DEVELOPMENT AND ANALYSIS OF A FRICTION STIR SPOT ...

DEVELOPMENT AND ANALYSIS OF A

FRICTION STIR SPOT WELDING

PROCESS FOR ALUMINIUM

By

MICHAEL STEPHEN

A DISSERTATION IN COMPLIANCE WITH THE FULL

REQUIREMENTS FOR THE DEGREE OF

Magister Technologiae: Engineering: Mechanical

In the

FACULTY OF ENGINEERING

NELSON MANDELA METROPOLITAN UNIVERSITY

January 2005

Promoter: Prof. DG Hattingh

Co-Promoters: Z. Georgeou and A. Els-Botes

Submitted on……………………

Signed…………………………..

ii

The copy of this dissertation has been supplied on condition that anyone

who consults it is understood to recognize that its copyright rests with the

author and no quotation from the dissertation and no information derived

from it, may be published without the author’s prior consent, unless

correctly referenced.

iii

ABSTRACT

Author: Michael Stephen

Title: Development and Analysis of a Friction Stir Spot Welding

Process for Aluminium

Friction Stir Spot Welding (FSSW) has been developed from the

conventional Friction Stir Welding (FSW) process, developed at The

Welding Institute (TWI). FSSWs have been done without the keyhole being

eliminated. Elimination of the keyhole would result in the process being

more commercially viable.

This dissertation focuses on an attempt of eliminating the keyhole using a

retractable pin tool as well as a comparison of the weld integrity of a FSSW

to that of a conventional Resistance Spot Weld (RSW). Welds were

conducted on aluminium alloy 6063 T4. Comparisons between different

weld procedures were done. Further analysis of the weld integrity between

FSSW and RSW were conducted, comparing tensile strengths,

microstructure and hardness.

For the above welding procedure to take place, the current retractable pin

tool, patented by PE Technikon, was redesigned. Problems associated

during the welding process and the results obtained are documented.

Reasons for the keyhole not being eliminated as well as recommendations

for future work in the attempt to remove the keyhole are discussed.

iv

ACKNOWLEDGEMENTS

I would like to extend my special thanks to both my promoter and co-

promoters:

• Professor Danie G. Hattingh, (Promoter) whose encouragement and

guidance throughout the year lead to the completion of my

dissertation. His understanding and friendly approach, regarding all

matters, made the journey travelled easier and enjoyable.

• Zacharias Georgeou, (Co-promoter) whose friendship and guidance

was invaluable. His insight into FSW as well as life was greatly

appreciated and all the lessons learnt will not be forgotten. He

guided me in all the right directions and our discussions during lunch

breaks and walks to the cafeteria proved to be relaxing and very

valuable.

• Annelize Els-Botes, (Co-promoter) whose help behind the scenes

and in the metallurgy lab was greatly appreciated.

I would like to thank the staff members of the Automotive Components

Technology Station (ACTS):

• Lucinda Lindsay for all her hard work and for being accommodating

at all times.

• Andrew Young, William Rall and Ian Wedderburn for their

contributions they made in the tool design process.

I would also like to thank my fellow researchers:

• Calvin Blignault and Grant Kruger for their help with regard to the

FSW process and the use of the machine.

• Basil Esterhuysen, for the good entertainment, as well as the games

of squash after work that helped relax and clear our minds.

v

• Greg North whose great sense of humour and friendship helped me

get through the hard times.

• Toa Hua who was always there and would be prepared to help with

anything when needed.

Further thanks also goes to:

• Ian Clarke whose good sense of humour kept me smiling and

motivated at all times.

• Gideon Gouws for his help concerning the tool manufacture and

laboratory testing.

• The National Research Foundation who funded my project and who

gave me the opportunity to broaden my horizons.

• My family who have supported me in everything I have done. To my

parents, Roger and Margaret, and my brother Graham, thanks for all

the support and encouragement.

vi

TABLE OF CONTENTS

Abstract iii

Acknowledgements iv

List of Figures xi

List of Table xix

Glossary of Terms xxii

CHAPTER 1

INTRODUCTION

1.1 Introduction ................................................................................1

1.2 The Friction Stir Spot Welding Process ......................................3

1.3 FSW and FSSW Application ......................................................5

1.4 Motivation..................................................................................7

1.5 Objectives ..................................................................................8

1.5.1 FSSW Tool Development ...............................................8

1.5.2 Pin Profile and Retraction Studies ...................................9

1.5.3 Material Flow Evaluation................................................9

1.5.4 Spot Weld Evaluation .....................................................9

1.5.5 Delimitating Research.....................................................10

1.6 Summary ....................................................................................11

CHAPTER 2


2.1 Introduction .............................................................................12

2.2 Development of Spot FSW Robot Systems for Automobile

Body Members ........................................................................12

2.3 A Preliminary Investigation on the Static Properties of

Friction Stir Spot Welds ..........................................................17

2.4 GKSS .....................................................................................21

vii

2.5 Lap Joints of Aluminium Alloys by Friction Stir Welding .......22

2.6 Force Characterisation on the Welding Pin of a Friction

Stir Welding Retractable Pin-Tool using Aluminium-

Lithium 2195 ...........................................................................22

2.7 Material Selection ...................................................................24

2.8 Distributions of Hardness and Microstructure in Friction

Stir Welds of Al Alloy 6063 ....................................................26

2.9 Temperature and Flow Patterns ...............................................28

2.9.1 Numerical and experimental study of the heat

transfer process in friction stir welding ........................29

2.9.2 Thermal modelling of friction stir welding in a

moving coordinate system and its validation ................30

2.9.3 Analysis of material flow around a retractable

pin in a friction stir weld ..............................................31

2.10 Resistance Spot Welding .........................................................32

2.11 Summary .................................................................................33

CHAPTER 3

FSSW TOOL DEVELOPMENT

3.1 Introduction .......................................................................34

3.2 Design Process for the Development of the FSSW Tool ..........35

3.2.1 Identification of Need ..................................................35

3.2.2 Problem Statement and Goal Definition .......................36

3.2.3 Research ......................................................................36

3.2.3.1 Auto-Adjustable Pin Tool for Friction Stir

Welding .........................................................37

3.2.3.2 Programmable Friction Stir Welding

Process ..........................................................39

3.2.3.3 Adjustable Pin for Friction Stir Welding ........40

3.2.3.4 Friction Stir Welding Tool for Welding

Variable Thickness Work Pieces ....................41

viii

3.2.3.5 Method and Device for Joining at least Two

Adjoining Work Pieces by Friction

Welding .........................................................42

3.2.3.6 Retractable Pin Tool for the use with the

Friction Stir Welding Process ........................43

3.2.4 Development of Specifications ....................................44

3.2.4.1 Pneumatics .....................................................44

3.2.4.2 Hydraulics ......................................................46

3.2.4.3 Pin Forces ......................................................47

3.2.5 Generation of Ideas and Concepts ................................56

3.2.6 Preparing the Preliminary Design and Working

Drawings .....................................................................56

3.2.7 Prototype and Laboratory Testing ................................56

3.3 Description of the FSSW Tool .................................................57

3.4 Summary ........................................................................60

CHAPTER 4

FSSW EQUIPMENT AND EXPERIMENTAL SET-UP

4.1 Introduction..............................................................................61

4.2 Assembly of the FSSW Tool ....................................................61

4.3 Calibration of the FSSW Tool ..................................................65

4.4 Thermocouple Calibration ........................................................69

4.5 Experimental Procedure ...........................................................71

4.5.1 The Weld Matrix ..........................................................73

4.5.2 Machine Operating Procedure.......................................78

4.5.3 Number of Welds done with Variables Constant ...........78

4.5.4 Data Recorded ..............................................................78

4.6 FSSW Evaluation Procedure ....................................................79

4.7 Summary..................................................................................80

ix

CHAPTER 5

DEVELOPMENT OF THE EXPERIMENTAL PROCEDURE FOR

FSSW

5.1 Introduction..............................................................................81

5.2 Machine Set-up ........................................................................81

5.3 FSW Operating System ............................................................83

5.3.1 Operating Procedure of The FSW Machine...................83

5.3.2 The Actual Weld...........................................................83

5.4 Temperature Calibration...........................................................84

5.5 Creating Friction Stir Spot Welds.............................................86

5.6 Methods for Testing the Weld Integrity ....................................111

5.6.1 Tensile Tests.................................................................112

5.6.2 Microstructure and Hardness ........................................113

5.7 Summary..................................................................................119

CHAPTER 6

PERFORMANCE EVALUATION OF FSSW

6.1 Introduction..............................................................................120

6.2 Keyhole Elimination ................................................................120

6.3 Weld Evaluation.......................................................................122

6.3.1 Shoulder Comparisons..................................................122

6.3.2 Different Dwell Times ..................................................124

6.3.3 Different Retraction Rates.............................................128

6.3.4 Different rpm................................................................130

6.3.5 Different Pin Lengths....................................................131

6.3.6 Different Pin Diameters ................................................136

6.3.7 Different Plunge Depths ...............................................140

6.3.8 Friction Stir Spot Welds compared to Resistance .........143

Spot Welds

6.4 Summary..................................................................................151

x

CHAPTER 7

CONCLUSIONS AND FUTURE WORK

7.1 Conclusions..............................................................................152

7.2 Future Work.............................................................................154

REFERENCES ..................................................................................157

APPENDIX

A Design Calculations and Drawings ..........................................163

B Weld Data ...............................................................................183

C Temperature Graphs ................................................................202

D Hardness Graphs .....................................................................206

E Microstructure .........................................................................209

xi

LIST OF FIGURES

1.1 Schematic illustration of a FSSW with the keyhole ....................4

1.2 Schematic illustration of a FSSW without the keyhole ...............4

2.1 Schematic illustration of spot FSW ............................................13

2.2 Schematic diagram of the spot FSW process ..............................13

2.3 Spot friction stir weld ................................................................14

2.4 Relationship between UTSS and force time at various

rotational speeds ........................................................................15

2.5 FSSW lap joints .........................................................................16

2.6 Friction stir spot welding unit and RIFTEC bobbing tool ...........18

2.7 Equivalent FSSW length ............................................................18

2.8 Force elongation curve / different spot diameters .......................19

2.9 Force elongation curve / different sheet thicknesses ...................19

2.10 Hardness profile of a friction stir spot weld, 1/3mm plate

configuration .............................................................................20

2.11 GKSS FSSW process .................................................................21

2.12 Proof of concept and phase I automated retractable pin tool .......23

2.13 Phase II and IIA retractable pin tools .........................................23

2.14 Cross section perpendicular to the welding direction and the

horizontal hardness profile .........................................................27

2.15 Micrographs at locations a, b and c ............................................28

2.16 Calculated isothermals versus microstructure morphology .........30

2.17 Calculated isothermals during the welding period, front view ....31

3.1 The keyhole ...............................................................................36

3.2 Ding and Oelgoetz, 1999 ...........................................................38

3.3 Holt and Lang, 1998 ..................................................................39

3.4 Wykes, 1997 ..............................................................................40

3.5 Adjustable pin tool ....................................................................40

xii

3.6 Colligan, 1998 ...........................................................................41

3.7 dos Santos and Schilling, 2002 ..................................................42

3.8 Georgeou and Hattingh, 2004 ....................................................43

3.9 Fluidic muscle ...........................................................................45

3.10 Fluidic muscle operating range MAS-20 ....................................46

3.11 Illustration of the pin and the shoulder .......................................48

3.12 Force versus shoulder area .........................................................52

3.13 Static plunge result of trial weld 1 ..............................................54

3.14 Exploded view of the FSSW tool ...............................................58

3.15 Cross-sectional view of the pin extended ...................................59

3.16 Cross-sectional view of the pin retracted ....................................60

4.1 Measurement of concentricity in the lathe ..................................62

4.2 Concentricity of the transducer ..................................................63

4.3 Positions for the concentricity measurements of the transducer ..63

4.4 Concentricity of the pneumatic adapter ......................................64

4.5 Concentricity of the tool ............................................................65

4.6 Schematic set-up of the pneumatic muscle ..................................65

4.7 Retraction calibration set-up ......................................................66

4.8 Voltage and pressure gauges ......................................................66

4.9 Muscle pressure compared to pin position ..................................67

4.10 Illustration of the lines visible corresponding to pin position ......68

4.11 Thermocouple set-up around a spot weld ...................................70

4.12 Thermocouples in ice water .......................................................71

4.13 Flow diagram of the evaluation process .....................................79

5.1 Machine set-up ..........................................................................81

5.2 Thermocouple software set-up ...................................................82

5.3 Welding clamps .........................................................................82

5.4 Effect of machine interference on temperature ...........................85

5.5 Weld 1 .......................................................................................87

xiii

5.6 Weld 2 .......................................................................................89

5.7 Cross sectional view of shoulders 1 and 2 ..................................89

5.8 Weld 3 .......................................................................................90

5.9 Weld 4 .......................................................................................90

5.10 Weld 5 .......................................................................................91

5.11 Pin tool 1 and 2 ..........................................................................92

5.12 Weld 6 .......................................................................................93

5.13 Weld 6 .......................................................................................93

5.14 Cross sectional view of shoulders 2 and 3 ..................................94

5.15 Shoulder 3 .................................................................................94

5.16 Weld 7 .......................................................................................95

5.17 Anti-swaying bracket installed....................................................96

5.18 Weld 8 .......................................................................................97

5.19 Weld 9 .......................................................................................97

5.20 Weld 10 .....................................................................................98

5.21 Weld 11 .....................................................................................98

5.22 Weld 12 and 13 ..........................................................................99

5.23 Weld 14 .....................................................................................100

5.24 Weld 15 .....................................................................................101

5.25 Weld 16 .....................................................................................101

5.26 Weld 17......................................................................................102

5.27 Weld 18 .....................................................................................102

5.28 Weld 19 .....................................................................................103

5.29 Weld 20 and 23 ..........................................................................104

5.30 Weld 25 and 26 ..........................................................................104

5.31 Weld 27 and 28 ..........................................................................105

5.32 Weld 29 .....................................................................................105

5.33 Weld 30 and 31 ..........................................................................106

5.34 Weld 32 and 33 ..........................................................................106

5.35 Weld 34 and 35 ..........................................................................107

5.36 Weld 36 and 37 ..........................................................................108

xiv

5.37 Illustration of how the plates were sectioned ..............................111

5.38 The Instron 8801 tensile tester ...................................................112

5.39 2mm and 3mm spacers ..............................................................112

5.40 Diamond cut-off-wheel ..............................................................113

5.41 Comparison of different etching times .......................................115

5.42 Comparison of different etching times .......................................116

5.43 Microstructure of weld 23 ..........................................................117

5.44 Vickers micro-hardness tester ....................................................117

5.45 Vickers micro-hardness set-up ...................................................118

5.46 Final Vickers micro-hardness set-up ..........................................118

5.47 Hardness set-up of the RSW ......................................................119

6.1 Forced out material during plunging ..........................................121

6.2 Tensile comparison between shoulders 1 and 2 ..........................123

6.3 Tensile comparison between an 8s and 16s dwell time with

a 2.85mm pin diameter ..............................................................125

6.4 Hardness comparison between a 8s and 16s dwell time with

a 2.85mm pin diameter ............................................................126



6.7 Microstructure of weld 13 (Slow retraction) ...............................129

6.8 Microstructure of weld 9 (Medium retraction) ...........................129

6.9 Microstructure of weld 10 (Fast retraction) ................................129

6.10 Tensile comparison between 300, 400 and 500rpm ....................130

6.11 Tensile comparison between a 3mm and 4mm pin length ...........131



6.14 Hardness comparison between a 3mm and 4mm pin length ........133

6.15 Comparison between a 3mm and 4mm pin length ......................134



xv

6.18 Tensile comparison between different pin diameters ..................136

6.19 Microstructure of weld 5 ............................................................137


6.21 Microstructure of weld 23...........................................................137

6.22 Hardness comparison of the 3mm plate for different pin

diameters ...................................................................................138

6.23 Hardness comparison of the 2mm plate for different pin

diameters ...................................................................................139



6.26 Microstructure of weld 6, x100 magnification ............................142

6.27 Microstructure of weld 9, x50 magnification ..............................142

6.28 Tensile comparison between a FSSW and a RSW ......................143

6.29 Cross section of the macrostructure of a RSW ...........................145

6.30 Microstructure of a resistance spot weld, x50 magnification ......145


6.32 Microstructure of a RSW, x500 magnification ...........................147

6.33 Microstructure of a FSSW, x500 magnification .........................147

6.34 Microstructure of a RSW, x50 magnification .............................148

6.35 Microstructure of a FSSW, x50 magnification ...........................148

6.36 Hardness comparison between a FSSW and a RSW ...................150

7.1 Friction stir spot welding by RIFTEC GmbH .............................155

A.1 Tool housing structure ...............................................................163

A.2 Euler formula .............................................................................164

A.3 Static loading of the tools housing structure ...............................169

A.4 Static loading of the pin ..............................................................170

A.5 Uniform loading .........................................................................174

A.6 Loading constraints ....................................................................174

xvi

A.7 General formulas for deflection, slope, moment, shear and

stress as a function of r ..............................................................176

A.8 Transverse bending stress ..........................................................177

A.9 Radial bending stress ..................................................................178

A.10 Static nodal stress of the locking ring..........................................178

A.11 Static displacement of the locking ring .......................................179

APPENDIX A

• Design Drawings: Shoulder (1)

• Design Drawings: Shoulders 1, 2 and 3 (1)

• Design Drawings: Shoulder Flange (2)

• Design Drawings: Pin 1, 2 and 3 (3)

• Design Drawings: Lower Ring (4)

• Design Drawings: Spacer (5)

• Design Drawings: Ratchet (6)

• Design Drawings: Column (7)

• Design Drawings: Locking Ring (8)

• Design Drawings: Extension Parts 1 and 2 (9)

• Design Drawings: Extension (9)

• Design Drawings: Ratchet Pin (10)

• Design Drawings: Moving Pin (11)

• Design Drawings: Extension Head (12)

• Design Drawings: Cam Pin (13)

• Design Drawings: Flange (14)

• Design Drawings: Muscle Ring (15)

• Design Drawings: Bush (16)

• Design Drawings: Upper Ring (17)

• Design Drawings: Outer Ring (18)

• Design Drawings: Assembled view

xvii

• Design Drawings: Exploded view

• Design Drawings: Sectioned view

• Design Drawings: Adapter Plate

• Design Drawings: Holder

• Design Drawings: Stabiliser Bracket (Part 1)

• Design Drawings: Stabiliser Bracket (Part 2a)

• Design Drawings: Stabiliser Bracket (Part 2b)

• Design Drawings: Stabiliser Bracket (Part 3)

• Design Drawings: Stabiliser Bracket Assembly

• Design Drawings: Plate Set-up

C.1 Temperature Graphs: Weld 1 .....................................................202




C.5 Temperature Graphs: Weld 5......................................................203

C.6 Temperature Graphs: Weld 20 ...................................................203






C.12 Temperature Graphs: Weld 26 ....................................................204









xviii



D.1 Hardness Graphs: Weld 5 ..........................................................206



D.4 Hardness Graphs: Weld 10 .........................................................206

D.5 Hardness Graphs: Weld 11 ........................................................207


D.7 Hardness Graphs: Weld 23 .........................................................207




D.11 Hardness Graphs: RSW .............................................................208

E.1 Microstructure: Weld 5 ..............................................................209



E.4 Microstructure: Weld 10 ............................................................209







E.11 Microstructure: RSW .................................................................211

xix

LIST OF TABLES

2.1 Chemical composition of aluminium alloy used, mass% ...........15

2.2 Static strengths of RSW lap joints ..............................................16

2.3 Static strengths of FSSW lap joints ............................................16

2.4 Spot weld configuration and groups ...........................................17

2.5 Chemical composition, maximum values ...................................25

2.6 Mechanical properties ................................................................25

2.7 Chemical composition limits % Al 6063, maximum values ........25

2.8 Mechanical Properties of Al 6063 ..............................................25

3.1 Material composition .................................................................48

3.2 Calculated forces .......................................................................49

3.3 Different tool combinations .......................................................50

3.4 Forces obtained by Johnson .......................................................51

3.5 Calculated shoulder areas and pressures .....................................51

3.6 Tool specifications .....................................................................53

3.7 Summarised results of the static welds .......................................54

3.8 Calculated forces for the 4mm pin .............................................55

3.9 Calculated pin and shoulder forces .............................................55

4.1 Pin position corresponding to number of lines visible ................68

4.2 Rate of retraction calibration sheet .............................................69

4.3 Initial welding procedure ...........................................................74

4.4 Welding matrix, test 1 ................................................................75



5.1 Welding parameters of weld 1 ...................................................86

5.2 Data from weld 1 .......................................................................88

5.3 Data from weld 1 .......................................................................88

5.4 Summary of weld parameters......................................................110

xx

B.1 Weld Data: Weld 1 ....................................................................183

B.2 Weld Data: Weld 2 ....................................................................183

B.3 Weld Data: Weld 3 ....................................................................184

B.4 Weld Data: Weld 4 ....................................................................184

B.5 Weld Data: Weld 5 ....................................................................185

B.6 Weld Data: Weld 6 ....................................................................185

B.7 Weld Data: Weld 7 ....................................................................186

B.8 Weld Data: Weld 8 ....................................................................186

B.9 Weld Data: Weld 9 ....................................................................187

B.10 Weld Data: Weld 10 ..................................................................187

B.11 Weld Data: Weld 11 ..................................................................188

B.12 Weld Data: Weld 12 ..................................................................188

B.13 Weld Data: Weld 13 ..................................................................189

B.14 Weld Data: Weld 14 ..................................................................189

B.15 Weld Data: Weld 15 ..................................................................190

B.16 Weld Data: Weld 16 ..................................................................190

B.17 Weld Data: Weld 17 ..................................................................191

B.18 Weld Data: Weld 18 ..................................................................191

B.19 Weld Data: Weld 19 ..................................................................192

B.20 Weld Data: Weld 20 ..................................................................192

B.21 Weld Data: Weld 21 ..................................................................193

B.22 Weld Data: Weld 22 ..................................................................193

B.23 Weld Data: Weld 23 ..................................................................194

B.24 Weld Data: Weld 24 ..................................................................194

B.25 Weld Data: Weld 25 ..................................................................195

B.26 Weld Data: Weld 26 ..................................................................195

B.27 Weld Data: Weld 27 ..................................................................196

B.28 Weld Data: Weld 28 ..................................................................196

B.29 Weld Data: Weld 29 ..................................................................197

B.30 Weld Data: Weld 30 ..................................................................197

B.31 Weld Data: Weld 31 ..................................................................198

xxi

B.32 Weld Data: Weld 32 ..................................................................198

B.33 Weld Data: Weld 33 ..................................................................199

B.34 Weld Data: Weld 34 ..................................................................199

B.35 Weld Data: Weld 35 ..................................................................200

B.36 Weld Data: Weld 36 ..................................................................200

B.37 Weld Data: Weld 37 ..................................................................201

xxii

GLOSSARY OF TERMS

A

Alloy – A metal comprised of two or more elements, at least one of which is

metallic.

C

Conventional friction welding – A solid-state welding process in which

coalescence is achieved by frictional heat combined with pressure.

Capillary action - A phenomenon associated with surface tension and

resulting in the elevation or depression of liquids in capillaries.

D

Deformation - An alteration of shape, as by pressure or stress.

E

Etching – Subjecting the surface of a metal to preferential chemical or

electrolytic attack in order to reveal structural details for metallographic

examination.

F

Finite Element Analysis (FEA) - The analysis of static and dynamical

systems using computational methods such as the finite difference method.

Friction Stir Weld (FSW) – A solid-state process in the field of non-fusion

welding. By rotating and plunging a non-consumable tool into the work

piece, the two pieces are “stirred” together resulting in a uniform bond.

Friction Stir Spot Weld (FSSW) – is derived from the FSW process and is

used to create spot welds on lap joints.

xxiii

Fusion welding – A welding process whereby heat is used to melt the base

metals.

Force time – The period that the welding force is applied, from the start of

the dwell period to when the tool is extracted.

H

Hardness – Hardness of a material is defined as its resistance to permanent

indentation.

Hardness profile – Making a series of hardness impressions at various

distances from the surface or a specific point to measure the hardness

variation within the sample.

Heat Affected Zone (HAZ) – The zone in which the metal experiences

temperatures below its melting point, yet high enough to cause

microstructural changes in the solid material.

L

Lap Joint – A joint between two boards or metal parts in which the ends are

overlapped and fastened together.

M

Mechanical properties – The properties of a material that reveal its elastic

and inelastic behaviour when force is applied, thereby indicating is

suitability for mechanical applications.

Microhardness – The hardness of a material as determined by forcing an

indenter into the surface of a material under very light load.

xxiv

Microstructure – The structure of an object as revealed through

microscopic examination.

N

Non-fusion – The process where the materials are not liquefied or nor

melted, to form a bond, however with heat applied to reduce the energy

required to cause plastic deformation.

P

Plastic deformation – The permanent distortion of materials under applied

stresses that strain the material beyond its elastic limit.

Parent Material - The original material that has not been affected, by heat

or pressure.

Polishing – Smoothing metal surfaces, often to a high lustre, by rubbing the

surface with a fine abrasive, usually contained in a cloth or other soft lap.

S

Sheet – A flat-rolled metal product of some maximum thickness and

minimum width arbitrarily dependent on the type of material.

Stress – Internal force exerted by either of two adjacent parts of a body

upon the other across an imagined plane of separation. When the forces are

applied parallel to the plane, the stress is called shear stress; when the forces

are normal to the plane, the stress is called normal stress; when the normal

stress is directed toward the part on which it acts, it is called compressive

stress; and when it is directed away from the part on which it acts, it is

called tensile stress.

xxv

Solid-state welding – A joining process in which coalescence results from

application of pressure alone or a combination of heat and pressure whereby

the temperature of the process is below the melting point of the metals being

welded.

T

Thermo-Mechanically Affected Zone – The metal has been plasticized

and mechanically "stirred," but not heated enough to cause significant

changes to metallurgical properties.

U

Ultimate Tensile Shear Strength - The highest load applied to a material

in the course of a tensile test, divided by the original cross-sectional area,

when the forces are applied parallel to the plane.

Ultimate Tensile Strength – The highest load applied to a material in the

course of a tensile test, divided by the original cross-sectional area, when

the normal stress it is directed away from the part on which it acts.

W

Weld time – the period between the contact and extraction of the welding

tool, from when the welding pin is in contact with the material to the time

when the tool extracted.

1. Introduction

1

CHAPTER 1

INTRODUCTION

1.1) INTRODUCTION

Welding is a materials joining process in which two or more

components are coalesced at their contacting surfaces (metallurgical

bond) by suitable application of heat and/or pressure. The assembled

parts joined by welding are called the weldment. Welding processes

vary by the types of techniques used. Some are accomplished by heat

alone where no pressure is applied, some by a combination of heat and

pressure, and others where only pressure is applied with no external

heat source. Some methods also require filler materials to be added to

facilitate the join (Groover, 1996).

The commercial and technological importance of welding derives

from the following:

Welding provides a permanent joint where the welded parts become

one. It is usually the most economical way to join components in

terms of material usage and fabrication costs. Alternative methods

require other alterations such as drilling of holes and the addition of

fasteners. The welded joint is often stronger than its parent material, if

correct welding techniques are used (Groover, 1996).

Some of the disadvantages of welding are the expensive costs

involved, the need for skilled tradesmen, high-energy consumption

levels and welding defects (Groover, 1996).

This brings us to a fairly new welding process that has only in the past

few years been introduced into production and has many benefits

compared to conventional welding techniques.

1. Introduction

2

Friction Stir Welding (FSW) is a solid state process in the field of

non-fusion welding and an offshoot of conventional friction welding,

developed in 1991 by The Weld Institute (TWI) (Thomas et al., 1995),

a research and technology centre based in the United Kingdom.

The FSW process has the advantage of achieving metallic bonding

below the melting point of the base material. It is thus a low heat input

welding technique and therefore has good joint properties such as low

distortion, low heat affection and is hot-cracking free (Matsumoto and

Sasabe, 2001). More recently, the FSW process has been further

developed to produce spot welds (i.e. Friction Stir Spot Welding,

FSSW) (Schilling et al., 2000).

The FSW process has made it possible to weld a number of materials

that previously would have been extremely difficult to reliably weld

without voids, cracking or distortion. All aluminium alloys can be

welded, including those that cannot be joined by conventional fusion

techniques. Even dissimilar alloys can be joined, such as series 2xxx

to 7xxx (Kallee and Mistry, 1999).

The FSW process is regarded as a very energy-efficient machine tool

technology. The attraction of this technique from a design and

production point of view, when compared to available fusion welding

techniques, is to produce a lighter and more consistent weld (Page,

2003). Fasteners can also be eliminated from many applications and

can lead to a reduction in both cost and weight. Other major

advantages include, excellent mechanical properties, the use of a non-

consumable tool, no spatter, no filler material required, no fillets, a

restricted heat affected zone (HAZ) and welding preparation is not

usually required.

1. Introduction

3

1.2) THE FRICTION STIR SPOT WELDING PROCESS

The FSW process produces high-strength, defect free joints in metallic

materials that are difficult to fusion weld. With the aid of a non-

consumable pin tool, which is rotated and plunged into the parent

material to “stir” the two pieces together, a uniform weld is formed

(National Center for Advanced Manufacturing, 2004). This FSW

process has lead to the development of a Friction Stir Spot Weld

(FSSW), using similar techniques and also sharing many benefits of

the FSW process.

The FSSW process involves a cylindrical tool with a shouldered pin

used to generate frictional heat at the point of immersion and induces

plastic deformation in the material. The process is applied to a lap

joint, which consists of upper and lower sheets. The tool, having a

profiled projecting pin, is rotated and slowly plunged into the surface

of the material. At the same time a backing plate contacts the lower

sheet from the under side to support the downward force. After

keeping the weld force and the tool rotational speed stable for an

appropriate time, the tool is extracted from the material. This results in

a solid phase bond between the upper and lower sheets (Sakano et al.,

2001). At the end of the weld the tool is extracted from the material,

resulting in a keyhole, as shown figure 1.1

1. Introduction

4

(a) Plunging (b) Bonding (c) Drawing Out

Figure 1.1: Schematic illustration of a FSSW with the keyhole

(reproduced from Sakano et al., 2001)

With aid of a retractable pin tool, an attempt will be made to close the

keyhole resulting in a FSSW, as shown in figure 1.2.

(a) Plunging (b) Bonding (c) Pin Retraction (d) Drawing Out

Figure 1.2: Schematic illustration of a FSSW without the keyhole

The FSSW process has significant advantages compared with

conventional fusion welding processes from a quality and an energy

consumption point of view. It is also easily automated and reduces the

need for highly skilled workers because the process is based on tool

technology. Investigations have confirmed that the FSSW lap joint has

equal or superior, mechanical properties and reproducibility, to the

conventional Resistance Spot Welding (RSW) process. In addition,

the comparative estimation per single spot weld shows that the cost of

1. Introduction

5

the FSSW system is 85% less than that of the RSW system. This cost

reduction is brought about by the cut in utility costs (electric power,

cooling water, etc.) and consumables (RSW electrodes, etc).

The FSSW system is also approximately 50% smaller than the RSW

system because several parts, which include a large electric power

supply, a cooling unit, an electrode dresser and others, are not

necessary in the FSSW system (Sakano et al., 2001).

This system is considered to be a very attractive welding technique

and with further research and development could take over the

conventional RSW processes in some aluminium production lines

(Sakano et al., 2001).

1.3) FSW AND FSSW APPLICATIONS

Friction stir welding is an innovative process that has been

successfully implemented in joining normally extremely difficult to

weld aluminium alloys reliably, without voids, cracking or distortion.

This process is still fairly new and not well understood but due to its

many benefits is finding an increasing number of uses in various

fields.

Friction Stir Spot Welding is derived from the FSW process and with

the elimination of the keyhole could have a major effect on the

automotive industry. The original application for friction stir welding

was the welding of long lengths of material for the aerospace,

shipbuilding and railway industries.

1. Introduction

6

In the aerospace field, Lockheed Martin, Boeing, and Airbus have

initiated projects to use the FSW process. Boeing uses FSW on the

fuel tanks of Delta Rockets II and IV, and in terms of reliability and

repeatability, have made 2.5km of continuous defect-free welds

(Hansen, 2003; Page, 2003). They also claim a 71% reduction in weld

cycle time in comparison with fusion welding techniques, as well as

an 81% reduction in labour costs. Boeing has also already

demonstrated curvilinear FSW of a complex landing gear door and the

FSW of sandwich assemblies for a fighter aircraft fairing. Airbus has

successfully used the FSW process to join foils of 0.3mm to 0.4mm in

laminate construction (Page, 2003). Eclipse Aviation Corporation of

Albuquerque, N.M., claims it has reduced costs up to one-fourth. They

claim that on the Eclipse 500 twinjet, more than 7000 fasteners were

replaced by 263 welds, thus leading to a saving in cost and process

time.

In shipbuilding, Hydro Marine Aluminium of Norway was the first

manufacturer to use FSW in production (Hansen, 2003). They created

light aluminium deck panels by sandwiching corrugated sections

between skins and using FSW to fabricate the sandwich. Aluminium

ship builders in the UK are fabricating hovercraft bulkhead and deck

panels using FSW. Panels for passenger rail vehicles are also

fabricated with FSW (Page, 2003), such as roof panels for Alstom

made by Hydro Marine Aluminium.

In the automotive industry, Mazda Motor Corporation is the first car

manufacturer to use FSW on body assemblies. With the new

technique used by Mazda, a spot-joining technique, the rear doors and

the bonnet of the 2004 RX-8 are FSSW. Mazda claims in large energy

consumption savings of up to 99% and that the equipment costs have

dropped by up to 40% (Machine Design, 2003).

1. Introduction

7

In the automotive industry, components such as suspension arms,

wheel rims, aluminium engine cradles and lightweight panels have all

been manufactured using the FSW process (Johnson and Kallee,

1999).

There is no doubt that in the future the automotive field will find an

increasing number of uses for this cost effective process as a result of

the ability to weld dissimilar material combinations and create

lightweight vehicle designs with minimal distortion (Johnson and

Kallee, 1999). The important approach to make vehicles lighter by

using aluminium and FSW and FSSW processes leads to enhanced

fuel efficiency and improvements in safety and dynamic performance

(Machine Design, 2003).

1.4) MOTIVATION

At the end of the weld, the tool is extracted from the material,

resulting in a keyhole. National Aeronautics and Space Administration

(NASA) developed a retractable pin tool (Ding et al., 1999; Ding,

2000), in cooperation with MTS and MCE Technologies Inc. in the

United States of America, to close the keyhole (NASA Techtrans,

2001; NASA Marshall Space Flight Centre, 2002). PE Technikon has

also developed a similar tool to close the keyhole (Georgeou and

Hattingh, 2004).

Further research should benefit the FSW research community by

improving the knowledge base in the field of friction stir welding and

friction stir spot welding. This should expand the Manufacturing

Technology Research Centres (MTRC) current FSW research base

and assist researchers to advance FSW technology both locally and

1. Introduction

8

internationally. Through further research and utilizing new processes

and methods, this technology should become acceptable to industry.

Further research opportunities are constantly identified due to present

research advances.

With this research and investigation into FSSW, a better

understanding of the process will be created. As this is relatively new

technology advancement, it should lead to faster development in the

field of FSSW, allowing local and international industry to benefit

sooner from this process.

1.5) OBJECTIVES

The primary objective of this research work is to perform a spot weld

on aluminium sheet using friction stir welding principals and to

compare the weld integrity to conventional welding techniques. This

will be achieved by performing several sub-objectives and creating

certain delimitations in support of the main objective.

1.5.1) FSSW Tool Development

This will be done by developing the existing FSW retractable pin tool,

Patent Application No. RSA 2004/0538 (Georgeou and Hattingh,

2004), to create a FSSW. This will require an in-depth study of the

forces needed to ensure tool penetration. Methods of actuation will

also be investigated. Currently pneumatics is employed for actuation

but other options such as hydraulics and electro-mechanical systems

need to be investigated. The design of the pin profile and the shoulder

will also be carefully evaluated so as to determine the best mechanical

properties of the weld. Necessary sensors also need to be added to the

1. Introduction

9

tool, such as pin position and temperature sensors. A clamping device

to hold the two plates together also needs to be designed.

1.5.2) Pin Profile and Retraction Studies

An investigation into the pin retraction process in FSSW in order to

create the best possible weld and to eliminate the keyhole effect will

be evaluated. The influence of pin retraction on the material

behaviour, mechanical and material characteristics, and resulting weld

quality will be researched. Welds will be done on aluminium alloy

plates, series 6000, as used by many manufacturers in the automotive

industry. Methods of changing the dwell time, pin retraction,

rotational speed and different tool designs will also be analysed. A

literature study into previous tests will be conducted.

1.5.3) Material Flow Evaluation

Analysis and evaluation of material flow patterns, to improve the

quality of the weld, will be investigated by the use of existing material

marker techniques. The weld integrity will be evaluated with the

objective of measuring the weld quality, through discussions with

experts and practitioners, as well as with information relevant from

literature.

1.5.4) Spot Weld Evaluation

Testing and analysing the material and mechanical properties of the

weld and comparing them to ASTM * standards will be carried out.

These tests will compare the mechanical properties of a FSSW to a

conventional spot weld for aluminium plate. (e.g. Ultimate tensile

shear strength (UTSS), microstructure and microhardness.)

* ASTM – American Society for Testing and Materials

1. Introduction

10

1.5.5) Delimitating Research

The following delimitation factors have been considered to make the

study effective and possible.

• The design and development of the retractable pin will

simulate pin penetration in a fixed position and retraction in

stepped stages, and will only be used for research purposes.

• The process control parameters will be investigated and tested

to obtain the best quality weld. Once the parameters are set,

further tests for consistency and other factors will be carried

out.

• Welding is limited to FSSW of lap joints in thin aluminium

plate.

The final aim of this research is to develop a retractable pin tool

system which will enable the current FSW machine set-up to create

friction stir spot welded joints. Also the FSSW process will produce a

joint with similar mechanical characteristics to that of a conventional

spot weld.

1. Introduction

11

1.6) SUMMARY

This dissertation will be divided into a number of chapters addressing

the following:

• Chapter 1 will be a brief introduction to the FSW and the

FSSW process.

• Chapter 2 will be an overview literature survey undertaken.

This will comprise of an investigation of previous studies and

methods that have been done.

• Chapter 3 will consist of the tool design and development.

• Chapter 4 will contain the experimental set-up.

• Chapter 5 will give an overview of the experimental procedure

undertaken.

• Chapter 6 will evaluate results obtained during the welding

process. Comparisons between FSSWs and RSWs will also be

done.

• Chapter 7 will discuss conclusions and recommendations for

future investigations and research.

2. Literature Survey

12

CHAPTER 2


2.1) INTRODUCTION

Currently, in the automotive industry, joining techniques on

aluminium structures are produced by resistance spot welds, riveting,

clinching and adhesives. The FSSW process is a new process that can

offer an alternative method of joining to these conventional methods.

Mazda Motor Corporation has already implemented the use of friction

stir spot welds on the rear doors and on the bonnet of the 2004 RX-8.

They claim that they have reduced energy consumption as well as

equipment costs.

Although the process is still relatively new, with further testing and

evaluation it will only be a matter of time before other automotive

manufacturers soon grasp this new technology in their quest for

improved quality and cost savings.

Evaluation of friction stir spot welds has been documented and the

next section will give insight into numerous different attempts in

detail.

2.2) DEVELOPMENT OF SPOT FSW ROBOT SYSTEMS

FOR AUTOMOBILE BODY MEMBERS

The process developed by Sakano et al., 2001, “Development of spot

FSW robot systems for automobile body members”, is applied to lap

joints as illustrated in figure 2.1.


13

Figure 2.1: Schematic illustration of spot FSW


The rotating tool is plunged into the material for a certain time to

generate heat. A backing tip is applied at the same time to support the

downward force applied. The heated and softened material adjacent to

the tool causes plastic flow while the shoulder gives a strong

compression force, resulting in a solid phase bond between the two

sheets (Sakano et al., 2001).

During the process, as soon as the rotating tool contacts the surface of

the material, the axial force increases up to a pre-determined welding

force. After keeping the welding force and the rotational speed stable

for a certain amount of time the tool is extracted. This procedure is

illustrated in the graphical schematic diagram of figure 2.2.

Figure 2.2: Schematic diagram of the spot FSW process

sequence (reproduced from Sakano et al., 2001)


14

The weld time is the period between the contact and extraction of the

welding tool, from when the welding pin is in contact with the

material to the time when the tool is extracted. The force time is the

period that the welding force is applied, from the start of the dwell

period to when the tool is extracted. The tool rotational speed remains

constant throughout the welding process.

Figure 2.3 illustrates a FSSW produced.

Figure 2.3: Spot friction stir weld


The capable RPM of the machine is between 100 and 3500 rpm and

can generate a welding force of up to 3.4kN, with high accuracy of

±10N. The axial loading sequence is controlled with an error of less

than 0.1 seconds. The tool is attached to a Kawasaki ZX-165U multi-

articulated robot. The robot has a programme specifically installed for

the FSSW process and governs both the motion of the robot and the

welding operation of the FSW gun.

Aluminium 6xxx series aluminium alloy was selected for evaluating

the joint performances of the FSSW process of which the chemical

composition was as follows in table 2.1.


15

Table 2.1: Chemical composition of aluminium alloy used, mass%

Si Mn Mg Fe Cu Ti Zn Cr Al

1 1 0.46 0.15 0.08 0.04 <0.01 <0.01 Bal.

The thickness of the upper/lower plates used was 1mm/1mm.

A shoulder diameter of 7mm and a pin diameter of 2.1mm were used

to perform the tests.

From the results and the graph in figure 2.4, it can be observed that at

any rotational speed the Ultimate Tensile Shear Strength (UTSS) first

increases according to an increase in force time and then decreases

after reaching a maximum value. The decrease of the UTSS seems to

be caused by excessive frictional heat generated (Sakano et al., 2001).

Figure 2.4: Relationship between UTSS and force time at various

rotational speeds (reproduced from Sakano et al., 2001)


16

a) Tensile shear test b) Tensile test

Figure 2.5: FSSW lap joints (reproduced from Sakano et al., 2001)

From the results it can be observed that the welding force, force time

and rotational speed are considered to be essential welding parameters

to obtain high static strengths in FSSW. Table 2.2 shows the UTSS

and Ultimate Tensile Strength (UTS) values of a RSW lap joint, while

table 2.3 shows the values from a FSSW process (Sakano et al., 2001).

Table 2.2: Static strengths of RSW lap joints

Thickness combination

of lap joint

UTSS

(Ultimate Tensile Shear Strength)

UTS

(Ultimate Tensile Strength)

0.8 mm/0.9mm 1.488 kN [average] 0.799 kN [average]

1.0 mm/0.9 mm 1.811 kN [average] 0.922 kN [average]

(adapted from Sakano et al., 2001)

Table 2.3: Static strengths of FSSW lap joints

Thickness combination

of lap joint

UTSS

(Ultimate Tensile Shear Strength)

UTS

(Ultimate Tensile Strength)

0.8 mm/0.8mm 1.844 kN [average] 0.892 kN [average]

1.0 mm/1.0 mm 2.017 kN [average] 0.998 kN [average]

(adapted from Sakano et al., 2001)


17

From the information in table 2.2 and 2.3 it would appear that FSSW

lap joints have equal or superior UTSS and UTS values compared to

conventional RSW lap joints (Sakano et al., 2001).

2.3) A PRELIMINARY INVESTIGATION ON THE STATIC

PROPERTIES OF FRICTION STIR SPOT WELDS

In a paper by Schilling et al., (2000), “A preliminary investigation on

the static properties of friction stir spot welds”, weld diameters of

6mm, 7mm and 10mm were carried out on various different lap joint

configurations. The thickness of combinations included the following:

1/3mm, 1.5/3mm, 2/3mm, 3/3mm. Each combination sheet thickness /

weld spot diameter represented a specimen group (A – J, see table

2.4).

Table 2.4: Spot weld configuration and groups

Spot Diameter

Configuration 6mm 7mm 10mm

1mm/3mm A B C

1.5mm/3mm D E F

2mm/3mm G H

3mm/3mm I J

Specimen Group

(adapted from Schilling et al., 2000)

Aluminium 6061-T4 alloy was used for both the top and bottom sheet.

No description of the tool used nor the welding parameters are

mentioned in the paper, only the evaluation of the results are

discussed. Microhardness profiles and shear strength tests have been

produced for these samples.


18

The developed Friction Stir Spot Welding Unit

Figure 2.6: Friction stir spot welding unit and RIFTEC bobbing

tool (reproduced from Schilling et al., 2000)

The Equivalent length of the spot welds was derived to produce the

same area as a spot diameter of resistance welds, in order for a

meaningful comparison to be made.

Figure 2.7: Equivalent FSSW length

Three shear tension tests for each spot diameter were conducted and

the results showed a homogeneous weld and consistent behaviour.

Results from the welds showed that due to increased heat input

required by bigger plate combinations the strength loss due to the heat

affected zone decreased from the 3mm/3mm plate to the 1mm/3mm

configuration. The shear strength of the friction stir spot welds were

reported greater than the shear strength for resistance spot welds

(Schilling et al., 2000).


19

Figure 2.8: Force elongation curve / different spot diameters

(reproduced from Schilling et al., 2000)

From figure 2.8, it can be observed that an increase in the size of the

spot weld causes an increase in the shear strength. The 10mm spot

diameter has the greatest shear strength and the 6mm spot diameter

the least. These tests were all done on the 1.5mm/3mm sheet

configuration. The parameters of the different letters in figure 2.8 and

2.9 can be referred to in table 2.4.

As for the different sheet thickness, the 1mm/3mm plate proved to be

the weakest with the other combinations being very similar. However

the 1.5mm/3mm configuration showed the best results.

Figure 2.9: Force elongation curve / different sheet thicknesses

(reproduced from Schilling et al., 2000)


20

The microhardness profiles of the FSSWs with a spot diameter of

7mm in different overlap configurations were selected to illustrate the

hardness results obtained. The hardness values were measured mid

thickness of the top sheet and 0.5mm below the interface of the

bottom sheet for the different sheet configurations.

There was a recorded loss of hardness in the heat affected zone

(HAZ), with the lowest values obtained representing a percentage of

the base material.

The joint efficiencies in terms of microhardness for the various

overlap configurations were as follows:

• Configuration 1mm on 3mm plate: 86%

• Configuration 1.5mm on 3mm plate: 82%



The greatest efficiency was obtained from the smallest combined

sheets - the 1/3mm configuration. This is as a result of lower heat

inputs required in combinations with shorter pin lengths.

Figure 2.10: Hardness profile of a friction stir spot weld, 1/3mm plate

configuration (reproduced from Schilling et al., 2000)


21

From the results presented in figure 2.10 it can be observed that the

top sheet experienced more softening than the bottom sheet. This is a

result of the top sheet being in direct contact with the shoulder, which

generates a large proportion of the heat input. The pin plunges

completely through the top sheet and only partially penetrates the

bottom sheet, resulting in the nugget being formed mostly in the top

sheet (Schilling et al., 2000).

2.4) GKSS (Freeman et al., undated)

GKSS, a German research institute, has been developing FSSW

technology. They have introduced a new bi-directional pin tool design

that has lent itself to take the place of rivets. In figure 2.11 an

illustration of the process can be seen. (Freeman et al., undated).

Figure 2.11: GKSS FSSW process

(reproduced from Freeman et al., undated)

This process involves the plunging of the tool into the work piece with

the pin flush with the shoulder. As the shoulder is plunged into the

material the pin is retracted into the shoulder creating a reservoir for

Material Reservoir


22

the plasticized material to be forced into. The sheets of material are

then stirred creating the bond. When the shoulder is removed, the pin

is extended forcing the plasticized material in the reservoir back into

the base material. This results in keyhole free FSSWs that are

favourable in appearance and performance.

2.5) LAP JOINTS OF ALUMINIUM ALLOYS BY FRICTION

STIR WELDING (Matsumoto and Sasabe, 2001)

In the paper by Matsumoto and Sasabe (2001), friction stir welds on

lap joint configurations of 1mm and 2mm were carried out. It was

reported that when the thicker plate, the 2mm plate, was on top of the

1mm plate, the tensile shear strength was greater than when the 1mm

sheet was on top of the 2mm sheet.

2.6) FORCE CHARACTERISATION ON THE WELDING

PIN OF A FRICTION STIR WELDING RETRACTABLE

PIN-TOOL USING ALUMINIUM-LITHIUM 2195 (Ding,

2000)

The retractable pin-tool developed by NASA/Marshall Space Flight

Center, the “Auto-Adjustable Pin Tool” for friction stir welding,

allows for the keyhole to be removed. This is achieved by retracting

the pin at the end of the weld without adding any extra material. The

weld joint depth can also be varied to accommodate different material

thicknesses (Ding, 2000).

Analysis of the forces was concentrated on the shoulder of the tool.

With the auto-adjustable tool the total force exerted by the pin can

further be evaluated by assessing the force behaviour of the welding


23

pin independently. This capability allows for unique analysis of the

reaction forces, which results from the plastic material beneath the pin

pushing upwards against the pin (Ding, 2000). Below are pictures of

the developed retractable pin tool.

Figure 2.12: Proof of concept and phase I automated retractable pin

tool (reproduced from Ding, 2000)

Figure 2.13: Phase II and IIA retractable pin tools

(reproduced from Ding, 2000)

Static weld trials were done, thus there was no travel after plunging.

This process consisted of a 24mm diameter shoulder and a 9.6mm

diameter pin and was plunged into aluminium lithium alloy 2195. Pin

length, pin forces, penetration ligament and rpm were recorded.


24

The initial pin length in the various tests was set between 7.7mm and

8.36mm and was plunged into the material until the shoulder was at

the correct plunge depth - approximately 0.254mm. The pin was then

extended and retracted to different depths within the material. The

welding pin could be extended and retracted by increments as small as

0.025mm and the device recording the force was accurate up to ±

11.34kg. Four static trial welds were completed (Ding, 2000).

The static plunge tests took about 2 minutes to conduct with the rpm

set at 225. Two minutes of the shoulder rotation in one spot generated

a tremendous amount of heat input, allowing the material to become

hotter and more plastic than in the actual weld.

The results of the applied forces can be seen in Chapter 3, where the

forces are analysed in detail to assist with the experimental design for

this research project.

2.7) MATERIAL SELECTION

Some of the above FSSWs were conducted with aluminium 6xxx

series aluminium alloy. From this it would be acceptable to use similar

materials during this research project.

In a technical report from a particular automotive company,

aluminium with the following material composition is being used for

areas where FSSW could have an impact.


25

Table 2.5: Chemical composition, maximum values

Si Fe Cu Mn Mg Cr Zn Ti V Other elements Al

0.2-0.6 0.35 0.25 0.2 0.3-0.9 0.2 0.15 0.1 0.2 0.05 0.15 Rest

Table 2.6: Mechanical properties

0.2% Proof

Strength (Mpa)

Ultimate Tensile

Strength (Mpa)

Elongation

%

200 - 245 215 - 265 ≥ 11

The material selected for the research experiments was aluminium

alloy 6063 T4.

Table 2.7: Chemical composition limits % Al 6063,

maximum values

Other elements Si Fe Cu Mn Mg Cr Zn Ti

Each Total Al

0.2-0.6 0.35 0.1 0.1 0.45-0.9 0.1 0.1 0.1 0.05 0.15 97.5

Table 2.8: Mechanical Properties of Al 6063

Tensile Yield

Strength (Mpa)

Ultimate Tensile

Strength (Mpa) Elongation %

T4 89.6 172 22

Characteristics

Corrosion Resistance Very Good

Anodising Very Good

Formability Good

Weldability Good

Brazeability Good


26

Other Mechanical Properties

Shear Strength: 110MPa

Hardness, Brinell: 46, 500kg load with 10mm ball

Melting Point: 616 - 654˚C

2.8) DISTRIBUTIONS OF HARDNESS AND

MICROSTRUCTURE IN FRICTION STIR WELDS OF

AL ALLOY 6063 (Sato et al., 2001)

Microstructure and Hardness

Several studies on microstructure and hardness have been done on

various aluminium alloys including the 6063 series. The frictional heat

and plastic flow produce a microstructural change in the material

during the weld. This change has an effect on the mechanical

properties of the material, therefore making the relationship between

microstructure and mechanical properties very important. Due to the

re-crystallisation arising from the above factors a fine grained

microstructure results. This fine-grained microstructure often results

in excellent mechanical properties in welds of non heat-treatable

aluminium alloys (Sato et al., 2002).

The maximum temperature reached rises with an increase in rpm, and

the re-crystallised grain size of the weld increases exponentially with

an increase in temperature. Thus the grain size increases with an

increase in rpm up until 800rpm, beyond which it is largely the same.

With regard to the hardness, tests revealed that the hardness around

the centre of the weld was reduced (Sato et al., 2002).


27

Figure 2.14: Cross section perpendicular to the welding direction and

the horizontal hardness profile

(reproduced from Sato et al., 2001)

a: Stir zone

b: Thermo-mechanically affected zone (TMAZ)

c: Unaffected base material region

As seen from figure 2.14 the minimum hardness values are just off the

centre of the weld, where as the outside region retains its base material

hardness.


28

Figure 2.15: Micrographs at locations a, b and c

(reproduced from Sato et al., 2001)

Figure 2.15 shows the coarse grain structure of the base material as

well as the fine grain structure in regions “a” and “b”. These fine

grains occupy the overall stir zone, while the deformed grains are

observed just outside the stir zone (Sato et al., 2002).

2.9) TEMPERATURE AND FLOW PATTERNS

Estimations of heat input in the FSW process have been studied by

several researchers. Shinoda measured the heat generation from the

shoulder and the pin separately and concluded that 60% heat input

yields from the pin, even if a simple straight pin is used. The

tendencies are not greatly affected by changing alloys (Shinoda et al.,

2001). Papers by other researchers have claimed otherwise, in terms of

heat generated by the shoulder and the pin. Their results can be seen in

the following sections.


29

2.9.1) Numerical and Experimental Study of the Heat Transfer

Process in Friction Stir Welding (Song, 2003)

In a paper by Song and Kovacevic (2003) a numerical and

experimental study of the heat transfer process in friction stir welding

was done. A detailed model of heat generation and conduction is

useful for FSW for the following reasons:

1. A heat transfer model is helpful in predicting the temperature

pattern during the welding;

2. A predicted temperature can be used in evaluating the heat

affected zone (HAZ), microstructure and hardness; and

3. A provided temperature field is necessary for determining the

temperature dependant viscosity of the material for flow

modelling.

The main heat supply from the weld is stated as being:

• Friction heat generated at the interface between the tool

shoulder and the work piece; and

• The material plastic deformation heat; generated near the tool

pin.

The friction heat input from the shoulder is believed to be the main

heat input in the FSW process (Sato et al., 2002).


30

Figure 2.16: Calculated isothermals versus microstructure

morphology

(reproduced from Song, 2003)

2.9.2) Thermal Modelling of Friction Stir Welding in a Moving

Coordinate System and its Validation (Song, 2002)

In another paper by Song and Kovacevic (2002), “Thermal modelling

of friction stir welding in a moving coordinate system and its

validation”, a three dimensional heat transfer model for FSW, is

presented. A comparison between calculated and actual values was

done on Al 6061-T6 plates. The tool was made from H13 steel and the

pin is 12mm in diameter and height. The tool shoulder is 100mm in

diameter and 50mm in height.

In this paper it states how other heat transfer models have been made

by numerous methods and that in all the models, the heat input was

calculated from the shoulder only. The heat generated by the tool

pin/work piece interface has not been included. The heat generated by

the pin was only estimated to be 2% of the total heat generated during

the FSW. However this ratio was estimated to be up to 20% by some

researchers (Song, 2002).


31

For this reason, they have included the effect into the modelling

equations, but how much of an effect it has, is still unknown.

As one can see from the above discussed, the heat input due solely to

the pin is still a vague point in terms of the FSW process as there are

some vast differences in opinion regarding different researchers.

Figure 2.17: Calculated isothermals during the welding period, front

view (reproduced from Song, 2002)

2.9.3) Analysis of material flow around a retractable pin in a

friction stir weld (Georgeou, 2003)

In the Masters dissertation Georgeou states that there is a

misconception that the shoulder generates enough heat to allow for pin


32

penetration in a FSW cycle. After a dwell period of 12-20s the heat

generated by the shoulder seems to decrease, with all parameters

remaining constant. Heat generated by the shoulder is said to be only

responsible for heat generated at the surface, to a depth of 2-3mm.

Tests conducted with the pin extended into a pilot hole resulted in

higher temperatures recorded (Georgeou, 2003).

As can be seen in section 2.9, it is still unclear which components are

responsible for what percent of heat generation during the FSW

process. This proves that research still needs to be done in order to

clarify the results obtained.

2.10) RESISTANCE SPOT WELDING

Resistance spot welding (RSW) is a process that involves the joining

of two or more metal parts together in a localised area by the

application of heat and pressure. The heat is generated within the

material being joined by the resistance to the passage of high current

through metal parts, which are held under a pre-set pressure. Heat is

developed mainly at the interface of the two sheets, causing the

material being welded to melt, forming a molten pool, the weld

nugget. The molten pool is contained by the pressure applied by the

electrode tip and the surrounding solid metal (The Welding Institute,

2004).

Resistance spot welding is the most commonly used joining technique

in the mass production of aluminium car bodies. Some of the

advantages are, low capital cost, ease of maintenance and high

tolerance to a poor part fit. However some of the disadvantages are,


33

limited electrode tip life and poor weld consistency due to the

degradation of the electrode tips (Schilling et al., 2000).

2.11) SUMMARY

In chapter 2 a literature survey into the friction stir spot welding

process was done. Different FSSW processes of other researchers

were discussed in detail. The properties of FSSW were evaluated and

compared to conventional spot welding techniques, such as resistance

spot welding. Tensile and microhardness comparisons were made. The

reasons for choosing to use Aluminium 6063-T4 in this research work

were mentioned and the material properties tabulated. Heat generated

by the different parts of the tool, the pin and the shoulder, were

mentioned. The values obtained by the different researchers are

however contradicting and proves that more research in this field is

needed. A brief description of a resistance spot welding and its

advantages and disadvantages was also discussed.

3. FSSW Tool Development

34

CHAPTER 3

FSSW TOOL DEVELOPMENT

3.1) INTRODUCTION

A design is the mechanism whereby a requirement is converted to a

meaningful and functional plan (Deutschman et al., 1975). In this

section the FSSW tool design process is looked at in detail.

During the design and development of any engineering project, a

system needs to be put in place, followed and carefully monitored.

The scope of manufacturing in engineering includes many activities

which need to be carefully evaluated and controlled throughout the

design and manufacturing process. The whole design falls under the

following main categories, which each has its own sub-divisions. The

headings below are guidelines for the process and need to be carefully

evaluated for a successfully designed component.

• Identification of need

• Problem statement and goal definition

• Research

• Development of specifications

• Generation of ideas and concepts

• Preparing the preliminary design and final drawings

• Prototype and laboratory testing

Machine design is the art of developing new ideas for the construction

of machines and expressing those ideas in the form of plans and

drawings. To design any component well, the designer must have a

working knowledge of the elements of machine construction; must


35

know how to analyse applied forces and calculate stresses, must

understand the influence of shape, method of assembling and the

working conditions of parts (Hartman and Maleev, 1962).

3.2) DESIGN PROCESS FOR THE DEVELOPMENT OF

THE FSSW TOOL

3.2.1) Identification of Need

Friction Stir Spot Welding

The first step in design is to indicate the need and to realize that a

need exists in order for the design process to begin. Once the need has

been identified, its requirements can be stated in the processes that

follow.

The need in this case is to be able to perform a FSSW with the attempt

to remove the keyhole in the weld after pin retraction.

Due to FSW being a relatively new process, there are still many areas

that are not yet fully understood. Further research into pin retraction

and material flow around the pin needs to be conducted for the

keyhole to be eliminated. With the keyhole removed at the end of the

weld, the FSSW process could be used in industry and could take over

conventional techniques such as resistance spot welding. The benefits

of this process are discussed within the dissertation.


36

3.2.2) Problem Statement and Goal Definition

Re-design of the Retractable Tool for FSSW

To state the problem means to write down all the available data and

requirements. This would indicate the nature of the problem and give

a clear overview of what needs to be achieved. The main aim is to be

able to develop and re-design the existing retractable pin tool, Patent

Application No. RSA 2004/0538 (Georgeou and Hattingh, 2004), used

for FSW to do FSSW. This will require an in-depth study of the forces

needed to insure tool penetration as well as the method of actuation.

Currently pneumatics is employed for actuation but other options such

as hydraulics and electro-mechanical systems need to be investigated.

Following the problem statement, there are a few general procedures

to follow which will cover the above aspects of the design in much

greater detail.

3.2.3) Research

FSW is a fairly new process and has only in the past few years been

implemented in manufacturing processes for industry. The reason for

this is that the process is still not fully understood and in addition at

the end of the weld, the tool is extracted from the material, resulting in

a keyhole. This leads to the process not being able to be used

effectively in many situations, e.g. it complicates the welding of pipes.

a) Schematic of the keyhole b) Photograph of the keyhole

Figure 3.1: The keyhole

Keyhole


37

There have been many attempts to design a Retractable Pin Tool

(RPT) that will eliminate the keyhole effect. National Aeronautics and

Space Administration (NASA) developed a retractable pin tool (Ding

et al., 1999; Ding, 2000) in cooperation with MTS and MCE

Technologies Inc. in the United States of America, to close the

keyhole. This technology has been successfully implemented on the

space shuttle’s external booster tank, and for the manufacture of

Eclipse aviation aircraft. There is however limited research

concerning the retractable pin tool technology and process (NASA

Marshall Space Flight Centre, 2002; NASA Techtrans, 2001).

A study into the operation of various different tool designs, that have

been manufactured and patented in an attempt to close the keyhole,

will be evaluated. This will help in the design of a FSSW tool, which

needs a retractable pin to be able to close the keyhole. FSSWs that are

produced still have the keyhole when the tool is removed. Therefore a

research tool needs to be designed and manufactured to conduct

further investigations of this problem, and to assist with making

FSSW more acceptable for large scale use. The keyhole at the end of

the weld is a major problem in FSW and FSSW and if removed will

be a huge benefit to many companies and will also lead to many

benefits in the field of fusion.

A few patents and concepts will be discussed in this section, to give a

brief overview of what has been done in the past, in order to help with

the design of a tool that will be able to perform FSSW.

3.2.3.1) Auto-Adjustable Pin Tool for Friction Stir Welding

(Ding and Oelgoetz, 1999)

The tool designed by R.J. Ding and P.A Oelgoetz (1999), Patent

Number 5,893,507, for NASA, is an auto-adjusting pin tool for


38

friction stir welding where the pin tool automatically adjusts for

welding materials of varying thickness. The pin can also be withdrawn

in increments thus eliminating the keyhole effect at the end of the

weld. The pin position in this tool is controlled by an electric motor

and hydraulics.

Figure 3.2: (Ding and Oelgoetz, 1999)

Figure 3.2 illustrates a cross section of the invention by Ding and

Oelgoetz. The method of actuation of the pin is by controlling the

hydraulic pressure in lines 20 and 27. Altering the pressure in these

lines, causes the piston 25 to move to the top or bottom end of the

cylinder 41, relative to the pressure applied. The piston is attached to

the pin 33 resulting in the pin moving in relation to the shoulder 37. A

major advancement in this tool design has been the development of

the control system to form a closed-loop feedback system to control

the pin.


39

3.2.3.2) Programmable Friction Stir Welding Process

(Holt and Lang, 1998)

The tool patented by Holt and Lang (1998), L.J. Patent Number

5,713,507, Programmable Friction Stir Welding Process, also uses

hydraulics to control the pin position.

The pin can be accurately adjusted relative to the shoulder. At the

beginning of the weld, the pin is slowly extended into the material,

and at the end of the weld is slowly retracted, thus no holes are left in

the material at the end of the weld. It also caters for welding different

thickness materials without having to change tools.

The operation of this process is shown below.

Figure 3.3: (Holt and Lang, 1998)

When hydraulic fluid is added to the top of the piston chamber

through the fluid entry tube 11, the pin is forced downwards into the

weld material. When fluid is added to the bottom chamber 12, the pin

is forced upwards. In this way, by controlling the hydraulic pressure

the pin position is controlled (Holt and Lang, 1998).


40

3.2.3.3) Adjustable Pin for Friction Stir Welding

(Wykes, 1997)

The adjustable pin tool for friction stir welding by D.H. Wykes (1997)

also reveals pin retraction, but has the option of varying the shank

diameter for different pin depths, as shown in figure 3.4.

Figure 3.4: (Wykes, 1997)

The pin and shanks positions are mechanically controlled by electric

motors as in figure 3.5.

Figure 3.5: Adjustable pin tool (reproduced from Wykes, 1997)


41

3.2.3.4 Friction Stir Welding Tool for Welding Variable

Thickness Workpieces (Colligan, 1998)

The invention by Colligan (1998), Patent Number 5718366 covers the

bases of welding different thickness materials but does not allow for

the keyhole at the end of the weld to be removed. This process uses a

spring to control the depth of the pin position.

Figure 3.6: (Colligan, 1998)


42

3.2.3.5) Method and Device for Joining at least Two Adjoining

Work Pieces by Friction Welding (dos Santos and

Schilling, 2002)

This method was invented by Schilling and dos Santos (2002), patent

no. US 2002/0179682 A1. This method is applied to lap joints.

In this method, the pin projection is simultaneous with the first sleeve

retraction. This allows the plasticized material to be pressed back into

the joint area. The displaced material is forced back into the void

created by the pin retracting, by the first sleeve. The rate of plunging

and retraction between the pin and the first sleeve needs to be

controlled in order to eliminate the keyhole. The process is shown

schematically in figure 3.7.

Figure 3.7: (dos Santos and Schilling, 2002)


43

3.2.3.6) Retractable Pin Tool for the use with the Friction Stir

Welding Process (Georgeou and Hattingh, 2004)

The design by Georgeou and Hattingh (2004), patent Application No.

RSA 2004/0538, allows for the pin to be accurately adjustable in

length relative to the shoulder. The pin can be extended and retracted.

The actuation of the pin is by means of a “Fluidic Muscle” ® Festo

AG. This method leads to the elimination of the keyhole and can be

used to weld different thickness materials.

Figure 3.8: (Georgeou and Hattingh, 2004)


44

3.2.4) Development of Specifications

Development of Specifications the for FSSW Tool

This section considers the requirements needed to design the FSSW

tool. Preliminary design requirements and specifications are listed

below.

• The tool needs to stimulate pin penetration, in a fully extended

position of 4mm, and must be able to be withdrawn 4mm on

pin retraction.

• The pin position needs to be locked on pin penetration.

• Maximum spindle speed of 500 rpm.

• Heat distribution around the shoulder and other components

needs to be carefully looked at in terms of close tolerances.

• The pin temperature needs to be known at all times.

• The tool must fit in the current experimental FSW machine.

• The tool should be modular in design.

Actuation of the Pin

In this section the advantages and disadvantages of both the pneumatic

muscle and hydraulics will be evaluated for methods of pin actuation.

3.2.4.1) Pneumatics

The current PE Technikon tool design, Patent Application Number

RSA 2004/05 38, uses a fluidic muscle and an attempt will be made to

keep this concept as it has some control advantages.

The fluidic muscle can be powerfully contracted at will. It is

based on a combination of flexible tubes and tightly woven


45

threads in a diamond shape pattern resulting in a three

dimensional grid structure. When internal pressure is applied,

the tubing extends in its peripheral direction, thus creating a

tensile force and a contraction in motion in the longitudinal

direction. When the pressure is released the muscle contracts

and extends in the longitudinal direction (Festo,

2003)

Figure 3.9: Fluidic Muscle (reproduced from Festo, 2003)

Advantages

• High dynamic response, even at high loads

• No moving mechanical parts

• No jerking even with slow movement

• Simple technology without decoders

• Use of air as a fluid medium

• Robust design

• Stroke of up to 25% of its nominal length


46

Disadvantages

• If the fluidic muscle is inflated with compressed air and the

volume blocked, the pressure in the muscle can increase

significantly if the external force applied to the muscle

configuration is varied.

• Air is compressible and this creates stability problems from a

control perspective.

The correlation between the supplied pressure and muscle length is

not linear, but there is a relationship between the two. In the absence

of external force, the pressure versus contraction can be calibrated.

Figure 3.10 contains the calibration information for the 20mm

diameter muscle, where a maximum force of 1200N is obtainable.

Figure 3.10: Fluidic muscle operating range MAS-20

(reproduced from Festo, 2003)

3.2.4.2) Hydraulics

Hydraulics can be defined as the physical behaviour of liquids at rest

and in motion. It includes the manner in which liquids act in tanks and

pipes, deals with their properties and explores ways to take advantage

of these properties. Hydraulics can consist of pressurized oil which


47

provides power for raising and lowering the 3PH and which can be

used to operate attached or towed implements having hydraulic

pistons and cylinders.

If a hydraulic system is well adapted to the work it is required to

perform it can provide smooth, flexible, uniform action without

vibration and is less affected by a variation of load than pneumatics. In

a case of overload, an automatic release of pressure can be installed

protecting the system from failure. Fluid systems are also fairly

economical to operate. Hydraulics is normally applied when the

application requires a great amount of pressure and where extremely

accurate control processes are needed. A pneumatic system is used

when the system requires a medium amount of pressure and fairly

accurate control. A hydraulic system also requires strong pipes and

containers. Leaks due to high pressures obtained could cause serious

problems.

3.2.4.3) Pin Forces

An important aspect in designing the tool is to know the forces

involved and to be able to design the tool within this specification.

Information obtained from papers on forces in friction stir welding,

gives a fair indication of what forces involved are achieved in the

process.

During FSW there are forces exerted by the pin tool upon the material.

Analysis of these forces has concentrated upon the shoulder of the pin

tool, as it is the shoulder force that is directly responsible for the

plunge depth of the pin tool into the surface of the work piece. The

total forces exerted on the pin tool can be evaluated further by looking

at the force behaviour of the pin independently to the shoulder (Ding,


48

2000). This will be discussed later in this chapter with regards to

NASA’s retractable pin tool.

Force Analysis

Before the forces can be analysed, the diameters of the shoulder and

the pin of the tool being designed needs to be known. It was decided

that a pin diameter of 4mm and a shoulder diameter of 12mm would

be used - a 3:1 ratio between the shoulder and the pin. This ratio is

very similar to most cases of FSW tools. Figure 3.11 illustrates this

ratio and other details related to the pin and the shoulder.

Figure 3.11: Illustration of the pin and the shoulder

In the paper by Sakano et al. (2001), “Development of Spot FSW

Robot System for Automobile Body Members” spot welds were

carried out on upper and lower sheets of Aluminium 6xxx series, 1mm

thick.

Composition of the material was as follows:

Table 3.1: Material composition

Si Mn Mg Fe Cu Ti Zn Cr Al

1 1 0.46 0.15 0.08 0.04 <0.01 <0.01 Bal.

Shoulder

area

Pin


49

The weld tool used had a shoulder diameter of 7mm and a probe pin

diameter of 2.1mm. These welds were carried out under a constant

weld force of 2.94kN (Sakano et al., 2001)

This spot weld technique did not have a retractable pin, therefore the

force applied was to the shoulder and to the pin. It is considered

however that the force applied to the shoulder is the greatest.

Therefore calculations were done using the area of the shoulder and

ignoring the forces on the pin in order to get a comparison.

Pressure = Force/Area

Shoulder Area = π(D2-d

2)/4

By calculation, the pressure using the above equations is 76.4MPa.

Using this pressure value, a force required for a 12mm diameter

shoulder and a 4mm diameter pin could be calculated. The values are

shown in table 3.2.

Table 3.2: Calculated forces

Shoulder

Dia.(mm)

Pin

Dia.(mm)

Pressure

(MPa)

Shoulder

Area (m2)

Force

(N)

7 2.1 83.95 3.502 x 10-5

2940

12 4 83.95 1.005 x 10-4

8439.5

The force for a tool shoulder design of 12mm in the extended position

would be approximately 8.44kN.

In the paper by Schilling et al. (2000), “A preliminary investigation on

the static properties of friction spot welds”, welds with pin diameters

of 6,7 and 10mm were produced on overlap joint configurations of

1/3mm, 1.5/3mm, 2/3mm, 3/3mm. The length of the spot weld in this


50

case varied from 8mm to 18mm in order to make an accurate

comparison between RSW. Aluminium Alloy 6061-T4 was used for

both the top and the bottom sheets. However no actual weld forces

were given.

In the paper by Johnson (2001), “Forces in friction stir welding of

aluminium alloys”, welds were created on different series alloys.

Aluminium Alloy series 2014-T6, 6082-T6, 5083-H111 and 7075-

T7351 were used. As it has already been decided that the FSS welds

are going to be performed on 6063-T4, only the forces from the 6000

series aluminium plate were considered. The tests were also carried

out at 500 RPM. The use of different shoulder and pin diameters in

this process is given in table 3.3.

Table 3.3: Different tool combinations

Tool

Shoulder

Diameter

(mm)

Pin Diameter

(mm)

Pin Length

(mm)

K1 25 10 6.1

K2 25 8 5.8

K3 20 8 5.6

K4 20 9 6.15

(adapted from Johnson, 2001)

Different processes which involved changing the dwell periods were

experimented with. The plunge depth was set to about 0.2mm into the

plate surface. The results for the different tools are shown in table 3.4.

The forces shown are assumed to be the forces applied to the shoulder

of the material.


51

Table 3.4: Forces obtained by Johnson

Tool Alloy Rotation RPM Downward Force

kN

Transverse Speed

mm/min

Torque

Nm

K1 6082-T6 500 13.5 80 55

K2 6082-T6 500 14 80 56

K3 6082-T6 500 11 80 41

K4 6082-T6 500 10.5 80 41

The pressure from the applied force can be calculated if the area of the

shoulder is known.

Table 3.5: Calculated shoulder areas and pressures

Tool Shoulder Dia.

(mm)

Pin Dia.

(mm)

Shoulder Area

(mm2)

Downward Force

kN

Pressure

(MPa)

K1 25 10 412.3x10-6

13.5 32.74

K2 25 8 440.6x10-6

14 31.77

K3 20 8 263.9x10-6

11 41.68

K4 20 9 250.5x10-6

10.5 41.92

From table 3.5 it can be observed that as the shoulder area decreases

the downward force follows the same trend.

Using the data obtained in table 3.5, a force versus area relationship

for the 12-4mm pin tool was established, as illustrated in figure 3.12.


52

Force vs. Shoulder Area

y = -0.5569x2 + 5.6056x

0

2

4

6

8

10

12

14

16

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

Area x10-4

(mm)

Fo

rce (

kN

)

Series1 Poly. (Series1)

Figure 3.12: Force versus shoulder area

From the polynomial trend line in figure 3.12, the predicted downward

force for a shoulder diameter of 12mm was calculated to be 5.32kN.

According to Ding (2000) in the paper “Force characterisation on the

welding pin of a friction Stir welding retractable pin-tool using

aluminium-lithium 2195” the total forces exerted on the pin tool could

be evaluated further by looking at the force behaviour of the pin

independently to the shoulder. The Hydraulic Controlled Auto-

Adjustable Pin Tool, RPT, developed by NASA, has been

instrumented with a load detection devise that allows the forces placed

on the welding pin to be detected during the welding process (Ding,

2000).

In this paper by R.J. Ding, the forces have been measured and

characterized using aluminium lithium alloy 2195.

Being able to measure the pin forces will allow for analysis of the

reaction forces created by the plastic material below the weld. Static

plunging trials and welding trials were performed. For evaluation


53

purposes of FSSWs only the static plunging forces will be looked at.

The welds were done at a rotational speed of 225rpm.

Table 3.6: Tool specifications

Tool

Shoulder

Diameter

(mm)

Pin

Diameter

(mm)

RPT 24 9.6

In the various tests, the shoulder remains constant with a plunge depth

of approximately 0.254mm and only the pin length is varied. Four

static plunge tests were completed.

Different tests were conducted by plunging into the material with

different initial pin lengths, and then the pin was extended at a

predetermined rate to a final length. These final pin lengths were

varied. The maximum pin forces given in the results are the pin forces

after plunging has taken place. They ignore the slightly higher forces

induced during plunging when the material is not up to welding

temperature. Figure 3.13 illustrates the first static weld. The forces for

weld trials 2 to 4 were recorded from graphs relevant to each weld.


54

Figure 3.13: Static plunge results of trial weld 1

(reproduced from Ding, 2000)

The summarised results from figure 3.13 and from figures in the paper

by Ding can be seen in table 3.7.

Table 3.7: Summarised results of the static welds

Static

Test no.

9.6mm Pin

Area

(mm2)

Max. Force

during Plunging

(MPa)

Max. Force

during Plunging

(kN)

Initial Pin

Length

(mm)

Final Pin

Length

(mm)

Max. Force

after plunging

(MPa)

1S-IT3 72.38x10-6

69.57 5.036 7.7 8.46 26.73

2S-IT4 72.38x10-6

69.06 4.999 8.13 8.61 28.958

3S-IT5 72.38x10-6

76.64 5.547 8.36 8.56 28.317

4S-IT6 72.38x10-6

79.76 5.773 7.62 8.56 28.999

The force for the 4mm pin was calculated using the same pressure

values from the static tests. The 4mm pin area was used to calculate

the force. The results are tabulated in table 3.8.


55

Table 3.8: Calculated forces for the 4mm Pin

Static

Test no.

4mm Pin

Area (mm2)

Max. Force

during

Plunging (N)

1S-IT3 12.566x10-6

874.24

2S-IT4 12.566x10-6

867.83

3S-IT5 12.566x10-6

963.09

4S-IT6 12.566x10-6

1002.29

From the results obtained in this chapter deductions could be made in

order to complete the tool design in terms of forces applied. From the

paper “Development of spot FSW robot system for automobile body

members” the maximum force on the 12mm shoulder is calculated to

be approximately 8439.5N. With use of data obtained and charts

plotted in the paper “Preliminary investigation on the static properties

of friction stir spot welds”, an approximated value of 5322N was

calculated for the 12mm shoulder.

The force on the pin itself was calculated by using the data from the

paper written by Ding (2000). With the aid of this document the force

was calculated to be a maximum of 1002N on the pin during plunging.

The above results enable a preset standard to be made to set design

criteria. The results of the forces on the shoulder and the pin are

tabulated in table 3.9.

Table 3.9: Calculated pin and shoulder forces

Area Calculated

Forces (kN)

Design

Forces

(kN)

Safety

Factor

Shoulder 8.44 & 5.32 25 3

Pin 1.00 5 5

Structure N/A 25 N/A


56

3.2.5) Generation of Ideas and Concepts

FSSW Tool Design

Information contained in this chapter and chapter 2 was used in order

to complete concepts that could perform the task of FSSW. In this

section no in-depth detail into completed concepts will be evaluated.

Different means of actuating the tool were looked at as well as locking

devices to keep the pin in a fixed position during plunging.

Approximately seven concepts were drawn up and advantages and

disadvantages were evaluated. The final concept used the fluidic

muscle as a means of actuation.

3.2.6) Preparing the Preliminary Design and Working Drawings

FSSW Tool Design

The next part of the design procedure was to transfer the concept into

actual working drawings. The calculations done to ensure that the

structure is both physically strong enough and safe, as well as the

working drawings are illustrated in Appendix A. A Finite Element

Analysis (FEA) of some of the components was done to verify the

calculations. These results are also illustrated in Appendix A.

3.2.7) Prototype and Laboratory Testing

This section can be seen in chapter 4 where the FSSW equipment and

experimental set-up is introduced. The assembly and calibration of the

tool is discussed in detail.


57

3.3) DESCRIPTION OF THE FSSW TOOL

The best concept selected, still used the fluidic muscle as a means of

actuation by operating a cam mechanism that in turn causes pin

retraction. The pin is locked during plunging and can be retracted in

small increments by means of a ratchet system. A spring attached to

the pin applies a required force for pin retraction. As seen in articles,

there is still quite a large force on the pin during welding, therefore it

is assumed that the pin will tend to extract by itself due to the capillary

action of the plasticized material below it. This force pushing up on

the pin, as well as the additional spring force will keep forcing the pin

upwards so that when the cam is operated, the pin will retract. A

clearer understanding of this process will be explained in figures 3.14

to 3.16.

Figure 3.14 illustrates an exploded view of the tool. The outer ring is

not shown but can be seen in Appendix A. Figures 3.15 and 3.16 are

cross sectional views illustrating the operation of the tool.


58

Parts List

Item Part Item Part

1 Shoulder 12 Extension Head

2 Shoulder Flange 13 Cam Pin

3 Pin 14 Flange

4 Lower Ring 15 Muscle Ring

5 Spacer 16 Bush

6 Ratchet 17 Upper Ring

7 Column 18 Outer Ring

8 Locking Ring 19 Grub Screw

9 Extension 20 Pin Spring

10 Ratchet Pin 21 Ratchet Nut

11 Moving Pin 22 Ratchet Spring

Figure 3.14: Exploded view of the FSSW tool


59

Figure 3.15: Cross sectional view of the pin extended

Figure 3.15 illustrates a cross sectional view of how the pin retraction

system is operated. The pneumatic muscle, not shown, is attached to

the muscle ring [15]. When pressure is applied to the muscle, it

retracts and causes the muscle ring [15] to move upwards. The muscle

ring [15] is attached via the flange to the cam pins [13]. Thus when

the muscle ring [15] moves upwards so do the cam pins [13]. When

the cam pins [13] move upward, they force the ratchet pins [10] to

move outwards. The ratchet spring [22] behind the ratchet pin [10] is

held in place by a grub screw [19] and applies an inward force on the

ratchet pin [10]. This spring allows for smooth movement of the

ratchet pin [10]. When the ratchet pins [10] are forced outwards, the

pin spring [20] forces the ratchet [6] to move upwards in steps of

0.5mm. The ratchet [6] is connected to the pin [3]. Therefore when the

ratchet is forced upward by the pin spring [20], so is the pin [3].

When the muscle is fully retracted, the cam pins [13] are pulled all the

way up, resulting in the ratchet pins [10] being forced all the way out,

22

20

1

21

19

15

10

13

3

6


60

allowing the ratchet [6] to move through all its steps causing the pin

[3] to fully retract as illustrated in figure 3.16.

Figure 3.16: Cross sectional view of the pin retracted

3.4) SUMMARY

In chapter 3, a design procedure is put in place and followed. Research

of other patented FSW tool designs and the processes used were

evaluated. Specifications for the tool were developed in this section.

The different methods of actuation were discussed: pneumatics and

hydraulics. An in-depth study of the forces involved on the shoulder

and the pin was analysed and calculated. The actual design of the tool

and its method of operation was also illustrated and explained. The

design calculations and working drawings can be referred to in

Appendix A.

4. Experimental Set-up

61

CHAPTER 4

FSSW EQUIPMENT AND EXPERIMENTAL SET-UP

4.1) INTRODUCTION

In chapter 4 the assembly and calibration of the tool will be discussed,

as well as the experimental procedure. The assembly of the tool

designed provides more insight into the operation of the FSSW tool.

This will create a better understanding of the operation and processes

involved in tool assembly. The layout of the pneumatic system is

illustrated. Calibration of the pneumatic muscle in relation to pin

position is illustrated and discussed. A weld matrix is also set-up in

this section and the reasons for the different process parameters are

discussed.

4.2) ASSEMBLY OF THE FSSW TOOL

The FSSW tool was assembled with high heat chrome compound, to

ensure that all moving parts moved freely. On assembly, it was

discovered that the pin only retracted 3.5mm as opposed to 4mm, this

being due to the changes that were made during machining. This issue

was discussed and it was decided that the tool would still be

operational for the required purpose.

Initially the concentricity of the tool was checked in a lathe, by

clamping the end that is attached to the milling machine, into the lathe

chuck. A dial gauge was used to measure the concentricity at various

points along the tool whilst the chuck was rotated by hand. This

procedure is illustrated in figure 4.1.


62

Figure 4.1: Measurement of concentricity in the lathe

The concentricity was found to be out by 0.2mm at the end of the tool,

and 0.03 at the beginning. Because of the weight of the tool it was

difficult to clamp the tool 100% true in the chuck. However a 0.2mm

run out on the lathe was still considered acceptable considering the

complexity of the design and the length of the tool.

The concentricity tests done in the lathe were done without the adapter

plate that measured a concentricity of 0.15mm, from its inner hole to

its outer diameter.

The next step was to measure the concentricity of the FSW machine to

which the FSSW tool would be attached. The spindle of the machine

was checked with a dial gauge and was running 100% true. The

transducer used for FSW was bolted to the spindle as illustrated in

figure 4.2. The concentricity measured a run-out of 0.1mm on the side

and 0.01mm underneath, as indicated in figure 4.3a and b.

FSSW Tool

Dial Gauge

Lathe Chuck


63

Figure 4.2: Concentricity of the transducer

a) Measurement on the side b) Measurement underneath

Figure 4.3: Positions for the concentricity measurements of the

transducer

The adapter plate was then bolted to the transducer, and turned in such

a way that the concentricity values of 0.1mm and 0.15mm were 180˚

opposite each other. From this it was calculated that the misalignment

should only be 0.05mm, which compared well to the measured

0.06mm on the side and 0.04 underneath.

Transducer

Spindle

Dial Indicator

Telemetry

System


64

The pneumatic adapter’s concentricity was also checked in the lathe

and had a run-out of 0.01mm. It was then bolted to the adapter plate

and measured a run-out of 0.8mm. The extended length of the

pneumatic adapter magnified the run-out of the adapter plate and

transducer. A 0.1mm shim was then placed between the adapter plate

and the pneumatic adapter. The concentricity was measured to be

0.02mm on the side and 0.01mm underneath.

Figure 4.4: Concentricity of pneumatic adapter

Then the tool was bolted to the pneumatic adapter and the

concentricity was measured to be 0.45mm at shoulder. However the

bottom of the lower ring also ran out and a 0.05mm shim was placed

between the pneumatic adapter and the tool. The concentricity was

measured to be 0.03mm on the shoulder, and 0.01 underneath the

lower ring. The milling machine was started and run at incrementing

speeds of 100, 200, 300, 400 and 500rpm to ensure the tool was stable

at equivalent welding speeds.

0.1mm Shim

Adapter

plate

Pneumatic

adapter

Telemetry

system

Dial

indicator


65

Figure 4.5: Concentricity of the tool

4.3) CALIBRATION OF THE FSSW TOOL

The next part of the set-up procedure was to calibrate the pneumatic

muscle to pin position. This required the pressure gauge to be

connected to a one-way pressure-regulating valve and to the tool

itself. The supply pressure was set to a maximum of 6 bar, the

maximum pressure the muscle could withstand. The voltage to the

digital pressure gauge was supplied from a variable voltage supplier,

with the voltage set at 20V and the amps at 0.05A. The schematic set-

up can be seen in figure 4.6.

Figure 4.6: Schematic set-up of the pneumatic muscle

0.03mm

0.01mm

Shoulder

Lower ring


66

Figure 4.7: Retraction calibration set-up

a) Voltage supply b) Pressure supply

Figure 4.8: Voltage and pressure gauges

The first two tests were done with no loading on the pin and the third

test was done with a loading of 9.8kg on the pin. The pressure

supplied to the tool was started at zero and the pin extension was

measured to be 4mm. The pressure was then slowly increased until the

pin retracted in 0.5mm increments. Each time the pin clicked a

position, the pressure was recorded. The procedure was repeated five

times with each different condition. In the first test with no loading,

the ratchet pin grub screw was set to a depth of 6mm, its full depth

position for maximum spring tension. In the second test the grub

FSSW Tool Control valve

Pressure

gauge

Pneumatic

pipes


67

screw was turned out 3mm and in the third test a load of 9.8kg was

placed on the pin to see what effect the different conditions would

have on the muscle pressure compared to pin position. The recorded

data are illustrated graphically in figure 4.9, where comparisons

between the different procedures are shown.

Muscle Pressure compared to Pin Position

0

1

2

3

4

5

6

4 3.5 3 2.5 2 1.5 1 0.5

Pin Position (mm)

Pre

ssu

re (

Bar)

Test 1 Test 2 Test 3

Figure 4.9: Muscle pressure compared to pin position

From these values, the pin position can be determined from the

pressure in the muscle.

The lines on the ratchet pin were also calibrated to pin position. In

table 4.1 the pin position with the corresponding number of lines

visible is shown. This is also illustrated in figure 4.10.


68

Table 4.1: Pin position corresponding to number of lines visible

Pin Position No of Lines

4.0 0

3.5 1

3.0 2

2.5 3

2.0 4

1.5 5

1.0 6

0.5 7

0.0 8

Figure 4.10) Illustration of the lines visible corresponding to pin

position

The next step in calibration was to determine how the different rates

of retraction could be obtained. Considering the supply pressure was

constant at 6 bar, the flow rate could be varied by opening the one-

8 lines

visible

Pin position

fully retracted

Ratchet

pin


69

way control valve different amounts. As seen in table 4.2, the control

valve was opened to 1 1/2, 1 3/8, 1 5/8 and 2 turns with the different

time values for pin retraction shown. The time was taken from when

the valve was opened to when the pressure in the muscle reached

5.7bar.

Table 4.2: Rate of retraction calibration sheet

5.7 Bar, Pin is fully retracted.

No of Turns Time Taken for air pressure to reach 5.7 Bar (s)

1 2 3 4 5 Average

1 1/4 Turns 18.86 20.02 21.1 21.97 20.9 20.57

1 1/2 Turns 6.52 6.52 6.88 7.1 6.97 6.798

1 5/8 Turns 4.44 4.4 4.39 4.31 4.4 4.388

2 Turns 1.6 1.7 1.8 1.85 1.9 1.77

This in turn Equates to the pin retracting at X mm/s

1 1/4 Turns 0.17mm/s

1 1/2 Turns 0.51mm/s

1 5/8 Turns 0.8mm/s

2 Turns 1.977mm/s

Once the tool was assembled and calibrated, it was assembled to the

FSW machine, where all the pneumatic pipes and tool parts were

fitted in preparation for welding to begin. The milling machine was

started and run at various speeds to test the software and to ensure that

the muscle retracted the pin in its dynamic environment.

4.4) THERMOCOUPLE CALIBRATION

Type K thermocouple wire using Waveview for Windows as an

interface and a data recording mechanism, were used so that the

temperature during the weld could be recorded. 2mm holes were

drilled into the 3mm plate as shown in Appendix A, where the


70

thermocouple wires were placed. These wires would only go through

the first plate were held in place by glue or prestik.

Figure 4.11: Thermocouple set-up around a spot weld

Channels 0, 1, 3 and 6 were used for recording as shown in figure

4.11. Channels 3 and 6 were placed 11.50mm away from the centre of

the weld were as channels 0 and 1 were placed 10mm away.

These channels were calibrated previously for other work and were

checked by placing the thermocouple ends into a glass of ice water.

This was done in order to make sure the correct readings were still

being recorded as well as determining the difference between the

various channels. The temperature was recorded for 130s and these

results can be seen in figure 4.12

0

6

1

3

Ø11 ±1mm


71

Calibration 1Thermocouples in Ice Water

Time (s)

0 20 40 60 80 100 120 140

Te

mp

era

ture

(°C

)

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

Channel 0

Channel 1

Channel 3

Channel 4

Figure 4.12: Thermocouples in ice water

The expected temperature of ice water is approximately 0˚C. Thus

there is a maximum variation of +1.25˚C and –0.75˚C. The

temperature difference across the four channels is only 2˚C and there

was minimal temperature fluctuation.

4.5) EXPERIMENTAL PROCEDURE

After the tool was assembled and calibrated, focus was turned towards

welding and the factors that need to be controlled during the welding

process. After deciding which factors were the most important, a weld

matrix needed to be set-up.


72

From the literature survey it was established that the main parameters

required to be monitored and controlled were:

• Shoulder plunge depth: The shoulder is responsible for a

percentage of heat generated during the welding process and

also applies a downward force on the material, causing the

capillary action of the plasticized material.

• Pin plunge depth: The pin is also responsible for heat

generation and the stirring of the two layers together. The

length of the pin will alter both of these factors.

• Dwell time: The dwell time is responsible for heat generation

by the shoulder. Varying the dwell time can have an effect on

the properties of the weld.

• Pin retraction rate: Different rates of retraction can have an

effect on the properties of the weld.

• Spindle speed: The spindle speed is also responsible for

causing frictional heat and in conjunction with dwell time can

have an effect on the heat supplied during the weld process.

• Temperature: By controlling the above parameters a trend for

process can be monitored

• Welding time: The time period is measured from when the pin

is in contact with the material to when the tool is retracted.

This includes the time taken for the pin to plunge, dwell time

and retraction of the pin

• Pin and shoulder design: The shape of the pin and the

shoulder can have a major effect on the properties of the weld

as well. Only through experimental procedure can the correct

tool be obtained.


73

These parameters needed to be accurately controlled in order for good

repeatability of the welds to be obtained.

Preliminary test were carried out to check the operation of the tool and

to evaluate the tool performance. Various different combinations of

process controlling factors and tool designs were analysed. This can

be seen in the weld matrix shown in section 4.5.1.

4.5.1) The Weld Matrix

A weld matrix was set-up in this section. The reason for this was to

establish a welding procedure so that different parameters could be

successfully evaluated. In table 4.3 a brief description of some

welding terms is covered. Initial starting parameters and variables are

also shown. In tables 4.4 to 4.6 the different processes with different

parameters are tabulated. This matrix is a guide to ensure that certain

parameters such as dwell time, pin length, rpm and rates of retraction

can be successfully evaluated at the end of the welding process. The

matrix also allows for the best parameters at the end of each section to

be carried over to the following procedure; thus theoretically creating

the best parameters for the FSSW at the end of the weld matrix.


61

Table 4.3: Initial welding procedure

Dwell Time:

Plunge Depth:

Rate of Plunging:

Pin Plunge Depth:

Rate of Retraction

T

S

dh1

dm1

dl1

rh2

rm2r2

sh1

sm1

sl1

Sample 1

Sample 1

each other by 100mm

The upper plate will be 3mm and the lower plate 2mm thick, overlapping

Spacing between weld centres will be 26 mm apart, with the centre of the

Sample 2Sample 2

Temperature

Shear strength

˚C

kN

For each variable, 7 welds will be done.

2 for microstructure analysis

3 for shear strengh testing

first spot being 22mm from the edge of the plate

Dwell time high

Dwell time medium

Dwell time low

1 for visual inspection

1 available for future use if necessary

Rate of retraction high

Rate of retraction mediumRate of retraction low

Spindle speed high

Spindle speed medium

Sample 1

Spindle speed low Sample 1

Distance the pin is plunged into the material

Time taken to retract the pin per 0.5mm

Sample 1

Sample 1

0.2mm

Rate of Retraction

400 RPM

Rate of Retraction

until plunge depth

WELDING PROCEDURE

Pin Plunge Depth

s

RPM

mm/s

mm

Time from when the shoulder is in contact with

Rate that the pin is plunged into the material

Distance the shoulder is plunged into the

material

the material.

Plunge Depth

Initial Settings

Rotational Speed

Plunge Depth

Dwell Time

Variables

Dwell Time

Rotational Speed

Constants

Sample 2

Rate of Plunging 0.25mm/s

0.2mm

Variable

0.5mm/3s(1.5 turns)


62

Table 4.4: Welding matrix, test 1

Constants (s)

Spindle Speed 400 RPM 24

Plunge Depth 0.2 mm 16

Retraction Rate 1 1/2 mm/s 8

Sample Number 1 2 3 4 5 6 7

Dwell Time 24 24 24 24 24 24 24

Temperature Tdh1 Tdh2 Tdh3 Tdh4 Tdh5 Tdh6 Tdh7

Shear Strength Sdh1 Sdh2 Sdh3 Sdh4 Sdh5 Sdh6 Sdh7

Dwell Time 16 16 16 16 16 16 16

Temperature Tdm1 Tdm2 Tdm3 Tdm4 Tdm5 Tdm6 Tdm7

Shear Strength Sdm1 Sdm2 Sdm3 Sdm4 Sdm5 Sdm6 Sdm7

Dwell Time 8 8 8 8 8 8 8

Temperature Tdl1 Tdl2 Tdl3 Tdl4 Tdl5 Tdl6 Tdl7

Shear Strength Sdl1 Sdl2 Sdl3 Sdl4 Sdl5 Sdl6 Sdl7


Dwell Time 24 24 24 24 24 24 24



Dwell Time 16 16 16 16 16 16 16



Dwell Time 8 8 8 8 8 8 8




Dwell Time 24 24 24 24 24 24 24



Dwell Time 16 16 16 16 16 16 16



Dwell Time 8 8 8 8 8 8 8



High

3mm.A high, medium and low dwell time will be tested.

Dwell Time

Variable

3mm Pin Plunge Depth


Three different dwell times will be tested, with the pin plunge pepth being 4, 3.5 and

Dwell Time Varied

3.5mm Pin Plunge Depth

Time

Medium

Low

TEST 1


63


Constants

Spindle Speed 400 RPM High

Plunge Depth 0.2 mm Medium

Dwell Time Low


Rate of Retraction High High High High High High High

Temperature Trh1 Trh2 Trh3 Trh4 Trh5 Trh6 Trh7

Shear Strength Srh1 Srh2 Srh3 Srh4 Srh5 Srh6 Srh7

Rate of Retraction Low Low Low Low Low Low Low

Temperature Trl1 Trl2 Trl3 Trl4 Trl5 Trl6 Trl7

Shear Strength Srl1 Srl2 Srl3 Srl4 Srl5 Srl6 Srl7















TEST 2

4, 3.5 and 3mm. The medium rate of retraction, 0.5mm/3s, was used in Test 1 and

Pin Rate of Retraction Varied

Three different pin rates of retraction will be tested, with the Pin Plunge Depth being

therefore need not be repeated as the results are already recorded for each dwell time.

Rate of 0.5mm/0.5s

mm/s

Retraction

Variable

3.5mm Pin Plunge Depth


0.5mm/3s

0.5mm/6s


Best From Test 1(X1)


64


Constants

Rate of Retraction High 450

Plunge Depth Medium 400

Dwell Time Low 350


Spindle Speed High High High High High High High

Temperature Tsh1 Tsh2 Tsh3 Tsh4 Tsh5 Tsh6 Tsh7

Shear Strength Ssh1 Ssh2 Ssh3 Ssh4 Ssh5 Ssh6 Ssh7


Temperature Tsl1 Tsl2 Tsl3 Tsl4 Tsl5 Tsl6 Tsl7

Shear Strength Ssl1 Ssl2 Ssl3 Ssl4 Ssl5 Ssl6 Ssl7

Three different spindle speeds will be tested. The best pin penetration

length will also be used (Z1). The medium spindle speed, 400rpm, was used

Pin Plunge Depth (Z1)

RPM

Best From Test 2(Y1)

in Test 2 and therefore need not be repeated as the results are already

Variable

TEST 3

RPM VARIED

Spindle Speed

Best From Test 1(X1)

0.2mm

recorded for all rates of retraction.


61

4.5.2) Machine Operating Procedure

• Start machine and set the rotational speed to the required rpm.

• Adjust the initial pressure in the muscle to 0 bar.

• Plunge the tool into the material to a depth of 0.2mm while the

pin is locked in full extension.

• After a certain dwell time, increase the pressure in the muscle,

as per the calibration, to obtain the desired rate of retraction.

• Once the pin is fully retracted the pressure applied by the tool

shoulder on the material may be released.

4.5.3) Number of Welds done with Variables Constant

Seven welds with the same settings will be done, even if any changes

need to be made to improve the weld. The reason for this is to check

for repeatability as well as to have two test samples for

microstructural analysis, three samples for shear testing, one sample to

keep as an example and one sample remaining available if further

testing at a later stage needs to be done.

4.5.4) Data Recorded

All data from all the welds needed to be recorded. This would create a

large experimental database, as well as the ability to be able to make

accurate and calculated assumptions at a later stage.

Rotational speed, plunge depth, welding force, dwell time, rate of

retraction and temperature will be recorded.


62

4.6) FSSW EVALUATION PROCEDURE

After the first weld was made, an evaluation needed to be done, to see

whether the tool had done its specific purpose of welding the materials

together and eliminating the keyhole. In the flow diagram shown in

figure 4.13, the evaluation process is illustrated.

Figure 4.13: Flow diagram of the evaluation process

Weld Evaluation

• Are the two sheets welded together?

• Is the keyhole eliminated?

• How is the appearance of the weld?

• Were there any weld problems?

Yes No

Eliminating the Problem

What are the problems?

How can they be eliminated?

Make recommended changes.

Weld Improvements

Were there any minor

problems?

What can be done to further

improve the weld?

Creating a FSSW

Creating a new

FSSW weld

The above questions need to be

asked, and with initiative a

decision needs to be made

whether the welds are acceptable

or not, to see whether the matrix

can be continued

Continue with

the weld matrix


63

The flow diagram illustrated in figure 4.13 is a guideline for the weld

process shown in the matrix set-up. The procedure of the matrix set-

up may be changed if deemed necessary. One thing that is aimed for

in the welding procedure is to change only one variable at a time. This

will eliminate any guesswork that may result from two parameters

being changed at once and causing a change in the weld properties.

4.7) SUMMARY

In chapter 4 the assembly of tool was discussed. This included

measuring the tool’s concentricity both in the lathe and on the FSW

machine. Calibration of the pneumatic muscle to pin position was also

done. The calibration of the thermocouples in ice water was discussed

and the set-up of the thermocouples around the spot weld was

illustrated. The experimental procedure in the form of a weld matrix

was tabulated. The evaluation procedure was also shown in the form

of a flow diagram.

5. Experimental Procedure

81

CHAPTER 5

DEVELOPMENT OF THE EXPERIMENTAL PROCEDURE

FOR FSSW

5.1) INTRODUCTION

Chapter 5 consists of the weld procedure and evaluates the

characteristics of the different process parameters used. A step-by-step

approach for each weld was used so that the process undertaken could

be clearly understood.

5.2) MACHINE SET-UP

The equipment to perform the FSSW was set-up. The air supply,

pressure regulator, pressure gauge and power supply were already

installed to the milling machine. The thermocouples and the software

for data-recording were set-up. Clamps were designed, using parts of

the existing FSW clamps, to hold the pieces of material in position.

The first two plates were placed in their positions, with the 3mm plate

being on top of the 2mm plate, and fastened down. The thermocouples

were then installed into the pre-drilled with aid of a hot melting glue

gun. The machine set-up is shown in figures 5.1 to 5.3.

Figure 5.1: Machine set-up

Telemetry

system

FSSW Tool

Clamps

Pressure gauge

Thermocouples

Safety shield

Air supply

Plates


82

Figure 5.2: Thermocouple software set-up

Figure 5.3: Welding clamps

Backing Plate

Holes for

Thermocouples

3mm Plate

Clamps

FSSW Tool

2mm Plate


83

5.3) FSW OPERATING SYSTEM

5.3.1) Operating Procedure of the FSW Machine

The following is a brief summary operation procedure of the FSW

machine.

The procedure is as follows:

• Turn on the main power supply

• Turn on the secondary power switch

• Turn on both computers (User interface)

• Open the FSWGUI interface on computer 1

• Open the X and Z channel commands on the main computer

• Use computer 1 to create a link to the main computer

• Turn on the FSW machine via the user interface

• Acknowledge faults

• Set-up process control parameters

Once all the parameters were set, the whole process was run in mid

air, before an actual weld was done. This was done to make sure the

plunge depth was correct and that the operation ran without problems.

The weld data recording telemetry was then cleared.

5.3.2) The Actual Weld

Before each weld was commenced, the temperature recording

telemetry was started before plunging occurred. The weld data

telemetry had to be cleared before each weld and then saved directly

after the weld was done.

By clicking, “start weld,” the spindle started turning at a low rpm, ±

200rpm. The bed was then raised until the pin just touched the

aluminium plate. This measurement was then also checked on the bed


84

measuring dial. The “continue weld” button was then clicked and the

pin plunged 4.2mm into the aluminium, having a shoulder plunge

depth of 0.2mm. This depth can obviously be altered in the parameter

set-up menu. After a specific dwell period, which was timed manually

from when the bed dial stopped moving, the pin was retracted by

opening the pressure regulator a certain amount. After the pressure

reached 5.7bar, which is the theoretical pressure when the pin should

be fully retracted, the “continue weld” button was clicked and the

shoulder retracted from the material. The milling machine software

has a delay of approximately 5s before the bed begins to move away

from the shoulder after the “continue weld” button is pushed.

Interference from the milling machine caused a slight increase in

temperatures that were recorded. This noise that affected the

thermocouples can be seen in the temperature noise calibration sheet.

5.4) TEMPERATURE CALIBRATION

With the thermocouples already attached in their specified places with

glue, the machine was turned off. The temperature logger was then

turned on to record data. Data was then recorded for approximately

20s for each different setting to see what effect each component on the

FSW machine had on the temperature.

The FSW machine was first:

A - Switched off

B - Switched on

C - X bed motor was turned on

D - X bed motor was switched off

E - The Z bed motor was switched on

F - The spindle was rotated at 100rpm

G - The Z bed motor was turned off


85

The reason for the Z-bed motor and the spindle being kept on at the

same time was due to the fact that when welding, they are the only

two motors operational. It is therefore important to know what effect

these two items will have on the temperature. The results can be seen

in figure 5.4.

Temperature CalibrationEffect of Machine Interference on Temperature

Time (s)

0 20 40 60 80 100 120 140 160

Tem

pera

ture

(°C

)

10

20

30

40

50

60

70

80

Channel 0

Channel 1

Channel 3

Channel 6

A B C D E F G

Conditions:

A Machine switched off

B Machine switched on

C X bed motor was turned on

D X Bed motor was switched off

E The Z bed motor was switched on

F The spindle was rotated at 100rpm

G The Z bed motor was turned off

Figure 5.4: Effect of machine interference on temperature


86

As can be seen from figure 5.4, when the Z-bed motor and the spindle

are rotated, the temperature rises to approximately 50˚C. This is about

30˚C higher than the ambient temperature of about 20˚C measured

when the machine was off. Thus the values obtained during welding

will be about 30˚C higher than the actual values recorded.

5.5) CREATING FRICTION STIR SPOT WELDS

With the set-up being completed and calibrated the welding tests

commenced. In this section of chapter 5, a rundown of each weld and

its associated problems is discussed. All weld parameters not

mentioned can be obtained from Appendix B, referenced to the weld

number.

Weld 1

The first weld was conducted with the following parameters.

Table 5.1: Welding parameters of weld 1

Parameters

Date 2004/09/08

Shoulder Number 1

Pin Number 1

Plunge Rotation Speed 400rpm

Weld Rotational Speed 400rpm

Rate of Plunging 0.25mm/s

Plunge Depth 0.2mm

Pin Length 4mm

Dwell Time 8s

Rate of Retraction 1 1/2 Turns

Spring Number 1

The weld was done according to the procedure discussed in section

5.2 of this chapter. The temperature was recorded and the data was

cleared before and saved after the weld.


87

After the weld was complete it was noticed that although the

pneumatic muscle was fully retracted the pin was still protruding from

the shoulder - the pin did not retract all the way. When measured

afterwards and looking for how many lines were visible, it was

recorded that the pin only retracted 1mm. On visual inspection it could

be seen that the welded material had formed a chamfer between the

shoulder and the pin as seen in figure 5.5a.

a) Shoulder and pin b) Spot weld

Figure 5.5: Weld 1

Due to the pin not retracting, the keyhole could not be closed. This

lead to solving the pin retraction process before the weld matrix could

be continued. Other data that were recorded during and after the weld

can be seen in table 5.2 and 5.3. The maximum temperature recorded

during the weld was 245.9˚C.

Chamfer

Thermocouple

holes

Thermocouples


88

Table 5. 2: Data from weld 1

Thermocouple installed Yes

Temperature Logged Yes

Data Cleared and Saved Yes

Pin Position after weld 3mm

Number of lines 2

Pin retracted X mm 1mm

Muscle Fully Retracted Yes

Shoulder Clean No

Pin Free No

Keyhole Visible Yes

Stabiliser No

Table 5. 3: Data from weld 1

Plunging Time 19.8s Measured Keyhole Depth 4.02mm

Welding Time 41.8s Measured Keyhole Diameter 3.94mm

Force Time 22s Measured Plunge Depth 0.23mm

The above data for all the welds can be seen in tables in Appendix B.

Temperature graphs can be seen in Appendix C.

Weld 2

One of the possibilities that the chamfer in weld 1 occurred was due to

the temperature of the weld being too low, the material not being

plasticized enough. Weld 2 would therefore be done with the same

parameters however increasing the dwell time to 16s.

Weld 2 however suffered the same problems as weld 1, the pin only

retracted 1mm before being jammed by the chamfer of welded

material between the pin and the shoulder. The temperature only

reached a maximum value of 253.9˚C, not substantially greater than in

weld 1.


89


Figure 5.6: Weld 2

Weld 3

Seeing that an increase in dwell time did not solve the problem of the

chamfer forming, alternative parameters needed to be changed. The

design of the shoulder was changed to a flat bottom, as the first

shoulder design promoted the material being forced between the

shoulder and the pin. This is clearly illustrated in figure 5.7. The dwell

time was kept at 16s. A stiffer retraction spring of 5.36N/mm replaced

the initial spring of 2.5N/mm to help with the retraction of the pin.

The thermocouples were also held in place by prestik rather than the

melted glue. This was done for easier use of the thermocouples

without damaging them on removal.

a) Shoulder 1 b) Shoulder 2

Figure 5.7: Cross sectional view of shoulders 1 and 2

Chamfer


90

Weld 3 gave the same results as weld 1 and 2, with the chamfer still

being formed but not as severe as in the first 2 cases. The pin only

retracted 0.5mm. The maximum temperature recorded was 257.5˚C

which is not significantly higher than with shoulder 1.


Figure 5.8: Weld 3

Weld 4

Weld 4 was done with the same parameters as weld 3, shoulder 2 and

the 5.3N/mm Spring. The difference being that the dwell time was

reduced back to 8s. Once again the chamfer between the pin and the

shoulder was formed and the pin only retracted 0.5mm. Maximum

temperatures of 275.2˚C and 250˚C were recorded.


Figure 5.9: Weld 4


91

Weld 5

After the first four welds the same problem still existed. For weld 5

the retraction spring was once again increased in stiffness to

9.93N/mm in order to help the pin retract. The dwell time was kept

constant at 8s. Due to sensor failure only temperature from channel 3

was recorded. The chamfer once again formed and the pin only

retracted 0.5mm. The maximum temperature recorded by channel 3

was 211˚C.


Figure 5.10: Weld 5

Weld 6

After the previous five welds the pin still did not retract as intended.

After evaluating the problem, it was decided that the shape of the pin

needed to be altered. From the previous welds it was thought that as

the pin started retracting, the thin film of aluminium that remained on

the pin had nowhere to go and formed an aluminium chamfer between

the pin and the shoulder on retraction. Therefore the pin was modified;

the welding part was stepped down to 3.2mm. This would allow for

the material in contact with the pin to be removed into a hole with a

diameter of 4mm when retracting. This can be seen in figure 5.11.

Chamfer


92

a) Pin 1 b) Pin 2

Figure 5.11: Pin tool 1 and 2

No temperature was recorded for weld 6, as emphasis was on getting

the pin to retract. Weld 6 was also done on a new sheet of material.

During the weld, after opening the control valve and allowing the

muscle to retract, the shoulder started swaying across the top of the

material. After inspection of the tool it was seen that the pin had in

fact retracted 2.5mm. This leads one to believe that the pin acts as a

guide, like a rudder for a boat, during the welding process and as soon

as the pin retracts there is no centre piece to guide the shoulder. This

allows the shoulder to move around on top of the work piece, as the

frictional forces of the shoulder are acting in different directions. On

inspection of the weld it also appeared that the plunge depth looked

shallow.


93


Figure 5.12: Weld 6

Figure 5.12a shows the pin position after the weld. The gap created by

the stepped pin allows the material on the pin to be retracted into the

gap as shown. Figure 5.12b shows the pin being pushed back to its

initial position. The residue aluminium can be seen on the pin.

Figure 5.13: Weld 6

The swaying action of the tool caused the appearance of the weld to

be poor. However, promising signs of pin retraction were shown in

this weld. No signs of the keyhole being closed were evident.

Weld 7

The shoulder was once again modified to try and force the material

away from the pin without forcing the material to the outside of the

shoulder. This was done by machining a circular groove around the

centre of the shoulder, as seen in figures 5.14 and 5.15.

Gap

Aluminium

residue


94

a) Shoulder 2 b) Shoulder 3

Figure 5.14: Cross sectional view of shoulders 2 and 3

Figure 5.15: Shoulder 3

After experiencing excessive swaying of the tool in weld 6, after pin

retraction, it was deemed important to recheck the concentricity of the

tool. The concentricity was measured to be 0.08mm, which was

considered acceptable. The pin length was also verified after changing

the shoulder. Chrome compound was placed on the moving parts

when the shoulder and pin were re-assembled. Weld 7 was done with

the same parameters as weld 6; the only difference being that the

correct plunge depth of 0.2mm was used.

The tool also swayed across the top of the material, in weld 7, once

the pin had been retracted. This swaying of the tool leads one to

believe that the pin retracted. Visual inspection showed that the pin

had retracted 3.5mm. This was the first weld where the pin retracted

as intended. However the swaying action of the tool was a concern

Circular

groove

Circular

groove


95

and weld 7 once again proved that the pin acts as a guide for the

shoulder and once removed causes the shoulder to wander around on

top of the welded material.


Figure 5.16: Weld 7

Figures 5.16a and 5.16b show the effects that the swaying action of

the tool had on the weld. The weld is of very poor appearance, looking

very rough and smothered. This can also be seen by the fact that the

keyhole looks closed, but with further inspection, it was the smearing

action by the wandering tool that made it appear closed. Underneath

the pin-sized hole as illustrated in figure 5.16b, the keyhole still

remained. The residue built up on the shoulder is also an indication

that the shoulder was swaying on the work piece.

Anti-Swaying Bracket Installed

Due to the fact that the tool started wandering about on the material

surface when the pin started to retract, an extra support bracket

between the milling machine and the tool was designed. This was a

safety item, which was installed to ensure safe use of the tool and to

prevent the tool bolts from shearing off.

Pin size hole


96

One of the problems with the weld process was the delay time from

when the computer “stop weld” command was pushed. The reaction

time of the machine to start removing the shoulder from the material is

approximately 5s, and this enhances the swaying effect. This is the

stage where the shoulder needs to be removed instantly to reduce the

wandering effect of the tool.

Figure 5.17: Anti-swaying bracket installed

Weld 8

Weld 8 was done with the same parameters as weld 7, however this

time the stabiliser bracket was in place. Although there was still a

small amount of movement when the pin retracted, it was deemed

acceptable in terms of safety. The pin only retracted 2mm in this weld

and no signs of the keyhole being closed were visible. An aluminium

chamfer appeared again.

Stabiliser

bracket

Bearing

FSSW Tool


97


Figure 5.18: Weld 8

Weld 9

Due to the aluminium build up, the shoulder was dipped in

hydrochloric acid (HCl), which cleaned the off aluminium. Weld 9

was done with the same parameters as weld 8, except the plunge depth

was approximately 0.1mm to deep. The pin retracted 2mm and no

swaying of the tool occurred.


Figure 5.19: Weld 9

Weld 10

Weld 10 was done with the same parameters as weld 9, except that the

retraction rate of the pin was increased to 2 turns, making pin

retraction faster. Because the keyhole had still not been removed, the


98

matrix set-up was changed slightly. With time constraints limiting the

number of welds being done, the matrix was also changed in terms of

number of welds per parameter setting being reduced. In weld 10 the

pin retracted 2mm. The shoulder was again covered in aluminium and

cleaned with HCl acid before the next weld was done.

Figure 5.20: Weld 10

Weld 11

Weld 11 had the same welding parameters as weld 10; however

chrome compound was placed on the moving parts when reassembled.

Some of this chrome compound was placed on the shoulder by

accident and could be seen by smoke that was generated by the

shoulder during welding. The pin retracted 3.5mm and the shoulder

was free of aluminium after the weld. The keyhole still remained at

the end of the weld.


As can be observed in figure 5.21, the chrome compound made the

visual appearance of the weld very poor.


99

Weld 12 and 13

Weld 12 and 13 were done with the same parameters as weld 9 except

that the retraction rate of the pin was reduced. By opening the control

valve to 1 ¼ turns, the retraction was slower. In both welds the pin

retracted 3.5mm. Some chrome compound once again was left on the

shoulder of weld 12. Weld 13 was done on a new sheet of material.

There were still no signs of the keyhole being removed.

a) Weld 12 b) Weld 13

Figure 5.22: Weld 12 and 13

As can be seen from figure 5.22, the chrome compound on the

shoulder left the visual appearance of the weld poor once again. More

care in the placement of chrome compound on the moving parts was

therefore taken.

Weld 14

With still no signs of the keyhole being closed, some parameters

needed to be changed to try and eliminate the keyhole. Considering

there were no signs of the keyhole being closed in the previous welds,

the question, “Why is the keyhole not being filled?” had to be asked.

It was thought that the material that was removed by the pin during

plunging escaped round the periphery of the shoulder during the dwell

time. Thus there was no material to fill the hole created when the pin

retracted. The dwell time was therefore reduced to 0 seconds. The pin


100

was retracted as soon as the shoulder came into contact with the

material. This would allow for the shoulder to push material into the

void, created by the pin retracting, as soon as the shoulder was in

contact with the material. This method of opening the air control valve

was done by reading the dial indicator on the milling machine bed. As

soon as the indicator reached 3.8mm pin plunge depth, the air was

opened. This process was done with a fast retraction rate of 2 turns on

the control valve. The pin retracted 2.5mm and the shoulder was

clogged.


No signs of the keyhole being eliminated are shown in figure 5.23.

Weld 15

Weld 15 was done with the same parameters as weld 14, except that

the plunge depth was increased to 0.3mm and the retraction of the pin

was started at 3.5mm of the pin plunge depth. The tool started

swaying during the weld and the pin retracted 3.5mm.


101


From studying figure 5.24 it appeared that the keyhole had been

removed. After inspection of the weld, it was found that the keyhole

still remained. The swaying action of the tool had smeared the surface

of the material again. The two materials also did not bond to each

other, leading one to believe that the pin was retracted too soon,

before creating a bond between the two materials.

Weld 16

Weld 16 was done with the same parameters as weld 14, no dwell

time, plunge depth 0.2mm, a fast pin retraction and the pin was

retracted at 3.8mm pin plunge depth. The RPM was reduced to 300

rpm, as per the matrix set-up. Just before the weld took place the main

computer froze and caused a delay in the welding process. The pin

only retracted 1.5mm.



102

Weld 17

Weld 17 was also done at 300rpm. The dwell time was changed back

to 8s and the pin retraction rate set back to a medium rate. The pin

retracted 2.5mm.


Weld 18

In weld 18 and 19 the rpm was increased to 500rpm. In weld 18 the

other parameters were the same as weld 17. The pin retracted 3mm.


Weld 19

With the spindle speed being set at 500rpm, the other parameters were

set the same as for weld 14, no dwell time and a fast pin retraction.

The pin retracted 3mm.


103


From the welds where the dwell time was decreased to 0s, and the rpm

varied between 300 and 500rpm, no signs of the keyhole being

eliminated were observed. When the dwell time was 8s and the

spindle speed changed, no success in eliminating the keyhole was

achieved.

Weld 20 - 26

Before weld 20 was done the welding pin was removed and the

diameter of the welding pin was reduced to 2.85mm. This was done in

order to compare the weld integrity of different pin diameters. It was

also reduced to see what effect it would have on the keyhole being

eliminated. With a smaller area to fill, elimination of the keyhole

might become successful. These welds were all done on a new sheet

of material.

Welds 20 to 26 were all done with the same parameters, the dwell

time being set to 8s, medium pin retraction and the pin plunge depth

being 4mm. The plunge depth of weld 21 and 22 was approximately

0.2mm too deep. When the stop weld button during weld 24 was

pressed, the shoulder did not remove, and stayed in contact with the

welded material for an extra 38s. Therefore this left weld 20, 23, 25


104

and 26 to be evaluated. Pin retraction varied between 2.5mm and

3.5mm and can be seen in Appendix C for individual weld details.





No elimination of the keyhole was seen. Maximum temperatures of

228.9, 224.7, 223.5 and 214.6°C were recorded. The two welds with

deeper plunge depths showed marginally higher temperatures of 230.5

and 237.1°C. When the shoulder was not removed in weld 24, the

temperature did not rise as high as one would have expected and only

reached 238.8°C.


105

Weld 27 - 29

Welds 27 to 29 were done with the same welding parameters as weld

20 – 26, however the dwell time was increased to 16s. This was in

order to create a comparison between the two different dwell times

with the 2.85mm diameter pin. Thermocouples were installed during

these welds. Weld 27 was also done on a new sheet of material. The

pin retracted 3.5mm, 3mm and 2.5mm respectively. The shoulder was

also clean in all three welds. The plunge depth of weld 29 was

measured to be 0.2mm to deep.




The maximum temperatures recorded respectively were 230.7°C,

232.4°C, 245.7°C. No signs of the keyhole being eliminated were

observed.


106

Weld 30 - 33

Welds 30 – 33 were done with the same parameters as welds 20 – 26

with the exception that the pin length was reduced to 3mm. Thus the

pin would only penetrate the underneath sheet by 0.2mm. Reducing

the pin length was done for two reasons: To try and eliminate the

keyhole and to compare the weld integrity compared to the 4mm pin

length. The idea being that the material might be more plasticized

closer to the top surface causing the capillary forces to be greater. The

“stop weld” button was pressed when the air pressure reached 4.5bar,

as this is the pressure corresponding to 3mm of pin retraction. The pin

retracted 1.5mm, 2mm, 3.5mm and 3.5mm respectively. The shoulder

was clean after each weld.






107

There was still no elimination of the keyhole observed. The plunge

depth of weld 32 was about 0.05mm too shallow. After separating the

welds on the sheet it was noted that weld 32 did not make a permanent

bond between the two layers. The maximum temperatures for the

above welds were 224.3°C, 150.4°C, 142.4°C and 211°C. The

temperatures recorded for weld 31 and 32 were substantially lower

than the welds 30 and 34. This could be due to the shallow plunge

depth in weld 32.

Weld 34 - 37

These 4 welds were done with the same parameters as weld 30 – 34,

with the exception that the dwell time was increased to 16s. The pin

length remained at 3mm. Weld 34 was done on a new sheet of

material. The pin retracted 3.5mm, 2mm, 3.5mm and 3.5mm

respectively. The temperature for weld 36 was not recorded. Swaying

of the tool during weld 34 and 35 was visible.




108



The maximum temperatures recorded from the above welds were

136.2°C, 187.8°C and 217.7°C for welds 34, 35 and 37. There were

still no signs of the keyhole being removed. Once again with the 3mm

pin length two of the welds did not create a successful bond, these

being weld 34 and 35.

Resistance Spot Welds

Resistance Spot Welds (RSW) were done on the same material by an

outside company. These welds were however done with both sheets

being 2mm thick. The RSWs were conducted with a welding tip of

16mm in diameter and a 5mm diameter flat section in contact with the

material. These RSWs were done in order to make comparisons

between FSSWs and RSWs. The results of these comparisons can be

seen in chapter 6.


109

Overview

From the above 37 welds a number of comparisons and conclusions

can be made. These results can be seen in chapter 6. There are

however a few points that need to be noted.

The keyhole was never eliminated throughout all the welds. Although

the matrix set-up was not followed exactly due to problems and time

constraints, comparisons of the varied parameters could still be done.

Even though the pin did not retract to its full extent in all the welds,

there was still pin retraction with no signs of the keyhole being filled

by plasticized material. Some faulty temperature recordings of certain

thermocouple channels were ignored and the data from the remaining

channels were used.


110

Table 5.4: Summary of the weld parameters

Summary of Weld Parameters

4mm Pin

All welds were done with pin tool 1

Shoulder Dwell

Time (s) RPM

Rate of

Retraction

Plunge depth

(mm)

Weld 1 1 8 400 Medium 0.2



Weld 2 1 16 400 Medium 0.2

Weld 3 2 16 400 Medium 0.2

3.2mm Pin

All welds were done with shoulder 3, except weld 6 was done

with shoulder 2.

Dwell

Time (s) RPM

Rate of

Retraction

Plunge

depth (mm) Pin Length

Weld 6 8 400 Medium 0.2 4mm




Weld 10 8 400 Fast 0.2 4mm

Weld 11 8 400 Fast 0.2 4mm

Weld 12 8 400 Slow 0.2 4mm

Weld 13 8 400 Slow 0.2 4mm

Weld 14 0 400 Fast 0.2 4mm

Weld 15 0 400 Fast 0.3 4mm

Weld 16 0 300 Fast 0.2 4mm



Weld 19 0 500 Fast 0.2 4mm

2.85mm Pin

Dwell

Time (s) RPM

Rate of

Retraction

Plunge

depth (mm) Pin Length

Weld 20-26 8 400 Medium 0.2 4mm

Weld 27-29 16 400 Medium 0.2 4mm

Weld 30-33 8 400 Medium 0.2 3mm

Weld 34-37 16 400 Medium 0.2 3mm


111

5.6) METHODS FOR TESTING THE WELD INTEGRITY

Before any testing could be done, the welded samples needed to be

separated from each other to allow for testing of each individual spot

weld. The welds were numbered on both sheets by inscribing the

number of the weld into the aluminium to ensure that there was no

confusion once the plate had been cut up. The samples were then

separated from each other by cutting in-between the welds on the band

saw. This left ±20mm strips of the top and the bottom sheet of

aluminium. An illustration of how the plates were cut up can be seen

in figure 5.37.

Figure 5.37: Illustration of how the plates were sectioned

Bottom plate,

2mm thick

Top plate,

3mm thick

FSSW

Section

lines

Section used for

microstructure

and hardness

Section used for

tensile testing


112

5.6.1) Tensile Tests

Tensile tests on welds 1, 2, 3, 4, 7, 8, 17, 18, 20, 25, 26, 27, 29, 31, 33

and 36 were done, as well as the RSWs. The edges of these samples

were cleaned up with a file in preparation for the tensile testing. The

samples were then placed in the Instron 8801 tensile testing machine.

The sample was first clamped in the bottom clamp, the forces cleared

and then clamped in the load cell. Small pieces of 2mm and 3mm

plate were used as spacers to ensure that the samples were pulled

straight. This can be seen in figure 5.38. The samples were pulled with

an extension rate of 3mm/min and the results were recorded.

Figure 5.38: The Instron 8801 tensile tester

Figure 5.39: 2mm and 3mm spacers

Results of the tensile tests can be seen in chapter 6.

Spacers


113

5.6.2) Microstructure and Hardness

Material Preparation

The samples that were required to be analysed, were cut into smaller

sections on the band saw. They were then further cross-sectioned

through the centre of the weld using the diamond cut-off-wheel. The

resistance spot welds were also sectioned. This wheel is used on non-

ferrous metals and plastics and provides a very fine cut. All the

samples sectioned were sectioned in the same direction.

Figure 5.40: Diamond cut-off-wheel

These sectioned samples were then mounted in clear thermosetting

resin powder and polished using standard metallographic techniques.

Etching

For the etching process, Sodium Hydroxide (NaOH) solution was

mixed in the chemical labs, 1g NaOH per 100ml of distilled water.

To reveal the general grain structure it was recommended to swab the

sample in the solution for 10s (Metallography Principles and

Procedures, 1992). This proved to be unsuccessful when looking at the

sample under the microscope, as nothing conclusive could be seen.


114

Another procedure in Metallography Principles and Procedures

(1992), recommended, for aluminium 6xxx series, to immerse the

sample for 15 minutes with a 10 minutes rinsing time in water to form

a film with hatching which varies with grain structure.

This process was done on a test sample and looking under the

microscope it was found that the etching time seemed too long, as

black pits in the material had started to appear. However, both the

Thermo-Mechanically Affected Zone (TMAZ) and more detail could

be seen with this method.

The sample was then re-prepared.

The sample was then etched for 5 minutes, rinsed for 5minutes,

inspected under the optical microscope and photographs taken.

The above process was repeated to have inspections of the

microstructure at 10, 15 and 20 minute etching times. The

photographs from the different etch times can be seen in figures 5.41

and 5.42.


115

Different Etching Times

a) 5min Etching Time b) 10min Etching Time

c) 15min Etching Time d) 20min Etching Time

Figure 5.41: Comparison of different etching times,

x500 magnification


116

Different Etching Times

a) 5min Etching Time b) 10min Etching Time

c) 15min Etching Time d) 15min Etching Time

Figure 5.42: Comparison of different etching times

x500 magnification

From figures 5.41 and 5.42 is was evident that pitting started

occurring after the 15 minute etching period. From the 10 minute to

the 15 minute etch it can also be seen that some of the material flow

patterns are destroyed, where as from the 5 to the 10 minute etch some

of the flow patterns were made more distinct.

The etching time of all the samples were set at 10 minutes, with a 9

minute rinse period under running water.

All the samples were studied under the optical microscope at 7.5x

zoom. Only half the weld could be photographed at a time and the


117

photographs were merged in Adobe Photoshop. Figure 5.43 shows a

photograph of weld 23 that has been merged.

Figure5.43: Microstructure of weld 23

Vickers micro-hardness

The indenting tool consists of a square based pyramid with the angle

between the opposite faces being 136 degrees. The indenter is

subjected to a load of 200grams for 15s. A microscope accurate up to

0.005mm was used to measure the diagonal lengths. The average of

the two measurements was then used to read the Vickers Hardness

from the tables supplied with the machine.

Figure 5.44: Vickers micro-hardness tester

For the first sample to be checked the hardness set-up was done

according to the drawing in figure 5.45 and thereafter analysed and the

procedure for the next samples determined.


118

Figure 5.45: Vickers micro-hardness set-up

The columns were spaced 0.5mm apart with the three rows being

1mm, 2mm and 4mm from the top plate. This allowed for 2 rows of

hardness to be made on the 3mm plate and one row on the 2mm plate.

7 hardness points were also taken below the pin, 0.5mm above the

bottom edge of the lower sheet. After inspection of the results, another

row a hardness tests was carried out, continued from the hardness

points under the pin.

Figure 5.46 illustrates how the remaining samples were done. Further

from the keyhole, hardness points were increased to 1mm in pitch.

Figure 5.46: Final Vickers micro-hardness set-up

The hardness of the resistance spot weld was also done across the

weld. Points at 0.5mm and 1mm intervals were taken at 0.5mm,

1.5mm and 2mm from the surface of the top plate as shown in figure

5.50.


119

Figure 5.47: Hardness set-up of the RSW

Hardness on both the 2mm and 3mm plates was also carried out to

ensure the hardness standards for the material were correct. Results of

the hardness profiles can be seen in chapter 6.

5.7) SUMMARY

In chapter 5 the final machine set-up was discussed. The method of

operating the FSW machine was briefly described as well as a

description of the operating procedure of the welds. Further

calibration of the thermocouples due to noise from the FSW machine

was done. A detailed rundown of how each weld was conducted,

problems associated and parameters changed were also discussed. A

summary of the welds and their parameters was also tabulated and

will be further discussed in chapter 6. Methods for testing the weld

integrity were also documented.

6. Performance Evaluation of FSSW

120

CHAPTER 6

PERFORMANCE EVALUATION OF FSSW

6.1) INTRODUCTION

In chapter 6 an in-depth investigation into the results obtained from

chapter 5 will be looked at. The weld integrity of the different welds

will be discussed in detail; including the tensile results, microstructure

and hardness. The reasons why the expected results were not obtained

will also be looked at. A comparison between friction stir spot welds

and resistance spot welds will be shown.

6.2) KEYHOLE ELIMINATION

At no stage during any of the welds discussed in chapter 5, did the

keyhole show signs of being filled due to the capillary action of

plasticized material. Although the pin did not retract all the way in

some welds, no signs of filling the keyhole to the extent of the

retraction were observed. At the beginning of the welding process it

was expected that the capillary forces of the plasticized material

would force the pin upwards and cause the keyhole to be closed. This

was not the case and reasons for the keyhole not being eliminated as

well as recommendations will be discussed later in section 6.2 as well

as in chapter 7. Initially the pin did not retract as required as explained

in detail in chapter 5. The pin profile had to be modified to prevent the

chamfer of aluminium forming between the pin and shoulder. The

keyhole still remained no matter what welding parameters were used.

Why was this the case? It is believed that during plunging the material

is forced upwards and to the outside of the shoulder, as illustrated by

figure 6.1.


121

Figure 6.1: Forced out material during plunging

Thus when the pin is retracted, this material that was forced out, has

no way of being forced back into the hole that is left. The procedure to

try and overcome this problem was to have no dwell time and

retracting the pin as soon as the shoulder came into contact with the

material. The problem with this method however was that the process

needed to be very accurately controlled, with greater accuracy than

what the current FSW machine and FSSW tool can accommodate. The

profile of the shoulder and the pin is also important. Another concern

is that with zero dwell time, there is no time for the shoulder to heat

up the material before initial retraction of the pin, resulting in the

material not being plasticized enough. The material forced out during

plunging needs somehow to be retained under the shoulder and then to

be forced back into the keyhole. Some recommendations are

formulated in chapter 7. Although the keyhole was not removed, a

bond was still created between the two materials, creating a successful

spot weld without the elimination of the keyhole. This allowed for

evaluations of the weld to continue.

Forced out

material

Keyhole


122

6.3) WELD EVALUATION

Comparisons between the weld integrity with different parameters

follow, as well as the comparison between the FSSW and the RSW.

All images regarding microstructure were done with a magnification

of 7.5 unless otherwise stated.

The evaluation was based on tensile strength, hardness comparisons

and microstructural evaluation. These evaluations were done in

sections of different weld parameters.

6.3.1 Shoulder comparison

6.3.2 Different dwell times

6.3.3 Different retraction rates

6.3.4 Different rpm

6.3.5 Different pin lengths

6.3.6 Different pin diameters

6.3.7 Different plunge depths

6.3.8 FSSW compared to RSW

6.3.1) Shoulder Comparisons

Only a tensile comparison between shoulders 1 and 2 was done using

the 4mm pin diameter. Welds with a 8s and 16s dwell time for each

shoulder were compared. These results can be seen in figure 6.2.


123

Tensile Shear Strength

Tensile Shear StrengthComparison between shoulders 1 and 2

4mm Diameter Pin

Extension (mm)

-1 0 1 2 3 4 5 6

Forc

e (

N)

-500

0

500

1000

1500

2000

2500

3000

3500

Weld 1 - 8s - Shoulder 1




Figure 6.2: Tensile comparison between shoulders 1 and 2

As illustrated by figure 6.2, the tensile shear strength increased with

the usage of the flat shoulder, shoulder 2, in both cases of dwell time.

The temperatures recorded for shoulder 2 were also higher than with

shoulder 1. With shoulder 2, the frictional surface contact area is

greater than with shoulder 1, resulting in the higher temperatures

recorded. A compression force directly above the stirred zone is also

created, due to the direct pressure applied from the shoulder above.

Shoulder 1 had a recess above the stirred zone as was shown in

section 5.5, figure 5.7. For detailed drawings consult Appendix A.


124

6.3.2) Different Dwell Times

As seen in figure 6.2, the different dwell times were tested as related

to the 4mm and the 2.85mm pin diameters.

4mm Pin Diameter


Comparison of the tensile shear strength of welds with dwell times of

8s and 16s were conducted, including the two different shoulders, 1

and 2.

As can be observed from figure 6.2, with the increase in dwell time to

16s, the maximum tensile strength decreased, with both shoulders 1

and 2. This could be due to more heat being supplied with an increase

in dwell time. An increase in heat supply to the material could change

the properties of the material resulting in a reduction of the tensile

shear strength properties. Due to a temperature cycle imposed by the

welding process some degree of over aging takes place in the HAZ of

heat treatable alloys (e.g. Al6061-T4). The growth of precipitates

results in a loss of strength (Schilling et al., 2000). The weld

temperatures in weld 2 with the 16s dwell time were higher than that

with the 8s dwell. Although channel 0 on weld 4 showed a higher

temperature than weld 3, all the other temperature channels showed

temperatures lower than that of weld 3 with a 16s dwell. It was

assumed that the thermocouple was damaged and the other three

channels readings were correct, as the difference between channel 0

and the other channels was immense.

No analysis of the microstructure and hardness for welds 1 – 4 were

conducted due to insufficient weld formation with the 4mm pin.


125

2.85mm Pin


Comparisons between dwell times using a pin diameter of 2.85mm

can be seen in figure 6.3.

Tensile Shear StrengthComparison between an 8s and 16s Dwell Time

4mm Pin Length2.85mm Pin Diameter

Extension (mm)

-0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8

Fo

rce

(N

)

-500

0

500

1000

1500

2000

2500

Weld 20 - 8s Dwell Time





Figure 6.3: Tensile comparison between an 8s and 16s dwell time

with a 2.85mm pin diameter

As can be seen from figure 6.3, the values for the 8s dwell are slightly

higher than those for the one 16s dwell time. Weld 29 showed results

greater than weld 27 and all the other 8s dwell periods. The plunge

depth of weld 29 was slightly too deep and this factor could have had

an effect on the tensile properties. Thus no decisive results could be

concluded. The average temperatures recorded for the welds done

with a 16s dwell were marginally higher than recorded for the welds

done with the 8s dwell time, being 9.6 °C higher. Channel 3 with weld


126

29 gave inaccurate readings, and the maximum temperature was

recorded from channel 2.

Hardness

The hardness between the two dwell times can be seen in figure 6.4.

Hardness from Weld CentreComparison between 8s and 16s Dwell Time

Pin Length 4mmWeld 23 and 28 plotted

Distance (mm)

-12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12

Vic

kers

Hard

ne

ss N

um

ber

24

26

28

30

32

34

36

38

40

42Heat Affected Zone

Pin

Row 1 - Weld 23 - 8s Dwell Time

Row 2 - Weld 23 - 8s Dwell Time

Row 1 - Weld 28- 16s Dwell Time

Row 2 - Weld 28- 16s Dwell Time

3mm Plate Average

Heat Affected Zone

Pin

Advancing SideRetreating Side

Figure 6.4: Hardness comparison between an 8s and 16s dwell time

with a 2.85mm pin diameter

From figure 6.4 it can be observed that the welds done with the 16s

dwell time suffered slightly more softening of the material than the 8s

dwell time. With the temperatures of the 16s dwell time being slightly

higher it could be assumed that the material suffered more softening

due to the increase in temperature. The 16s dwell time in row 1 of the

top sheet represented a lowest hardness value of 70.1% to that of the

parent material, while the 8s dwell represented a value of 76.8%. In


127

row 2 hardness values of 79.4% and 81.1% for the 16s and 8s dwell

were recorded. It could be seen that further away from the shoulder

the influence on hardness is less. The effect of the different dwell

times on the bottom sheet did not seem to have a significant effect.

Minimum values of 80.2% and 79.8% of the parent material, for an 8s

and 16s dwell time, were obtained.

Microstructure

Figure 6.5: Microstructure of weld 23


From figure 6.5 and 6.6, it can be seen that the area of the Thermo-

Mechanically Affected Zone (TMAZ) for the 16s dwell is slightly

greater than that of the 8s dwell for the top and bottom sheet, with the

effect being greater on the top piece. This would be expected

considering the temperatures recorded.


128

Overview

The average temperatures for the 8s dwell time for the 2.85mm pin of

222.8 °C was about 26.7 degrees lower than that of the 4mm pin with

the same dwell time. A maximum temperature of 249.5 °C was

recorded for the 4mm pin with shoulder 2.

For the 16s dwell time the same was true, with the average for the

2.85mm pin being 234.8 °C, and being 22.7 °C lower than the

maximum recorded temperature of 257.5 °C of weld 3.

The above results show the importance of the pin diameter with

respect to the weld temperature. It shows that the pin has a significant

influence on the weld temperature, with the same dwell time. This

underlines the theory that the pin plays a significant part in heat

generated during the spot weld.

6.3.3) Different Retraction Rates

For this section only the microstructure was analysed. The results

shown in figures 6.7 to 6.9 were used to compare the visual changes

taking place when the pin retraction rate was altered.


129

Microstructure

Figure 6.7: Microstructure of weld 13 (Slow retraction)

Figure 6.8: Microstructure of weld 9 (Medium retraction)

Figure 6.9: Microstructure of weld 10 (Fast retraction)

As can be observed from figures 6.7, 6.8 and 6.9, the rate of pin

retraction does not really have an effect on the TMAZ. No distinct

difference in the shape or size of the TMAZ is apparent. If the keyhole

were to be filled, the retraction rate would probably be of more

significance.


130

6.3.4) Different rpm

Four welds were conducted with the 3.2mm diameter pin at 300, 400

and 500rpm. Only tensile tests on these samples were evaluated.


Tensile Shear StrengthComparison between 300, 400 and 500rpm

3.2mm Pin Diameter8s Dwell

4mm Pin Length

Extension (mm)

0 1 2 3 4 5

Fo

rce

(N

)

-500

0

500

1000

1500

2000

2500

3000

Weld 7 - 400rpm

Weld 8 - 400rpm

Weld 17 - 300rpm

Weld 18 - 500rpm

Figure 6.10: Tensile comparison between 300, 400 and 500rpm

From figure 6.10 it can be observed that the original rpm of 400 shows

the best results. Weld 18 conducted at 500rpm also shows good

results. When the rpm was reduced to 300rpm, the strength was

substantially less than the other two settings. Not only was the shear

force less but the elongation of the weld was also less. This is

attributed to not enough heat being generated for a dwell time of 8s.

Perhaps with a lower rpm, to obtain the same shear strength values,

the dwell time needs to be increased.


131

6.3.5) Different Pin Lengths

These tests were done with a pin diameter of 2.85mm. Two

comparisons between pin lengths were done, the one using a dwell

time of 8s while the other comparison was done with a dwell time of

16s.

8s Dwell Time


Tensile Shear StrengthComparison between a 3mm and 4mm Pin Length

8s Dwell Time2.85mm Pin Diameter

Extension (mm)

0.0 0.2 0.4 0.6 0.8 1.0 1.2

Forc

e (

N)

-400

-200

0

200

400

600

800

1000

1200

1400

1600

1800

Weld 20 - 4mm Pin Length





Figure 6.11: Tensile comparison between a 3mm and 4mm pin length

From figure 6.11 it can definitely be seen that the welds done with the

4mm pin length have a greater maximum shear force as well as

increased elongation before fracture. The consistency of the 4mm pin

length welds was also better. There was inconsistency with the welds

done with the 3mm pin length, weld 31 failing at 200N while weld 33


132

failed at approximately 1350N. Weld 32 as mentioned in chapter 5 did

not bond during the welding process. It is apparent that the plunge

depth during the 3mm welds is more critical than when the pin length

is 4mm. If the plunge depth is shallow on a 3mm pin length weld,

such as weld 32, pin penetration into the lower sheet is not as good.

Microstructure



Although the TMAZ in both cases extends into the bottom sheet,

stirring of the material in weld 23 is greater. In weld 23, the bottom

sheets TMAZ area is greater than in weld 30. Stirring takes place

around the pin on both sheets in weld 23 whereas in weld 30 only a

small margin of stirring takes place at the bottom sheet. It is deemed

important that the pin penetrate through the bottom sheet to create the

desired stirring action between the two materials. The keyhole was

still clearly visible with the pin length being 3mm. The average

temperatures recorded for the 3mm pin length were lower than those


133

recorded with the 4mm pin length. See Appendix C for temperature

data.

Hardness

Hardness from Weld CentreComparison between 3mm and 4mm Pin Length

8s Dwell TimeWeld 23 and 30 plotted

Distance (mm)

-12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12

Vic

ke

rs H

ard

ne

ss N

um

ber

30

32

34

36

38

40

42

44

Heat Affected Zone

Pin

Row 3 - Weld 23 - 4mm Pin Length

Row 3 - Weld 30 - 3mm Pin Length

2mm Plate Average

Heat Affected Zone

Pin


Figure 6.14: Hardness comparison between a 3mm and 4mm pin

length

From the information obtained it is clear that pin penetration into the

bottom plate is essential as it increased the shear strength of the weld

but more important is its contribution to the more consistent

performance behaviour of the FSSW process.


134

16s Dwell Time


Tensile Shear StrengthComparison between a 3mm and 4mm Pin Length

16s Dwell Time2.75 Pin Diameter

Extension (mm)

-0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8

Forc

e (

N)

-500

0

500

1000

1500

2000

2500




Figure 6.15: Comparison between a 3mm and 4mm pin length

From figure 6.15 the values obtained from the 4mm pin length welds

are inconsistent in terms of the shear strength, thus making

comparison of results very difficult. However, the weakest 4mm pin

weld has very similar tensile strengths to the 3mm pin weld. The

elongation of the 4mm pin length was greater as well. Considering the

fact that 2 of the 3mm pin length welds did not create a bond, as

discussed in chapter 5, it can be concluded that the 4mm pin length

has improved weld properties to that of the 3mm pin length.


135

Microstructure



As mentioned for the 8s dwell time, the average temperatures for the

3mm pin length were lower than those with the 4mm pin length. The

change in shape of the TMAZ shown in figure 6.16 and 6.17 support

the philosophy that the pin has an effect on heat generation. Weld 37

shows very little penetration of the TMAZ into the bottom sheet,

while with the 4mm pin the TMAZ around the pin in the bottom sheet

is clearly visible, as indicated by the arrow in figure 6.16. The TMAZ

of weld 28 is greater than in weld 37 indicating also that the 4mm pin

is doing more stirring.

TMAZ in

bottom sheet


136

6.3.6) Different Pin Diameters

This section compares the tensile shear strength, microstructure and

hardness of welds with different pin diameters but with the other

welding parameters being the same.


Tensile Shear StrengthComparison between a 2.85mm, 3.2mm and 4mm Pin Diameter

8s Dwell Time

Extension (mm)

0 1 2 3 4 5

Forc

e (

N)

0

500

1000

1500

2000

2500

3000

3500

Weld 4 - 4mm Diameter Pin-Shoulder 2

Weld 7 - 3.2mm Diameter Pin - Shoulder 3





Figure 6.18: Tensile comparison between different pin diameters

From the graph in figure 6.18 it is clear that with an increase in the pin

diameter, the maximum shear strength increases substantially. This is

in conjunction with the paper written by Shilling et al. (2002) where

similar trends were recognised. This appears logical, as the welding

area increases with an increase in pin diameter, as the material is

joined around the perimeter of the pin. Thus, the greater the pin

diameter, the greater the stirring effect, resulting in an increased

maximum shear strength obtained.


137

Microstructure




As observed in figure 6.19 to 6.21, the TMAZ decreases in size with a

decrease in pin diameter. This can be noticeably observed in the

bottom plate of each weld. The TMAZ in this region is definitely

larger with the greater pin diameter. This supports the theory that the

pin once again has a great influence on heat supplied during the


138

welding process, especially at a distance further away from the

shoulder. Unfortunately no temperatures were recorded when welding

with the 3.2mm pin. However the vast difference in temperatures can

be seen between the 4mm and 2.85mm pin. The maximum

temperature recorded for weld 4, at 10.5mm from the weld centre, was

249.5°C. The average maximum temperature for 2.85mm welds was

222.8°C. This maximum average is substantially lower with the

smaller pin diameter and substantiates the results obtained above

concerning microstructure.

Hardness

Hardness from Weld CentreComparison between 4mm, 3.2mm and 2.85mm Pin Diameters

Weld 5, 9 and 23 Plotted

Distance (mm)

-12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12

Vic

ke

rs H

ard

ness N

um

be

r

24

26

28

30

32

34

36

38

40

42

Heat Affected Zone

Pin

Row 1 - Weld 6 - 4mm Diameter Pin


Row 1 - Weld 9 - 3.2mm Diameter Pin




3mm Plate Average

Heat Affected Zone

Pin

Retreating Side Advancing Side

Figure 6.22: Hardness comparison of the 3mm plate for different pin

diameters


139

Hardness from Weld CentreComparison between 4mm, 3.2mm and 2.85mm Pin Diameters

Weld 5, 9 and 23 Plotted

Distance (mm)

-12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12

Vic

kers

Ha

rdness N

um

be

r

25

30

35

40

45

50Pin




2mm Plate Average

Pin


Figure 6.23: Hardness comparison of the 2mm plate for different pin

diameters

Although it is difficult to see an exact trend in the hardness values in

figure 6.23, the minimum hardness values for the different pin

diameters expressed as a percentage of the base material can be

extracted. This data is shown in table 6.1.

Table 6.1: Hardness expressed as a % of the base material

Pin Diameter Hardness as a %

Row 1 Row 2 Row 3

4mm 67.65 69.7 64.5

3.2mm 78.2 71.2 71.3

2.85mm 76.8 81.1 80.1

From table 6.1 it is observed that the 4mm pin shows the most

softening of the material. This is in conjunction with the higher


140

temperatures recorded during welds made with the 4mm pin. The

minimum hardness values in the bottom plate correspond to the top

plate. The bottom plate minimum hardness values were recorded just

below the pin, where the material experiences increased plasticization.

For the rest of the hardness in the lower sheet, the percentages are

closer to that of the parent material compared to the top layer. Thus

the top layer experiences a larger degree of softening under the

influence of the shoulder.

The above results compare well to Schilling et al. (2000), where for

similar plate configurations a lowest hardness value of 79% of the

base material was recorded. They also claimed the top layer suffers

more softening than the bottom layer.

6.3.7) Different Plunge Depths

In this section only the microstructure and hardness were compared.

Two welds with the same parameters were done, except plunge depth

varied. The pin diameter, dwell time and rotational speed were all the

same. The only variable was the plunge depth. The plunge depth for

weld 6 measured to be approximately 0.1mm in comparison to

0.34mm in weld 9. In figure 6.24 and 6.25 the difference in the TMAZ

can be seen.


141

Microstructure



With the shallow plunge depth in weld 6, the TMAZ is much smaller

than in weld 9. Although the plunge depth in weld 6 was shallow, a

bond between the two materials was still obtained. Previously in this

chapter it was mentioned that the pin plays an important role in heat

generated. This section shows that the shoulder with a correct plunge

depth also plays a major role in the weld. From previous papers read,

it appeared that different variations for calculating temperatures of the

weld relied only on the shoulder. The results above indicate that the

pin and the shoulder are vital when these calculations are considered

and neither should be neglected.


142

Merged Photographs at a Greater Magnification

Figure 6.26: Microstructure of weld 6, x100 magnification

Figure 6.27: Microstructure of weld 9, x50 magnification

As seen from figure 6.26 and 6.27, when the plunge depth was

shallow, the two materials did not bond well. It was observed how the

material was pushed up during plunging by the curved line created

between the two materials interface. In figure 6.27 the joint line is

horizontal, fades into jagged lines and then disappears. The material

next to the pin is shown as one layer. This represents the area where

the material has been stirred and bonded. The forces applied by the

shoulder when the correct plunge depth is used, as well as the heat

Curved

joint line

Line

Joint

line

Jagged

lines No visual

separation


143

supplied, creates a more uniform bond as illustrated in figure 6.26 and

6.27.

6.3.8) Friction Stir Spot Welds compared to Resistance Spot

Welds

One of the final objectives is to compare the weld integrity of a FSSW

to that of a conventional RSW. In this section the tensile shear

strengths, microstructure and hardness will be compared in detail.

This will give a good indication of what progress has been made with

FSSW.


Tensile Shear StrengthComparison between Friction Stir Spot Welds and

Resistance Spot Welds

Extension (mm)

-1 0 1 2 3 4 5 6 7

Forc

e (

N)

-500

0

500

1000

1500

2000

2500

3000

3500

Weld 4 - 4mm Pin

Weld 7 - 3.2mm Pin

Weld 8 - 3.2mm Pin

Weld 20 - 2.85mm Pin



RSW 1 (Single Spot)

RSW 2 (Single Spot)

RSW 3 (Double Spot)

RSW 4 (Single Spot)

Figure 6.28: Tensile comparison between a FSSW and a RSW


144

As illustrated in figure 6.28, the maximum shear force obtained for the

RSWs, is less than that obtained by the FSSWs. This is true for all

diameters of the FSSWs. The double RSW shows similar shear forces

to that of a 3.2mm diameter weld. A single resistance spot weld shows

results similar to that of a 2.85mm diameter pin, only slightly lower.

The only advantage the RSW seems to have is that the curve is more

rounded and uniform, indicating that the welded joint is less brittle

than that of a FSSW.

It is very difficult to compare these welds to each other as no real joint

area in the friction stir spot weld is obtainable and thus the stress

values cannot be calculated. Only shear force values of RSWs with a

welding tip diameter of 16mm and a flat diameter of 5mm, which was

in contact with the material, were compared to various FSSWs. These

results also link the literature survey results where Schilling et al.

(2000) reported that FSSWs were superior to resistance spot welds.

Sakano et al. (2001) also stated that friction stir spot welded lap joints

have equal or superior UTSS and UTS values compared to

conventional RSW lap joints.


145

Microstructure

Because the tensile properties of the 2.85mm pin diameter were

closest to that of the RSW, their microstructure properties were

compared to that of a RSW.

Figure 6.29: Cross section of the macrostructure of a RSW

Figure 6.30: Microstructure of a resistance spot weld, x50

magnification

As observed in figures 6.29 and 6.30, the heat affected zone is in the

shape of an eye, with a weld nugget formed in the middle. The size of

the weld nugget was measured to be 1.32mm wide and 0.45mm high,

while the eye measured to be 4.45mm wide and 1.98mm high.

Nugget Parent Material

HAZ


146


The differences between the two welding processes are clearly

illustrated in figures 6.29 to 6.31.

Micrographs with a magnification of x50 and x500 were used to

compare the grain structure in the HAZ, TMAZ, nugget and parent

material.

TMAZ Parent

plate


147

500X Magnification

Resistance Spot Weld

A) Eye B) Nugget

C) Parent Material D) Nugget/Eye

Figure 6.32: Microstructure of a RSW, x500 magnification

Friction Stir Spot Weld

A) TMAZ B) Parent Material

Figure 6.33: Microstructure of a FSSW, x500 magnification


148

50X Magnification

Resistance Spot Weld

A) Nugget and Eye B) Parent Material

Figure 6.34: Microstructure of a RSW, x50 magnification

Friction Stir Spot Weld

A) Heat Affected Zone B) Parent Material

Figure 6.35: Microstructure of a FSSW, x50 magnification

From the micrographs that were taken at a greater magnification, the

different grain structures in the various areas could be clearly seen.

The comparison between the FSSW and the RSW can be seen clearly.

The grain structure of the nugget in the RSW appears very fine. This

part of the material is melted during the weld process and is caused by

the current and pressure applied. The eye reveals another pattern

completely different to the parent material. The structure of the

material undergoes a dramatic change. The grains change shape and


149

appear to have lines representing the direction of the current flow

through the material, as illustrated in figure 6.32a.

With regard to the FSSW, in the TMAZ the grain structure is a lot

finer compared to that of the parent material. The structure is also

uniform.

Thus the FSSW has less of an effect on changing the original material

grain structure to that of a RSW. This is due to the two processes

being completely different. The RSW relies on heat, caused by the

current flowing through the material, and the pressure to melt the two

materials together. The FSSW is more of a plasticized forging process

where the two materials are stirred together, below melting point. The

welding temperature of the material during a FSSW is therefore lower

than that of a RSW.


150

Hardness

The Hardness of the FSSW and the RSW were also compared.

Hardness from Weld CentreComparison between a FSSW and a RSW Weld

Weld 23 plotted against a RSW

Distance (mm)

-10 -8 -6 -4 -2 0 2 4 6 8 10

Vic

kers

Hard

ne

ss N

um

be

r

20

25

30

35

40

45

RSW - On joint Line

RSW - 1.5mm from Top of the Plate

RSW - 0.5mm From Top of the Plate

Weld 23 - Row 1

Weld 23 - Row 2

Weld 23 - Row 3

3mm Plate Average

2mm Plate Average

Figure 6.36: Hardness comparison between a FSSW and a RSW

The nugget of the RSW showed a minimum hardness value of 36.7.

The average hardness in this region was measured to be 39.1. The

minimum value represents a figure 90% of the hardness of the parent

material. In the eye the lowest recorded hardness was 58.3% of the

parent material. In the other rows the minimum recorded hardness

represented a value of 53.7% of the parent material. As cited by

Schilling et al. (2000), Thornton et al. (1996) reported similar results

obtained in RSWs using aluminium alloy Al6011-T4 (2mm welded on

2mm sheet). The lowest hardness values reported for the resistance

spot welds were 62% (T4) in comparison to the base metal.


151

Results from FSSW 23 showed a minimum hardness value of 76.8 %

of the base material in the TMAZ and 80.1% under the pin.

From figure 6.36 it can be seen that the hardness in the heat affected

zone of a RSW is less than that of a FSSW. This can be due to the

lower heat input characteristics of the FSW process. Temperatures

recorded during the FSSW process show that the material does not

reach melting point, unlike in a resistance spot weld.

The results presented confirm the statement by Schilling et al. (2000),

that due to the limited extent of microstructural changes caused by the

low heat input FSSW process, improved mechanical properties in

comparison to RSW are obtained. One of the benefits a RSW has over

a FSSW is that the weld is less brittle when conducting the tensile

shear test and the fracture occurs over a greater elongation.

6.4) SUMMARY

Chapter 6 consists of an in-depth evaluation of the FSSW. Reasons for

the keyhole not being eliminated as well as attempts that were made to

solve the problem were discussed. Evaluation of the weld integrity for

the different welding parameters was conducted. The evaluation was

based on tensile shear strength, hardness comparison and

microstructural evaluation. A comparison between FSSWs and RSWs

were also made.

7. Conclusions and Future Work

152

CHAPTER 7

CONCLUSIONS AND FUTURE WORK

7.1) CONCLUSIONS

The purpose of this study and its relevance to friction stir spot welding

should be used to obtain a better understanding of the weld process,

the elimination of the keyhole as well as comparing the weld integrity

to that of conventional spot welding techniques.

The re-design of the tool patented by Georgeou and Hattingh (2004)

was successful in using the pneumatic muscle actuator to operate the

cam pins that allowed the pin to retract in increments of 0.5mm during

retraction. The pin penetrating the material in its extended position

was successful however retraction during the weld was problematic,

relying on the spring and the capillary forces of the plasticized

material to force the pin upwards. Only with modification of the

welding pin did the pin retract. Because of the above problems the

calibration of the muscle pressure to that of the pin position could not

be used to its full potential. Material between the pin and the shoulder

formed a chamfer during pin retraction causing problems.

Although in some of the welds the pin did not retract fully, retraction

was enough to promote the closing of the keyhole. Unfortunately in

none of the welds did the keyhole show any signs of elimination. This

is due to previous thinking that during pin retraction the forces

involved would force the plasticized material upwards, eliminating the

keyhole.


153

During plunging the material forced out by the pin, eventually ends up

around the perimeter of the shoulder, thus leaving no material to fill

the keyhole. Different methods of attempting to close the keyhole

were practised with little success. One method was with no dwell time

and retracting the pin as soon as the shoulder came into contact with

the material. For this method to possibly work, more control of the pin

and of the timing between pin retraction and the shoulder coming into

contact with the surface is needed. The movement of the pin manually

to the surface of the material also proved problematic, as the plunge

depth did not always remain exactly the same.

Although the welds did not remove the keyhole, a permanent joint

between the two sheets was still achieved, producing a friction stir

spot weld with a keyhole. Both the shoulder and the pin were

responsible for heat input into the material and neither can be

neglected concerning temperature calculations. The shoulder and pin

are both responsible for creating a successful spot weld. Although the

temperature was only measured at a distance away from the weld,

comparisons between different procedures and tool designs could still

be made.

A comparison between Friction Stir Spot Welds and Resistance Spot

Welds was also made, with the outcome similar to those discussed in

the literature survey. FSSWs have equal or superior properties to that

of RSWs. Lower heat inputs during the FSSW process has less of an

effect on mechanical properties of the material as the process is

operated below melting point, unlike in resistance spot welding. The

tensile shear strength of a FSSW is also greater than that of the RSW

it was compared to, although the weld is more brittle.


154

With further development of the FSSW process this fairly new method

of joining lap joints could be used more widely in industry.

7.2) FUTURE WORK

This section is divided into two parts, the first part discussing the

recommended changes to improve the tool and the control parameters,

and the second part discussing the future work required to better

understand the friction stir spot welding process.

Firstly, a change in design of the tool is needed. The pin needs to

rotate in the opposite direction or at a different speed to the shoulder,

to prevent a bond being made between the two. This will keep the

material between these two components in a plasticized state during

the welding process. A method whereby the material that is forced out

by the pin during plunging needs to be forced into some sort of

reservoir and then forced back into the keyhole when the pin is

retracted.

In a RIFTEC GmbH brochure, Meyer and Schilling (2004) illustrate

that they have created successful friction stir spot welds with the

removal of the keyhole. Their tool is similar to that of the patent

submitted by dos Santos and Schilling (2002). Their method involves

a stationary shoulder, a rotating sleeve and a pin rotating in the

opposite direction. The tool is applied to the area at a pre-set pressure.

The frictional heat created by the sleeve and the pin causes the

material to become plasticized. The pin is then pressed into the work

piece to a certain depth. At the same time the sleeve is retracted

creating a reservoir for the material to flow into between the pin and

the stationary shoulder. When the pin is retracted the sleeve presses


155

back the still plasticized material into the void creating a successful

keyhole free weld (Meyer and Schilling, 2004). The process can be

seen in figure 7.1.

Material being Pin plunging, Sleeve plunging, Tool Removal

plasticized Sleeve retracting Pin retracting

Figure 7.1: Friction stir spot welding by RIFTEC GmbH

(adapted from Meyer and Schilling, 2004)

For successful elimination of the keyhole in FSSWs, whether using

similar methods as above or not, more control of the pin and its

position is needed. The pin needs to be able to be plunged or retracted

at any stage during the weld, especially in trial welds when ideal

parameters are not known. The position of the pin also needs to be

more accurately known. The tool in this dissertation only allowed for

pin retraction after it was plunged in its extended position. This

proved to be a delimitating factor. The means of actuation would have

to be reconsidered with the possible use of hydraulics. The

temperature of the welding pin itself would possibly give readings of

more relevance to the actual weld temperatures recorded. The weld

process also needs to be fully automated to eliminate problems

obtained by manually operating some of the weld processes. Plunge

depth, dwell and weld time all need to be controlled with greater

accuracy.


156

Finally, as proposed in this research, the development and analysis of

the FSSW process needs to be continued. More data needs to be

obtained and material flow techniques, tool design and control

parameters need to be further evaluated. A process whereby the

keyhole is eliminated also needs to be thought of, considering the

above recommendations.

References

157

REFERENCES

Colligan, K.J. (1998) Friction Stir Welding Tool for Welding Variable

Thickness Workpieces. US Patent number 5,718,366

Deutchman, A.D., Michels, W.J. and Wilson, C.E. (1975) Machine

Design: Theory and practice. Macmillan Publishing Co.,

United States of America

Ding, R.J. (2000) Force Characterisation on the Welding Pin of a

Friction Stir Welding Retractable Pin-Tool using Aluminum-

Lithium 2195. 2nd

International Symposium on Friction Stir

Welding, Gothenburg, Sweden.

Ding, R.J. and Oelgoetz, P.A. (1999) Auto-Adjustable pin Tool

for Friction Stir Welding. US Patent Number 5,893,507

dos Santos, J. and Schilling, C. (2002) Method and device for joining

at least two adjoining work pieces by friction stir welding. US

Patent application publication number US 2002/0179682 A1

Drotsky, J.G. (1994) Strength of Materials for Technicians. Second

edition. Butterworth publishers (pty) Ltd., Isando.

Festo (2003) Fluidic Muscle MAS, Info 501. Festo AG & Co. KG.

URL: http:/www.festo.com/Info_501_en.pdf

Freeman, J., Koch, N., Moore, C., Podraza, D. and Trujillo, A.

(undated) Friction Spot Welding. [Online]. Available:

http://ampcenter.sdsmt.edu/fric_spot_weld_pro.html [2004,

February 10]

References

158

Georgeou, Z. (2003) Analysis of Material Flow around a Retractable

Pin in a Friction Stir Weld. MTech Thesis, PE Technikon

Georgeou, Z. and Hattingh, D.J. (2004) Retractable Pin Tool for the

use with the Friction Stir Welding Process. RSA Patent

Application Number 2004/05 38

Groover, M.P. (1996) Fundamentals of Modern Manufacturing:

Materials, Processes and systems. Prentice Hall, Upper Saddle

River, New Jersey, pp 712-713.

Hansen, M. (2003) A cooler weld, In Mechanical Engineering Design,

[Online], Available:

http://www.memagazine.org/medes03/coolweld/coolweld.html

[2003, August 19]

Hartman, J.B. and Maleev, V.L. (1962) Machine Design. Third

edition. International Textbook Company, Scranton,

Pennsylvania.

Holt, E.S. and Lang, L.J. (1998) Programmable Friction Stir Welding

Process. US Patent Number 5,713,507

Johnson, R. (2001) Forces in Friction Stir Welding of Aluminium

Alloys. 3rd

International Friction Stir Welding Symposium,

Kobe, Japan.

Johnson, R. and Kallee, S. (1999) Friction Stir Welding. [Online],

Available: http://ww.azom.com/details.asp?ArticleID=1170

[2004, January 30]

References

159

Kallee, S. and Mistry, A. (1999) Friction Stir Welding in the

Automotive Body in White Production. First International

Symposium on Friction Stir Welding, Thousand Oaks, CA.

Machine Design (2003) Car bodies first to be friction stir welded.


http://www.machinedesign.com/asp/viewSelectedArticle.asp?s

trArticleID/56348/strSite/MDSite [2004, January 1]

Matsumoto, K. and Sasabe, S. (2001) Lap Joints Aluminium Alloys by

Friction Stir Welding. 3rd

International Friction Stir Welding

Symposium, Kobe, Japan

Matweb.com, The Online Materials Database. Alumnium 6063-T4.

(Copyright 1996-2004). [Online]. Avalaible:

http://www.matweb.com/search/SpecificMaterialPrint.asp?bas

snum=MA6063T4 [2004, November 26]

Metallography Principles and Procedures (©1992). Leco®

Corporatation, MI 49085, U.S.A.

Meyer, A. and Schilling, C. (2004) Innovative Friction Spot Welding

for Point-Shaped Lap Joints. Friction Spot Welding, Robotic

Friction Welding Application and Technology Company,

Germany.

NASA Marshall Space Flight Centre (2002) NASA Marshall’s

Friction Stir Welding Technology successfully commercialised

by two companies. [Online]. Available:

http://www1.msfc.nasa.gov/NEWSROOM/news/releases/2002

/02-009.html [2004, January 29]

References

160

NASA Techtrans (2001) Friction Stir Welding. [Online]. Available:

http://techtan.msfc.nasa.gov/pdf/FSW11.20.01.pdf

[2004, January 29]

National Center for Advanced Manufacturing ( 2004) Friction Stir

Welding. [Online]. Available:

http://www.ncampl.org:2004/technology/FSW.htm

[2004, March 1]

Page, M. (2003) Friction stir welding broadens applications base.


http://www.manufacturingtalk.com/news/mij/mij107.html

[2004, January 29]

Sakano, R. Murakami, K., Yamashita, K., Hyoe, T., Fujimoto, M.,

Inuzuka, M., Nagao, Y. and Kashiki, H. (2001) Development

Of Spot FSW Robot System for Automobile Body Members. 3rd

International Symposium on FSW, Hobe, Japan

Sato, Y.S., Kokawa, H., Enomoto, M., Jogan, S. and Hashimoto, T.

(2001) Distribution of Hardness and Microstructure in

Friction Stir Weld of Al Alloy 6063. 3rd

International

Symposium on FSW, Hobe, Japan.

Sato, Y.S., Urata, M. and Kokawa, H. (2002) Parameters Controlling

Microstructure and Hardness during Friction-Stir Welding of

Precipitation-Hardenable Aluminum Alloy 6063. In

Metallurgical and Materials Transactions A, Volume 33A.

References

161

Schilling, C., von Strombeck. A., dos Santos. J. and von Heesen, N.

(2000) A Preliminary Investigation on the Static Properties of

Friction Stir Spot Welds. 2nd

International FSW Symposium,

Gothenburg, Sweden.

Song, M. and Kovacevic, R. (2003a) Numerical and experimental

study of the heat transfer process in friction stir welding. In

Proc. Instn Mech. Engrs Vol. 217 Part B: J. Engineering

Manufacture, IMechE, pp. 73-85

Song, M. and Kovacevic, R. (2003b) Thermal modelling of friction

stir welding in a moving coordinate system and its validation.

International Journal of Machine Tools & Manufacture 43,

605-615

Silyn-Roberts, H. (2000) Writing for Science and Engineering: Papers

Presentaions and Reports. Butterworth Heinemann, Oxford.

The Welding Institute (©2004) Resistance Spot Welding. [Online].

Available:

http://www.twi.co.uk/j32k/ptotected/band_3/kssaw001.html

[2004, December 18]

Thomas, W.M., Nicholas, E.D., Needham, J.C., Murch, M.G.,

Temple-Smith, P. and Dawes, C.J. (1995) Friction Welding.

US Patent Number 5,460,317

Thornton, P.H., Krause, A.R. and Davies, R.G. (1996) The aluminium

spot weld. Welding Journal, Research Supplement, 101s-08s.

References

162

Wykes, D.H. (1997) Adjustable Pin for Friction Stir Welding Tool.

US Patent number 5,697,544

Young, W.C. (1989) Roark’s Formula’s for Stress and Strain. 6th

edition. McGraw-Hill, Inc., United States of America.

Appendix A

163

APPENDIX A

DESIGN CALCULATIONS AND DRAWINGS

A.1) INTRODUCTION

This section deals with taking the concept and transferring the idea

into actual drawings. This also involves making the necessary

calculations to insure that the structure is both physically strong

enough and safe.

A.2) DESIGN CALCULATIONS OF THE FSSW TOOL

The design of the main components of the tool structure and working

parts are discussed in this section. The FEA of some components are

also illustrated.

A.2.1) Housing Structure

Figure A.1: Tool housing structure

Columns

Upper Ring

Lower Ring

Appendix A

164

As illustrated in figure 1, the four columns are attached at both ends to

the upper and lower ring plates. The force that is applied to the

shoulder during welding is transferred through these columns and the

outer ring, which is not shown in this diagram. Calculations were done

in order to obtain dimensions for the diameter of the columns for a

defined maximum design load.

The term strut is used for a member subjected to an axial compressive

load. If the member is long in comparison to its cross sectional

dimensions, the strut is slender and will fail due to buckling before the

yield strength of the material is reached. The buckling load is defined

as the axial load that will keep the strut in its bent form. Thus the cross

sectional area is based on the buckling load and not the yield strength

of the material.

The Euler theory is used to calculate the cross sectional areas of the

struts/columns and the following assumptions need to be considered.

• Initially the strut is perfectly straight

• The load is applied axially

A factor of safety is also included.

Figure A.2: Euler formula

P - The compressive force keeping the strut bent

Let the deflection at a distance x from M be equal to y.

Appendix A

165

Mdx

ydEI =

2

2

(A.1)

PyM −=

(A.2)

Then by substitution A.2 into A.1

02

2

=+ yEI

P

dx

yd (A.3)

2l

EIP

E

π= (A.4)

PE is the Euler Buckling Load for a pin-jointed strut.

The effective length (e

l ) for a pin-jointed strut in this case is equal to

l , where l is the length of the column.

For a pin jointed strut:

lle

=

For a strut with both ends fixed:

2

ll

e=

Hence equation A.3, now becomes:

2

e

E

l

EIP

π= (A.5)

Where 64

4d

Iπ

= (Moment of Inertia for a circular cross section)

Appendix A

166

Slenderness ratio:

The slenderness ratio is the ratio k

l where l is the length of the strut

and k the radius of gyration about the axis at which buckling will

occur. The effective slenderness ratio k

le results in a general

expression so that struts of the same material but with different end

fixings may be compared with each other.

Validity:

The Euler formula will not be valid if the limit of proportionality is

exceeded, since it depends on the modulus of elasticity. In some cases

the yield stress is used as an approximation of the proportional limit

stress. Let σ c be the proportional limit (critical stress) caused by the

application of a load Pc on the strut.

APcc

σ= (A.6)

Where A is the cross sectional area of the strut.

4

2dA

π= (A.7)

The effective slenderness ratio at this point is termed the validity limit

for the Euler formula.

cc

eE

k

l

σ

π2

= (A.8)

Where 4

dk =

Struts used in industry usually have an effective slenderness ratio of

less than 200.

Appendix A

167

Table A.1: Calculation criteria

No of

Columns(C) 4

Pe 25000N

E 209x109Pa

L 206mm

Le 103mm

Safety Factor 3

The factor of safety is introduced by multiplying the applied force by

the safety factor. There are also 4 columns therefore the load is

divided by the number of columns to achieve the unit load per column.

Formula A.5 is now shown as in A.9, where C is the number of

columns and SF represents a safety factor.

SFl

EICP

e

E

×

=2

π (A.9)

By rearranging the formula in terms of inertia:

EC

xSFlPI

eE

π

2×

= (A.10)

410209

5103.0250009

2

×

×=

x

xI

π

410102315.3 mxI−

=

64

4d

Iπ

= (A.11)

By rearranging the formula in terms of diameter:

464

π

×=

Id

4

10 64100493.5

π

×=

−

xd

Appendix A

168

=d 0.01007m

=d 10.07mm

=d 12mm (Chosen column size)

The slenderness ratio from formula A.8 is now used to check the

validity limit.

cc

eE

k

l

σ

π2

=

4

dk =

Table A.2: Calculation criteria

σc 220Mpa

A 1.131x10-4

m2

Pc 2488.14N

E 209x109Pa

K 3

6

92

10220

10209

x

x

k

le

π=

83.96=

k

le (Which is less than 200)

The diameter of the columns will therefore be 12mm. The use of

12mm diameter columns allows for the tool to be used for future

research work in friction stir welding, where larger forces are required

due to larger shoulder diameters. This will prevent the need for a

completely new tool to be manufactured.

Appendix A

169

The outer ring which is attached to the upper and lower rings (not

illustrated) can withstand compression, which was not included in the

calculation. This outer ring allows for concentricity between the upper

and lower plates to be accurate. It is merely a support for the structure,

to prevent twisting, and the forces it can withstand were not included

in the design calculations.

The above calculations for the columns and the ring were both

evaluated with (FEA), using a COSMOS Designstar software. The

upper plate was constraint to be rigid and a 25kN compression force

was applied normal to the lower ring.

a) Static Nodal Stress b) Static Displacement

Figure A.3: Static loading of the tools housing structure

From figure A.3a is can be seen that the maximum stress induced in

the column is approximately 95MPa. This value is less than the yield

stress of the material used. A maximum deflection of about 0.07mm

can be seen from figure A.3b.

Static

Displacement:

Units: mm

Scale 1:10

Static Nodal

Stress:

Units: Mpa

Scale: 1:1

Upper

ring

Lower

ring

Maximum

stress

Appendix A

170

A.2.2) Shoulder and Pin

The shoulder and the pin, the areas responsible for welding, are the

most important part of the tool. These components were designed to

be modular, for two reasons:

1. To facilitate the development program; and

2. To facilitate easy replacement because of high temperatures and

loads which they are subjected to during the welding process.

The profiles of both the pin and the shoulder were kept simple, as the

process is not as yet well understood in terms of what effect tool

profile has on FSSW. These parameters will however be looked at

during the welding trials and evaluation. FEA on the welding pin was

also done. A force of 5kN was exerted on the pin in the direction as

indicated in figure A.4. A fixed restraint was applied as illustrated.

a) Static nodal stress b) Static displacement

Figure A.4: Static loading of the pin

A maximum stress of 500MPa was illustrated which is less than a

third of the yield stress of H13 tool steel, which is the material

selected for the pin. A maximum deflection of 0.043mm was obtained.

Static Nodal Stress

Units: MPa

Scale 1:1

Static Displacement:

Units: mm

Scale 1:10

Fixed

restraint

Applied

load

Maximum

stress

Appendix A

171

A.2.3) Moving Pins

The moving pins are used to operate the cam pins. These pins are

subjected to the force from the muscle and are not influenced by the

forces from the actual welding process. These pins however transfer

the torque to the retractable pin. They are subjected to shear in an

upward pulling force, a downward pushing force to reset the pin, and a

rotational torque.

The Maximum force created by the muscle is 1200N.

Thus for an upward pulling force, the following applies.

A

F=σ (A.12)

Table A.3: Moving pin design criteria

Max Vertical Force 1200N

No of Pins 4

Safety Factor (SF) 2

Material EN 19 Condition T: Yield Stress = 740Mpa

Formula A.12 with a safety factor included:

A

SFF ×=σ (A.13)

Formula A.13 rearranged in terms of area:

σ

SFFA

×= (A.14)

610740

21200

x

xA =

6102432.3 −

= xA m2

Appendix A

172

Area Formula:

44 2d

Aπ

= (Four Pins)

Area formula in terms of diameter:

π

Ad =

π

6102432.3 −

=x

d

=d 0.001016m

=d 1.02mm

Power rating of the electric motor for the milling machine

The power is rated at 7.5kW for the electric motor on the milling

machine.

Power Formula:

60

2 NTP

π= (A.15)

Power formula in terms of Torque:

N

PT

π2

60= (A.16)

Table A.4: Torque from the electric motors

RPM Torque (Nm)

250 286.5

500 143.2

Torque formula:

rFT ×= (A.17)

Torque formula in terms of force:

r

TF = (A.18)

Design for Maximum conditions:

Material EN 19 Condition T: Yield Stress = 740Mpa

Appendix A

173

Table A.5: Moving pin design criteria

Factor of Safety 2

Radius at shear 0.01m

Max Torque 358.1Nm

No. of Pins 4

01.0

5.286=F

=F 28650N

σ

SFFA

×=

610740

228650

x

xA =

5107432.7 −

= xA m2

π

Ad = (For four pins)

π

5107432.7 −

=x

d

=d 0.004965m

=d 4.965mm

=d 5mm

Therefore the diameter of the retracting pins needs to be 5mm in

diameter.

Appendix A

174

A.2.4) Bending Force on the Locking Ring, created by the

Locking Pins

A formula for a uniform annular line load applied (w) to the inner ring

at radius (ro) was used from “Roark’s Formulas for Stress and Strain”

(Young, 1989). Calculations using EN 19 condition T were done.

Figure A.5: Uniform loading (reproduced from Young, 1989)

Figure A.6: Loading constraints (reproduced from Young, 1989)

Appendix A

175

Table A.6: Locking ring properties

PLATE DIMENSIONS

Thickness t 0.2362in. 6mm

Outer Radius a 0.5512in. 14mm

Inner Radius b 0.1575in. 4mm

Applied Load w 2807.68lbf/in 318.31N/mm

Modulus of Elasticity E 30.24x106lbf/in

2 209Gpa

Poisson's Ratio v 0.3 0.3

Radical Location of

Applied Load ro 0.17717in. 4.5mm

The calculations were done using “Roarks formula’s for stress and

strain”, on a computer program, a standard formula where the values

are inserted and the results are calculated.

Appendix A

176

General formulas and graphs for deflection, slope, moment, shear

and stress as a function of r

Figure A.7: General formulas and graphs for deflection, slope,

moment, shear and stress as a function of r

(reproduced from Young, 1989)

As illustrated in figure A.7, which was calculated on the roark’s

computer programme, the maximum deflection is 7.444x10-5

inches,

which equates to 0.00189mm.

Appendix A

177

Transverse Bending Stress

Figure A.8: Transverse bending stress

(reproduced from Young, 1989)

The maximum stress value is also read off the graph. For the inner

edge guided the maximum stress in the material is 4843.33psi. This

relates to a value of 33.48Mpa, which is below the yield stress of EN

19 condition T, which has a yield stress of 740Mpa.

Appendix A

178

Radial Bending Stress

Figure A.9: Radial bending stress (reproduced from Young, 1989)

The maximum stress value is also read off the graph. For the inner

edge guided, the maximum stress in the material is 1.614x104psi. This

relates to a value of 111.55Mpa, which is well below the yield stress

of EN 19 condition T. These values were also checked in Cosmos

Designstar, as illustrated in figure A.10.

Figure 10: Static nodal stress of the locking ring

Appendix A

179

Figure A.11: Static displacement of the locking ring

For both analyses done on the locking mechanism, a restraint

underneath the locking ring was placed and a compressive force of

5kN applied to the ratchet. The arrows represent the direction of the

force.

The maximum stress and deflection values obtained compare

favourably to the calculated values in section A.2.4.

Ratchet

Locking

ring

Fixed

restraint

Appendix A

180

A.2.5) Spring Calculations

A.2.5.1) Ratchet Pin Spring

The spring needs to overcome the forces of friction between the pin

and the housing in the locking ring.

Table A.7: Ratchet pin spring properties

Pin Mass (m) 12 grams

Normal Force (N) 0.11772N

Deflection 10mm

Coefficient of Friction, static

(u) [steel on steel] 0.6

The frictional force can be represented by the following formula:

(A.19)

6.011772.0 ×=F

NF 0706.0=

This calculated force is required to keep the spring in its fully

extended position. A force of 2N/mm was therefore selected to be

used for the spring, allowing for no sticking when the pin is able to

return to its original position. The spring also prevents the centrifugal

forces, caused by the rotation of the tool, forcing the ratchet pin to

move outwards.

NuF =

Appendix A

181

Table 8: Obtained ratchet spring values

Free Length 27mm

ID (min) 5mm

OD (max) 8mm

Spring Stiffness at Free Length 2N

Deflection 12mm

Specifications obtainable

Wire Dia 0.9

TC 16

Achieved Rates 1.992N/mm

Od 7

A.2.5.2) Welding Pin Spring

The mass of the welding pin and its components causing a downward

force was calculated at being ± 238grams.

Thus a total downward force of 2.335N needs to be overcome by the

spring in order to push the pin upwards.

A spring with a spring force of 2.5N/mm was selected. The maximum

outer diameter of the spring could be 30mm with the minimum inner

diameter being 21mm. The spring requires a free length of 19mm and

must have a deflection of 9mm to allow for slight tensioning of the

spring if required.

The spring specification was calculated by an outside spring company

and can be seen in table 9.

Table 9: Obtained pin spring values

Free Length 19mm

ID (min) 21mm

OD (max) 30mm

Spring Stiffness at Free Length 2.5N

Deflection 9mm

Specifications obtainable

Wire Dia 1.8

TC 5

Achieved Rates 2.503

od 25.6

Appendix A

182

A.3) SUMMARY

This section has looked at the design calculations of the main

components of the FSSW tool. FEA on the important components was

done. The final working drawings that include dimensions, tolerances,

material selection and quantity of parts needed have been included.

Transferring the detailed drawings into a prototype that can be used to

do the intended laboratory testing will be discussed in chapter 4 and 5.

Calibration and testing will also be done.

Appendix B

183

APPENDIX B WELD DATA

Table B.1: Weld 1

Weld 1

Parameters:

Date 2004/09/08 Thermocouple installed Yes

Shoulder Number 1 Temperature Logged Yes

Pin Number 1 Data Cleared and Saved Yes

Plunge Rotational Speed 400rpm Pin Position after weld 3mm

Weld Rotational Speed 400rpm Number of lines 2

Rate of Plunging 0.25mm/s Pin retracted X mm 1mm

Plunge Depth 0.2mm Muscle Fully Retracted Yes

Pin Length 4mm Shoulder Clean No

Dwell Time 8s Pin Free No

Rate of Retraction 1 1/2 Turns Keyhole Visible Yes

Spring Number 1 Stabiliser No

Comments:

Material formed a solid chamfer between the pin and the shoulder.

Recorded Data:




Max. Temperature 245.9°C Max Tensile Shear Force 2.520kN

Table B.1: Weld 2

Weld 2

Parameters:

Dwell Time increased to 16s












Comments:


Recorded Data:



Force Time 27.5s Measured Plunge Depth 0.23mm


Appendix B

184

Table B.3: Weld 3

Weld 3

Parameters:

Shoulder 2 used, Spring 2 used

Dwell Time 16s




Plunge Rotational Speed 400rpm Pin Position after weld 3.5mm


Rate of Plunging 0.25mm/s Pin retracted X mm 0.5mm






Comments:


Recorded Data:





Table B.4: Weld 4

Weld 4

Parameters:

Dwell time changed back to 8s












Comments:


Recorded Data:



Force Time 26.3s Measured Plunge Depth 0.2.mm


Appendix B

185

Table B.5: Weld 5

Weld 5

Parameters:

Spring 3 used, only temperature from channel 3.



Pin Number 1 Data Cleared and Saved No









Comments:


Recorded Data:

Plunging Time NA Measured Keyhole Depth 4.00mm

Welding Time NA Measured Keyhole Diameter 3.97mm

Force Time NA Measured Plunge Depth 0.21mm

Max. Temperature 211°C Max Tensile Shear Force NA

Table B.6: Weld 6

Weld 6

Parameters:

New sheet used, Pin Tool 2

Date 2004/09/09 Thermocouple installed No

Shoulder Number 2 Temperature Logged No










Comments:

As soon as the pin started retracting, the shoulder started wondering on the

surface of the material. The plunge depth was slightly shallow.

Recorded Data:




Max. Temperature NA Max Tensile Shear Force NA

Appendix B

186

Table B.7: Weld 7

Weld 7

Parameters:

Shoulder 3 used










Rate of Retraction 1 1/2 Turns Keyhole Visible Pin Hole


Comments:

As soon as the pin started retracting, the shoulder started wondering on the

surface of the material. This smothered the weld and it appeared that the keyhole was gone, only a small pin hole visible.

Recorded Data:


Welding Time 42.8s Measured Keyhole Diameter NA


Max. Temperature NA Max Tensile Shear Force 2.562kN

Table B.8: Weld 8

Weld 8

Parameters:

Stabilizer bracket installed









Dwell Time 8s Pin Free Yes


Spring Number 3 Stabiliser Yes

Comments:

Recorded Data:





Appendix B

187

Table B.9: Weld 9

Weld 9

Parameters:

Same as weld 8

Shoulder cleaned with HCL acid.












Comments:

Start plunge depth approximately 0.1mm to deep.

Recorded Data:





Table B.10: Weld 10

Weld 10

Parameters:

Same as weld 8, except with a faster pin retraction, 2 turns

Shoulder was cleaned with HCL acid.










Rate of Retraction 2 Turns Keyhole Visible Yes


Comments:

Recorded Data:





Appendix B

188

Table B.11: Weld 11

Weld 11

Parameters:

Same as weld 10, assembled with chrome compound

Shoulder cleaned with HCL acid.








Pin Length 4mm Shoulder Clean Yes




Comments:

Smoke generated when welding, presumably some chrome compound ended

up on the shoulder

Recorded Data:





Table B.12: Weld 12

Weld 12

Parameters:

Same as weld 8, except retraction rate slow, 1.25 turns












Comments:

Recorded Data:

Plunging Time 20.9s Measured Keyhole Depth NA


Force Time 36.3s Measured Plunge Depth NA


Appendix B

189

Table B.13: Weld 13

Weld 13

Parameters:

Same as weld 12

New sheet of aluminium.












Comments:

Recorded Data:





Table B.14: Weld 14

Weld 14

Parameters:

No dwell time

Retraction rate fast












Comments:

The air was opened at 3.8mm pin plunge depth.

Recorded Data:


Welding Time 44s Measured Keyhole Diameter 3.05mm



Appendix B

190

Table B.15: Weld 15

Weld 15

Parameters:

Same as weld 15, except 0.3mm plunge depth

Air was opened at 3.5mm pin plunge depth












Comments:

Keyhole was smeared closed, but further inspection showed the keyhole was

not eliminated. Swaying of the tool occurred.

Recorded Data:





Table B.16: Weld 16

Weld 16

Parameters:

Same as weld 14, Rpm changed to 300rpm












Comments:

Recorded Data:





Appendix B

191

Table B.17: Weld 17

Weld 17

Parameters:

Dwell time 8s, Retraction rate medium

Rpm 300












Comments:

Recorded Data:





Table B.18: Weld 18

Weld 18

Parameters:

Dwell time 8s, retraction rate medium

Rpm 500












Comments:

Recorded Data:





Appendix B

192

Table B.19: Weld 19

Weld 19

Parameters:

Same as weld 14, Dwell time 0s, Retraction rate fast

Except Rpm 500












Comments:

Recorded Data:





Table B.20: Weld 20

Weld 20

Parameters:

New sheet of aluminium, weld 20-26 done with same parameters

Dwell time 8s, Retraction rate medium, Rpm 400, Pin diameter 2.85mm












Comments:

Recorded Data:





Appendix B

193

Table B.21: Weld 21

Weld 21

Parameters:

Same as weld 20












Comments:

Initial starting point 0.2mm to deep, therefore plunge depth to deep.

Recorded Data:





Table B.22: Weld 22

Weld 22

Parameters:

Same as weld 20












Comments:

Initial starting point 0.2mm to deep, therefore plunge depth to deep.

Recorded Data:





Appendix B

194

Table B.23: Weld 23

Weld 23

Parameters:

Same as weld 20












Comments:

Recorded Data:




Max. Temperature 224.7°C Max Tensile Shear Force NA

Table B.24: Weld 24

Weld 24

Parameters:

Same as weld 20












Comments:

Shoulder did not remove after weld.

Recorded Data:





Appendix B

195

Table B.25: Weld 25

Weld 25

Parameters:

Same as weld 20












Comments:

Recorded Data:





Table B.26: Weld 26

Weld 26

Parameters:

Same as weld 20

No Spring












Comments:

Friction of the tool components still pulled the pin out.

Recorded Data:

Plunging Time 19.8s Measured Keyhole Depth 3.84s




Appendix B

196

Table B.27: Weld 27

Weld 27

Parameters:

New sheet of aluminium, weld 27-29 done with the same welding parameters

Dwell Time increased to 16s












Comments:

Recorded Data:





Table B.28: Weld 28

Weld 28

Parameters:

Same as weld 27












Comments:

Recorded Data:





Appendix B

197

Table B.29: Weld 29

Weld 29

Parameters:

Same as weld 27












Comments:

Recorded Data:

Plunging Time 19.8s Measured Keyhole Depth 3.83




Table B.30: Weld 30

Weld 30

Parameters:

Weld 30-33 done with the same parameters

Pin Length 3mm, Dwell time 8s












Comments:

Shoulder was removed when air pressure reached 4.5bar.

Recorded Data:





Appendix B

198

Table B.31: Weld 31

Weld 31

Parameters:

Same as weld 31












Comments:

Recorded Data:





Table B.32: Weld 32

Weld 32

Parameters:

Same as weld 30












Comments:

Recorded Data:


Welding Time 34.1s Measured Keyhole Diameter 2.65



Appendix B

199

Table B.33: Weld 33

Weld 33

Parameters:

Same as weld 30












Comments:

Recorded Data:




Max. Temperature 211°C Max Tensile Shear Force 1.346kN

Table B.34: Weld 34

Weld 34

Parameters:

Weld 34-37 done with the same parameters

Pin Length 3mm, Dwell time 16s,












Comments:


Recorded Data:





Appendix B

200

Table B.35: Weld 35

Weld 35

Parameters:

Same as weld 34












Comments:

Tool Swayed

Recorded Data:





Table B.1: Weld 36

Weld 36

Parameters:

Same as weld 34












Comments:


Recorded Data:





Appendix B

201

Table B.37: Weld 37

Weld 37

Parameters:

Same as weld 34












Comments:


Recorded Data:

Plunging Time 16.5 Measured Keyhole Depth 2.97mm




Appendix C

202

APPENDIX C TEMPERATURE GRAPHS

Temperature vs TimeWeld 1

Time (s)

0 20 40 60 80 100 120 140 160

Tem

pera

ture

(°C

)

0

50

100

150

200

250

300

Channel 0

Channel 1

Channel 3

Channel 6


Time (s)

0 20 40 60 80 100 120

Te

mp

era

ture

( °

C)

0

50

100

150

200

250

300

Channel 0

Channel 1

Channel 3

Channel 6

Figure C.1: Weld 1 Figure C.2: Weld 2


Time (s)

0 20 40 60 80 100 120

Te

mpe

ratu

re (

°C)

0

50

100

150

200

250

300

Channel 0

Channel 1

Channel 3

Channel 6


Time (s)

0 20 40 60 80 100 120 140 160

Tem

pera

ture

(°C

)

0

50

100

150

200

250

300

Channel 0

Channel 1

Channel 3

Channel 6


Appendix C

203


Time (s)

0 50 100 150 200 250

Te

mp

era

ture

(°C

)

0

20

40

60

80

100

120

140

160

180

200

220

Channel 0

Channel 1

Channel 3

Channel 6


Time (s)

0 20 40 60 80 100 120 140 160

Te

mpe

ratu

re (

°C)

0

50

100

150

200

250

Channel 0

Channel 1

Channel 3

Channel 6



Time (s)

0 20 40 60 80 100 120 140 160 180

Tem

pe

ratu

re (

°C)

0

50

100

150

200

250

Channel 0

Channel 1

Channel 3

Channel 6


Time (s)

0 20 40 60 80 100 120 140 160 180

Te

mpe

ratu

re (

°C)

0

50

100

150

200

250

Channel 0

Channel 1

Channel 3

Channel 6


Temperature vs Time

Weld 23

Time (s)

0 20 40 60 80 100 120 140

Tem

pera

ture

(°C

)

0

50

100

150

200

250

Channel 0

Channel 1

Channel 3

Channel 6


Time (s)

0 20 40 60 80 100 120 140

Te

mpera

ture

(°C

)

0

50

100

150

200

250

Channel 0

Channel 1

Channel 3

Channel 6


Appendix C

204

Temperature vs Time

Weld 25

Time (s)

0 20 40 60 80 100 120 140

Tem

pe

ratu

re (

°C)

0

50

100

150

200

250

Channel 0

Channel 1

Channel 3

Channel 6


Time (s)

0 20 40 60 80 100 120 140

Tem

pera

ture

(°C

)

0

50

100

150

200

250

Channel 0

Channel 1

Channel 3

Channel 6


Temperature vs Time

Weld 27

Time (s)

0 20 40 60 80 100 120 140 160

Te

mpe

ratu

re (

°C)

0

50

100

150

200

250

Channel 0

Channel 1

Channel 3

Channel 6


Time (s)

0 20 40 60 80 100 120 140

Tem

pera

ture

(°C

)

0

50

100

150

200

250

Channel 0

Channel 1

Channel 3

Channel 6


Temperature vs Time

Weld 29

Time (s)

0 20 40 60 80 100 120 140

Tem

pera

ture

(°C

)

0

50

100

150

200

250

300

Channel 0

Channel 1

Channel 3

Channel 6


Time (s)

0 20 40 60 80 100 120 140

Tem

pe

ratu

re (

°C)

0

50

100

150

200

250

Channel 0

Channel 1

Channel 3

Channel 6


Appendix C

205

Temperature vs Time

Weld 31

Time (s)

0 20 40 60 80 100 120 140 160

Te

mpera

ture

(°C

)

0

20

40

60

80

100

120

140

160

Channel 0

Channel 1

Channel 3

Channel 6


Time (s)

0 20 40 60 80 100 120 140

Tem

pera

ture

(°C

)

0

20

40

60

80

100

120

140

160

Channel 0

Channel 1

Channel 3

Channel 6


Temperature vs Time

Weld 33

Time (s)

0 50 100 150 200

Te

mpera

ture

(°C

)

0

50

100

150

200

250

Channel 0

Channel 1

Channel 3

Channel 6


Time (s)

0 20 40 60 80 100 120 140

Tem

pera

ture

(°C

)

20

40

60

80

100

120

140

160

Channel 0

Channel 1

Channel 3

Channel 6


Temperature vs Time

Weld 35

Time (s)

0 20 40 60 80 100 120 140 160 180

Te

mpe

ratu

re (

°C)

0

20

40

60

80

100

120

140

160

180

200

Channel 0

Channel 1

Channel 3

Channel 6


Time (s)

0 20 40 60 80 100 120

Te

mpe

ratu

re (

°C)

0

50

100

150

200

250

Channel 0

Channel 1

Channel 3

Channel 6


Appendix D

206

APPENDIX D HARDNESS GRAPHS

Hardness from Weld CentreWeld 5

Distance (mm)

-12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12

Vic

ke

r H

ard

ness N

um

ber

20

25

30

35

40

45

50

55Pin

Heat Affected Zone

Row 1

Row 2

3mm Plate Average

Row 3

2mm Plate Average

Plot 1 Pin

Plot 1 Heat Affected Zone



Distance (mm)

-12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12

Vic

kers

Hard

ness N

um

ber

20

25

30

35

40

45

50

55

60

Heat Affected Zone

Pin

Row 1

Row 2

3mm Plate Average

Row 3

2mm Plate Average

Heat Affected Zone

Pin


Figure D.1: Weld 5 Figure D.2: Weld 6


Distance (mm)

-12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12

Vic

ke

rs H

ard

ness N

um

ber

25

30

35

40

45

50

55

Heat Affected Zone

Pin

Row 1

Row 2

3mm Plate Average

Row 3

2mm Plate Average


Plot 1 Pin


Hardness From Weld CentreWeld 10

Distance (mm)

-12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12

Vic

ke

rs H

ard

ness N

um

ber

28

30

32

34

36

38

40

42

44

46

Heat Affected Zone

Advancing SidePinRetreating Side

Row 1

Row 2

3mm Plate Average

Row 3

2mm Plate Average

Heat Affeted Zone

Pin


Appendix D

207


Distance (mm)

-12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12

Vic

ke

rs H

ard

ness N

um

ber

25

30

35

40

45


Pin

Row 1

Row 2

3mm Plate Average

Row 3

2mm Plate Average

Heat Affected Zone

Pin



Distance (mm)

-12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12

Vic

kers

Hard

ness N

um

ber

20

25

30

35

40

45

50

Heat Affected Zone

Pin

Row 1

Row 2

3mm Plate Average

Row 3

2mm Plate Average

Heat Affected Zone

Pin




Distance (mm)

-10 -8 -6 -4 -2 0 2 4 6 8 10

Vic

ke

rs H

ard

ness N

um

be

r

26

28

30

32

34

36

38

40

42Pin

Row 1 - 3mm Plate

Row 2 - 3mm Plate

Row 3 - 3mm Plate

3mm Plate Average

2mm Plate Average

Pin



Distance (mm)

-12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12

Vic

ke

r H

ard

ness N

um

ber

25

30

35

40


Pin

Row 1

Row 2

3mm Average

Row 3

2mm Plate Average


Plot 1 Pin



Appendix D

208


Distance (mm)

-12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12

Vic

ke

rs H

ard

ness N

um

ber

20

25

30

35

40

45

Pin

Heat Affected Zone

Row 1

Row 2

3mm Plate Average

Row 3

2mm Plate Average

Pin

Heat Affected Zone



Distance (mm)

-12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12

Vic

ke

rs H

ard

ness N

um

ber

28

30

32

34

36

38

40

42


Pin

Row 1

Row 2

3mm Plate Average

Row 3

2mm Plate Average

Heat Affected Zone

Pin



Hardness from Weld Centre

Restistance Spot Weld

Distance (mm)

-8 -6 -4 -2 0 2 4 6 8

Vic

kers

Ha

rdn

ess N

um

be

r

20

25

30

35

40

45

Middle Row

1.5mm From Top Surface

0.5mm From Top Surface

2mm Plate Average

Figure D.11: RSW

Appendix E

209

APPENDIX E MICROSTRUCTURE

Figure E.1: Weld 5

Figure E.2: Weld 6

Figure E.3: Weld 9

Figure E.4: Weld 10

Appendix E

210

Figure E.5: Weld 11

Figure E.6: Weld 13

Figure E.7: Weld 23

Figure E.8: Weld 28

Appendix E

211

Figure E.9: Weld 30

Figure E.10: Weld 37

Figure E.11: RSW

DEVELOPMENT AND ANALYSIS OF A FRICTION STIR SPOT ...

Documents