Plastic pipe extrusion process improvement

Master`s Thesis in Automotive Engineering

FLORIAN MORICE


Department oI Applied Mechanics
Division of Jehicle Engineering and Autonomous Svstems
CHALMERS UNIVERSITY OF TECHNOLOGY
Gteborg, Sweden 2011
Master`s Thesis 2011:55










MASTER`S THESIS 2011:55

Plastic pipe extrusion process improvement

Master`s Thesis in Master`s Thesis in Automotive Engineering
FLORIAN MORICE












Department oI Applied Mechanics
Division of Jehicle Engineering and Autonomous Svstems
CHALMERS UNIVERSITY OF TECHNOLOGY
Gteborg, Sweden1


Plastic pipe extrusion process improvement

Master`s Thesis in Automotive Engineering
FLORIAN MORICE

FLORIAN MORICE1


Master`s Thesis 2011:55
ISSN 1652-8557
Department oI Applied Mechanics
Division oI Vehicle Engineering and Autonomous Systems
Chalmers University oI Technology
SE-412 96 Gteborg
Sweden
Telephone: 46 (0)31-772 1000












Cover:
leIt side: Iive-layer extrusion head, Rivoli, Italy
right side, top: PA6/EVOH/ADH/PA12 multilayer structure
right side, bottom: Ilaming treatment unit

Department oI Applied Mechanics
Gteborg, Sweden1

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 I
Plastic pipe extrusion process improvement

Master`s Thesis in Automotive Engineering
FLORIAN MORICE
Department oI Applied Mechanics
Division oI Vehicle Engineering and Autonomous Systems
Chalmers University oI Technology

Abstract

This thesis is the introductive part oI a study on extrusion process improvement where extruded
products are multilayer smooth thermoplastic pipes. The study Iocuses specially on PA9T/PA12 and
PA6/EVOH/ADH/PA12 structures used Ior Iuel transIer applications.

The Iirst reason Ior working on process improvement is to reduce production cost which depends on
process reliability and line speed in the case oI extrusion. A long-term study Iocussed on pipe
dimensional properties will be carried out in 2012 to improve process reliability. The Iirst purpose oI
this thesis is to prepare this study. Another reason is when some essential characteristics oI the product
do not meet customer speciIication requirements. For plastic pipes, elongation at break is an important
characteristic. On new projects in development, minimum value required Ior pipe elongation at break
has been increased and is not reached with current process conIiguration. The purpose oI this thesis is
also to improve this characteristic.

Polymer main characteristics in relation with pipe properties and extrusion process are presented as a
prerequisite oI this study. Then, an analysis oI the mechanisms occurring Ior each operation oI the
process allows identiIying the main Iactors that inIluence pipe essential characteristics then identiIying
extrusion line parameters which allow controlling these Iactors. Impact oI these parameters on pipe
elongation at break is studied later in this thesis.

BeIorehand, pipe monitored essential characteristics and their respective controls are identiIied. A Iirst
study is carried out to improve current dimensional controls in anticipation oI the study on process
reliability improvement. Thus a concept oI cutting machine and a program to automate measures is
presented but their development is out oI the scope oI this thesis. A second study is carried out to
improve tensile test reliability beIore working on pipe elongation at break improvement. New test
conIiguration allows reducing signiIicantly the quantity oI test results rejected.

Finally a design oI experiment is carried out on PA9T/PA12 and PA6/EVOH/ADH/PA12 structures to
improve pipe elongation at break. Target value oI elongation at break is 200. Best results are obtained
Ior extrusion line speeds lower than current ones which would impact process cost. Moreover, even
with optimized parameters, some pipes are statistically below 200 oI elongation at break that is not
acceptable. A solution is been to equip extrusion line with a Ilaming treatment that allows achieving
target value oI 200 and in the meantime, improving process perIormance by increasing line speed.
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 II

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 III
Contents

ABSTRACT............................................................................................................................................ I
CONTENTS......................................................................................................................................... III
FIGURES ............................................................................................................................................... V
TABLES ............................................................................................................................................... VI
PREFACE........................................................................................................................................... VII
1. INTRODUCTION......................................................................................................................... 1
2. POLYMER CHARACTERISTICS ............................................................................................ 3
2.1. CHEMICAL COMPOSITION ....................................................................................................... 3
2.2. WATER ABSORPTION.............................................................................................................. 3
2.3. THERMAL CONDUCTIVITY ...................................................................................................... 4
2.4. THERMAL STABILITY.............................................................................................................. 4
2.5. GLASS TRANSITION TEMPERATURE AND MELTING POINT........................................................ 4
2.6. VISCOUS PROPERTIES AT MELTED STATE................................................................................ 5
2.6.1. Sensibilitv to temperature................................................................................................. 5
2.6.2. Sensibilitv to shear rate .................................................................................................... 5
2.6.3. Jiscous heat generation.................................................................................................... 7
2.7. SPECIFIC VOLUME VARIATION BETWEEN MELTED AND SOLID STATES .................................... 7
2.8. MECHANICAL PROPERTIES AT SOLID STATE............................................................................ 8
2.9. CONCLUSION.......................................................................................................................... 8
2.9.1. Flow properties at melted state ........................................................................................ 9
2.9.2. Specific volume variation between melted and solid states .............................................. 9
2.9.3. Mechanical properties at solid state................................................................................. 9
3. EXTRUSION MECHANISMS.................................................................................................. 11
3.1. EXTRUSION .......................................................................................................................... 12
3.1.1. Solids conveving ............................................................................................................. 13
3.1.2. Melting............................................................................................................................ 14
3.1.3. Melt conveving................................................................................................................ 16
3.1.4. Mixing............................................................................................................................. 17
3.2. FORMING.............................................................................................................................. 19
3.3. SIZING AND COOLING ........................................................................................................... 20
3.3.1. Heterogeneous cooling rate............................................................................................ 22
3.3.2. Draw down and violent cooling rate .............................................................................. 22
3.3.3. Friction........................................................................................................................... 25
3.4. CONCLUSION........................................................................................................................ 26
3.4.1. Extrusion factors............................................................................................................. 26
3.4.2. Si:ing factors .................................................................................................................. 27
4. PIPE ESSENTIAL CHARACTERISTICS MONITORING.................................................. 29
4.1. DIMENSIONAL CONTROLS..................................................................................................... 29
4.1.1. Non destructive controls ................................................................................................. 30
4.1.2. Destructive controls........................................................................................................ 33
4.2. STRUCTURAL CONTROLS...................................................................................................... 43
4.2.1. Burst test ......................................................................................................................... 43
4.2.2. Cold impact test .............................................................................................................. 44
4.2.3. Tensile test ...................................................................................................................... 44
4.3. CONCLUSION........................................................................................................................ 52
4.3.1. Dimensional controls...................................................................................................... 53
4.3.2. Structural controls.......................................................................................................... 53
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 IV
5. EXTRUSION PERFORMANCE IMPROVEMENT: RESULTS AND DISCUSSION....... 55
5.1. CURRENT PROCESS OPTIMIZATION ....................................................................................... 55
5.2. FLAMING TREATMENT ADDITION.......................................................................................... 59
6. CONCLUSION ........................................................................................................................... 63
6.1. PIPE DIMENSIONAL PROPERTIES: IMPROVEMENT OF PROCESS RELIABILITY .......................... 63
6.2. PIPE MECHANICAL PROPERTIES: IMPROVEMENT OF PIPE ELONGATION AT BREAK................. 64
APPENDIX 1: PLASTIC PIPE PRODUCT PRESENTATION...................................................... 67
HUTCHINSON PRODUCTS..................................................................................................................... 67
MULTILAYER STRUCTURES................................................................................................................. 68
APPENDIX 2: PRODUCT QUALITY MANAGEMENT................................................................ 69
Characterisation ........................................................................................................................... 69
Design........................................................................................................................................... 69
Qualitv control .............................................................................................................................. 69
APPENDIX 3: CUTTING MACHINE BILL OF MATERIALS..................................................... 71
REFERENCES..................................................................................................................................... 73

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 V
Figures

Figure 1 Example oI multilayer structures ........................................................................................................................................1
Figure 2 Extrusion process cost model ..............................................................................................................................................1
Figure 3 Six sigma method applied to the thesis ...............................................................................................................................2
Figure 4 Chemical Iormula oI polyamide 12.....................................................................................................................................3
Figure 5 Chemical Iormula oI amide group.......................................................................................................................................3
Figure 6 Water absorption according to ambient air relative moisture oI diIIerent polyamides |CARI05|.....................................3
Figure 7 Power law curve oI a polymer |RAUW10|.........................................................................................................................5
Figure 8 Fluid shearing between two plates ......................................................................................................................................6
Figure 9 EIIect oI shear rate on a rectangular Iluid element .............................................................................................................6
Figure 10 Semi-crystalline material shrinkage during cooling process ............................................................................................7
Figure 11 Molecular structure oI semi-crystalline materials.............................................................................................................8
Figure 12 Five-layer extrusion line, Rivoli, Italy ............................................................................................................................11
Figure 13 Monolayer extrusion line |CARI05| ...............................................................................................................................11
Figure 14 Five-layer extrusion head, Rivoli, Italy...........................................................................................................................12
Figure 15 InIluence oI melt pressure in the extruder on screw motor load.....................................................................................12
Figure 16 Screw geometry and Iunctional zones |MIAN09| ..........................................................................................................12
Figure 17 Material movement along screw channel........................................................................................................................13
Figure 18 Melting mechanism in single screw extruders ................................................................................................................14
Figure 19 EIIect oI barrel temperature on viscous and conductive heating....................................................................................14
Figure 20 Hutchinson extruder screw..............................................................................................................................................15
Figure 21 Melting mechanism in Hutchinson extruder ...................................................................................................................15
Figure 22 Plastic melt element path in screw channel.....................................................................................................................16
Figure 23 Velocity and shear rate proIiles in screw channel...........................................................................................................17
Figure 24 - Plastic melt element residence time in screw channel.....................................................................................................18
Figure 25 Total shear strain in screw channel .................................................................................................................................18
Figure 26 Extruded pipe Iorming tools............................................................................................................................................19
Figure 27 Die Ilow channel zones ...................................................................................................................................................19
Figure 28 InIluence oI Iorming tool dimensions on melt pressure inside the extruder ..................................................................20
Figure 29 Hutchinson sizing system................................................................................................................................................21
Figure 30 Hutchinson sizing sleeve .................................................................................................................................................21
Figure 31 Residual stress generation mechanism............................................................................................................................22
Figure 32 Schematic oI sizing operation .........................................................................................................................................22
Figure 33 InIluence oI Iriction Iorces inside sizing sleeve on draw down distribution..................................................................23
Figure 34 InIluence oI sizing operation on pipe molecular structure..............................................................................................24
Figure 35 EIIect oI thermo-mechanical load on pipe elongation....................................................................................................24
Figure 36 InIluence oI vacuum level on Iriction |CARI05|............................................................................................................25
Figure 37 PA9T/PA12 10x1 structure drawing................................................................................................................................29
Figure 38 Monitoring instruments on extrusion line.......................................................................................................................30
Figure 39 Schematic oI ultrasonic system working principle |ZUMB07|......................................................................................30
Figure 40 Schematic oI laser system working principle |ZUMB07| ..............................................................................................31
Figure 41 Example oI electronic signal sent by the data acquisition system Ior PA9T/PA12 structure ........................................31
Figure 42 Calculation oI internal diameter Irom external diameter and thickness values..............................................................32
Figure 43 Example oI dimensional analysis ....................................................................................................................................33
Figure 44 Basic outline oI the problem ...........................................................................................................................................34
Figure 45 SpeciIic cutting machines developed in the past ............................................................................................................34
Figure 46 Interaction diagram..........................................................................................................................................................35
Figure 47 Function diagram.............................................................................................................................................................39
Figure 48 Pipe cutting Iorce.............................................................................................................................................................40
Figure 49 InIluence oI die thickness on pipe guiding......................................................................................................................41
Figure 50 Cutting machine concept description ..............................................................................................................................42
Figure 51 Cutting machine working principle.................................................................................................................................42
Figure 52 Example oI measurement with a Iirst version oI the program........................................................................................43
Figure 53 Hutchinson burst test bench ............................................................................................................................................43
Figure 54 Hutchinson impact test apparatus....................................................................................................................................44
Figure 55 Tensile curve....................................................................................................................................................................44
Figure 56 Hutchinson dynamometer................................................................................................................................................44
Figure 57 Basic tensile test conIiguration........................................................................................................................................45
Figure 58 SelI-closure mechanical gripping system........................................................................................................................45
Figure 59 Tensile test result validity................................................................................................................................................46
Figure 60 Pipe radial stress distribution in gripping zone...............................................................................................................46
Figure 61 Jaw conIiguration studied................................................................................................................................................46
Figure 62 Pipe stress distribution in gripping zone .........................................................................................................................47
Figure 63 Pneumatic gripping system .............................................................................................................................................47
Figure 64 Pipe radial stress distribution in gripping zone...............................................................................................................48
Figure 65 Gripping zone model .......................................................................................................................................................48
Figure 66 Tip edge rounding radius variation .................................................................................................................................49
Figure 67 Pin length variation..........................................................................................................................................................49
Figure 68 Variation oI Von Mises stress in Iunction oI FEM analysis variables ...........................................................................49
Figure 69 Stress distribution in gripping zone according to tip edge rounding radius ...................................................................50
Figure 70 Tensile test tool design ....................................................................................................................................................50
Figure 71 Final test conIiguration....................................................................................................................................................51
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 VI
Figure 72 Test result interpretation (break elongation at break) .............................................................................................52
Figure 73 InIluence oI vacuum level on water Ilow........................................................................................................................56
Figure 74 InIluence oI experiment parameters on pipe elongation at break...................................................................................56
Figure 75 InIluence oI experiment parameters on vacuum level ....................................................................................................57
Figure 76 InIluence oI sizing sleeve internal diameter on vacuum level ........................................................................................57
Figure 77 Results oI the experiment ................................................................................................................................................58
Figure 78 Schematic oI Ilaming treatment principle .......................................................................................................................59
Figure 79 Flaming unit layout on extrusion line, Rivoly, Italy.......................................................................................................60
Figure 80 Pipe surIace temperature along extrusion line ................................................................................................................61
Figure 81 InIluence oI pipe surIace temperature aIter Ilaming treatment on pipe elongation at break..........................................61
Figure 82 Experiment results ...........................................................................................................................................................62
Figure 83 Example oI Hutchinson products ....................................................................................................................................67
Figure 84 Example oI pipe conIigurations ......................................................................................................................................67
Figure 85 - Product quality management procedure...........................................................................................................................70
































Tables

Table 1 Glass transition temperature and melting point oI diIIerent polymers ................................................................................4
Table 2 Comparison between viscosities oI diIIerent Iluids (order oI magnitude)...........................................................................5
Table 3 InIluence oI material characteristics on optimal screw design ..........................................................................................26
Table 4 EIIect oI increasing extrusion line parameters on process Iactors.....................................................................................26
Table 5 Monitored parameters on extrusion line.............................................................................................................................27
Table 6 EIIect oI increasing process Iactors on extruded pipe characteristics................................................................................27
Table 7 EIIect oI increasing extrusion line parameters on process Iactors.....................................................................................28
Table 8 Function description.............................................................................................................................................................35
Table 9 Test results with mechanical gripping system....................................................................................................................47
Table 10 Model parameters..............................................................................................................................................................48
Table 11 Test parameter combination..............................................................................................................................................55
Table 12 Air / gas mass ratio ...........................................................................................................................................................59
Table 13 Test parameter combinations............................................................................................................................................60
Table 14 EIIect oI increasing extrusion line parameters on process Iactors...................................................................................63
Table 15 EIIect oI increasing extrusion line parameters on process Iactors...................................................................................64
Table 16 Structure layer Iunctions...................................................................................................................................................68

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 VII
Preface

This thesis is the introductive part oI a study on extrusion process improvement where extruded
products are multilayer smooth thermoplastic pipes. This study was perIormed Irom September 2010 to
February 2011 in collaboration with Hutchinson company which core business is polymer processing.
Department involved belongs to the division 'low pressure Iluid transIer system specialized in plastic
pipe production Ior automotive sector. Technical centre is located at Rivoli, Italy whereas principal
production plant is located at Zywiec, Poland. Tests required Ior this study have been carried out in
both sites. The thesis has been supervised by Dr. Katia Rossi, responsible Ior structure development
and Eng. Luciano Bertero, technical director.

First, I would like to thank Mr Ronconi, Vice President oI Hutchinson Company, that gave me the
possibility to carry out this thesis. I would like to thank also Dr. Katia Rossi, Eng. Luciano Bertero and
all the team, especially Daria Cereghino, Romain Jeanson, Salvatore Rizzuto, Luca Barberis and
Claudio Ramaro Ior their help during this study. Lastly, I would like to thank my parents and my
Iriends Ior their support during this diIIicult period.

Rivoli, September 2011
Florian MORICE
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 VIII



CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 1
1. Introduction

This thesis is the introductive part oI a study on extrusion process improvement where extruded products are
multilayer smooth thermoplastic pipes. The study Iocuses specially on PA9T/PA12 and PA6/EVOH/ADH/PA12
structures used Ior Iuel transIer applications. These structures are illustrated Figure 1 and presented more in
details in Appendix 1.



Layer Material Thickness
Inner PA9T 0,2 mm
2 PA12 0,8 mm


Layer Material Thickness
Inner PA6 0,45 mm
2 EVOH 0,15 mm
3 Bonding 0,1 mm
4 PA12 0,3 mm

a. PA9T/PA12 structure

b. PA6/EVOH/ADH/PA12 structure

Figure 1 Example oI multilayer structures

The Iirst purpose oI working on process improvement is to reduce production cost. Extrusion cost depends on
process reliability (scrap mass) and line speed (production time) as it can be deduced Irom the model Figure 2.
Set-up cost has been voluntarily excluded Irom this model as set-up cost reduction is out oI the scope oI this
study.


Figure 2 Extrusion process cost model

A long-term study Iocussed on pipe dimensional properties will be carried out in 2012 to improve extrusion
process reliability. The Iirst purpose oI this thesis is to prepare this study.
Granules
Input
Ouput 1
Ouput 2
Pipe Extrusion
Scrap
Material (cost / mass)
Operation (cost / time)
Where:

Extrusion cost
machine operator extrusion
t t t cos cos cos + =
Production cost
extrusion scrap pipe production
t t t t cos cos cos cos + + =
Process eIIiciency
production
pipe
t
t
cos
cos
=

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 2
Another reason Ior working on process improvement is when some essential characteristics oI the product do not
meet customer speciIication requirements. For plastic pipes, elongation at break is an important characteristic.
On new projects in development, minimum value required Ior pipe elongation at break has been increased and is
not reached with current process conIiguration. The purpose oI this thesis is also to improve this characteristic.

Thesis structure has been built according to six-sigma method as presented Figure 3. This method was created
and implemented by Japanese people many years ago. Nowadays it is still considered as a reIerence Ior people
working on continuous improvement.


Figure 3 Six sigma method applied to the thesis

First chapter oI this report presents polymer main characteristics in relation with pipe properties and extrusion
process. Second chapter describes the mechanisms occurring Ior each operation oI the process and identiIies the
main Iactors that inIluence pipe essential characteristics then extrusion line parameters which allow controlling
these Iactors. Third chapter identiIies pipe monitored characteristics and presents control method and instrument
Ior each characteristic monitored. This chapter includes also a study to improve current dimensional controls in
anticipation oI the study on process reliability improvement and a study to improve tensile test reliability beIore
working on pipe elongation at break improvement. Lastly, Iourth chapter presents and discusses the results
obtained Irom the designs oI experiment carried out to improve pipe elongation at break. Explanations provided
in this report try to reIer to basic physical concepts to be easily understandable.
Defining
Measuring
Analyzing
Improving
Controlling
1. Polymer characteristics:
Identifying polymer main characteristics in relation with pipe properties and extrusion

2. Extrusion mechanisms:
Understanding the mechanisms occurring for each operation of the process
Identifying the main factors that influence pipe essential characteristics
Identifying extrusion line parameters which allow controlling previous factors


3. Extrusion performance monitoring:
Identifying pipe monitored essential characteristics
Identifying control method and instrument for each characteristic monitored
Improving current controls: tensile test, dimensional controls





4. Extrusion performance improvement: results and discussion:
Carrying out two designs of experiment to improve pipe elongation at break:
optimizing current process
adding a flaming treatment
Analyzing the results




Implementing a flaming treatment on extrusion line: currently in progress
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 3
2. Polymer characteristics

BeIore working on plastic pipe extrusion process and as a prerequisite oI this study, it is necessary identiIying
polymer main characteristics in relation with pipe properties and extrusion process. They are presented in this
chapter.

2.1. Chemical composition

Plastic pipes Ior automotive application are essentially composed oI semi-crystalline polyamides. For instance,
PA9T/PA12 and PA6/EVOH/ADH/PA12 structures contain both a layer oI polyamide 12 that is the most widely
used polyamide. Its chemical Iormula is presented Figure 4.


Figure 4 Chemical Iormula oI polyamide 12

A particularity oI polyamides is the presence oI hydrogen bonds between macromolecular chains at amide group
location. As presented Figure 5, these electrostatic interactions occur between:
N-H group: highly electropositive
CO group: highly electronegative


Figure 5 Chemical Iormula oI amide group

These bonds are very energetic because dipolar moments are high. It gives to polyamides better characteristics in
terms oI mechanical and chemical resistance than that oI other polymers like polyethylene or polypropylene.

2.2. Water absorption

Polyamides are Iormed by poly-condensation oI an amino-acid group. Poly-condensation is a balance reaction
between polymerisation that releases water molecules and hydrolysis by incorporation oI water molecules
between macromolecular chains: the shorter the chain, the higher the sensibility to hydrolysis. For instance,
quantity oI water absorbs by polyamide 12 in humid environment is lower than polyamide 6 whose carbon chain
is shorter. This property is important because hydrolysis reaction damages partially material cohesion increasing
molecule mobility. As a consequence, mechanical characteristics can be signiIicantly lowered. Water absorption
oI diIIerent polyamide is presented Figure 6.


Figure 6 Water absorption according to ambient air relative moisture oI diIIerent polyamides |CARI05|
-C-N-
|| |
O H
-[-N-(CH
2
)
11
-C-]
n
-
| ||
H O
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 4
Pipe mechanical characteristics are sensible to air moisture content. Thus when controlling pipe mechanical
characteristics, air moisture content in test room should be controlled to ensure reproducibility oI the results Ior
diIIerent atmospheric conditions outside.

High moisture content in plastic pellets would result in water degassing inside the extruder that would perturb
process operating conditions then degrade Iinal product mechanical characteristics. Thus pellets must be dried
beIore being processed.

2.3. Thermal conductivity

Thermal conductivity deIines the ability to conduct heat. Polymers are oIten considered to be thermal insulators
because oI their low thermal conductivity. ThereIore heating or cooling the material inside the extruder is a slow
process.

2.4. Thermal stability

Polymers can degrade when exposed to high temperatures: the higher the temperature, the Iaster the degradation.
Thus the two main Iactors involved in degradation mechanism are:
temperature level
time oI exposure

Degradation mechanism inside extrusion machine can result in:
low viscosity oI plastic melt
inclusions oI degraded material in the pipe: dimensional, mechanical, visual

Thermal stability oI polymers can be improved by adding thermal stabilizers.

2.5. Glass transition temperature and melting point

Glass transition temperature corresponds to a change oI molecule conIormation that aIIects polymer mechanical
properties. Polymers are soIter above this temperature than below. Melting point is an important characteristic oI
semi-crystalline materials. It corresponds to crystal melting temperature. Polymers behave like Iluids above this
temperature and like solids below. Glass transition temperature and melting point oI diIIerent polymers are given
Table 1.

Table 1 Glass transition temperature and melting point oI diIIerent polymers
Material Glass transition
temperature |C|
Melting point |C|
PA6 45 195
PA12 50 165
PA9T 140 255
EVOH 65 175

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 5
2.6. Viscous properties at melted state

Viscosity oI a Iluid deIines its resistance to Ilow uniIormly: the higher the viscosity, the higher the resistance. It
depends on Iluid molecule entanglement: the lower the entanglement, the lower the viscosity. Thus viscosity is
inIluenced by Iluid molecular weight: the higher the weight, the higher the viscosity. A Iluid which Ilow is not
uniIorm is called viscous Iluid. Polymers behave like viscous Iluids at melted state. Viscosities oI diIIerent Iluids
are compared Table 2.

Table 2 Comparison between viscosities oI diIIerent Iluids (order oI magnitude)
Fluid Viscosity |Pa.s|
Air 10
-5

Water 10
-3

Oil 10
-1

Plastic melts 10
2
to 10
6


2.6.1. Sensibility to temperature

Molecule mobility depends on Iluid temperature: the higher the temperature, the higher the mobility. Moreover
viscosity depends on molecule mobility: the higher the mobility, the lower the entanglement, the lower the
viscosity. ThereIore viscosity depends on Iluid temperature: the higher the temperature, the lower the viscosity.
For semi-crystalline polymers, viscosity varies Irom 2 to 3 per degree centigrade.

2.6.2. Sensibility to shear rate

Pseudo-plasticity oI a Iluid deIines viscosity sensibility to shearing. It depends on the capacity oI Iluid entangled
molecules to rearrange when exposed to shearing. The rearrangement increases with increasing shear rate: the
larger the rearrangement, the lower the entanglement, the lower the viscosity. A Iluid which viscosity depends on
Ilow shear rate is called non-Newtonian or non-linear Iluid. Polymers which are composed oI long entangled
molecules behave like non-linear Iluids at melted state. Non-linear behaviour can be characterized by plotting
the power law curve oI the material as shown Figure 7.


Figure 7 Power law curve oI a polymer |RAUW10|

Log (shear rate)
Log (viscosity)
Consistency index
Material viscosity level
Log (power law index)
Actual behaviour
Power law approximation
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 6
Considering two elements oI a viscous Ilow, shear rate is the ratio between their velocity diIIerence and their
normal distance as illustrated Figure 8. It corresponds also to velocity proIile slope also called velocity gradient.
Thus it depends on:
Flow velocity: the higher the velocity, the higher the shear rate
Ilow channel width: the smaller the channel, the higher the shear rate


Figure 8 Fluid shearing between two plates

EIIect oI shear rate on a rectangular Iluid element is illustrated Figure 9: the higher the shear rate, the higher the
shear strain and the higher the shear stress at given viscosity.




a. no shear strain b. low shear strain c. high shear strain

Figure 9 EIIect oI shear rate on a rectangular Iluid element

Shear stress is given by the Iollowing Iormula:
& &). , (T =
With:
Shear stress
) , ( & T
Material viscosity
&
Shear rate
T
Material temperature

High shear stress inside extrusion machine prevents material accumulation in dead zones: the higher the stress,
the higher the removal. This is important:
To avoid material degradation (see 2.4.)
To ensure eIIicient cleaning oI the machine (during cleaning phase)

W
V
Mobile plate
Fix plate
Shear rate =
W
V
&

&

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 7
2.6.3. Viscous heat generation

A viscous material generates heat when sheared. This mechanism is called viscous heat generation. Heat amount
is calculated Irom the product oI viscosity and shear rate squared:
2
). , ( & & T H
viscous
=
With:
viscous
H
Viscous heat
) , ( & T
Material viscosity
&
Shear rate
T
Material temperature

2.7. Specific volume variation between melted and solid states

SpeciIic volume deIines a volume per unit oI mass. It depends on material molecule organization and mobility.
Thus it is inIluenced by:
pressure: the lower the pressure, the higher the speciIic volume
temperature: the lower the temperature, the lower the speciIic volume

SpeciIic volume oI semi-crystalline materials decreases during cooling process as illustrated Figure 10.


Figure 10 Semi-crystalline material shrinkage during cooling process

Two phenomena related to speciIic volume variation can be observed during extrusion process:
pipe swelling during plastic melt discharge Irom the die (released pressure)
pipe shrinkage during cooling process inside sizing sleeve (cooling temperature)

Specific volume
Cooling
Shrinkage at molten state
Solidification shrinkage
Shrinkage at solid state
Start solidification
End solidification
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 8
2.8. Mechanical properties at solid state

Two types oI organisation can be identiIied when looking at the molecular structure oI semi-crystalline materials
at solid state as illustrated Figure 11:
amorphous phase: random and irregular structure
crystalline phase: highly regular structure composed oI crystals


Figure 11 Molecular structure oI semi-crystalline materials

Two Iactors related to the molecular structure inIluence material mechanical properties:
Crystal rate (below 25)
Molecular orientation

When crystal rate increases, material internal energy decreases and thereIore mechanical energy required Ior
displacing molecules increases. When material molecules are pre-orientated, stretching capacity oI molecular
chains is reduced in orientation direction and mechanical energy required increases also.

A break is caused by propagation oI a crack through the material. A crack initiation is typically a deIect into the
material. This crack propagates only iI the load applied to the material exceeds a limit that depends on deIect
location, size and orientation in the material.

When mechanical energy required Ior displacing molecules increases, the load applied to the material increases
at given elongation. ThereIore crack propagation probability increases and elongation at break decreases Iatally.
As a consequence, material elongation at break decreases with increasing crystal rate and molecular orientation.

Thus pipe mechanical properties, especially elongation at break, depend on:
Material crystal rate: the higher the crystal rate, the lower the elongation at break
Material molecular orientation: the higher the molecular orientation, the lower the elongation at break
SurIace state: the higher the quantity and size oI the deIects, the lower the elongation at break

2.9. Conclusion

BeIore working on plastic pipe extrusion process, it is necessary identiIying polymer main characteristics in
relation with pipe properties and extrusion process.

Plastic pipes Ior automotive application are essentially composed oI semi-crystalline polyamides. An important
characteristic oI semi-crystalline materials is their melting point that corresponds to crystal melting temperature.
Polymers behave like viscous Iluids above this temperature and like solids below.

Crystalline phase
Amorphous phase
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 9
2.9.1. Flow properties at melted state

Plastic melt viscosity and temperature inside extrusion machine are closely linked. They are important Iactors
when working on extrusion as it will be explained next chapter.

Plastic melt viscosity depends on:
shear rate: the higher the shear rate, the lower the viscosity
temperature: the higher the temperature, the lower the viscosity

Plastic melt temperature depends on:
viscous heat generation:
shear rate (predominant Iactor): the higher the shear rate, the higher the heat generation
plastic melt viscosity: the higher the viscosity, the higher the heat generation
conductive heat transIer: limited by polymer low thermal conductivity

2.9.2. Specific volume variation between melted and solid states

SpeciIic volume oI semi-crystalline materials, which deIines a volume per unit oI mass, decreases during cooling
process.

2.9.3. Mechanical properties at solid state

Pipe mechanical properties, especially elongation at break, depend on:
Material integrity: no inclusions oI non melted or degraded material in the pipe
Material crystal rate: the higher the crystal rate, the lower the elongation at break
Material molecular orientation: the higher the molecular orientation, the lower the elongation at break
Pipe surIace state: the higher the quantity and size oI the deIects, the lower the elongation at break

When controlling these characteristics, room air moisture content should be controlled to ensure reproducibility
oI the results Ior diIIerent atmospheric conditions outside.
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 10
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 11
3. Extrusion mechanisms

Hutchinson has several extrusion lines which diIIer by the number oI layers that can be extruded simultaneously:
monolayer line: monolayer pipe
three layer line: till three layer pipe
Iive-layer line: till Iive layer pipe

This study Iocuses specially on the Iive-layer line presented Figure 12 on which are produced PA9T/PA12 and
PA6/EVOH/ADH/PA12 structures.


a. extrusion head sizing system b. cooling bath c. caterpillar cutting unit

Figure 12 Five-layer extrusion line, Rivoli, Italy

Components oI a monolayer line are presented Figure 13. On a Iive-layer line, the single extruder is substituted
by Iive extruders connected to a head with inside a multi-maniIold that combines material Ilows just beIore
Iorming tools. Head and multi-maniIold are presented Figure 14. Process operations are also speciIied Figure 13.


Figure 13 Monolayer extrusion line |CARI05|

Extruder screw
Extruder barrel
Feed hopper
Barrel heating band
Sizing system
Cooling tank
Caterpillar
Cutting unit
Die and mandrel
Extrusion
Forming
Sizing
Cooling
hauling
cutting
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 12


a. multi-maniIold components b. back view oI the head (multi-maniIold removed)

Figure 14 Five-layer extrusion head, Rivoli, Italy

BeIore working on extrusion process improvement, it is necessary identiIying, Ior each operation oI the process,
the main Iactors that inIluence pipe essential characteristics then identiIying extrusion line parameters which
allow controlling these Iactors. BeIorehand, it requires understanding the mechanisms occurring Ior each
operation oI the process. All these aspects are presented in this chapter.

3.1. Extrusion

The main Iunction oI an extruder is to generate suIIicient pressure to Iorce the material through the die. Thus an
extruder works like a pump. In the case oI plastic pipe extruders, pumping action is created by the rotation oI the
screw as shown Figure 15 screw where motor load increases with increasing melt pressure in PA12 and PA9T
extruders.
20
25
30
35
40
45
50
55
60
150 200 250 300 350 400 450 500
MeIt pressure (bar)
S
c
r
e
w

m
o
t
o
r

I
o
a
d

(
%
)
PA12-1 PA9T

Figure 15 InIluence oI melt pressure in the extruder on screw motor load

The screw is the principal component oI an extruder. Its rotation speed determines extruder output and material
shear rate inside the extruder. As the extruder is Ied by a rotating screw, extrusion process is continuous by
deIinition and instabilities are very oIten caused by irregular material Ieeding. ThereIore Ieeding regularity is an
important Iactor to control.

The proIile oI a screw dedicated to polyamides extrusion is presented Figure 16 where Iunctional and operating
zones have been identiIied.


Figure 16 Screw geometry and Iunctional zones |MIAN09|
Feeding zone Compression zone Metering zone
Solids conveying
Melting
Melt conveying
Pumping effect
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 13
Functional zones are geometrically deIined whereas operating zones are not determined and are interdependent.
In Iact, melting process starts where material temperature Iirst exceeds melting point and ends when the last
granule is melted whatever the zone oI the screw. However iI end oI this process is postponed until the end oI
metering zone, there could be some inclusions oI non melted material in extruded pipe. Moreover iI beginning oI
this process is anticipated until Ieeding zone, Ieeding process could be disrupted aIIecting Ieeding regularity.
ThereIore melting rate in compression zone oI the screw is an important Iactor to control.

In a Iive-layer extrusion machine, when materials Irom the diIIerent extruders are combined, Ilow properties
should be compatible to limit distortion phenomenon at pipe layer interIace. ThereIore plastic melt viscosity at
the exit oI extrusion machine is an important Iactor to control.

3.1.1. Solids conveying

Solids conveying process starts in the Ieed hopper and ends when the last granule is melted.

3.1.1.1 Gravity induced conveying

Conveying mechanism in the Ieed hopper is called gravity induced conveying because it is generated by plastic
pellet weight. Hopper is designed to ensure a steady Ilow oI material. A vibrating device can be added Ior high
Ilow resistance pellets.

3.1.1.2 Drag induced conveying

Conveying mechanism along screw channel is called drag induced conveying as it results Irom Iriction Iorces
acting on the material. This mechanism determines machine output. It is illustrated Figure 17.


a. excluding barrel interaction: radial trajectory b. including barrel interaction: spiral trajectory

Figure 17 Material movement along screw channel

This mechanism is generated by the combined eIIect oI barrel Iriction Iorces acting on the material that limits its
rotation and screw helical Ilight rotation that pushes the material Iorward: the lower the material rotation, the
greater the conveying. In Iact, considering only screw interaction, the material would not have any axial
movement as it would rotate together with the screw around the shaIt. An analogy can be made with a nut Iree oI
rotating around a threaded shaIt.

From previous analysis, Ieeding regularity in Ieeding zone oI the screw depends on:
material characteristics:
pellet Ilow resistance
screw design:
helix angle: the larger the angle, the lower the Iriction on the screw
machine parameters:
extruder barrel temperature: the lower the temperature, the higher the Iriction on the barrel

Barrel
Screw
Helix angle
pool
Material
Screw
Helix angle
pool
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 14
3.1.2. Melting

Melting process starts when material temperature Iirst exceeds melting point and ends when the last granule is
melted.

3.1.2.1 Contiguous solids melting

In single screw extruders like Hutchinson ones, melting mechanism is called contiguous solids melting. This
mechanism is illustrated Figure 18.


Figure 18 Melting mechanism in single screw extruders

While moving along screw channel, pellets are compacted, heated up and melted at the contact oI barrel wall
which creates a melted Iilm. Then melting process is principally located at the interIace between solids bed and
melted Iilm.

Conductive heat transIer Irom barrel wall to solids bed is limited by polymer low thermal conductivity (see 2.3.):
the thinner the melted Iilm, the higher the transIer. Thus viscous heat generation is the main source oI heat inside
the extruder as illustrated Figure 19 and it depends on shear rate in the Iilm (see 2.6.3.). ThereIore melting rate
depends on shear rate in the Iilm: the higher the shear rate, the higher the melting rate. However shear rate
depends on Iilm thickness (see 2.6.2.): the thinner the Iilm, the higher the shear rate. ThereIore melting rate
depends also on Iilm thickness: the thinner the Iilm, the higher the melting rate. Film thickness is primarily
determined by Ilight clearance: the smaller the clearance, the thinner the Iilm. Thus melting rate decreases
irreparably when screw or barrel get worn.


Figure 19 EIIect oI barrel temperature on viscous and conductive heating

Increasing barrel temperature increases conductive heat transIer Irom barrel wall to solids bed. It increases also
melted Iilm temperature and as a consequence, reduces material viscosity then viscous heat generation. Thus,
over a given barrel temperature, total heating decreases and melting rate decreases.

Shear rate: -

Shear rate: +

Barrel temperature
Heating
Conductive heating
Viscous heating
Total heating
Solids
bed
Melt
pool
Screw
Barrel
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 15
From previous analysis, melting rate in compression zone oI the screw depends mainly on:
material characteristics:
melting point
viscous properties: power law curve oI the material
screw design:
helix angle: the lower the angle, the higher the shear rate, the higher the melting rate
compression ratio: the higher the ratio, the higher the shear rate, the higher the melting rate
machine parameters:
screw speed: the higher the speed, the higher the shear rate, the higher the melting rate
barrel temperature: depending on temperature level

3.1.2.2 Drag induced melt removal

Newly melted material is continuously dragged away to the melt pool by the combined eIIect oI barrel Iriction
Iorces acting on plastic melt and screw helical Ilight rotation. This mechanism called drag induced melt removal
limits melted Iilm expansion and as a consequence, avoids melting rate decreasing during melting process (see
3.1.2.1).

Hutchinson extruder screw is composed oI two threads in Ieeding and melting zones as presented Figure 20:
one principal thread that conveys the material
one variable barrier thread that separates physically solid Irom molten material


Figure 20 Hutchinson extruder screw

Melt channel width increases as quantity oI melt material increases and quantity oI solid material decreases. End
oI the solids channel is closed. Working principle is illustrated Figure 21.


Figure 21 - Melting mechanism in Hutchinson extruder

The advantage oI this conIiguration is that probability that none melted particles be at the end oI melting zone is
reduced. In Iact, while particles are accumulated at the end oI solids channel, pressure increases in the channel
that accelerates melting mechanism. ThereIore process operating window is larger with this type oI screw than
with a standard one.
Solids
bed
Melt
pool
Barrel
Melt channel Solids channel
Barrier flight
Main flight
Screw
Barrier undercut
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 16
3.1.3. Melt conveying

Melt conveying process starts as soon as melting process starts. Thus both solids conveying and melt conveying
occur in the compression zone oI the screw with quantity oI solid material decreasing while quantity oI melted
material increases gradually. Melt conveying mechanism is viscous drag and it is Iundamentally identical to
solid conveying. It is generated by viscous Iorces acting on barrel surIace and limited by viscous Iorces acting on
screw surIace (see 3.1.1.2).

Plastic melt viscosity depends on shear rate inside extrusion machine and plastic melt temperature (see 2.9.1.).
InIluence oI barrel temperature on plastic melt temperature is limited by polymer low thermal conductivity (see
2.3.). Thus, Ior instance, plastic melt temperature decreases when screw speed decreases. In Iact, shear rate
decreases when screw speed decreases and, although residence time oI the material inside the machine increases,
viscous heat generation decreases more than conductive heat transIer increases.

From previous analysis, plastic melt viscosity in metering zone oI the screw depend mainly on:
material characteristics:
viscous properties: power law curve oI the material
screw design in metering zone:
helix angle: the lower the angle, the higher the shear rate, the lower the viscosity
compression ratio: the higher the ratio, the higher the shear rate, the lower the viscosity
machine parameters:
screw speed: the higher the speed, the higher the shear rate, the lower the viscosity
barrel temperature: inIluence limited on plastic melt temperature and viscosity

Plastic melt Ilow relative to the screw can be illustrated by unrolling screw and barrel and considering them to be
respectively a straight trough and a Ilat plate as presented Figure 22. Barrel movement direction makes an angle
with screw channel that corresponds to screw Ilight helix angle.



a. unrolled view b. view section B-B

Figure 22 Plastic melt element path in screw channel

Considering an element Irom the upper region oI the channel:
It travels Iollowing barrel movement until the pushing side oI the Ilight.
It moves downward until the bottom oI the channel.
It travels across the channel until the trailing side oI the Ilight.
It moves upward until the top oI the channel

This pattern is repeated numerous times along screw channel. ThereIore an element Iollows a helical path. Flow
close to barrel wall is induced by viscous drag whereas Ilow across screw channel is generated by the negative
pressure gradient between pushing side and trailing side oI the Ilight.
Screw
Barrel
Screw
Fixed
pool
Barrel
B
B
Helix angle
Screw axis
A
A
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 17
Figure 23 presents velocity and shear rate proIiles in down-channel (section A-A) and cross-channel (section B-
B) oI the screw.


a. velocity proIile view section A-A b. shear rate proIile - view section A-A

c. velocity proIile view section B-B d. shear rate proIile - view section B-B

Figure 23 Velocity and shear rate proIiles in screw channel

Pressure gradient in screw channel is positive as the screw works like a pump (see 3.1.). ThereIore there is a
backIlow also called recirculation at the bottom oI down-channel as shown Figure 23 a. which intensity increases
with increasing melt pressure.

Mass Ilow balance inside the machine is given by the Iollowing Iormula:
ion recirculat output screw
Q Q Q + =
With:
Q
screw
Mass Ilow provided by the screw
Q
output
Mass Ilow provided by the machine
Q
recirculation
Mass Ilow recirculation inside the extruder

When recirculation intensity increases, plastic melt average residence time inside the machine increases and
temperature increases either. ThereIore average melt temperature increases when melt pressure increases.

3.1.4. Mixing

Shear rate in screw channel is not uniIorm as shown Figure 23. Thus viscous heating is not uniIorm either and
melt temperature in screw channel is likely to be heterogeneous because oI polymer low thermal conductivity
(see 2.3.).

II melt temperature is heterogeneous in die channel, plastic melt Ilow will be heterogeneous either. It will result
in a distorted pipe. ThereIore it is necessary homogenising the material beIore it reaches the die by providing an
intensive mixing during melt conveying.

Screw
Barrel
Shear rate
pool
Screw
Velocity
pool
Barrel
Screw
Barrel
Shear rate
pool
Screw
Barrel
Velocity
pool
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 18
Mixing intensity is deduced Irom shear strain that is the product oI shear rate and time oI exposure to shearing:

&
& t =
With:

Shear strain
&
Shear rate
&
t
Time oI exposure to shearing

Thus mixing distribution in screw channel depends on:
shear rate distribution
residence time distribution:
Ilow path
velocity proIile

Part oI mixing mechanism occurs in screw channel. Shear distribution is presented Figure 23. Residence time is
deduced Irom Ilow path and velocity proIile presented respectively Figure 22 and 23. It is long close to barrel
and screw walls and short Iurther away Irom the walls as illustrated Figure 24.


Figure 24 - Plastic melt element residence time in screw channel

Thus total cross-channel shear strain is approximately the sum oI shear strain in lower and upper regions oI the
channel. Total down-channel shear strain is deduced the same way. Total shear strain is the sum oI total cross-
channel shear strain and total down-channel shear strain. ThereIore it depends also strongly on element location
in channel depth as illustrated Figure 25.


Figure 25 Total shear strain in screw channel

Elements close to barrel and screw walls experience a high level oI shear strain whereas elements Iurther away
Irom the walls experience a low level. Thus mixing mechanism in screw channel is not uniIorm and plastic melt
temperature could be not Iully homogeneous at the exit oI extrusion machine. However mixing mechanism can
be improved by adding mixing sections along material channel inside the machine. In Hutchinson Iive-layer
machine, these sections are included in the design oI multi-maniIold component presented Figure 14 a.
Screw
Barrel
Residence time
pool
Screw
Barrel
Total shear strain
pool
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 19
3.2. Forming

Forming operation gives extruded product shape to plastic melt. Forming tools are located at the exit oI extrusion
machine. Pipe products are Iormed with annular die and annular mandrel. Examples oI annular die and annular
mandrel are presented Figure 26.


a. die b. mandrel

Figure 26 Extruded pipe Iorming tools

Flow channel is designed so that Ilowing material achieves a uniIorm velocity across die exit. It can be divided
in three zones as illustrated Figure 27:


Figure 27 Die Ilow channel zones

Two Iactors can characterize extruded pipe at the exit oI extrusion machine. These Iactors are important because
they inIluence sizing operation conditions:
Diameter ratio
Thickness ratio

Their Iormula is given below:
pipe outer
die
ratio
D
D
D

=
pipe inner pipe outer
mandrel die
ratio
D D
D D
T

=
With:
D
die
Extruded pipe outer diameter ( Die diameter)
D
mandrel
Extruded pipe inner diameter ( Mandrel diameter)
D
outerpipe
Final pipe outer diameter
D
innerpipe
Final pipe inner diameter

Mandrel
Die
Inlet channel Land region Manifold
Material
Dmandrel
Ddie
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 20
Melt pressure inside the machine is strongly inIluence by Iorming tool dimensions as illustrated Figure 28 since
it depends on pressure discharge along die channel given by the Iollowing Iormula:

Q K P
apparent
. . = (Poiseuille law)
With:
AP Pressure discharge
K Geometrical and dimensional Iactor
q
apparent
Material apparent viscosity
Q Material Ilow

200
250
300
350
400
450
500
50 100 150 200
Discharge section (mm2)
M
e
I
t

p
r
e
s
s
u
r
e

(
b
a
r
)
PA12-1 PA9T

Figure 28 InIluence oI Iorming tool dimensions on melt pressure inside the extruder

Material viscosity close to tooling wall increases when tooling temperature decreases. Thus material apparent
viscosity increases and pressure discharge at given material Ilow increases either. ThereIore melt pressure inside
the machine increases. However average melt temperature increases when melt pressure increases (see 3.1.3.).
Moreover plastic melt viscosity decreases when plastic melt temperature increases (see 2.9.1.). ThereIore plastic
melt viscosity can be decreased by decreasing Iorming tool temperatures.

From previous analysis, plastic melt viscosity at the exit oI extrusion machine depends also on:
Iorming tool dimensions: the smaller the discharge section, the lower the viscosity
Iorming tool temperature: the lower the tool temperatures, the lower the viscosity

3.3. Sizing and cooling

Sizing operation Iixes extruded pipe Iinal dimensions, especially external diameter and thickness.

Extruded pipe needs to be supported otherwise it would collapse under its own weight. It means that aIter sizing
operation, pipe surIace should be stiII enough to not be damaged by guiding devices. ThereIore it has to be
cooled down rapidly and in this condition, air cooling is not suIIicient and water cooling is necessary.

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 21
Hutchinson vacuum-sizing system is presented Figure 29.


a. principle b. picture

Figure 29 Hutchinson sizing system

In this system, the pipe is sized Irom outside while passing through the sizing sleeve located inside the vacuum
tank. The vacuum tank is Iilled with cooling water. Vacuum level is typically between 60 and 300 mbar. Water
lubrication is required to reduce Iriction at the interIace between pipe and sleeve. This water is provided by the
annular chamber located at the entrance oI the sizing sleeve. Water Ilow is typically between 40 and 80 L/h.

An example oI sizing sleeve is presented Figure 30. Sizing sleeve is made oI high thermal conductivity material
such as copper, bronze or brass. It has a cylindrical geometry like the pipe but its internal diameter is higher than
Iinal pipe external diameter, typically between 1.07 and 1.1 times, to take into consideration material shrinkage
during solidiIication (see 2.9.2.). Its minimum length depends mainly on line speed: the higher the speed, the
longer the sleeve. Holes drilled along the sleeve allow vacuum drawing the material on sleeve inner wall and Iix
pipe external diameter: the higher the vacuum, the larger the external diameter at given sizing sleeve. ThereIore
vacuum is an important Iactor to control.


Figure 30 Hutchinson sizing sleeve

Three major phenomena are involved in sizing process These phenomena are important because they aIIect pipe
mechanical properties:
heterogeneous cooling rate: residual stress in pipe material
draw down and violent cooling rate: crystal rate and molecular orientation oI pipe material
Iriction: surIace state oI the pipe, draw down

Annular chamber
Channel at 45
Sizing sleeve
Water under vacuum
Water outlet
Water inlet
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 22
3.3.1. Heterogeneous cooling rate

While entering into the sizing sleeve, the pipe is violently cooled down at the contact oI annular chamber water.
This violent cooling creates a high thermal gradient in the external layer oI the pipe due to polymer low thermal
conductivity (see 2.3.). This gradient depends initially on water temperature. Then it decreases steadily along
cooling bath as heat Irom the core is conduced outside the pipe and dissipated in the water.

Heterogeneous cooling rate in pipe external layer generates residual stress in the material due to the variation oI
material speciIic volume during solidiIication process (see 2.9.2.). This mechanism is illustrated Figure 31. This
residual stress aIIects pipe mechanical properties, especially pipe elongation at break.


Figure 31 Residual stress generation mechanism

3.3.2. Draw down and violent cooling rate

Pipe thickness depends on pipe section and pipe external diameter determined by vacuum level (see 3.3.). Pipe
section depends on pipe draw down which is the ratio between extrusion line speed and plastic melt speed inside
the die given by screw rotation speed. ThereIore pipe thickness is determined by pipe draw down and vacuum
level. Pipe draw down ratio (DDR) is given by the Iollowing Iormula:

2

2 2
pipe inner pipe outer
mandrel die
melt
line
D D
D D
J
J
DDR

= (Mass flow conservation)

Draw down mechanism occurs both between die and sizing sleeve and inside sizing sleeve as illustrated Figure
32. ThereIore previous Iormula can also be written:

air sleeve
melt
sleeve
sleeve
line
melt
line
DDR DDR
J
J
J
J
J
J
DDR = = =


Figure 32 Schematic oI sizing operation
Douter_pipe Dmandrel
Ddie
Mandrel
Pipe
Die
Vmelt
Vline
Sizing system
Draw down
Water Air
Vsleeve
Cooling water
Pipe external layer
Pipe internal layer
Fast solidification: free shrinkage
Slow solidification: shrinkage limited
by external layer already solidified

Generation of residual stress:
Traction of external layer
Compression of internal layer
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 23
Pipe draw down inside sizing sleeve increases with increasing Iriction as illustrated Figure 33. Friction Iorces are
concentrated at the beginning oI the sleeve because oI pipe shrinkage during cooling process (see 2.9.2.).


Figure 33 InIluence oI Iriction Iorces inside sizing sleeve on draw down distribution

Friction Iorces inside sizing sleeve counteract caterpillar hauling Iorce. Thus when Iriction Iorces increase, pipe
hauling Iorce upstream the sleeve decreases. ThereIore air draw down decreases and draw down inside the sleeve
increases in the meantime.

3.3.2.1 Between die and sizing system

Between die and sizing system, the material is not solidiIied yet. As a result, draw down does not aIIect the
process as long as no cracks appear in the Ilow. These cracks appear iI plastic melt viscosity is too high or not
homogeneous (inclusions oI non melted or degraded material). II viscosity is too low, melted pipe will collapse
beIore reaching the sleeve and it will not be possible to size it.

3.3.2.2 Inside sizing sleeve

Inside the sleeve, draw down mechanism occurs while solidiIication process is in progress. In Iact, solidiIication
starts at the contact oI annular chamber water and continues along the sleeve. The combined eIIect oI draw down
and violent cooling rate generates a high thermo-mechanical load on the pipe that results in a plastic deIormation
oI the material being solidiIied. This phenomenon is limited to pipe external layer since solidiIication is Iirstly
limited to pipe external layer (see 3.3.1.).

Hauling force
Hauling force
Friction forces
Sleeve
Pipe profile
DDR
air
DDR
sleeve

No friction force
Die
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 24
Previous studies conIirm this analysis. In Iact, the crystalline structure oI a pipe has been analysed aIter sizing
operation and two singularities have been noticed in the external layer (about 200 m) which reveal the presence
oI plastic deIormation mechanism:
high crystal rate whereas cooling is violent
high anisotropy whereas internal layer is isotropic: illustrated Figure 34





a. sizing operation b. pipe molecular structure

Figure 34 InIluence oI sizing operation on pipe molecular structure

Violent cooling results usually in low crystal rate because crystallisation kinetic is too slow. As a consequence,
plastic deIormation has accelerated and orientated crystallisation by providing mechanical energy.

Pipe mechanical properties, especially pipe elongation at break, depend on material crystal rate and molecular
orientation (see 2.9.3.). ThereIore they depend on draw down and cooling inside sizing sleeve: the higher the
thermo-mechanical load, the lower the elongation at break. EIIect oI thermo-mechanical load on pipe elongation
is illustrated Figure 35.


Figure 35 EIIect oI thermo-mechanical load on pipe elongation

From previous analysis, draw down conditions during sizing operation depends on:
plastic melt viscosity at the exit oI extrusion machine
Iorming tool dimensions: the higher the dimensions, the higher the draw down
water temperature: the lower the temperature, the more violent the cooling rate while drawing down
Iriction: the higher the Iriction, the higher the draw down inside the sleeve

Anisotropic zone: high
molecular orientation
Isotropic zone
Pipe
Plastic deformation
Sizing sleeve
Elongation
Stress
c
residual

o
residual

Cooling
No cooling
Plastic deformation
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 25
3.3.3. Friction

During sizing operation, extruded pipe is Iorced through the sizing sleeve. Interaction between pipe and sleeve
generates Iriction Iorces. When Iriction Iorces are high, some deIects can appear on pipe surIace, especially iI
there is adhesion between pipe and sleeve. These deIects are potential crack initiations during pipe elongation.

Friction Iorce intensity along the sleeve depends on:
extrusion line speed: the higher the speed, the higher the Iriction
water lubrication: the higher the lubrication, the lower the Iriction
vacuum: the higher the vacuum, the lower the lubrication
sleeve internal diameter: the larger the diameter, the lower the vacuum
sleeve surIace state: the better the surIace state, the lower the Iriction
pipe material shrinkage: Iriction Iorces concentrated at the beginning oI the sleeve

3.3.3.1 Extrusion line speed

A previous study carried out with a speciIic sizing system equipped with a stress gauge |CARI05| shows that
Iriction increases when extrusion line speed increases. In Iact, pipe and sleeve interaction through the water Iilm
generates a viscous Iriction, typically speed dependant.

3.3.3.2 Water lubrication

Using the same equipment, previous study demonstrates also that Iriction increases when vacuum increases as
shown Figure 36.

0
2
4
6
8
10
12
14
50 100 150 200 250 300
Vacuum (mbar)
F
r
i
c
t
i
o
n

f
o
r
c
e

(
N
)

Figure 36 InIluence oI vacuum level on Iriction |CARI05|

In Iact, while travelling along the sleeve, lubricating water Iilm is dissipated through the holes by vacuum eIIect:
the higher the vacuum, the shorter and the thinner the lubrication.

3.3.3.3 Sleeve surIace state

Sleeve surIace state inIluences two mechanisms that occur at the interIace between pipe and sleeve:
capillary water Ilow: the higher the roughness, the higher the lubrication
Iriction: the higher the roughness, the higher the Iriction

A study could be perIormed to quantiIy the eIIect oI sleeve roughness on both mechanisms. Such a study is out
oI the scope oI this thesis. For some applications, Hutchinson sleeves are coated with chromium to get a smooth
surIace that is easier to control in production.

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 26
3.4. Conclusion

BeIore working on extrusion process improvement, it is necessary identiIying, Ior each operation oI the process,
main Iactors that inIluence pipe essential characteristics, then identiIying extrusion line parameters which allow
controlling these Iactors.

3.4.1. Extrusion factors

Process Iactors that inIluence pipe essential characteristics during extrusion operation are the Iollowing:
Feeding regularity
Melting rate
Melt viscosity

Feeding regularity determines extrusion machine output stability. Melting rate is adjusted so that melting process
occurs exclusively in the compression zone oI the screw. II end this process is postpone, there could be some
inclusions oI non melted material in extruded pipe. II beginning oI this process is anticipated, solids conveying
process could be disrupted aIIecting Ieeding regularity. Plastic melt viscosity is adjusted to have compatible Ilow
properties when materials Irom the diIIerent extruders are combined to limit distortion phenomenon at pipe layer
interIace. Plastic melt viscosity should be also compatible with sizing operation. II viscosity is too high or not
homogeneous, some cracks could appear while melted pipe is drawn down. II viscosity is too low, melted pipe
will collapse beIore reaching the sleeve and it will not be possible to size it.

These Iactors depend on:
material characteristics
screw design
operating conditions (machine parameters)

The screw is the principal component oI an extruder. It is originally designed Ior a unique combination machine-
material. ThereIore its design is optimized according to the characteristics oI the material as presented Table 3.

Table 3 InIluence oI material characteristics on optimal screw design
Screw design (eIIect)
Characteristics ()
Helix angle
Compression
ratio
Pellet Ilow resistance
+
Sensibility to shearing
- +

In practice, the same screw is used Ior several materials when it is possible. However operating conditions may
have to be adapted to keep process Iactors matching with pipe required characteristics. Machine parameters that
can be adjusted on extrusion line are presented Table 4 with their respective inIluence on process Iactors.

Table 4 EIIect oI increasing extrusion line parameters on process Iactors
Factors (eIIect)
Parameters ()
Feeding
regularity
Melting rate Melt viscosity
Screw speed
+ -
Barrel temperature in Ieeding zone
-
Barrel temperature in compression zone
+ / -
Barrel temperature in metering zone

Head temperature

Forming tool dimensions
+
Forming tool temperature
+

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 27
Feeding regularity depends on solids conveying mechanism, especially Iriction oI extruder barrel on the pellets,
which is inIluenced by barrel temperature in Ieeding zone. Melting rate and plastic melt viscosity depend on
shear rate inside the extruder determined by screw speed. Melting rate depends also on barrel temperature in
compression zone: over a given temperature, it decreases with increasing barrel temperature because viscous
heating in the melted Iilm decreases more than conductive heating increases. Plastic melt viscosity depends also
on Iorming tool dimensions and temperature that inIluence melt pressure inside extrusion machine then melt
temperature because oI recirculation phenomenon. InIluence oI barrel temperature in metering zone and head
temperature on plastic melt viscosity is limited by polymer low thermal conductivity.

Impact oI each parameter on process Iactors and pipe essential characteristics will be studied in detail during the
study on process reliability improvement that will be carried out in 2012. BeIorehand, extrusion line parameters
to monitor process Iactors have been identiIied and are presented Table 5.

Table 5 Monitored parameters on extrusion line
Factors
Parameters
Feeding
regularity
Melting rate Melt viscosity
Motor load X X
Melt temperature X
Melt pressure X X

Feeding regularity aIIects melt pressure stability whereas plastic melt viscosity aIIects melt temperature and melt
pressure levels. Motor load is equivalent to melt pressure. Melting rate can be monitored only by checking the
presence oI inclusions on melted pipe at the exit oI extrusion machine.

3.4.2. Sizing factors

Process Iactors that inIluence pipe essential characteristics during sizing operation are presented Table 6 with
their respective inIluence on pipe characteristics.

Table 6 EIIect oI increasing process Iactors on extruded pipe characteristics
Pipe characteristics (eIIect)
Factors ()
External
diameter
Thickness
Elongation at
break
Cooling rate
-
Draw down
- -
Friction
-
Vacuum
+ - -

Pipe external diameter and pipe thickness depend on vacuum inside water tank that draws the material on sleeve
inner wall. Pipe thickness depends also on pipe section determined by pipe draw down. Pipe elongation at break
depends on material residual stress in pipe external layer which results Irom a heterogeneous cooling rate in pipe
external layer and the variation oI material speciIic volume during solidiIication process. Pipe elongation at
break depends also on the thermo-mechanical load on pipe external layer which results Irom the combination oI
a violent cooling rate oI pipe external layer and pipe draw down inside sizing sleeve. Lastly, pipe elongation at
break depends on Iriction inside sizing sleeve that can generates some deIects on pipe surIace which are potential
crack initiations.

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 28
Machine parameters that can be adjusted on extrusion line to keep process Iactors matching with pipe required
characteristics are presented Table 7 with their respective inIluence on process Iactors.

Table 7 EIIect oI increasing extrusion line parameters on process Iactors
Factors (eIIect)
Parameters ()
Cooling rate Draw down Friction Vacuum
Screw speed*
-
Forming tool dimensions
+
Water temperature
-
Extrusion line speed**
+ +
Water Ilow
-
Vacuum level
+ +
Sleeve dimensions
- -
* at constant extrusion line speed
** at constant screw speed

Pipe cooling rate depends on annular chamber water temperature. Pipe draw down depends on the ratio between
pipe section at the exit oI extrusion machine determined by Iorming tool dimensions and pipe Iinal section. It
depends also on the ratio between pipe speed at the exit oI extrusion machine determined by screw speed and
extrusion line speed. Vacuum depends on the Iorce required to draw the material on sizing sleeve internal wall
and as a consequence, it depends on sleeve internal diameter. Friction depends on the thickness oI the lubricating
Iilm inside sizing sleeve which decreases along the sleeve as water is dissipated by vacuum eIIect. Viscous
Iriction in lubricating Iilm depends on extrusion line speed.

It can be noticed that many Iactors are interdependent. For instance, plastic melt viscosity at the exit oI extrusion
machine and pipe draw down inside sizing sleeve depend both on Iorming tool dimensions.

Impact oI each parameter on process Iactors and pipe elongation at break will be studied chapter 5.
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 29
4. Pipe essential characteristics monitoring

BeIore working on extrusion process improvement, it is necessary being able to monitor extruded pipe essential
characteristics.

Monitored essential characteristics oI a product are identiIied Irom speciIication requirements and are speciIied
into the control plan. Product quality management is presented more in detailed in Appendix 2. For extruded
pipes, monitored characteristics and their respective controls are divided in two categories:
dimensional controls:
non destructive controls:
ultrasonic system: total thickness
laser system: external diameter
destructive controls:
microscope instrument: internal diameter, external diameter, intermediate layer
thickness, total thickness
structural controls:
burst test: burst pressure
cold impact test: cold impact resistance
tensile test: elongation at break

This chapter presents control method and instrument Ior each characteristic monitored. It includes also a study
to improve current dimensional controls in anticipation oI the study on process reliability improvement that will
be carried out in 2012 and a study to improve tensile test reliability beIore working on pipe elongation at break
improvement.

4.1. Dimensional controls

Dimensional controls allow characterising pipe dimensional properties. Dimensions and tolerances required Ior
the structure PA9T/PA12 10x1are presented Figure 37.


Figure 37 PA9T/PA12 10x1 structure drawing

There are two categories oI dimensional controls:
non destructive controls
destructive controls

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 30
4.1.1. Non destructive controls

Extrusion line is equipped with two monitoring instruments presented Figure 38:
ultrasonic system: total thickness
laser system: external diameter



a. Ultrasonic system b. laser system

Figure 38 Monitoring instruments on extrusion line

These instruments are used:
in starting phase to adjust extrusion parameters: set-up time reduction
in production phase to monitor process deviations: product characteristic control

They can only detect broad deviations on pipe dimensions, and not local ones. In Iact, control is made according
to a given Irequency and as a consequence, it is not continuous. Local deIects can be controlled using a dedicated
mistake-prooIing system. Instruments are separated on the line since their operating conditions are incompatible.
In Iact, ultrasonic system works in liquid environment whereas presence oI water droplets on window can distort
laser system measurement.

4.1.1.1 Ultrasonic system

The ultrasonic system is positioned around the circumIerence oI the pipe. It is composed oI piezoelectric crystals
which convert reversibly electrical energy into mechanical energy. Its working principle is presented Figure 39.



a. ultrasonic wave generation and reIlection b. electrical signal measured

Figure 39 Schematic oI ultrasonic system working principle |ZUMB07|

The electronic unit transIers a short electrical pulse to the crystals that are excited and generate sound waves.
When sound waves encounter a diIIerence oI material density, Ior instance passing Irom water to pipe material,
part oI them are reIlected to the crystals that transIer an electrical signal to the electronic unit. When a pipe is
measured, its outer wall creates a Iirst echo and its inner wall a second echo. Wall thickness is deduced Irom the
time diIIerence between the two echoes.

Propagation speed oI sound waves varies according to material density and as a consequence, according also to
material temperature. Thus measurement accuracy oI ultrasonic system depends on pipe layer materials and on
pipe temperature at the checking point which is diIIicult to control on the line. As a consequence, this system can
be used only to monitor process deviations and not as a measurement instrument like a microscope.

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 31
4.1.1.2 Laser system

The laser system is positioned around the circumIerence oI the pipe system. It is composed oI two laser diodes
that generate laser beams and scan the object measured. Its working principle is presented Figure 40.


Figure 40 Schematic oI laser system working principle |ZUMB07|

Laser beams are deIlected by the rotating mirror. Two deIlection mirrors direct the rays into the optic emitters
Irom which they are sent to the measuring Iield as two absolutely parallel beams at a right angle to each other.
Receiver lenses Iocus the luminous beams onto photodiodes by means oI deIlection mirrors. The photodiodes do
not receive any light when laser beams are outside oI the measuring Iield or iI there is an object in their path.
Resulting light shadow signals are detected by the photodiodes and converted into electronic signals to be sent to
the data acquisition system. The duration oI the shadow cast by the object is measured by the data acquisition
system which converts it into a dimension.

Next two sections present a Ieasibility study to substitute partly destructive controls by non destructive controls
extending use oI ultrasonic and laser systems. First section concerns the measurement oI intermediate layer
thickness oI multilayer pipes using ultrasonic system and second section concerns the calculation oI pipe internal
diameter Irom the data oI pipe external diameter given by laser system and pipe thickness given by ultrasonic
system.

4.1.1.3 Pipe intermediate layer measurement

Measuring intermediate layer thickness oI multilayer pipes is not Ieasible with current system resolution. In Iact,
densities oI layer materials are very close together. Thus reIlected signal at the interIace between two layers is
very low and cannot be detected by the system as shown Figure 41.


Figure 41 Example oI electronic signal sent by the data acquisition system Ior PA9T/PA12 structure

Inner diameter Outer diameter
No signal at the interface
between PA9T and PA12 layers
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 32
4.1.1.4 Pipe internal diameter calculation

Calculation oI pipe internal diameter Irom the data oI pipe external diameter given by laser system and pipe total
layer given by ultrasonic system can be deduced Irom the model presented Figure 42.



a. measurement direction oI current devices b. unrolled view: linear model

Figure 42 Calculation oI internal diameter Irom external diameter and thickness values

Thus pipe internal diameter is calculated Irom the Iollowing Iormula:

) (
45 12 1 1 int
e e D D
ext
+ =
) (
61 34 2 2 int
e e D D
ext
+ =
With:
4
. 3
1 2
12
E E
E
+
=
4
. 3
4 3
34
E E
E
+
=
4
. 3
4 5
45
E E
E
+
=
4
. 3
1 6
61
E E
E
+
=

This model is based on the assumption that measurement triggering between ultrasonic and laser systems can be
synchronized. However, aIter having consulted the supplier, synchronisation between the two systems is not
possible. Thus calculating pipe internal diameter using ultrasonic and laser systems is not Ieasible.
E
3

45 15
E
34
E
4

Linear model
D
ext2

D
ext1

E
6

E
1

E
4


E
3
E
5

45
15
Pipe
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 33
4.1.2. Destructive controls

Destructive controls comprise two steps:
sample preparation: cutting the pipe
sample analysis: using an optical instrument

Two types oI instruments can be used Ior dimensional analysis:
proIile projector: internal diameter, external diameter, total thickness
microscope: internal diameter, external diameter, total thickness, layer thickness

Use oI proIile projector is limited to monolayer pipes as internal layers cannot be measured. ThereIore this study
will be interested only in microscope instrument.

4.1.2.1 Current method

Sample preparation consists oI cutting by hand the pipe with a pipe-cutter. The sample is positioned in Iront oI a
microscope equipped with a video-camera which picture is displayed on a computer screen. Luminosity and
sharpness are adjusted and the picture is analysed with a dedicated program. An example oI analysis is presented
Figure 43.


Figure 43 Example oI dimensional analysis

Thus measurement quality depends on the Iollowing parameters:
sample cutting
section not altered to be able to distinguish clearly layer borderlines
section perpendicular to pipe axis to measure circles and not ellipsis
section not deIormed to measure circle and not ellipsis
sample positioning
section parallel to optical plane to be able to get a sharp picture
picture aspect
luminosity adjusted to maximize contrast between layers (sample length)
sharpness adjusted to maximize contrast gradient between layers
vision accuracy
high discrimination capacity to distinguish precisely layer borderlines

As each operation is perIormed manually, measurement quality relies on operator skills. As a consequence, it can
vary Irom a controller to another one. According to quality standards, uncertainty on the value measured must be
less than ten percent oI dimension tolerance otherwise measurement method is considered to be not capable. This
criterion is checked carrying out a Repeatability and Reproducibility study (R&R study). In the case oI current
dimensional control, study result indicates that method is acceptable but can be improved to reduce measurement
variability, especially Ior multilayer structures which tolerances on layer dimensions are smaller than those oI
monolayer structures. ThereIore it is necessary improving this aspect beIore the study on process reliability
improvement that will be carried out in 2012.

Two axes oI improvement can be deduced Irom previous analysis:
developing a speciIic cutting machine
developing a program to automate measures

Thus next sections present a concept oI cutting machine then a program to automate measures. However their
development is out oI the scope oI this thesis.
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 34
4.1.2.2 Cutting machine

A basic outline oI the problem is presented Figure 44.


Figure 44 Basic outline oI the problem

Several cutting machines were developed in the past but they did not completely meet the requirements speciIied
in section 4.1.2.1:
Spring-drive cutting machine:
section altered by the cutting depending on material stiIIness (blade marks)
Razor blade cutting machine:
section altered locally by the cutting (blade impact zone)
section not perpendicular to pipe axis
section deIormed

Two examples oI machines are presented Figure 45.


a. spring-drive cutting machine b. razor blade cutting machine

Figure 45 SpeciIic cutting machines developed in the past

Thus it is necessary developing a new machine which design should be based on a detailed Iunctional analysis.
Functional analysis then machine design are presented in this section.

Cutting machine
Quality operators
Who takes advantage of the system? What does it acts on?
What is the goal?
Pipe
Cutting the pipe for microscope dimensional analysis
C
H
A
L
M
E
R
S
,

A
p
p
l
i
e
d

M
e
c
h
a
n
i
c
s
,

M
a
s
t
e
r
`
s

T
h
e
s
i
s

2
0
1
1
:
5
5

3
5

F
i
r
s
t
,

e
x
p
e
c
t
e
d

I
u
n
c
t
i
o
n
s

o
I

t
h
e

s
y
s
t
e
m

h
a
v
e

b
e
e
n

i
d
e
n
t
i
I
i
e
d
:

O
p
e
r
a
t
i
n
g

I
u
n
c
t
i
o
n
s

(
O
F
)

C
o
n
s
t
r
a
i
n
i
n
g

I
u
n
c
t
i
o
n
s

(
C
F
)

I
n
t
e
r
a
c
t
i
o
n
s

o
I

t
h
e

s
y
s
t
e
m

w
i
t
h

i
t
s

e
n
v
i
r
o
n
m
e
n
t

a
r
e

p
r
e
s
e
n
t
e
d

F
i
g
u
r
e

4
6
.

F
u
n
c
t
i
o
n
s

a
s
s
o
c
i
a
t
e
d

a
r
e

d
e
s
c
r
i
b
e
d

T
a
b
l
e

8
.



F
i
g
u
r
e

4
6

I
n
t
e
r
a
c
t
i
o
n

d
i
a
g
r
a
m


T
a
b
l
e

8

F
u
n
c
t
i
o
n

d
e
s
c
r
i
p
t
i
o
n

N

D
e
s
c
r
i
p
t
i
o
n

C
r
i
t
e
r
i
o
n

L
e
v
e
l

O
F
1

C
u
t
t
i
n
g

t
h
e

p
i
p
e

T
w
o

s
e
p
a
r
a
t
e
d

p
a
r
t
s

a
t

o
p
e
r
a
t
i
o
n

e
n
d

Y
e
s

C
F
1

N
o
t

a
l
t
e
r
i
n
g

c
u
t
t
i
n
g

s
e
c
t
i
o
n

P
r
e
s
e
n
c
e

o
I

m
a
r
k
s

t
h
a
t

c
o
u
l
d

i
n
t
e
r
I
e
r
e

w
i
t
h

l
a
y
e
r
s

N
o

C
F
2

C
u
t
t
i
n
g

p
e
r
p
e
n
d
i
c
u
l
a
r

t
o

p
i
p
e

a
x
i
s

A
n
g
l
e

b
e
t
w
e
e
n

c
u
t
t
i
n
g

s
e
c
t
i
o
n

a
n
d

p
i
p
e

a
x
i
s

(

)

M
i
n
i
m
u
m

C
F
3

N
o
t

d
e
I
o
r
m
i
n
g

t
h
e

p
i
p
e

I
n
t
e
r
n
a
l

/

e
x
t
e
r
n
a
l

d
i
a
m
e
t
e
r

v
a
r
i
a
t
i
o
n

(

)

M
i
n
i
m
u
m

P
i
p
e

t
y
p
e

S
m
o
o
t
h

/

c
o
r
r
u
g
a
t
e
d


P
i
p
e

d
i
a
m
e
t
e
r

(
m
m
)

4

d
i
a
m
e
t
e
r

3
0


C
F
4

B
e
i
n
g

c
o
m
p
a
t
i
b
l
e

w
i
t
h

e
x
i
s
t
i
n
g

p
i
p
e
s

P
i
p
e

m
a
t
e
r
i
a
l

P
l
a
s
t
i
c

/

r
u
b
b
e
r

O
p
e
r
a
t
i
o
n

n
u
m
b
e
r

M
i
n
i
m
u
m

C
F
5

B
e
i
n
g

u
s
e
r
-
I
r
i
e
n
d
l
y

O
p
e
r
a
t
i
o
n

c
o
m
p
l
e
x
i
t
y

B
a
s
i
c

(
n
o

s
p
e
c
i
I
i
c

s
k
i
l
l
)

Q
u
a
l
i
t
y

o
p
e
r
a
t
o
r
s

C
u
t
t
i
n
g

m
a
c
h
i
n
e

P
i
p
e

M
i
c
r
o
s
c
o
p
e

C
F
5

C
F
4

C
F
1

C
F
2

C
F
3

O
F
1

C
H
A
L
M
E
R
S
,

A
p
p
l
i
e
d

M
e
c
h
a
n
i
c
s
,

M
a
s
t
e
r
`
s

T
h
e
s
i
s

2
0
1
1
:
5
5

3
6

T
h
e
n
,

t
e
c
h
n
i
c
a
l

s
o
l
u
t
i
o
n
s

I
o
r

e
a
c
h

I
u
n
c
t
i
o
n

h
a
v
e

b
e
e
n

d
e
I
i
n
e
d
.

T
h
e
y

a
r
e

p
r
e
s
e
n
t
e
d

F
i
g
u
r
e

4
7

a
n
d

j
u
s
t
i
I
i
e
d

b
e
l
o
w
.



a
.

F
u
n
c
t
i
o
n

d
i
a
g
r
a
m

(
O
F
1

a
n
d

C
F
1
)

C
u
t
t
i
n
g

t
h
e

p
i
p
e

N
o
t

a
l
t
e
r
i
n
g

c
u
t
t
i
n
g

s
e
c
t
i
o
n

B
l
o
c
k
i
n
g

t
h
e

p
i
p
e

V
-
b
l
a
d
e

(
X
2
)

G
u
i
l
l
o
t
i
n
e

b
l
a
d
e

C
o
n
t
r
o
l
l
i
n
g

c
u
t
t
i
n
g

p
r
o
g
r
e
s
s

C
o
n
t
r
o
l
l
i
n
g

c
u
t
t
i
n
g

t
o
o
l

d
i
s
p
l
a
c
e
m
e
n
t

H
a
n
d
l
e

C
o
n
v
e
r
t
i
n
g

h
u
m
a
n

p
o
w
e
r

i
n

m
e
c
h
a
n
i
c
a
l

p
o
w
e
r

D
r
i
v
i
n
g

c
u
t
t
i
n
g

t
o
o
l

B
l
a
d
e
-
h
o
l
d
e
r

S
c
r
e
w

W
a
s
h
e
r

T
h
r
e
a
d
e
d

s
h
a
f
t

B
l
a
d
e
-
h
o
l
d
e
r

t
h
r
e
a
d
e
d

h
o
l
e

C
o
n
v
e
r
t
i
n
g

r
o
t
a
t
i
n
g

m
o
v
e
m
e
n
t

i
n

t
r
a
n
s
l
a
t
i
n
g

m
o
v
e
m
e
n
t

S
c
r
e
w
-
n
u
t

s
y
s
t
e
m

C
H
A
L
M
E
R
S
,

A
p
p
l
i
e
d

M
e
c
h
a
n
i
c
s
,

M
a
s
t
e
r
`
s

T
h
e
s
i
s

2
0
1
1
:
5
5

3
7


b
.

F
u
n
c
t
i
o
n

d
i
a
g
r
a
m

(
C
F
2
)


C
u
t
t
i
n
g

p
e
r
p
e
n
d
i
c
u
l
a
r

t
o

p
i
p
e

a
x
i
s

G
u
i
d
i
n
g

t
h
e

p
i
p
e

D
i
e

p
i
p
e

h
o
l
e

G
u
i
d
i
n
g

c
u
t
t
i
n
g

t
o
o
l

F
r
a
m
e

g
r
o
o
v
e

D
i
e

b
o
d
y

F
r
a
m
e

s
h
a
f
t

(
X
2
)

V
-
b
l
a
d
e

(
X
2
)

B
l
a
d
e
-
h
o
l
d
e
r

b
o
r
e

(
X
2
)

B
l
o
c
k
i
n
g

t
h
e

p
i
p
e

V
-
b
l
a
d
e

(
X
2
)

G
u
i
d
i
n
g

b
l
a
d
e
-
h
o
l
d
e
r

G
u
i
d
i
n
g

t
h
e

b
l
a
d
e

S
l
i
d
i
n
g

c
o
n
t
a
c
t

C
H
A
L
M
E
R
S
,

A
p
p
l
i
e
d

M
e
c
h
a
n
i
c
s
,

M
a
s
t
e
r
`
s

T
h
e
s
i
s

2
0
1
1
:
5
5

3
8


c
.

F
u
n
c
t
i
o
n

d
i
a
g
r
a
m

(
C
F
3

a
n
d

C
F
4
)

N
o
t

d
e
f
o
r
m
i
n
g

t
h
e

p
i
p
e

M
i
n
i
m
i
z
i
n
g

n
o
r
m
a
l

c
u
t
t
i
n
g

f
o
r
c
e

R
o
t
a
t
i
n
g

t
h
e

p
i
p
e

w
h
i
l
e

c
u
t
t
i
n
g

C
o
n
i
c
a
l

g
e
a
r

(
X
2
)

C
h
a
n
g
i
n
g

r
o
t
a
t
i
o
n

a
x
i
s

d
i
r
e
c
t
i
o
n

L
i
m
i
t
i
n
g

p
i
p
e

d
e
f
o
r
m
a
t
i
o
n

D
i
e

p
i
p
e

h
o
l
e

P
i
p
e
-
h
o
l
d
e
r

V
-
b
l
a
d
e

(
X
2
)

D
r
i
v
i
n
g

t
h
e

p
i
p
e

B
e
i
n
g

c
o
m
p
a
t
i
b
l
e

w
i
t
h

e
x
i
s
t
i
n
g

p
i
p
e
s

D
i
e

r
o
t
a
t
i
n
g

d
i
s
c

P
i
p
e

h
o
l
d
e
r

A
d
a
p
t
i
n
g

p
i
p
e

g
u
i
d
i
n
g

a
n
d

d
e
f
o
r
m
a
t
i
o
n

l
i
m
i
t
i
n
g

s
y
s
t
e
m

A
d
a
p
t
i
n
g

p
i
p
e

d
r
i
v
i
n
g

s
y
s
t
e
m

A
d
a
p
t
i
n
g

b
l
a
d
e

i
n
i
t
i
a
l

p
o
s
i
t
i
o
n

T
h
r
e
a
d
e
d

s
h
a
f
t

B
l
a
d
e
-
h
o
l
d
e
r

t
h
r
e
a
d
e
d

h
o
l
e

D
i
e

p
i
n

V
-
b
l
a
d
e

(
X
2
)

S
c
r
e
w
-
n
u
t

s
y
s
t
e
m

C
H
A
L
M
E
R
S
,

A
p
p
l
i
e
d

M
e
c
h
a
n
i
c
s
,

M
a
s
t
e
r
`
s

T
h
e
s
i
s

2
0
1
1
:
5
5

3
9


d
.

F
u
n
c
t
i
o
n

d
i
a
g
r
a
m

(
C
F
5
)


F
i
g
u
r
e

4
7

F
u
n
c
t
i
o
n

d
i
a
g
r
a
m


B
e
i
n
g

u
s
e
r
-
f
r
i
e
n
d
l
y

L
i
m
i
t
i
n
g

o
p
e
r
a
t
i
o
n

c
o
m
p
l
e
x
i
t
y

H
a
n
d
l
e

G
e
a
r

(
X
2
)

D
i
e

r
o
t
a
t
i
n
g

d
i
s
c

L
i
m
i
t
i
n
g

c
u
t
t
i
n
g

o
p
e
r
a
t
i
o
n

n
u
m
b
e
r

V
-
b
l
a
d
e

(
X
2
)

G
e
a
r

(
X
2
)

L
i
m
i
t
i
n
g

a
d
a
p
t
i
n
g

o
p
e
r
a
t
i
o
n

n
u
m
b
e
r

D
i
e

r
o
t
a
t
i
n
g

d
i
s
c

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 40
Function.
Cutting the pipe

Two solutions were considered:
laser system
guillotine blade

Laser system was abandoned because cutting section could be altered by melted material. Moreover such system
is very expensive compare to a simple guillotine blade.

Function.
Blocking the pipe
Driving the pipe
Guiding the pipe
Adapting driving system
Limiting adapting operation number

Two solutions were considered:
Drill tightening system
Crossed V-blades

Drill tightening system requires putting a pin inside the pipe otherwise the pipe would collapse. This pin should
be changed Ior each pipe diameter. Crossed V-blade solution is easier to design and manuIacture. Moreover its
geometry ensures pipe selI-centring and Ilexibility in terms oI dimensions. But Ieasibility has to be conIirmed,
especially concerning pipe insertion Iorce.

Function.
Rotating the pipe while cutting

Blade relative movement has been Iixed at 1 mm per revolution. In Iact, considering that most pipes are 1 mm
thick, it is a compromise to limit both normal cutting Iorce which deIorms the pipe and blade marks on cutting
section (one unique mark on the section) as illustrated Figure 48. However minimum thread Ior standard screw-
nut system oI this order oI dimension is 4 mm. ThereIore diameter ratio between pipe gear and blade gear should
be 1:4.



a. pipe immobile b. Pipe rotating

Figure 48 Pipe cutting Iorce
F
cN
= F
c

Blade
Pipe
F
c

F
cN

F
cT

Blade
Pipe
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 41
Function.
Rotating movement in translating movement
Adapting blade initial position

Two solutions were considered:
Cam system
Screw-nut system

A cam system is more complex to develop than a screw-nut system. In Iact, cam geometry should be speciIically
designed according to rod movement proIile. Moreover rod spring which purpose is to maintain a continuous
contact between came and rod should be well sized to ensure smooth operating conditions. Lastly, blade initial
position cannot be adjusted. ThereIore this solution was abandoned.

Function.
Guiding the pipe

Die thickness has been Iixed at 30 mm (maximum pipe diameter) to minimize the angle between cutting section
and pipe axis inside pipe hole as illustrated Figure 49: the thicker the die, the smaller the angle.


Figure 49 InIluence oI die thickness on pipe guiding

Function.
Guiding the blade

Guiding the blade is necessary to avoid it bending during cutting. This Iunction is perIormed by die body and a
groove machined in the Irame.

Function.
Driving cutting tool

Friction Iorce generated by screw tightening pressure through the washer allows blade holder driving directly the
blade. Thus screw works only in traction and not in shearing.

Function.
Limiting pipe deIormation

Pipe shape under cutting stress is an ellipsis. It is assumed that pipe deIormation during cutting operation can be
limited by stopping radial expansion inside die pipe hole. II necessary, an internal pin can still be added.

Function.
Adapting pipe guiding and deIormation limiting system
Limiting operation complexity
Limiting adapting operation number

This Iunction is perIormed by die rotating disc. It avoids having extra tooling in the case oI smooth pipes.

L
Pipe
Die
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 42
Lastly, cutting machine has been designed using some standard mechanical elements chosen Irom a catalogue.
Machine concept is presented Figure 50.


Figure 50 Cutting machine concept description

Working principle oI the machine is presented Figure 51.



a. blade in initial position b. blade in cutting position

Figure 51 Cutting machine working principle

Handle rotation drives simultaneously in rotation the pipe and in translation the blade which cuts the pipe.

Next step consists oI completing the development oI the machine then manuIacturing it. However it is out oI the
scope oI this thesis. Cutting machine bill oI material is presented in Appendix 3.
Die
Frame
Die pin
Pipe
Screw + washer
Blade
Threaded shaft
Blade holder
Handle
Pipe gear
Blade gear
V-blade (X2)
Pipe holder
Frame shaft (X2)
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 43
4.1.2.3 Automatic measurement program

Measurement steps are the Iollowing:
Conversion oI the picture in black and white colour
Measurement oI luminosity contrast gradients to identiIy layer boundaries
Calculation oI layer diameters using mathematical Iormula

Boundaries between layers can be identiIied since they are zones oI high contrast gradients. An example oI
measurement is presented Figure 52.


a. picture analysed b. layer diameters calculated

Figure 52 Example oI measurement with a Iirst version oI the program

Result accuracy depends on the requirements speciIied in section 4.1.2.1, especially on sample cutting. ThereIore
it is necessary completing cutting machine development beIore working on program development.

4.2. Structural controls

Structural tests allow characterising pipe mechanical properties. During validation phase, the purpose oI these
tests is to check that pipe characteristics match with target application. Requirements are based on pipe mounting
and operating conditions. In serial production, the purpose oI these tests is to monitor raw material and process
quality deviations. Tests are carried out in parallel with production activity. Requirements are based on structure
typical perIormance determined during validation phase.

4.2.1. Burst test

Burst test characterizes pipe resistance to radial load. It consists oI increasing the pressure in the pipe tested until
it bursts. Burst pressure value is recorded. It must be over a given limit, typically Iour times operating pressure.
A picture oI Hutchinson burst test bench is presented Figure 53.


Figure 53 Hutchinson burst test bench

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 44
4.2.2. Cold impact test

Cold impact characterizes the ductility oI pipe material at high energy level. It consists oI cooling down the pipe
tested until a given temperature, typically -40C, then impacting it with a suspended mass with a Iixed geometry.
At the end oI the test, no crack should appear on pipe surIace. A picture oI Hutchinson impact test apparatus is
presented Figure 54.


a. impact machine b. suspended mass

Figure 54 Hutchinson impact test apparatus

4.2.3. Tensile test

Tensile test characterizes the ductility oI pipe material at low energy level. It is perIormed on a dynamometer. It
consists oI elongating the pipe tested until it breaks. Instantaneous values oI bench load and sample length are
recorded during the test as shown Figure 55. For production control, traction Iorce is orientated in the direction
oI extrusion and monitored parameter is elongation at break. A picture oI Hutchinson dynamometer is presented
Figure 56.


Figure 55 Tensile curve Figure 56 Hutchinson dynamometer

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 45
4.2.3.1 Test conIiguration

Test conIiguration is deIined by customer speciIications. Two conIigurations are possible as presented Figure 57.
Gripping zone is by deIinition where jaws, pipe and pin interIere. Pin role is to avoid pipe collapsing under grip
tightening pressure.


Figure 57 Basic tensile test conIiguration

ISO samples (injected or cut) are speciIically designed Ior tensile test. Their section is smaller in the middle than
at the extremities to get higher stress level. Thus ISO sample elongation is mostly located in measurement zone
contrary to pipe elongation which propagates till jaw area. This phenomenon is unavoidable since pipe geometry
is uniIorm. ThereIore gripping system should be able to compensate pipe section diminution otherwise the pipe
will slide out oI the jaws as tightening pressure will decrease during the test.

4.2.3.2 SelI-closure mechanical gripping system

SelI-closure mechanical gripping systems, especially those originally designed Ior non ductile materials like
metals, are very sensitive to sample section reduction. ThereIore grips are equipped with teeth to reduce sliding
phenomenon. In Iact, tooth presence increases jaw and sample Iriction Iorces because it increases locally contact
pressure: the higher the pressure, the higher the Iorces. However pressure peaks which level depends on tooth
geometry generate inevitably stress peaks in pipe material which are also potential crack initiations: the sharper
the geometry, the higher the peaks. As a consequence, many test results are rejected due to anticipated breaks
caused by the grips. ThereIore it is necessary improving test reliability beIore working on pipe elongation at
break improvement. A picture oI a selI-closure mechanical gripping system is presented Figure 58.


Figure 58 SelI-closure mechanical gripping system

Sample length = 150 mm
Gripping zone = 30 mm
Pipe sample
Metal pin
Dynamometer jaw
Configuration 1
Configuration 2
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 46
A compromise has to be Iound between limiting sliding phenomenon to be able to perIorm the test until the end
and avoiding undesirable anticipated breaks as presented Figure 59. It depends mainly on:
grip and pin relative positions
jaw teeth pattern
grip initial tightening










Pipe sliding Pipe anticipated break caused bv the grips Pipe self break
Rejected Rejected Accepted

Figure 59 Tensile test result validity

Concerning grip and pin positions, quantity oI tests rejected is lower in conIiguration 2 than in conIiguration 1.
In Iact, last tooth marks on sample surIace are not Iree compare to conIiguration 1 as illustrated Figure 60. Thus
anticipated breaks will be postponed and as a consequence, quantity oI tests rejected is likely to be lower.



a. conIiguration 1 b. conIiguration 2

Figure 60 Pipe radial stress distribution in gripping zone

However, the problem is not solved, especially with stiIIer materials which load at given elongation is higher. In
Iact, sliding phenomenon is accelerated and as a consequence, or the sample slid out oI the grips or the break is
anticipated depending on grip initial tightening: the higher the strength, the earlier the anticipated break. Thus
two new designs oI jaws have been studied. They are presented Figure 61.


a. original jaw b. threaded jaw c. ringed jaw

Figure 61 Jaw conIiguration studied

Tooth interference
Jaw edge effect
Jaw edge effect
Pipe radial stress level
Tooth interference
Pin edge effect
Tooth interference
Jaw edge effect
Pipe radial stress level
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 47
Threaded pattern purpose is to maximize grip and sample interaction to reduce sliding phenomenon. This sharp
geometry results systematically in anticipated break. ThereIore thread teeth has to be enlarged to reduce material
stressing as illustrated Figure 62. However no signiIicant improvement is noticed with ring pattern compared to
original one.



a. threaded jaw b. ringed jaw

Figure 62 Pipe stress distribution in gripping zone

All tests perIormed and their respective results are compiled Table 9.

Table 9 Test results with mechanical gripping system
Original jaw Threaded jaw Ringed jaw
ConIiguration 1 - - - - - - - -
ConIiguration 2 - - - - -
Success rate: - low, - - - - very low

Finally, use oI selI-closure mechanical gripping system is not appropriate Ior plastic pipe samples as quantity oI
tests rejected is always high whatever adjustments are made. The reason is that a mechanical system is not able
to Iollow large sample section reduction. This issue can be solved using a pneumatic gripping system. A picture
oI a pneumatic gripping system is presented Figure 63.


Figure 63 Pneumatic gripping system

Tooth interference
Jaw edge effect
Jaw edge effect
Pipe radial stress level
Tooth interference
Jaw edge effect
Jaw edge effect
Pipe radial stress level
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 48
4.2.3.3 Pneumatic gripping system

Contrary to mechanical system, pneumatic system is able to maintain a constant pressure on the sample. When
compressed air inlet is equipped with a section reducer, average contact pressure can be signiIicantly higher than
with a manual closure. Thus sliding phenomenon is strongly reduced and teeth are not necessary anymore. Pipe
stress distribution in gripping zone Ior smooth jaws is illustrated Figure 64.



a. conIiguration 1 b. conIiguration 2

Figure 64 Pipe radial stress distribution in gripping zone

Quantity oI tests rejected is signiIicantly lower than with mechanical system, particularly in conIiguration 1. In
Iact, pin edge eIIect is lower than jaw edge one because the angle is smaller and can be rounded. Jaw edge does
not have any impact on test success when located at pipe extremity.

A Iinite element model analysis has been perIormed to corroborate previous explanations. Criterion chosen is
Von Mises stress calculated Irom the Iollowing Iormula:

2 2 2 2 2 2

. 6 . 6 . 6 ) ( ) ( ) (
2
1
xv :x v: x : : v v x Mises Jon
+ + + + + =

Gripping zone model is presented Figure 65.



a. schematic model b. FEM model

Figure 65 Gripping zone model

Model parameters are presented Table 10.

Table 10 Model parameters
element Behavour law Young modulus (MPa) Yield stress (MPa) Hardening
Pin Elastic 210000 - -
Pipe Elastic-plastic 300 28 100
Grip Elastic 210000 - -

Pipe sample
Metal pin Grip
Pipe sample
Metal pin Grip
Jaw edge effect
Jaw edge effect
Pipe radial stress level
Pin edge effect
Jaw edge effect
Pipe radial stress level
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 49
Two variables have been considered in this study:
tip edge rounding radius: see Figure 66
pin length: see Figure 67


a. R 0.2 mm

b. R 0.5 mm c. R 1 mm

d. R 2 mm e. R 5 mm I. R 10 mm

Figure 66 Tip edge rounding radius variation


a. pin end grip end b. pin end grip end 5 mm c. pin end grip end 5

Figure 67 Pin length variation

Results are presented in the graphs Figure 68.

50
55
60
65
70
75
0 1 2 3 4 5 6 7 8 9 10
Tip edge rounding radius (mm)
V
o
n

M
i
s
e
s

s
t
r
e
s
s

(
M
P
a
)

50
55
60
65
70
75
-10 -5 0 5 10
pin Iength variation (mm)
V
o
n

M
i
s
e
s

s
t
r
e
s
s

(
M
P
a
)

a. tip edge rounding radius b. pin length

Figure 68 Variation oI Von Mises stress in Iunction oI FEM analysis variables

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 50
Maximum Von Mises stress in gripping zone decreases when tip edge rounding radius increases and pin length
decreases. These results corroborate previous explanations. Figure 69 shows stress distribution in gripping zone
Iunction oI tip edge rounding radius.



a. R 0.2 mm

b. R 0.5 mm c. R 1 mm

d. R 2 mm e. R 5 mm I. R 10 mm

Figure 69 Stress distribution in gripping zone according to tip edge rounding radius

Finally, jaw and pin designs are presented Figure 70.



a. jaw b. pin c. assembly

Figure 70 Tensile test tool design

Jaw dimensions are compatible with pipe diameters ranging Irom 4x6 to 13,5x16 that covers pipe dimensions.
As load at given elongation is higher Ior larger diameters, Iriction Iactor between plastic and aluminium is too
low to avoid the pipe sliding beIore the end oI the test. A solution could be to change jaw material.

Pin external diameter matches with pipe minimum internal diameter to limit the interIerence between pin and
pipe. Tip edge has a radius oI 60 mm to minimize edge eIIect. Pin base is slightly higher than pipe maximum
internal diameter to ensure keeping the pin inside the pipe during test set-up.

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 51
4.2.3.4 Measurement method

Elongation is deduced Irom the Iollowing Iormula:

100

=
i
i f
l
l l

With:
c Elongation
l
i
Sample actual initial length
l
f
Sample actual Iinal length

Pipe elongation at break can be measured using an optical or a mechanical extensometer. Optical extensometer
measures the distance between two laser reIlecting markings stuck to the sample. Elongation is then calculated
automatically by the system. Mechanical extensometer is composed oI two mobile Iorks carried by the sample
during the test. Elongation is deduced Irom the distance between the Iorks. Contact pressure on the pipe is light
enough so that Iorks are not sensible to material elastic relaxation aIter the break. Optical extensometer is the
most accurate system.

4.2.3.5 Result interpretation

Final test conIiguration is presented Figure 71.


Figure 71 Final test conIiguration

30 mm
40 mm
50 mm
Laser reflecting marking
150 mm
Jaw end marking
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 52
Maximum elongation gradient is located around break zone. Moreover measured value is the average oI sample
local elongations in measurement zone. Thus when a sample breaks out oI measurement zone, elongation at
break is underestimated and iI the break occurs out oI grip end initial positions, rupture mechanism is probably
inIluenced by pin and jaw interaction with the pipe. In the last case, the break is anticipated compare to pipe selI
rupture and as a consequence, elongation at break is highly underestimated. Test result interpretation is presented
Figure 72.


Figure 72 Test result interpretation (break elongation at break)

4.3. Conclusion

BeIore working on extrusion process improvement, it is necessary being able to monitor extruded pipe essential
characteristics.

Monitored essential characteristics oI extruded pipes and their respective controls are divided in two categories:
dimensional controls:
non destructive controls:
ultrasonic system: total thickness
laser system: external diameter
destructive controls:
microscope instrument: internal diameter, external diameter, intermediate layer
thickness, total thickness
structural controls:
burst test: burst pressure
cold impact test: cold impact resistance
tensile test: elongation at break

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 53
4.3.1. Dimensional controls

Feasibility oI substituting destructive controls by non destructive controls extending use oI ultrasonic and laser
systems has been studied in anticipation oI the study on process reliability improvement that will be carried out
in 2012. However neither intermediate layer thickness oI multilayer pipes nor internal diameter can be controlled
with current system characteristics. As a consequence, these characteristics can only be controlled by cutting the
pipe then analysing its section with a microscope. Current control method was acceptable but could be improved
to reduce measurement variability. Thus a concept oI cutting machine and a program to automate measures have
been presented. However their development was out oI the scope oI this thesis.

4.3.2. Structural controls

Current tensile test has been improved to have a reliable method to evaluate pipe elongation at break in the next
chapter. In Iact, many test results were rejected because oI anticipated breaks caused by the grips. Thus current
mechanical gripping system has been substituted by a pneumatic gripping system and new jaws have been
speciIically designed to limit anticipated breaking phenomenon. Finally new test conIiguration allows reducing
signiIicantly the quantity oI test results rejected.
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 54
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 55
5. Extrusion performance improvement: results and discussion

A reason Ior working on process improvement is when some essential characteristics oI the product do not meet
customer speciIication requirements. For plastic pipes, elongation at break is an important characteristic. On new
projects in development, minimum value required Ior pipe elongation at break has been increased and is not
reached with current process conIiguration. Thus the purpose oI this thesis is to improve this characteristic.

5.1. Current process optimization

A design oI experiment has been carried out on PA9T/PA12 and PA6/EVOH/ADH/PA12 structures to improve
elongation at break. Target value oI elongation at break is 200. This experiment Iocuses on sizing operation
and parameters have been deduced Irom chapter 3 (see 3.4.). They are the Iollowing:
Iorming tool dimensions: D ratio, T ratio, DDR
water temperature
extrusion line speed
sizing sleeve internal diameter: inIluence vacuum

Parameter combination Ior each test is presented Table 11.

Table 11 Test parameter combination

a. PA9T/PA12 structure
Test N D-ratio T-ratio D sleeve
|mm|
Line speed
|m/min|
6 2,05 1,45 10,85 20
4 2,05 3,35 10,85 20
7 1,78 2 10,85 20
8 1,78 2 10,95 20
12 1,42 1,45 10,85 20
13 1,42 1,45 10,95 20
9 1,42 2 10,85 20
10 1,42 2 10,95 20
15 1,42 2 10,85 10
16 1,69 1,55 10,85 20
17 1,69 1,55 10,85 10

b. PA6/EVOH/ADH/PA12 structure
Test N D-ratio T-ratio Water temp. |C| Line speed
|m/min|
1 1,73 1,8 12 40
2 1,73 1,8 6 40
3 1,51 1,77 12 40
4 1,73 1,8 12 30
5 1,73 1,8 12 20

Vacuum level and water Ilow have been voluntary excluded Irom the design oI experiment. In Iact, vacuum level
determines pipe dimensions (see 3.4.2.) that are Iixed during the experiment otherwise it would not be possible
to compare elongation at break. ThereIore vacuum level should be adjusted Ireely to keep pipe dimensions
constant between the diIIerent tests. Moreover vacuum level inIluences lubrication thickness inside sizing sleeve
at constant water Ilow (see 3.3.3.2). II vacuum level increases, lubrication will be scarce at the beginning oI the
sleeve where material is not solidiIied yet and as a consequence, the pipe could stick locally on the sleeve
aIIecting process stability. ThereIore water Ilow should be adjusted Ireely according to vacuum level to keep a
stable process as shown Figure 73 where water Ilow increases with increasing vacuum level.
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 56
40
50
60
70
80
90
100
110
120
130
140
100 150 200 250 300 350
Vacuum (mbar)
W
a
t
e
r

c
o
I
u
m
n

(
m
m
)

Figure 73 InIluence oI vacuum level on water Ilow

At given vacuum level, large deviations can be observed on water Ilow because water Ilow is adjusted according
to operator Ieeling. The eIIect oI water Ilow adjustment is limited to the beginning oI the sleeve since increasing
vacuum level reduces lubrication length (see 3.3.3.2). Thus average Iriction along the sleeve increases when
vacuum level increases and as a consequence, pipe elongation at break decreases anyway (see 3.3.3.2). ThereIore
correlation between water Ilow and elongation at break is not as obvious as Ior vacuum level as it can be
observed Figure 74 e and Figure 74 I.

140
150
160
170
180
190
200
210
220
230
240
1,4 1,6 1,8 2
D ratio
E
I
o
n
g
a
t
i
o
n

a
t

b
r
e
a
k

(
%
)
20 m/min 10 m/min

140
150
160
170
180
190
200
210
220
230
240
1,4 1,9 2,4 2,9 3,4
T ratio
E
I
o
n
g
a
t
i
o
n

a
t

b
r
e
a
k

(
%
)
20 m/min 10 m/min

a. D ratio b. T ratio
140
150
160
170
180
190
200
210
220
230
240
2,0 3,0 4,0 5,0 6,0 7,0
DDR
E
I
o
n
g
a
t
i
o
n

a
t

b
r
e
a
k

(
%
)
20 m/min 10 m/min

140
150
160
170
180
190
200
210
220
230
240
5 10 15 20 25
Line speed (m/min)
E
I
o
n
g
a
t
i
o
n

a
t

b
r
e
a
k

(
%
)

c. DDR d. extrusion line speed
140
150
160
170
180
190
200
210
220
230
240
40 60 80 100 120 140
Water coIumn (mm)
E
I
o
n
g
a
t
i
o
n

a
t

b
r
e
a
k

(
%
)

140
150
160
170
180
190
200
210
220
230
240
100 150 200 250 300 350
Vacuum (mbar)
E
I
o
n
g
a
t
i
o
n

a
t

b
r
e
a
k

(
%
)

e. water Ilow I. vacuum level

Figure 74 InIluence oI experiment parameters on pipe elongation at break
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 57
It can be deduced by comparing the diIIerent graphs Figure 74 that vacuum is the best indicator when working
on pipe elongation at break improvement: the lower the vacuum, the higher the elongation at break.

Vacuum level is inIluenced by extrusion line speed. In Iact, vacuum level required to size the pipe is lower at
lower speed as shown Figure 75 a. ThereIore Iriction is lower and pipe elongation at break is higher.

Vacuum level is also inIluenced by Iorming tool dimensions, especially D ratio. In Iact, vacuum is adjusted to Iix
pipe external diameter (see 3.3.). ThereIore when extruded pipe external diameter is closer to Iinal one, vacuum
level required is lower as shown Figure 75 b and pipe elongation at break is higher. Other ratios have a minor
inIluence.

100
120
140
160
180
200
220
240
260
5 10 15 20 25
Line speed (m/min)
V
a
c
u
u
m

(
m
b
a
r
)

100
150
200
250
300
350
1,4 1,6 1,8 2
D ratio
V
a
c
u
u
m

(
m
b
a
r
)

a. extrusion line speed b. D ratio

Figure 75 InIluence oI experiment parameters on vacuum level

Increasing sleeve internal diameter decreases vacuum level as shown Figure 76. However over a given diameter,
sizing operation is not stable anymore.

100
150
200
250
300
350
400
10,7 10,75 10,8 10,85 10,9 10,95 11
SIeeve diameter (mm)
V
a
c
u
u
m

(
m
b
a
r
)
l
n
s
L
a
b
l
e

Figure 76 InIluence oI sizing sleeve internal diameter on vacuum level

Lastly, water temperature does not seams to have a signiIicant inIluence on pipe elongation at break.
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 58
Results oI the experiment are summarized Figure 77.

155
154
145
175
179
231
155
205
172
159 159
209
185
239
178
223
166
156
152
192
183
234
170
216
130
150
170
190
210
230
250
4 6 7 8 9 10 12 13 13 16 17
Test number
E
I
o
n
g
a
t
i
o
n

a
t

b
r
e
a
k

(
%
)
min max average
l
n
s
L
a
b
l
e
l
n
s
L
a
b
l
e
l
n
s
L
a
b
l
e

a. PA9T/PA12 8x1 structure

163
171
153
152
208
171
184
189
176
238
167,0
176,0
166,0
164,0
227,0
130
150
170
190
210
230
250
1 2 3 4 3
Test number
E
I
o
n
g
a
t
i
o
n

a
t

b
r
e
a
k

(
%
)
min max average
l
n
s
L
a
b
l
e

b. PA6/EVOH/ADH/PA12 8x1 structure

Figure 77 Results oI the experiment

For PA9T/PA12 structure, best results are obtained Ior a line speed oI 10 m/min and a D ratio oI 1,42 as it could
be expected Irom previous explanations. Line speed is lower than current one oI 40 m/min which would impact
process cost. Moreover, even with optimized parameters, some pipes are statistically below 200 oI elongation
at break that is not acceptable. For PA6/EVOH/ADH/PA12 structure, conclusion is similar but Ior a line speed oI
20 m/min.

Pipe elongation at break decreases when material molecular orientation and quantity and size oI deIects on pipe
surIace increase (see 2.9.3.). High molecular orientation is concentrated in pipe external layer, typically 200 m
(see 3.3.2.2). Thus a solution to improve elongation at break would be to remove pipe external layer molecular
orientation and deIects by melting the surIace oI the pipe again aIter sizing operation, in other terms, to equip
extrusion line with a Ilaming treatment.
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 59
5.2. Flaming treatment addition

Flaming treatment purpose is to remove pipe external layer molecular orientation and deIects by melting the
surIace oI the pipe again aIter sizing operation as illustrated Figure 78.


Figure 78 Schematic oI Ilaming treatment principle

Two gases can be used Ior this treatment: methane (natural gas) and propane. Heating value oI methane gas is
higher than that oI propane gas. Combustion equation Ior each gas is given below:

Methane.

CH
4
2 O
2
CO
2
2 H
2
O

Propane.
C
3
H
8
5 O
2
3 CO
2
4 H
2
O

Flaming treatment main parameters are:
gas nature: treatment intensity
gas Ilow: treatment intensity
air /gas mass ratio: treatment eIIiciency (see Table 12)

Treatment intensity corresponds to the thermal power provided to the pipe. At given gas Ilow, treatment intensity
depends on extrusion line speed and pipe dimensions.

Table 12 Air / gas mass ratio
Gas Stoichiometric During trial
Propane 15.5 16
Methane 17.2 not deIined

Zone isotropic
Zone melted during
flaming treatment
= 0.2 mm
Pipe outer surface
Cooling
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 60
Impact oI Ilaming treatment on pipe elongation at break has been evaluated through a new design oI experiment
always carried out on PA9T/PA12 and PA6/EVOH/ADH/PA12 structures. Experiment parameters are:
Ilaming intensity
extrusion line speed

Parameter combination Ior each test is presented Table 13.

Table 13 Test parameter combinations

a. PA9T/PA12 structure
Test N Flaming treatment* Line speed |m/min|
1A No 20
2.1 20
2A No 40
2.2 40
3A No 60
2.3 60
4A No 100
2.4 100
C1 40
D1 40
E1 40
*Flaming intensity: low, medium, high, maximum

b. PA6/EVOH/ADH/PA12 structure
Test N Flaming treatment* Line speed |m/min|
1A No 20
2.1 20
2A No 40
2.2 40
3A No 60
2.3 60
4A No 100
2.4 100
*Flaming intensity: maximum

Three burners with their control unit have been rent specially Ior the trial. Flaming unit layout on extrusion line
is presented Figure 79.


b. burners







a. schematic c. cooling bath

Figure 79 Flaming unit layout on extrusion line, Rivoly, Italy
Burner
pipe
Water tank
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 61
Flaming treatment eIIiciency depends by deIinition on the capacity to heat up pipe surIace over material melting
point, typically 165C Ior polyamide 12. Pipe surIace temperature along extrusion line has been measured Ior
diIIerent test conditions as presented Figure 80. Measurement has been made with an inIrared detector.


Figure 80 Pipe surIace temperature along extrusion line

Temperature aIter cooling bath increases when line speed increases because cooling time decreases. Temperature
aIter Ilaming unit decreases signiIicantly when line speed exceeds 60 m/min. It means that the thermal power
provided to the pipe is too low over 60 m/min and as a consequence, that an additional burner should be installed
on the line. Temperature aIter Ilaming unit cooling bath increases when line speed increases. It shows that the
bath could be extended. Lastly temperature between cooling bath and caterpillar increases because external layer
oI the pipe is heated up by the core oI the pipe (see 3.3.1).

InIluence oI pipe surIace temperature aIter Ilaming treatment on pipe elongation at break is presented Figure 81.

100
120
140
160
180
200
220
240
260
0 50 100 150 200
Pipe surface temperature after fIaming treatment T2 (C)
E
I
o
n
g
a
t
i
o
n

a
t

b
r
e
a
k

(
%
)

Figure 81 InIluence oI pipe surIace temperature aIter Ilaming treatment on pipe elongation at break

Pipe elongation at break drops down when pipe surIace temperature aIter Ilaming treatment is under 160C. This
result corroborates previous explanations.
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 62
Results oI the experiment are summarized Figure 82.

203
234
122
234
110
166
130
177
206
238
130
254
140
193
150
184
205
236
127
243
124
182
140
181
100
120
140
160
180
200
220
240
260
280
1A 2.1 2A 2.2 3A 2.3 4A 2.4
Test number
E
I
o
n
g
a
t
i
o
n

a
t

b
r
e
a
k

(
%
)
min max average

a. PA9T/PA12 8x1 structure

226
246
136
251
110
218
144
231
229
257
150
274
178
241
166
245
227
253
145
266
147
232
155
238
100
120
140
160
180
200
220
240
260
280
300
1A 2.1 2A 2.2 3A 2.3 4A 2.4
Test number
E
I
o
n
g
a
t
i
o
n

a
t

b
r
e
a
k

(
%
)
min max average

b. PA6/EVOH/ADH/PA12 8x1 structure

Figure 82 Experiment results

Without Ilaming treatment, results are similar to those oI the Iirst design oI experiment: pipe elongation at break
decreases strongly over 20 m/min. Adding Ilaming treatment allows increasing pipe elongation at break up to
100 and achieving target value oI 200. Treatment eIIect is reduced over 40 m/min which conIirms that an
additional burner should be installed on the line. However extrusion line equipped with Iour burners could be run
until 80 m/min keeping pipe mechanical properties meeting requirements which would allow in the meantime
increasing production rate and as a consequence, improving process perIormance. Lastly, pipe mechanical
properties are less dependent on sizing conditions with a Ilaming treatment. Thus Iorming tool dimensions can
be chosen independently Irom sizing operation Iocussing exclusively on process reliability. This last point is
important in anticipation oI the study on process reliability improvement that will be carried out in 2012.
speed
speed
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 63
6. Conclusion

This thesis was the introductive part oI a study on plastic pipe extrusion process improvement. Plastic pipes Ior
automotive application are essentially composed oI semi-crystalline polyamides. An important characteristic oI
semi-crystalline materials is their melting point that corresponds to crystal melting temperature. Polymers
behave like viscous Iluids above this temperature and like solids below.

6.1. Pipe dimensional properties: improvement of process reliability

The Iirst purpose oI working on process improvement is to reduce production cost which depends on process
reliability and line speed in the case oI extrusion. A long-term study Iocussed on pipe dimensional properties
will be carried out in 2012 to improve process reliability. The Iirst purpose oI this thesis was to prepare this
study.

The main Iactors oI the process that inIluence pipe dimensional properties have been identiIied and justiIied.
They are the Iollowing:
Ieeding regularity inside the extruder:
no Iluctuations oI pipe dimensions
melting rate inside the extruder:
no inclusions oI non melted material in the pipe
Ieeding regularity
melt viscosity at the exit oI extrusion machine:
no distortions at pipe layer interIace
draw down inside sizing sleeve:
pipe thickness
vacuum inside water tank:
pipe external diameter
pipe thickness

Extrusion line parameters that allow controlling above Iactors have also been identiIied and are presented Table
14 with their respective inIluence on process Iactors.

Table 14 EIIect oI increasing extrusion line parameters on process Iactors
Factors (eIIect)
Parameters ()
Feeding
regularity
Melting rate Melt viscosity Draw down Vacuum
Screw speed
+ - -`
Barrel temperature in Ieeding zone
-
Barrel temperature in compression zone
+ / -
Barrel temperature in metering zone

Head temperature

Forming tool dimensions
+
Forming tool temperature
+
Extrusion line speed
+``
Vacuum level
+
* at constant extrusion line speed
** at constant screw speed

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 64
Impact oI each parameter on process Iactors and pipe dimensional properties will be studied in detail during the
study on process reliability improvement.

BeIorehand, extrusion line parameters to monitor process Iactors have been identiIied and are presented Table 5.
Moreover essential characteristics to monitor pipe dimensional properties and their respective controls have also
been identiIied and are the Iollowing:
non destructive controls:
ultrasonic system: total thickness
laser system: external diameter
destructive controls:
microscope instrument: internal diameter, external diameter, intermediate layer thickness, total
thickness

A study has been carried out to improve current dimensional controls. It shows that is not possible to substitute
destructive controls by non destructive controls with current ultrasonic and laser system characteristics. As a
consequence, pipe dimensional properties can only be controlled by cutting the pipe then analysing its section
with a microscope. Current control method was acceptable but it could be improved to reduce measurement
variability. Thus a concept oI cutting machine and a program to automate measures have been presented. Their
development will be completed during the study on process reliability improvement.

6.2. Pipe mechanical properties: improvement of pipe elongation at break

Another reason Ior working on process improvement is when some essential characteristics oI the product do not
meet customer speciIication requirements. For plastic pipes, elongation at break is an important characteristic.
On new projects in development, minimum value required Ior pipe elongation at break had been increased and
was not reached with current process conIiguration. The purpose oI this thesis was to improve this characteristic.

The main Iactors oI the process that inIluence pipe mechanical properties, especially elongation at break, during
sizing operation have been identiIied and justiIied. They are presented Table 6.

Extrusion line parameters that allow controlling above Iactors have also been identiIied and are presented Table
15 with their respective inIluence on process Iactors.

Table 15 EIIect oI increasing extrusion line parameters on process Iactors
Factors (eIIect)
Parameters ()
Cooling rate Draw down Iriction Vacuum
Forming tool dimensions
+
Water temperature
-
Extrusion line speed
+
Water Ilow
-
Vacuum level
+ +
Sizing sleeve dimensions
- -

It was noticed that many Iactors are interdependent. For instance, plastic melt viscosity at the exit oI extrusion
machine and pipe draw down inside sizing sleeve depend both on Iorming tool dimensions.

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 65
A design oI experiment has been carried out on PA9T/PA12 and PA6/EVOH/ADH/PA12 structures to improve
pipe elongation at break. Target value oI elongation at break was 200. This experiment Iocused on sizing
operation and parameters were deduced Irom Table 15.

BeIorehand, current tensile test has been improved to have a reliable method to evaluate elongation at break. In
Iact, many test results were rejected because oI anticipated breaks caused by the grips. New test conIiguration
allows reducing signiIicantly the quantity oI test results rejected.

Best results were obtained Ior extrusion line speeds lower than current ones which would impact process cost.
Moreover, even with optimized parameters, some pipes were statistically below 200 oI elongation at break that
was not acceptable. A solution was to equip extrusion line with a Ilaming treatment to remove pipe external layer
molecular orientation and deIects by melting the surIace oI the pipe again aIter sizing operation. Adding Ilaming
treatment allows increasing pipe elongation at break up to 100 and achieving target value oI 200. Moreover
extrusion line can be run until 80 m/min keeping pipe mechanical properties meeting requirements which would
allow in the meantime increasing production rate and as a consequence, improving process perIormance. Lastly
pipe mechanical properties are less dependent on sizing conditions with a Ilaming treatment. Thus Iorming tool
dimensions can be chosen independently Irom sizing operation Iocussing exclusively on process reliability. This
last point is important in anticipation oI the study on process reliability improvement that will be carried out in
2012.
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 66
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 67
Appendix 1: Plastic pipe product presentation

Hutchinson products

The principal Iunction oI a pipe is to transIer a Iluid. Pipe characteristics depend on vehicle application:
Assembly conditions:
Ergonomics: mounting eIIort, Ilexibility.
Operating conditions:
Fluid: nature, temperature, pressure.
Environment: temperature, vibrations, aggressive elements.

Hutchinson produces plastic pipes Ior diIIerent applications. Main ones are:
Fuel transIer: diesel liquid, gasoline liquid, gasoline vapour
cooling: water liquid, water vapour (degassing)
valve piloting: air vacuum

Examples oI Hutchinson products are presented Figure 83.


a. Iuel transIer pipe b. cooling pipe c. valve piloting pipe

Figure 83 Example oI Hutchinson products

A pipe also called structure can be smooth or corrugated as presented Figure 84 and composed oI one or several
layers oI material. On Iinal product, it can be assembled with other components such as:
Connector
Rubber
Protective sleeve
Tape
Clip



a. smooth pipe b. corrugated pipe

Figure 84 Example oI pipe conIigurations

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 68
Multilayer structures

Multilayer structures are more complex to extrude than monolayer ones:
extrusion oI several materials which processing conditions can diIIer signiIicantly
tolerances on intermediate layer dimensions are smaller than those oI monolayer structures

These structures were developed Iew years ago Ior two main reasons:
Technical: new environmental norms
Economical: cost reduction

New environmental norms constrained car manuIacturers to reduce drastically vehicle Iuel emissions. However
parts oI these emissions depend directly on permeability oI the materials used Ior Iuel transIer and storage. Thus
raw material suppliers had to develop new structures including an intermediate 'barrier layer to stop Iuel
particle migration. Concerning economical aspect, the possibility to dissociate internal and external Iunctions oI
the pipe allows selecting the cheapest material Ior each layer. Thus, Ior instance, Iunctions oI PA9T/PA12 and
PA6/EVOH/ADH/PA12 structure layers are presented Table 16.

Table 16 Structure layer Iunctions

a. PA9T/PA12 structure
Material Technical Iunction Cost
PA9T Compatibility with the Iluid oI the application (E85, 110C, 3.5 bars relative.)
PA12 Compatibility with the environment oI the application (110C, engine vibrations.)

b. PA6/EVOH/ADH/PA12 structure
Material Technical Iunction Cost
PA6 Compatibility with the Iluid oI the application (E10, 110C, 3.5 bars relative.)
EVOH Permeability to the Iluid oI the application: 'barrier layer
Bonding Adhesion between EVOH and PA12 layers
PA12 Compatibility with the environment oI the application (110C, engine vibrations.)

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 69
Appendix 2: Product quality management

Product quality management procedure is illustrated Figure 85. It is divided in three steps.

Characterisation

Product required characteristics are identiIied Irom speciIications then they are quantiIied or qualiIied depending
on their nature. Thus an exhaustive list oI parameters is drawn up with all criteria that manuIactured product
should meet to ensure that each requirement is IulIilled.

Design

Product and manuIacturing process are designed in parallel since they are interdependent. Product deIinition is
based on product required characteristics identiIied in the previous section.

Quality control

A Failure Mode and EIIects Analysis (FMEA) is carried out Irom product deIinition, process description and
product required characteristics. Risks relative to process deviations are preIerentially eradicated by mistake-
prooIing. When it is not possible, product characteristics are controlled during serial production or iI it is not
possible to measure them directly, equivalent monitored parameters are identiIied. Lastly, risks depending on
product design are removed by carrying out validation tests during development phase. First version oI FMEA
document is created early in development phase then it is continuously enriched by experience until the end oI
serial liIe.
CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 70

Figure 85 - Product quality management procedure

Monitored parameters, measurement method and control Irequency are speciIied into the control plan. A batch oI
production is blocked until all items oI the control plan are completed.
Extraction
Specification requirements
Technical and functional characteristics
Nominal value, tolerances
OK / NOK
OR
Definition Definition
Related to:
Mistake-proofing Validation plan
Design
C
H
A
R
A
C
T
E
R
I
S
A
T
I
O
N

Identification
Monitored essential characteristics
Nominal value, tolerances
OK / NOK
OR
Process Product
Q
U
A
L
I
T
Y

C
O
N
T
R
O
L

Test
Results
If negative result
Control plan
Failure Mode and Effects Analysis (FMEA)
Design
Product / Manufacturing process
Analysis
D
E
S
I
G
N

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 71
Appendix 3: Cutting machine bill of materials

Cutting machine components are listed below:

Frame

Frame part1

Frame shaIt (X2)

Frame part2

Screw M8x16 (X3)

Handle

Pipe holder

V-blade Iemale

V-blade male

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 72
Pipe holder gear

Blade holder gear

Threaded shaIt

Blade holder

Screw M8x12

Washer

Guillotine blade

Die

Die pin


CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 73
References

CARI05] Etude experimentale et modelisation thermomecanique de letape de calibration dans le procede
dextrusion de tubes en polvamide 12, Aurelien Carin, 2005
HENS97] Plastics extrusion technologv, Hensen, 1997
MIAN09] Lestrusione con parole mie, Mario Miani, 2009
MICH91] Extrusion dies for plastics and rubber, walter Michaeli, 1991
PILL05] Appliquer la maitrise statistique des processus MSP/SPC, Maurice Pillet, 2005
RAUW01] Polvmer extrusion, Chris Rauwendaal, 2001
RAUW10] Understanding extrusion, Chris Rauwendaal, 2010
ZUMB07] Zumbach instrument instruction manual, 2007

CHALMERS, Applied Mechanics, Master`s Thesis 2011:55 74