization is the nucleation and growth of new, strain-ree crystals from the strained crystals of a cold-worked material. Recryvialiiaution Temperatures.” The phenomenon, occurs over a wide temperature range with the length of time required for complete recrystallization in- versely related to the temperature and to the degree of strain present. For practical purposes, recrystalli- zation temperatures, such as shown in Table 4-1, are temperatures which will permit complete recry- stallization in a time period of approximately 1 hour for metals that, have been fully hardened by previous cold work. TABLE 4-1 Recrystallization Temperatures for Some Common Motals and Alloys Materiel ‘Aluminum (pure Aluminum alloys... Copper (pure) Copper alloys tron (pure) ‘Low carbon steal. Magnesium (pure). Magnesium alloys Zine Tin Lead ‘The table shows that zinc, tin, and lead re- crystallize at temperatures below room temperature. ‘This means that these metals in the pure state cannot, at ordinary temperatures, maintain a work-hardened condition. The normal use of deformation processes ‘on these materials would be hot working rather than cold working since it would be performed above their recrystallization temperatures. Examination of the table also reveals that contamination of a pure metal with other elements makes it more difficult for re- crystallization to occur, and the temperatures must be increased for completion to occur in a reasonable Aength of time. ‘Theory of Recrystallization! It is believed that recrystallization takes place by the nucleation of new gains mainly about the high energy points of dislo- cation in a work-hardened grain. They then appear to grow until they fill the old grain space and eliminate the existing strain by realignment of the atoms into a new crystal lattice. Recrystallization can thus be a grain-refining process as well as a method for recovery ‘of ductility, if it is discontinued as soon as complete recrystallization has taken place. ‘The new grains formed during recrystallization are likely to take positions with preferred orientations, Directional properties caused by preferred orientation are objectionable for most manufacturing operations, ‘This tendency can be reduced and more random orientation obtained by the addition of small amounts of an alloying element or by recrystallizing before maximum work hardening has been per- formed. Seldom ARereninif. in a few ‘eases, reerystallization may be used as an end process to leave a product in its most ductile condition or with its best electrical and chemical properties, but ‘more often it is an in-process treatment for ductility improvement or for grain refinement. In many cold deformation processes, such as deep drawing, the ductility of the material may be reduced by ‘cold working to the point where fracture failure is immi nent, Ductility may be retumed to the material any number of times by repeated recrystallization be tween steps of the forming operation. In most cases the last forming operation will not be followed by reerystallization, in order that the higher hardness and strength of the cold-worked material may be re- tained in the product. Although heating for recovery is a stress-elieving process, recrystallization at a higher temperature is sometimes aso called stress relieving. The same proc: 98 may be relerrod to as process annealing, particular- ly when performed in conjunction with deformation processes, GRAIN GROWTH © If a metal is kept heated at or above its recry. stallization temperature after the new, unstrained grains have formed, the tendency is for some of the new grains to absorb others and grow to larger size. Large grains are more stable than smell grains because of the higher grain-to-boundary-area ratio, which is a lower energy state. If fine grain structure is desired after the recrystallization process, it is necessary to reduce the temperature quickly to provent sub. sequent grain growth. This is usually performed by some kind of quench. Grain-Size Controf During processing, small grain size is not always wanted because large grains usually exhibit greater ductility, better machinability, and require less pressure to be deformed. The’ finalproduct usually should be of relatively fine structure, though, in order that the material will exhibit its best properties. Grain size for materials that do not go through allotropie phase changes is controlled pri- marily during the solidification process for cast metals and by recrystallization for wrought (defor- mation worked) metals. Allotropic metal (existing in more than one crystalline form) grain size can be controlled by a more effective and satisfactory method discussed later in the chapter. AMERERDENING, Some metal alloys display a variable solid state solu- bility of one metal in another with change of tempera- ture. If the solubility increases with increase of tem- perature above room temperature and if return to the normal room temperature state can be prevented by sudden cooling, the alloy may be susceptible to age hardening. ‘Theoty of Age Hardening“ Exact explanation of this hardening phenomenon is not available with pre- sent knowledge, but from close study it has been theor- ized that the precipitant from a supersaturated solu- tion first appears as a transition lattice widely dis- persed and closely associated with the solid solution lattice. Close association causes lattice distortion with accompanying increase of hardness, much as the dis: tortion by cold working increases hardness. With suffi cient time, which decreases with higher temperature, the transition particles combine to form a larger, more widely spaced, and more stable equilibrium precipi tant, as in the annealed structure, For hardening pur- poses, the intermediate phase must be present, and when it disappears because of the complete formation, of the final phase, the material is considered to be over- aged with loss of the special properties present during the intermediate, or transition, stage. ‘The need for hardness and strength is often not, present at the time of metal solidification. Com- mercial practice handles age hardening, precipitation hardening, or solution hardening (all names used to describe the same process) as a treatment separate from solidification when there is a need for develop- ment of hardness properties or strength properties, or both. Solution Hewe‘Preatasen®, The first step is solution treatment (heating) to dissolve a maximum amount of equilibrium precipitant. in the solid solution and freeze it in place by sudden cooling to eliminate the necessary time at temperature for precipitation to reoccur. The solution temperature used should be low enough to prevent excessive grain growth but high enough to insure maximum diffusion o? the preci- pitant to saturate the a phase in a minimum amount: of time. The time required depends upon the metal alloy and may vary from @ few minutes to several hours of soaking at the increased temperature. After ‘Materials and Solid State Changes in Metals. 37 saturation of the @ phase, the metal is quenched to exeate the supersaturated solid solution at room temperature. High energy points in the crystal lattice set up by the nonequlibrium situation of supersatura- tion causes the alloy to be harder than its annealed condition. ‘Transition Stage by Precipitation. The full hard- ness, however, is developed during the second stage of treatment when the excess metallic component is partially precipitated from the solid solution, This step is usually referred to as aging and may be natural or artificial. If the surplus materiel goes into the initial transition stage of precipitation of its own accord at room temperature, full hardness will develop naturally with the passage of time. If an increase of temperature is necessary, as is true with many alloys, to release the unnaturally held metal, this heat-treating step is called artifical aging. Too high an aging temperature or too much time with this stage, or both, causes the precipitant to reach its final equilibrium state in which the hardness and strength properties are low and similar to those of the an- nealed alloy. Process Valuable for Aluminum Alloys. One of the greatest uses for precipitation hardening is for improvement of properties of some aluminum alloys. ‘The system can be used for either cast or wrought shapes and can be of particular value in some in- stances because of the time that is necessary for full hardening to develop. For example, it has been ‘common practice in the aircraft industry to solution treat aluminum rivets and hold them under refrigera- tion after their quench to retard precipitation. Before precipitation starts, they are relatively ductile and easy to form plastically. In this condition they can be headed to join riveted assemblies and develop their full strength by aging after being upset in place. WEAN, A few metals change lattice struc- ture upon heating and cooling to exist in different forms through various temperature ranges. Such metals are classed as allotropic. Allotropic changes are ‘very similar to the phase changes from liquid to solid, although they occur completely in the solid state with a slower reaction, In addition to a significant change of properties, heat is given up or absorbed as the metal phase change occurs in the solid state but to a much lower degree than in freezing or melting. With some metals special methods are necessary to detect heat changes that accompany the solid-state phase change. Iron combined with carbon and sometimes small amounts of other elements is by far the most used metal for manufacturing. Iron is an allotropic ma terial that changes upon heating to 912° C (1,674°F)38 Materials and Processes for NDT Technology from a body-centered cubic (BCC) lattice to a face- centered cubic (FCC) lattice. A second phase change occurs with further heating to 1,394" C (2641° F), where the Inttice structure retums to the body- centered cubic form. The reverse transformation occurs on cooling through the same temperatures. Iron in the temperature range up to 912" C is called alpha iron; from 912° C to 1,94° C, gamma iron; and above 1,394° C to 1,538° C (2,800° F), the melt- ing point, delta iron. Little attention is given to delta iron because the changes that occur in this range have little or no effect in commercial practice of treatment for properties. The changes that take place between alpha and gamma iron at 912° C, however, are ex- ‘zemely important. The most effective change is the difference of carbon solubility in the two phases, which serves as the basis for all heat-treat hardening and most grain-size control for steel HRAT OROUTREENT OF STERY Steel has been treated by heating and cooling methods to vary its properties ever since its discovery, bbut even today the exact mechanism by which these variations take place cannot be completely explained by fully accepted theories. Most of the treatments have been developed empirically. Various theoretical explanations have been used to describe the mechan- ism, but it has been only in recent years thet the theory has advanced to the point that it is a prime source of new development of commercial heat-treat- ing methods. MBB Uee-Other Conteel Methods. Change of pro- perties of steel can be accomplished by cold working, by precipitation hardening, and by allotropic changes. Cold working changes are important in most of the cold deformation processes and, in some cases, may be the only treatment received by the metal. Precipi- tation hardening is seldom used intentionally, except for stainless steels, although it may be arr‘edced6ittal occurrenée with some of the processing treatments. Causing allotropie changes by heat treating proce- dures is the most effective and most easily accom- plished method of varying mechanical properties of steel and therefore is the most frequently used way of obtaining the desired properties. Heat treating is often defined as intentional heating ‘and cooling for control of properties. Such a defini- tion is perfectly good, but it must be remembered that the effects of temperature changes are no less Important when they are caused by unintentional heat transfer during a process such as fusion welding or during a service use in high environmental tempera- tures such as in a fumace or gas turbine. Assessment of thermal treatment, whether inten tonal or not, is often amenable to nondestructive test- ing techniques that are capable of measuring, subtle changes in electrical conductivity. The heat treatment processes described in this chapter produce various physical property changes including electrical com Guetivity. Both eddy current and thermo-electric meth- ‘ds are capable of indicating changes in electrical con- ductivity and to some extent ean provide absolute measures of electrical conductivity. However, both methods only probe relatively small volumes of the test material essentially at an exposed surface. During heat treatment, exposed surfaces tend to heat and cool ata different rate from the interior. Thus, measire- ments of surface characteristics do not necessarily characterize the condition of the interior, but in many Practical cases can provide adequate information for process control purpose APPROXIMATE EQUILIBRIUM FO a Several heat-treating processes place the material in either a complete or an approximate equilibrium ‘energy condition. These processes include austenitiz- ing, annealing, normalizing, and spheroidizing. Except for the first, all are finalized at room temperature, but since austenitizing consists of diffusion of carbon into face-centered cubic iron that exists at a mini- mum temperature of 727° C (eutectoid composition only, all others higher), stability, or equilibrium, in this state can be maintained only’ at the higher temp- eratures. -Austonitization is therefore not a final process but only a step in one of several heat-treating Procedures. For these approximate equilibrium Processes, it is possible to predict the material be- havior from the equilibrium phase diagrams, EEE TION 4 When steel is heated to or above its critical temp- erature (transformation temperature range), the value ‘of which is dependent upon the alloy percentages, and held at temperature for some period of time, carbon unites in solid solution with iron in the gamma or face-centered cubic lattice form. Th this, Phase, as much as 2% carbon can dissolve at the eutectic temperature of 1,148°C at which the widest range of gamma composition exists, GPTNEOiittO. Tt is important that the austenitization temperatures not be exceeded mare than necessary to accomplish the work in a reason- able length of time because grain growth can occur rapidly as the temperature is increased. One of the important features of austenitization is grain refine- ment that occurs with the formation of the new face- centered cubic lattice. These new small grains are nucleated with the raising of the metal temperature through the austenite range and will remain small if the temperature is not raised too high or maintained too long. With lowering temperature and decompo-sition of austenite into the room temperature phase, the grain size changes little. Grain sizes are affected only by increasing temperature through this range and not by decreasing temperature, However, because tnetal grains must be of a certain eritial size before they can maintain themselves alone, practically al the sain refinement that is possible can be acquired by ‘one or two austenitization treatments, providing grain trowth is not allowed at the higher temperature. werner Objectives of Annealing, ‘The word anneal has been used before to describe heat-treating processes for softening and regaining ductility in connection with cold working of material. It has a similar mean- ing when used in connection with the heat treating of allotropic materials. The purpose of full annealing is to decrease hardness, increase ductility, and some- times improve machinability of high carbon steels that might otherwise be difficult to cut. The treat- ment is also used to relieve stresses, refine grain size, ‘and promote uniformity of structure throughout the material. ‘The purpose of normalizing is somewhat similar to that of annealing with the exceptions that the steel is not reduced to its softest condition and the peaslite is left rather fine instead of coarse, Pearlite is @ crystal: line structure with layers of soft, ductile ferrite (iron containing small amounts of dissolved carbon) and hard, brittle cementite firon carbide which is a mechan- ical mixture of iron with greater amounts of carbon). Refinement of grain size, relief of internal stresses, at improvement of structural uniformity together with recovery of some ductility provide high toughness gualities in normalized steel. The process is frequently used for improvement of machinability and for stress relief to reduce distortion that might occur with partial machining or aging. An attempt is made during nor- malizing to dissolve all the cementite to eliminate, as far as possible, the settling of hard, brittle iron carbide in the grain boundaires. ‘The desired decomposition products are small-grained, fine pearlite with a mini mum of free ferrite and free cementite. Minimum hardness and maximum ductility of steel can be produced by a process called spheroidizing, which causes the iron carbide to form in small spheres or nodules in a ferrite matrix. In order to start with tunall grains that spheroidize more readily, the process is usually performed on normalized steel. Several variations of processing are used, but all require the holding of the steel near the A, temperature (usually slightly below) for a number of hours to allow the Materials and Solid State Changes in Metals 39 iron carbide to form in its more stable and lower energy state of small, rounded globules. ‘The main need for the process is to improve the machinability quality of high carbon steel and to pre- ‘treat hardened steel to help produce greater structural uniformity after quenching. Because of the lengthy ‘treatment time and therefore rather high cost, spheroidizing is not performed nearly as much as annealing or normalizing, pa, Semertemwer — AWWNNNR Most of the heat treatment hardening processes for steel are based on the production of high percentages of martensite. The first step, therefore, is that used for most of the other heat treating processes — treatment to produce austenite, The second step involves cooling rapidly in an attempt to avoid peatl- ite transformation. The cooling rate is determined by the temperature and the ability of the quenching media, to carry heat away from the surface of the material be- ing quenched and by the conduction of heat through the material itself. Table 4-2 shows some of the com- monly used media and the method of application to remove heat, arranged in order of decreasing cooling ability, TABLE 4-2 Heat-Treating Quenching edie Wethod Brine T Blast 2. Water 2. Violent agitation 3. Light oil 3. Slow agitation 4, Heavy oll 4. sein 6. Air OTT retetinpeetctiootig: ih temperature gradients contribute to high stresses that ‘cause distortion and cracking, so the quench should be only as extreme as is necessary to produce the desired structure, Care must be exercised in quench- ing that heat is removed uniformly to minimize thermal stresses. For example, a long slender bar should be end-quenched, that is, inserted into the quenching medium vertically so that the entire sec- tion is subjected to temperature change at one timo. Ifa shape of this kind were to be quenched in a way that caused one side to drop in temperature before the other, change of dimensions would likely cause high stresses producing plastic flow and permanent distortion. racks created by either heating or quenching can be detected by various NDT surface examinations includ- ing eddy current, magnetic particle, and the penetrant methods. For parts subject to fatigure failure, such cracks can be serious,40, Materials and Processes for NDT Technology ae feeem Several special types of quench are conducted 10 minimize quenching stresses and decrease the ten- dency for distortion and cracking. One of these, is called marcempering and consists of quenching an austenitized steal in a salt bath at a temperature about ‘that needed for the start of martensite formation. The steel being quenched is held in this bath until it is of ‘uniform temperature but is removed before there is time for the formation of bainite to start. Completion of the cooling in air thon causes the same hard marten- site that would have formed with quenching from the high temperature, but the high thermal or “quench” stresses that are the primary source of cracks and warping will have been eliminated Heenan A similar proc- ess performed at a slightly higher tomperature is called austempering. In this case the steel is held at the bath temperature for a longer period, and the result of the isothermal treatment is the formation of hainite. The bainite structure is not as hard as the martensite that could be formed from the same com- position, but in addition to reducing the thermal shock to which the stecl would be subjected under normal hardening procedures, it is unnecessary to perform any further treatment to develop good impact resistance in the high hardness range. A third step usually required to condition a hard ened steel for service is tempering, or as it is some- times referred to, drawing. With the exception of austempered steel, which is frequently used in the aschardened condition, most steels are not serviceable “as quenched”, The drastic cooling to produce mar- tensite causes the steel to be very hard and to contain both macroscopic and microscopic intemal stresses with the result that the material has little ductility and extreme brittleness. Reduction of these faults is accomplished by reheating the steel to some point below the lower transformation temperature. The structural changes caused by tempering of hardened ‘steel are functions of both time and temperature, with ‘temperature being the most important. It should be emphasized that tempering is not a hardening process, Dut is, instead, the reverse. A tempered steel is one that has been hardened by heat treatment and then stress rolieved, softened, and provided with increased ductility by reheating in the tempering or drawing pro- cedure. In general, corrosion is the deterioration of metals by the chemical action of some ‘surrounding or contracting medium which may be Ii- quid, gas, oF some combination of the two. To some de- ‘reo, corrosion can influence all metals, but ite effect varies widely depending upon the combination of metal and corrosive agent. ‘The term ‘‘corrosion'’ is used to describe action that is normally considered to be detrimental, but the prin ciple is actually used for benefit in some cases. For ex- ample, acids and alkalies are used to corrode metal away in the manufacturing process of chemical milling. Also, aluminum alloys are frequently anodized to pro- duce an oxide coating that resists further oxidation and, in addition, may serve as an improved surface for paint adhesion Corrosion attacks metals by direct chomicel action, by electrolysis (electrochemical action), or commonly by a combination of the two. The subject is complex and many persons have devoted their lives to its study. ‘This discussion will summarize some of the known facts concerning the subject in order to develop some understanding of corrosion, its detection, and preven- tion, PIRROPOHNEORT ACTION: ‘Theoretically, all corrosion. phenomena ‘are electro- mechanical because a transfer of electrons takes place but the term direct chemical action is used to deseribe those reactions where coupled anodes and cathodes ex- isting in an electrolyto ate not identifiable. The chem cal milling mentioned above is direct chemical action. Another example is pickling of steel, a process in ‘which heated dilute sulpheric acid baths are used to dissolve surface scale withut leaving a residue and pro- ducing only minor chemical attack on the steel proper. Figure 4-6 illustrates another example of direct chemi- cal action. Figure 4.5 Diroot chemical attack of nitric acid In which @ ‘magnesium alloy product is Immersed. Direct, ‘chemical reaction is usually evident from bubbles formed by gas evolution. SMMMETIes. Table 4-3 shows a list of metals ar- ranged in order oftheir decreasing chemical activity in sea water. This is special arrangement of the electro- mechanical and the electromotive force series. ItTABLE 43 Galvanic series of some metals in sea water Anodic (Most Corrodibley) Wagnasium Aluminum ‘Aluminum—Cu Atoy Zinc feo St Tin Lead Nicke Brass—Cuzn Bronze—CuSn Copper Stainless Steel sliver Gols Platinum Cathodic (Least Corrodible) should be noted that most of the list is made up of pure metals and indicates their relative resistance to sea ‘water corrosion. If the metals are alloyed or if the cor- roding medium is different, the arrangement of such a list might change somewhat. In general, a metal high in the series will displace from solution a metal lower in the series. | cnemoereelintebiememmmaneesll waren The electrochemical type corrosion also involves chemical change but involves the flow of an electric current between two electrodes, an anode (posit where electrons leave and negative ions are discharged) and a cathode (negative, where electrons enter and neg- ative ions are formed). An electrical contact must exist in addition to electron flow through the electrolyte to complete the circuit. The system is analogous to a plating system in which the anode supplies the metal to be deposited. The anode eventually is depleted thereby. SmmsidictaOARCEMR. Although there are other factors that influence corrodibility, at least theoretically the metals high in a galvanic series, which are anodic to any metal below them, when connected electrically both by contact and through an electrolyte will dis- solve while the cathode is protected. This is the basic use of zine coatings on steel The zine is attacked and sacrificed in order to protect the steel. Protection will continue as Jong as exposed areas of steel do not grow large enough to develop their own galvanic cells to ‘cause corrosion, The larger the anode area, the better is the protection. ‘The same principle is used when magnesium rods are ‘hung in hot water heaters to lower corrosion of the tank, Large anodes of magnesium, aluminum, or zine Matariale and Solid State Changes in Metals 41 may be attached to the steel hull of a ship to pro protection below the waterline, as shown in Figure 4 Buried steel pipe also may be protected by attaching anodes as shown in Figure 4-7. 5 = PRZNCANOUES() _STEELMULLI=) PAINTED SURFACE Figure 46 [A ship's stee! null may be protected by attachment Of sacrificial anodic plates to the sides under the waterline EARTH'S SURFACE cue % v cle be Wa Figure 47 Proferential corrosion reaulting in protection for buried stee! pipe by electrical attachment of ‘anodle material to the steo! Except when sacrificial corrosion protection is planned, it is normally not good practice to design pro- ducts with contacting metals of radically different gel- vanie position if there is likelihood of exposure to any corrosive medium. See Figures 4-8 and 4-0 which illus- trate a possible lack of good design judgement. Figure 4-8 Electrochemical corrosion of an aluminum part that was assembled against a steel washer in an environment containing moisture. Being anodic to steel, the aluminum dissolved as shown.42 Materials and Processes for NDT Technology N ge 2* se capone MATERIAL Datormad Carton St Figure 4.9 EXPOSURE: Mula Aton Corrosion likely when dissimilar metals are in intimate contact GaprEEAE Cela. ‘Tho system described above causes the flow of electrical current that in turn causes and ac- colerates corrosion. A type of battery called a galvanic call can be made of electrodes of two different metals immersed in an electrolyte. A similar result (current flow) is produced when two similar metals are joined, or even a single metal, when contact is made with an. clectrolyte that is not chemically uniform. This type of call is known as a concentration cell and is particularly detrimental when the chemical variation of the electro- lyte is in its oxygen concentration. Figure 4-10 illus- ‘trates the results from this type reaction. ‘Some common examples of corrosion in materials, that have localized high stresses 2, Oxygen content—Oxygen particularly harmful in corrosion of iron. 8. Acidity—In general, the higher the acid cofitent, the higher the corrosion rate. anianion oF 4, Motion—Velocity of a flowing electrolyte may {__Bivorung oll, Lion content move corrosion products exposing new metal to at- saumon on gee [7S }_—_ tack. Movement of electrolyte also may prevent forma: ‘OXYGEN CONTENT tion of concentration cells, thus reducing corrosion. 5, Temperature—Increase usually accelerates corro- sion, Figure 4-10 6. Stray electrical currents—Localized currents Electrochemical corrosion can occur with contacting from leaks, grounds, or eddy currents usually acceler- ‘similar metals when the concentration of the a ‘eloctrolyte varies, Most likely under MEE Moisture is usually stagnant conditions. blamed for atmospheric corrosion and although mois- ture may be present, pure water has relatively small e fect. The combination of moisture with impurities, cceuenes especially salts of chlorine and sulphur, accelerates 1-ssyaltmospheric corrosion greatly. Metal or Metals of a Corrosion System. 1. Position. in. the electrochemical serios.—The higher tae rebtn ndgcy ter cori bn mult? TROOP CONNDIROR metal systems the farther apart, the greater the elec: gjqperni-Cerrenian. The most common type corre: ‘trochemical action, sion fe that appearing relatively uniformly over the en- 2. The presence of residual stresses suchas shown in tice surface of the exposed mmetal, The bluish green cal Figure 41 for of a copper roof or the dulling of polished aluminum and brass are examples of general corrosion. Some of 1. Concentration—High concentration usually in- this type corrosion is selflimiting because the pro- creases corrosion. ducts of early corrosion inhibit further corrosionMEENA Fitting is « localized corrosion by which pits that extend deep into the metal develop. This is a more serious carrasion than the stower general type be- cause the pits may decrease the material strength and also be the nuclei for fatigue failure, With some materi- als pitting rate may increase with time. Steal which normally rusts uniformly upon exposure to atmo- sphere may, with sufficient time, develop pits. Figure 412 illustrates pitting. Figure 412 Pit type corrosion can be observed in the cylindrical machined surface of this ‘aluminum casting inmeepnatinonstiemmenines A serious type of corre sion is created when the attack is against the grain boundaries. Following the grain boundaries from the woetal surface, a cracklike discontinuity develops. Such cracks can cause material failure under static loading by reduction of load supporting cross-section. in the case of dynamic loading, they are likely to be the boginning source of fatigue failure. Because those cracks are seldom visually apparent on the surface, NDT may be called upon for their detection. The sketches of Figure 4-13 illustrate the three main types ‘of corrosion attack, LUNIEORM DISCOLORATION or Abs oF House ‘GENERAL SoMenmes accom, BY DISCOLORATION. Crack uke otsconnNumiEs STOR Shah BOUNDARIES POLISHED, ETCHED, AND tagniied CROSS SECTION Figure 4-13 Principal types of corrosion aterials and Sod State Changes in Met Some variations of intercrystalline corrosion are known by the names of season cracking, stress corro- sion (see Figure 4-14), and fretting, all of which are cor- rosion systems in which corrosion is accelerated by the metal being under load at the same time corrosion is occurring. Season cracking ie associated with brass and some other copper-beering alloys and occurs most frequently when the maveral has undergone cold work- ings. Season cracking is much accelerated when the corrosive atmosphere contains ammonia. Fretting is corrosion-assisted wear resulting from small oscilla tory movements between mating surfaces under load. ‘Stress corrosion cracking is of major concern because of its effect on a fairly large number of common alloys of various metals used in chemically aggressive envir. conments. In high strength steels and martensitic stain less steels, stress corrosion cracking is usually inter- granular; in austenitic stainless steel, usually trans granular. Control of stress corrosion cracking necess: tates controlling the four equitial requirements for stress corrosion to occur: a susceptible alloy: an ag gressive, conrrosive environment: applied or residual stress; and time. Acoustic emission monitoring tech- niques have been used in-situ to detect and record the progression of cracking due to stress corrosion. Figure 4-14 ‘Stress.corrosion residue shows on the surtace of this ‘magnesium part which has been under constant ‘static load in the presence of a corroding atmosphere CORNOtreTEPROSECTION: ‘There is no simple answer to preventing serious problems from the attack of corrosion. There is no cure-all because of the variety of metals, possible envir- onments, and corrosive media. The general combat ‘methods include: selection of the most suitable metals, treatment of or controlling the presence of the corrod- ing media, coating the metal with a protective layer, and occasionally stress relieving parts containing high stress arcas,“44. Materials and Processes for NDT ‘Technology ‘The first is basically a design problem, The second is ‘usually a manufacturing or service problem such as troating coolant water used in a machine tool with a chemical corrosion inhibiter or decerating boiler feed water to remove oxygen. The third is the most. com- mon approach and includes: coating with anodic mate- rials to promote preferential corrosion, developing a coating to retard corrosion, and application of a coat- ing to exclude the corrosion medium, The coatings used are metals, chemical compounds, and organic ma- terials and plastics. SAEZ. Coating of metal with another metal can be accomplished by electroplating, dipping in molten metal, metal spraying, cladding by rolling thin layers over the base metal and by heating the pro- ‘duct in fine metallic powders. Comme CONPTAEs, Most coatings that consist of chemical compounds are made by treating the base metal to change the chemistry of its surface, Anodiz- ing of aluminum is the artificial formation of alumi ‘num oxide to a controlled depth on the surface of an aluminum alloy. Stoel can be given @ protective coat- ing of iron phosphate by soaking the product in hot solution of manganese phosphate. NemamntnlineSentings, Point, enamel, varnishes, ‘greases, plastics, and many other materials are used to coat objects for corrosion protection. Most of these materials are used to exclude the corrosive environ- ment but some contain chemical inhibitors to exert greater control. Some are for only temporary protec tion such as for a few days or weeks, but others may hhave a useful life of several years. Jim all cases of corrosion protection regardless of the type, suitable preparation and cleaning of the original metal surface is essential, Where control of coating thickness is important, several methods of NDT are available. Eddy current liftoff techniques are most readily applied, but depending upon the type of coat- ing and substrate, beta-backscatter, magnetic field and radioisotopic tagging procedures have been used effectively. SPORT MMET As apparent from the foregoing discussion of corrosion, its effects are almost always detrimental to the serviceability of criti- ‘eal components, assemblies, and structures. The NDT ‘specialist must understand the effects of the various types of corrosion in order to properly select and direct the nondestructive tests most effective in detecting and assessing the extent of corrosion. For corroded surfaces that are accessible, penetrant. magnotic particle, and eddy current tests are particu lary useful in detecting the effects of corrosion that re- sult in small surface cracks or pits. Very small corro- sion cracks have been detected and recorded by mague- tic rubber techniques. For corrosion on the inside of pipes, vossels, and assemblies, other toheniquos are ap- plied. Ultrasonic techniques are particularly effective in the detection and accurate measurement of overall thinning that results from corrosion. Radiography is ‘commonly applied to detect corrosion and corrosion thinning in interior and otherwise inaccessible regions of assemblies, insulated components, and the like. New tron radiography has been used to detect interior cor rosion by virtue of the corrosion products having large neutron cross-sections as well as actually imaging cor- rosion in exceptionally dense materials like lead and uranium. Acousic emission monitoring has been used to monitor the initiation and growth of stress corrosion and hydrogen embrittlement cracks, ‘Visual means are also important in the detection of corrosion. Both corrosion discontinuities and corrosion products leave telltale signs by virtue of visible ‘changes in texture, coloration, topography, and goo- metry. Some corrosion products fluoresce when ill minated by ultraviolet light. Further study of carro: sion sites and corrosion products by spectrographic analyses can reveal otherwise elusive evidence as to the cause of corrosion.Ferrous Metals 5 CHOOSING METALS AND ALLOYS In Chapter 4 metals were discussed primarily on the basis of their atomic configurations. While it is true that this basis gives a more precise definition in the chemist's or physicist’s terms, of greater practical interest in manufacturing are the metallic properties, of relatively high hardness and strength, ability to undergo considerable plastic flow, high density, dur- ability, rigidity, luster. A distinction is sometimes mace between the word metal, meaning a pure chemi- cal element, and the word alloy, meaning a combina. tion of materials, the predominant one of which is 1a metal. The term metal in this text will be taken to mean any metallic material, whether pure or alloyed Availability of Ores, Among all the possible rea- sons for the choice and use of a material, one of very prime importance is availability, Table 5-1 shows the composition of the earth's crust, Of the first twelve elements in occurrence, aluminum, iron, magnesium, and titanium are used as the base metals of alloy systems, For the other metals, although the total tonnage in the earth’s crust may be considerable, the potential use is much more restricted. Some of them, such as copper, are found in relatively pure deposits but frequently in remote locations, and the total use is dependent on relatively few of these rich deposits. Most other metals are recovered only in relatively small quantities, either as byproducts of the recovery of the more predominant metals or as products of low-yield ores after extensive mining and concen- tration in which many tons of material must be handled for each pound of metal recovered. The United States has only marginal deposits of anti mony, chromium, cobalt, manganese, and nickel and imports the major quantity of these metals. It is almost totally dependent on imports for its supply of‘46 Matorils and Processes for NDT Technology mercury, tungsten, and tin. ‘The location and the availability of these materials have a marked influence on both the risk and cost of choosing these materials for large-use applications. TABLE 5-1 Elements In the Earth's Crust Element Percent | Element Snyoer 46.71 | Magnesturn Sitican 27,69 | Thanium .. Aluminum ....-+, 807 | Hydrogen ron... 5.05 | Phosphorus Catcium sees 365 | Carbon + 275 | Others Rase Metals. Approximately seventy of the ele- ments may be classed as metals, and of these, about forty are of commercial importance. Historically, copper, lead, tin, and iron are metals of antiquity because they are either found free in nature or their ores are rolatively easy to reduce. These four metals together with aluminum, magnesium, zine, nickel, and titantum are presontly tho most important metale for use as base metals for structural alloy systems, Most other commercially important metals either are metals used primarily as alloying metals or noble metals, such as gold, silver, or platinum, that aze important only for special uses or because of their rarity. Material Choice Affected by Process. The method of manufacture will frequently affect the alloy type chosen even after the base metal has been chosen. Although nearly all metals are cast at some time dur- ing their manufacture, those that are cast to approx mate finished shape and finished without deforma- tion are specifically referred to as casting alloys When the metal is fabricated by deformation processes, an alloy designed to have good ductility is specified ‘and referred to as a wrought alloy. Some alloys can be either wrought or cast, most wrought alloys can be cast, but many casting alloys have insufficient ductility for even simple deformation processing. Final Choice Dependent on Many Factors, The choice of a material is usually a stepwise process. Sales requirements, raw material cosis, equipment availability, or specific product requirements will frequently narrow the choice betwoon the fields of metals and plastics. With the choice of either metals or plastics, some may be eliminated on the basis of properties, although a considerable number of plastics ‘or metal alloys will still satisfy the functional require- ments for the great majority of products. The life to bbe expected from the product may also eliminate some matarials from considerstion. Finally, however, the choice usually becomes one based on costs. From. the various materials that would produce a function- ally acceptable product with sufficient life and from the various processing methods that are available to a manufacturer, the best combination must be found, ‘Obviously, many combinations will be rather quickly eliminated, but of those remaining, costs of some may not be entirely predictable without actual oxper- fence in producing the product. Consequently, the first choice is not always the final choice, and for this reason, as well as for reasons of sales appeal and product redesign, materials and processes frequently are changed on a trial and error basis. Importance of Ferrous Materials. ‘The role that ferrous materials play in the economy is evident from annual production figures, Approximately 100 mil- lion tons of ferrous products are made each year in the United States. For all nonferrous metals, the total is about 10 million tons per year. Even though much of the steel tonnage goes into heavy products such as rails and structural steel shapes that require hittle secondary work, ferrous metals are still the predomi nant materials of manufacturing. The wide variety of ferrous products is based largely on the economy of producing them; an attempt will be made to discuss ferrous metals in the economic order of their pro- duction in the section to follow. Generally, as better properties are required, more costly processes are necessary. FERROUS RAW MATERIALS Ore Reduction. Both iron and steel have their startin the blast furnace. Although other methods for reduction have been proposed and will likely be developed, the tremendous investment in equipment and trained personnel that would be required for the replacement of present facilities almost insures that the blast furnace method will remain for some time. ‘This device is a tall, columnar structure into whieh is fed, through a top opening, a mixture of iron ore (oxides of iron — Fe;05, hematite, or Fe,0,, mag- netite), coke, and limestone. A blast of hot air is supplied through the mixture from near the bottom to provide oxygen for combustion of the coke, ‘Temperatures in the neighborhood of 3000" F are developed in the melting zone. The iron ore is re. duced by chemical reactions with carbon monoxide gases and by high temperature contact directly with the carbon in the coke as well as with other impurity elements in the mixture, Near the bottom of the fur- nace, the iron and the slag, which is made up of other metallic oxides combined with limestone, melt and accumulate in a woll; the lighter slag floats on top of the melted iron. The molten iron and stag are tapped off periodically through separate holes. The slag is disposed of, either as trash or for byproduct wae, and the iron is run into open molds to solidify as pies. unless it is to be further processed immediately In ganlarge installations, the molten iron is frequently trans- ported in large ladles to other equipment for carbon reduction in the manufacture of stecl. Pig Iron, The product of the blast furnace, whether higuid or solid, is called pig iron. The dis. tinction between the terms pig and pig iron should be noted. The tarm pig refers to a crude casting, conveni- cent for transportation, storage, and remelting of any ‘otal: the term pig iron refers to the composition of the metal tapped from the blast fumace, whether in liquid or solid state, Although this composition varies with ore, coke, blast fumace conditions, and other factors, the blast furnace is controllable only within broad limits, Pig izon as a natural result of the con- ditions within the fumace always contains 3% to 4% ‘of carbon and smaller amounts of silicon, sulfur, Phosphorus, manganese, and other elements. Pig Irom Requires Further Processing. In the solid state, pig ton is weak, is too hard to be machined, and has practically no ductility to permit deforms. tion work. It must therefore be treated to improve some of its properties by one of the methods shown in Figure 5-1, ‘The simplest of these treatments are those shown on the left of Figure 5-1; the treatments invalve remelting with only moderate control of composition, in particular with no attempt to remove the carbon Figure 5-1 General relationship of ferrous materials CAST IRONS ‘These simplest ferrous materials are produced by causing the molten metal to solidify into approximate final product form. The result is known as a casting. ‘The processes of making castings is discussed in Chap- ter 8, Some of the relationships between common cast. irons are shown in Table 5-2. Ferrous Motels 47 STEEL One of the largest and most influential manufactur. ing operations today is the steel industry, which makes some finished products but is primarily con- cerned with the making of raw material for further processing. The annual production of more than L00 million tons exceeds by far the total production of all other metals and plastics combined. Comparison of Steel with Cast Iron. Pound for pound, castings of cast iron are cheaper than those of steel, and for those products that can be made with suitable chapes and strengths as castings, the cost of the finished product often will be lower in this form. However, all cast irons, because of their high carbon content, are subject to the definite processing timita- tions of casting. Thin sections, good finishes, and dimensional contro! are obtained at reasonable cost. ‘only by deformation processing instead of casting. Deformation can be performed only on materials having relatively high ductility. For ferrous materials, this requires reduction of carbon from the cast iron range to the extont that a material with an entirely new set of properties is produced, All cast irons are essentially pig iron with, at most, only minot modifications of composition. The essen: tial component of pig iron in addition to the iron is 3% to 4% carbon, When this carbon content is re duced to less than 2%, the resulting new material is called steel. WROUGHT IRON Prior to the introduction of currently used ‘methods for making steel, a method of reducing the cearbon content of pig iron had been tsed since before 1600. The prodvet, although called wrought iron, was actually the first low carbon steel to.be manufactured ‘in quantity. Forly Furnace Limitations. In the early manu- facture of wrought iron, molten pig iron was sub- jected to oxidizing agents, normally air and iron ‘oxide, and the silicon and carbon content of the melt was reduced. The furnaces used were incapable of maintaining the iron at temperatures greater than about 1480” C (2700" F}, Reference to the iron- carbon equilibrium diagram will show that at this temperature pig iron would be well above the liquidus, ine. However, as the carbon content was reduced, at constant temperature, the iron began to solidity; consequently, to keep the reaction proceeding within the melt, it was necessary to stir or puddle the ‘material in the fumace. Wrought Iron Contains Slag. Because this material included slag, which floated on top as long as the metal was liquid, the slag was mixed with the purified iron. ‘The resulting product was withdrawn from the furnace as a pasty ball on the end of the stirring rod46 Materials and Proceases for NDT Technology TABLE 5.2 ‘Common cast irons Relative Type tron How Produced Characteristics Cost White Rapid coating Hard, tittle 1 Lowe + Si Unmachinable Melleabie Hest treated TS, 358 x 10" Pa 4 White iron (60—120 ks) Good malleability and ductility Dvctite Ladle addition TS.4-10X 10" Pa (60-160 ks) 3 ‘Similar to malleable Gry ‘Slow cooling TS.1.468.1 10° Pa (20~60 ksi} 2 High ¢ + si Good machinability Brittle Crittea Fast surface chill Hard surface (white iron) 3 Soft core (gray iron) and, while low in carbon and silicon, contained from 3% to 4% slag, mostly Si0;. These balls were then deformation processed by repeated rolling, cutting, stacking, and rerolling in the same direction. The resulting product consisted of relatively pure iron with many very fine slag stringers running in the direction of rolling. Although cheaper methods have been developed for reducing the carbon from pig iron without ineor- pomting the slog in the product, a demand for ‘wrought iron continues, based primarily on ite eputa- tion for corrosion and fatigue resistance, It is presently manufactured by pouring molten refined iron into separately manufactured slag with subse quent rolling. Properties of Wrought Iron. Wrought iron has a tensile strength of about 350 MPa (50,000 psi) and good ductility, although the material is quite aniso- tropic (properties vary with orientation or direction of testing) because of the slag stringers. Its principal use is for the manufacture of welded pipe. While wrought iron originally referted to this product or to its composition, the term has fre- quently been extended to refer to any worked low carbon steel product, particularly a product shaped or worked by hand, such as ormamental iron railings and grillwork. STEEL MAKING Early Steel, The oldest known method of making higher carbon steel consisted of reheating wrought iron and powdered charcoal together in the cemen- tation process. According to the iron-carbon equil- brium diagram, at 1148° C (2098" F) carbon is soluble in tron’ up to 2%. At this temperature the carbon slowly diffused into the solid material; the process required a total cycle time, including heating, of about 2 weeks. Much of the slag in the wrough? fron migrated to the surface and formed surface Dilisters, which resulted in the term blister steel. Even after this lengthy treatment, the carbon was not uniformly dispersed throughout the material, and multiple cutting and rerolling procedures were ro quired to produce a high quality product. Crucible Steel. Further reduction of the slag, greater uniformity of the carbon, and closer control were later achieved by a secondary operation known as the crucible process. Bars made by the cementation process were remelted in a clay or graphite crucible in which the slag floated to the surface. This crucible process produced steel of very high quality, and ‘modifications of the method are still used today, but, it was made possible only by furnace developments that permitted higher temperatures to be achieved than were needed in the manufacture of wrought iron. Open-Hearth Steel. Both the modern open-hearth furnace and the Bessemer converter were developed in the 1850s. These two developments greatly increased the speed with which pig iron could be refined. The modem era of industry can be tied to these develop- ments that led to the production of large quantities of high quality, low-cost steelFigure 52 Cross-section of open-nearth furnace Figure 5-2 shows the construction of an open- hearth fumace as was used for the majority of steel produced until recently in the United States, Various proportions of pig iron (either solid or molten), steel serap, limestone for flux, and iron ore are charged on the hearth of the fumace, The principal reducing ‘action takes place between the iron ore and the carbon of the pig iron, the final carbon content of the steel being controllable by the proper proportions of the charged materials. The principal difference be- tween this furnace and that used previously in the manufacture of wrought iron lies in the preheating of the entering combustion air. In the open-hearth fur- nace for steel making, the air enters through a brick checkerwork that has been previously heated by the exhausting flue gases, Two similar checkerworks are used, one for the exhaust side and one for the enter- ing. air side of the furnace. After a relatively short period of operation in this manner, the airflow through the checkerworks is reversed. Preheating of the air permits higher temperatures to be developed in the fumace, and the bath of metal may be kept molten as the carbon content is reduced. Bessemer Steel. ‘The Bessemer converter is shown in Figure 5-3. The charge consists of molten pig iron. Steel scrap may be added to help control the temp- erature. After charging in the horizontal position, the air blast is turned on through the tuyeres and the Converter tumed upright so that the ait bubbles through the melt, oxidizing and buming out fist sili- con, then carbon. The process can be used to reduce the ‘carbon content to about 0.05%, Althotigh less expensive to operate than the basic-lined open-hearth furnace, the inability of the acid-lined Bessemer converter to reduce the phosphorus content of the retal has restricted its use to the production of only about 5% of the steel made in the United States. Some steel is produced by initial refining in the Bessemer converter followed by further refining in the open-hearth furnace. Ferrous Metals 49 Figure 53, Bessemer converter Electric Fumace Steel. Blectrie furnace stoel is produced in a variation of the older crucible process with the fumace heated by electric arc or induction. ‘The atmosphere can be well controlled in the electric fumace, and careful control of composition can be maintained. Steel of the highest quality is producod by this method. Basic Oxygen Steel. A steel making srocess known as the basic oxygen process was developed in ‘Switzerland and Austria after World War I and first ‘used in 1962. By 1957 the method was producing 1% of the world production. In 1966 the growth of use was to 26% and currently more than 50% of the world’s steel is made by the basic oxygen process. ‘The Basic Oxygen Process. There are a number of variations in the equipment and methods for making basic oxygen steel. Fundamentally they all operate much as follows: ‘a, Serap as great as 30% of the heat is charged into the refining vessel, as shown schematically in Figure 5-4 b, Molten pig iron is charged on top of the scrap. ¢. The lance is positioned, and a high velocity jet of oxygen is blown on top of the molten mix. ture for about 20 minutes. During this period, lime and various fluxes are added as aids for control of the final composition. 44, The metal is then sampled, and, if it meets speci- fications, poured through the tap hole into a ladle by tilting the vessel. e. Finally, the vessel is inverted to empty the slag. and then is ready for reuse. With careful use, the vessel lining may last for a8 many as 400 heats, ‘The total time for producing a heat by this method is 30 to 45 minutes. This compares very favorably with the 4 to 6 hours necessary for the open-hearth methods using oxygen. Basic Oxygen Process Provides a Number of Advan- tages. Steal made by this method can start from any grade of pig iron, The finish quality is similar to that50 Matarils and Processes for NDT Technology Zo % a ar ee LLLLLZEZDLZLLTZEL SreeL'Pouns rough Tar nove PROM Figure 4 Basle oxygan furnace ‘made in open-hearth furnaces. Scrap is usable in large ‘quantities so that the process becomes the cheapest current method for remelting and reusing scrap. "The largest size unit presently available is slightly greater than 300 tons. A 300-ton unit ean produce 3 million tons of steel per year. Basic Oxygen Process Limited by Huge Investment Needs, The growth of the basic oxygen process has deen extremely fast as industrial processes go but would probably have been even faster except for the large investments required. The immense quantities of oxygen and its use demand much special equipment. In such a conversion to a facility including a rolling mill, one steel manufacturer invested over $600 mil- lion. Practically All Steel Made Today by Use of Oxy- gen. The development of oxygen-making facilities and the reduction of cost of the gas has changed nearly all steel making. Even when the complete basic ‘oxygen process is not used, oxygen is used to speed steel making. Both open-hearth and Bessemer con- verters are likely to be supplied with oxygen to speed ‘combustion and refining. An open-hearth furnace fit- ted with oxygen lances can approximately double production with less than one-half the fuel of earlier methods, without use of pure oxygen. The making of Bessemer steel is speeded by use of oxygen combined with air but also is improved in composition, mainly by reduction of nitrogen impurities left in the steel. Little Bessemer steel is made in the United States, however. PLAIN CARBON STEEL Any steel-making process is capable of producing a product that has 0.05% or less carbon. With this small amount of carbon, the properties approach those of pure iron with maximum ductility and minimum strength. Maximum ductility is desirable from th: standpoint of ease in deformation processing and service use. Minimum strength is desirable for defor- mation processing. However, higher strengths than that obtainable with this low carbon are desirable from the standpoint of product design. The most practical means of increasing the strength is by the addition or retention of some carbon. However, it, should be fully understood that any increase of strength over that of pure iron can be obtained only at the expense of some loss of ductility, and the final choice is always a compromise of some degree. Figure 5-5 shows typical ferrous material applications in relation to carbon content. Because of the difficulty of composition control or the additional operatiori of increasing carbon content, the cost of higher carbon, higher strength steel is greater than that of low carbon. Plain Carbon Steels Most Used. Because of their low cost, the majority of steels used are plain carbon steels. These consist of iron combined with carbon HEAT TREATED ati ‘ NOT HEAT TREATED ay tow - 4 * aL.eane 2 | roowsrieon ee & WHITE IRON to01 stet 2 sean sre. zg | z ren ste, nacre re Forcing sree | ee canauriting STEEL abana ee Se —A-— ol WROUGHT RON Figure 55 Ferrous materials concentrated in three ranges classed as low carbon, medium carbon, and high carbon, With the exception of manganese used to control sulphur, other elements are present only in small enough quantities to be considered as impurities, though in some cases they ‘may have tninor effect on properties of the material Low Carbon. Steels with approximately 6 to 25 points of carbon (0.06% to 0.25%) are rated as low earbon steels and are rarely hardened by heat treat- ment because the low carbon content permits so little formation of hard martensite that the process is rela: tively ineffective. Enormous tonnages of these low