Improvement of Punch and Die Life and Part Quality in Blanking of Miniature Parts

Home , Blanking and piercing, Extrusion, Perforation, Polishing (metalworking), Punching, Punch (tool)

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

By Soumya Subramonian Graduate Program in Mechanical Engineering

The Ohio State University 2013

Dissertation Committee: Dr.Taylan Altan, Advisor Dr.Blaine Lilly, Advisor Dr.Gary L.Kinzel Dr.Jerald Brevick

Abstract

Blanking or piercing is one of the most commonly used sheet metal manufacturing processes in the industry. Having a good understanding of the fundamentals and science behind this high deformation shearing process can help to improve the tool life and blanked edge quality in various ways. Finite Element Modeling of the blanking process along with experimental testing is used in this study to study the influence of various process parameters on punch and die life and blanked edge quality. In high volume blanking and blanking of high strength materials, improving the tool life can save not only tool material but also change over time which can take up to a few hours for every change over.

The interaction between punch, stripper plate and sheet material is first studied experimentally since a fundamental understanding of the behavior of these components at different blanking speeds is very essential to design robust tooling for high speeds. A methodology is developed using the experimentally obtained blanking load and FEM of blanking to obtain flow stress data of the sheet material at high strains and strain rates. This flow stress data is used to investigate the effects of various process parameters on tool stress and blanked edge quality. The influence of all these parameters on tool stress, blanking load and blanked edge quality are studied. Some factors are found to influence the tool stress and blanked edge quality more than others.

Parameters like punch-die clearance, punch corner radius, application of stripper pressure and blanking velocity affect the blanked edge quality. As punches and dies wear, the punch-die clearance and punch corner radius increase, causing the blanked edge quality to deteriorate by increasing the rollover and burr. Punch-die clearance along with other factors like lubrication conditions, sheet material, tool material and coating also affects the rate of tool wear. The punch

ii tip geometry significantly affects the blanking load. For a given sheet material, tool material and punch geometry, can selecting the right punch-die clearance minimize punch wear? Since small radii in the punch geometry wear or chip earlier and more often than the straight edges, there is also non-uniform wear pattern observed in the punch due to the non-uniform stress on the punch.

The effect of punch geometry on punch wear is studied by conducting FE simulations of blanking and correlating the punch stress obtained from FEM and punch wear obtained via experiments.

The effect of sheet material and thickness on punch stress is also studied. In addition, the effect of punch-die clearance for different geometries is investigated. After having a good understanding of the relation between punch geometry and punch wear for different sheet materials, a guideline for selecting the most suitable punch-die clearance for a given punch geometry to have more uniform wear on the punch is suggested.

The performance of geometry dependent variable punch-die clearance and commonly used uniform punch-die clearance is compared by conducting blanking experiments and comparing the wear patterns for both cases. It was observed that the tooling with variable punch-die clearance could punch almost three times more parts (350,000 parts) with significantly less but more uniform wear on them than the tooling with uniform clearance (126,000 parts).

Methods to improve part edge quality by using the optimum stripper pressure and to improve tool life using geometry-dependent variable punch-die clearance are suggested in this study.

iii

Acknowledgements

I am very grateful to my advisor, Dr.Taylan Altan for giving me the opportunity to pursue my graduate studies under his guidance. His intellectual support, encouragement, and guidance are the main reasons for this research work to even be possible. I sincerely thank my co-advisor,

Dr.Blaine Lilly, for all his inputs and support. I thank my committee members Dr. Gary L. Kinzel and Dr. Jerald Brevick for their inputs, discussions and support.

I would like to thank Bogdan Ciocirlan and Craig Campbell from Tyco Electronics Corporation for providing all the support, especially for providing the facility for conducting experiments required for my study.

I thank my colleagues of the CPF, Adam Groseclose, Eren Billur, Tingting Mao and Xi Yang for the discussion and assistance during the course of my study.

I thank my family and friends for all their support in this endeavor.

Vita

2006…………………………………………B.E Mechanical Engineering, Anna University, India

2009…………………………………………M.S Mechanical Engineering, The Ohio State University

2010 to present…………………………….PhD Mechanical Engineering, The Ohio State University

Publications

1. Subramonian, S. (2012). Blanking. In A. E. T. Taylan Altan (Ed.), Sheet metal forming:

Processes and applications (1st edition ed., pp. 1). USA: ASM International.

2. Subramonian, S and Kardes.N. (2012). Material for Sheet Forming. In A. E. T. Taylan Altan

(Ed.), Sheet metal forming: Fundamentals (1st edition ed., pp. 73). USA: ASM International.

3. Subramonian, S., Kardes, N., Demiralp, Y., Jurich, M., and Altan, T., "Evaluation of Stamping

Lubricants in Forming Galvannealed Steels for Industrial Application", Journal of

Manufacturing Science and Engineering, 2011, vol. 133, issue 6.

4. Subramonian, S., and Altan, T., “Punch Wear in Blanking – Part I”, Stamping Journal,

July/Aug 2012, p. 14

5. Subramonian, S., and Altan, T., "Punch Wear in Blanking - Part II", Stamping Journal,

Sept/Oct. 2012, p.10

Fields of Study

Major Field: Mechanical Engineering

Table of Contents Abstract ...... ii

Acknowledgments ...... iv

Vita ...... v

List of Tables ...... xii

List of Figures ...... xiv

CHAPTER 1: Introduction ...... 1

1.1. Blanking Process ...... 2

1.2. Part Edge Characteristics ...... 3

1.3. Punch/Die Wear ...... 4

CHAPTER 2: Research Objectives and Approach ...... 5

CHAPTER 3: State-of-the-art review ...... 7

3.1. Fundamentals in blanking – effect of various process parameters on punch life and

blanked edge quality...... 7

3.1.1. Effect of punch-die clearance ...... 7

3.1.2. Effect of punch corner radius [Picas, 2010] ...... 12

3.1.3. Effect of stripper pressure on blanked edge quality of thin parts ...... 13

3.2. Punch materials and coatings ...... 15

3.2.1. Different Punch Materials and Coatings Used in Blanking ...... 15

3.2.2. Comparison of WC punches with Ceramics (Y-TZP) [DiRuggiero 2000] ...... 19

3.2.3. Environmentally Benign Tribo-systems for Metal Forming [Bay, 2010] ...... 20

3.2.4. Improving the wear resistance of tools for stamping [Straffelini 2010] ...... 20

vii

3.3. Material model and fracture model used in simulations ...... 21

3.3.1. Material Model ...... 21

3.3.2. Fracture Model ...... 21

3.4. Punch failure – simulation modeling and experiments ...... 22

3.4.1. Punch Failure Mechanisms ...... 22

3.4.2. Factors affecting tool wear and galling [Billur, 2009] ...... 26

3.4.3. Wear Models ...... 26

3.4.4. Effect of tool wear on part edge quality ...... 29

3.5. High speed blanking ...... 31

3.6. Press stability and reverse loading ...... 33

3.7. Flanging...... 35

3.8. Research Needs ...... 38

CHAPTER 4: Interaction between punch, sheet material and stripper plate in blanking ...... 41

4.1. Approach ...... 41

4.2. Experimental studies on punch loading during blanking at various speeds ...... 42

Blanking experiments are conducted at various speeds to understand punch loading and

punch-stripper interaction...... 42

4.2.1. Experimental Setup ...... 42

4.2.2. Experimental Procedure ...... 44

4.2.3. Method used to analyze experimental results ...... 45

4.2.4. Results ...... 47

4.3. FE simulations of blanking ...... 58

4.3.1. Simulation setup ...... 58

4.3.2. Comparison of experimental and simulated results ...... 65

viii

4.4. Development of high strain rate material model ...... 68

4.4.1. Description of the methodology...... 69

4.4.2. Evaluation of the methodology for C51100 (using simulation and experimental

results) 71

4.4.3. Flow stress curve for C5100 at quasi-static conditions ...... 72

4.4.4. Flow stress curve for C5100 at higher strain rates ...... 75

4.4.5. Effect of temperature on flow stress of C51100 ...... 78

4.4.6. Compare experimental and simulated force curves at high strain rate...... 79

4.5. Summary and conclusions ...... 80

CHAPTER 5: Influence of various parameters on blanked edge quality and punch load/stress

5.1. Effect of punch-die clearance on part quality and punch stress ...... 83

5.2. Effect of corner radius on part quality and punch stress ...... 86

5.3. Effect of punch tip geometry on punch load ...... 87

5.4. Effect of friction on punch and part ...... 90

5.5. Effect of stripper pressure on part edge quality ...... 92

5.6. Effect of punch misalignment on part edge quality ...... 94

5.7. Summary and Conclusions ...... 97

CHAPTER 6: Relation between punch-die clearance, punch stress and part geometry in blanking 98

6.1. Simulation Setup ...... 100

6.2. Effect of punch geometry ...... 101

6.2.1. Influence on contact stress on the punch ...... 102

6.2.2. Influence on part edge quality ...... 106

6.3. Effect of material properties on punch stress ...... 109

6.3.1. Effect of sheet material (strength) ...... 109

6.3.2. Effect of sheet thickness ...... 111

6.4. Effect of clearance on punch stress ...... 112

6.5. Punch-die clearance to obtain least punch stresses for different geometries ...... 114

6.5.1. Straight Edge...... 116

6.5.2. Round ...... 118

6.5.3. Oblong ...... 120

6.5.4. Corner of rectangle ...... 120

6.5.5. Variable punch-die clearances for different geometries ...... 121

6.6. Emperical relation to determine variable clearance as a function of radius of curvature,

material strength and thickness ...... 122

6.7. Summary and Conclusions ...... 123

CHAPTER 7: Case Studies showing the effect of variable punch-die clearance on tool wear

125

7.1. Influence of variable punch-die clearance on a rectangular punch ...... 125

7.1.1. Experimental Setup [Högman 2004] ...... 125

7.1.2. Experimental Results [Högman, 2004] ...... 127

7.1.3. Simulation Setup ...... 129

7.1.4. FEA Results ...... 129

7.1.5. Comparison of experimental and FE results ...... 131

7.2. Influence of variable punch-die clearance on complex geometries ...... 132

7.2.1. Selecting variable punch-die clearances for different geometries ...... 135

7.2.2. Experimental Procedure ...... 137

7.2.3. Experimental Results ...... 137

CHAPTER 8: Summary and Conclusions ...... 140

8.1. Summary ...... 140

8.2. Conclusions ...... 141

8.3. Research contributions ...... 143

The following reseacrh contributions came out of this study...... 143

8.4. Future work ...... 143

The following areas are identified for future work of this study...... 143

REFERENCES ...... 145

APPENDIX A: Alternate Flow Stress to Demonstrate the methodology to Obtain Flow Stresss

……………..……………………………………………………………………………………………...151

List of Tables

Table 3.1 Case Study: Comparison of WC and ceramic punches in precision stamping of copper lead frames ...... 19

Table 3.2: Wear Models used in numerical simulations ...... 27

Table 4.1: Tooling Parameters Used in Experiments ...... 44

Table 4.2: Experimental parameters used in measuring forces during blanking ...... 45

Table 4.3 Parameters and values used in simulations ...... 61

Table 4.4: Simulation Parameters and Boundary Conditions ...... 62

Table 4.5: Comparison of experimental and simulated blanked part edge ...... 66

Table 4.6: Parameters used in simulations – methodology to obtain high strain rate flow stress data using blanking tests...... 73

Table 4.7: Factor used to scale the flow stress curve for higher strain rate ...... 77

Table 4.8: Factor used to scale the flow stress curve for higher temperatures ...... 79

Table 5.1: Parameters Used in Simulations to study the influence of various parameters ...... 82

Table 5.2: Advantages and Disadvantages of different punch tip geometries ...... 88

Table 5.3: Blanked edge quality and maximum punch stress determined with various misalignments of punch (simulations) ...... 96

Table 6.1: Parameters used in FEA of blanking in studying the effect of various parameters on punch stress ...... 101

Table 6.2: Simulation matrix used to study the effect of geometry of the pierced shape ...... 102

xii

Table 6.3: Parameters Used in FE Simulations to determine the optimal punch-die clearance for different geometries (sheet material AISI 1010, 0.8128mm thick) ...... 115

Table 6.4: Geometry-specific variable punch-die clearance for blanking sheet material AISI 1010,

0.8128mm thick ...... 122

Table 7.1: Tool geometries used by [Högman 2004] ...... 127

Table 7.2: Parameters used in simulation (using Högman’s experimental data) ...... 128

Table 7.3: Geometries used in FE simulation to determine punch stress ...... 134

Table 7.4: Geometry-specific variable punch-die clearance ...... 135

Table A.1: Calculating the factor to ‘modify’ the flow stress………………………………………...146

xiii

List of Figures

Figure 1.1: Different blank and scrap for blanking and piercing ...... 1

Figure 1.2: Schematic of the blanking tool setup ...... 2

Figure 1.3: Phases of the blanking process [Schuler Handbook, 1998] ...... 3

Figure 1.4: Different zones of blanked part edge ...... 4

Figure 3.1: Effect of blanking clearance on part edge as predicted with finite element simulations on 0.58mm Cu alloy [Husson et al, 2008] ...... 8

Figure 3.2: Effect of blanking clearance on part edge in DP590 steel of 1.4mm thickness

[Wiedenmann et al, 2009] ...... 8

Figure 3.3: Influence of punch-die clearance on shear edge length for different blanking velocities

[Grünbaum et al, 1996] ...... 9

Figure 3.4: Effect of punch-die clearance on tool wear [Bell 2006] ...... 10

Figure 3.5: Influence of punch-die clearance on punch wear at the corner of the rectangle (left) with 10% and (right) with 20% punch-die clearance [Högman, 2004] ...... 11

Figure 3.6: (a) Maximum Von Mises stress shown for 10%, 15% and 20% clearances at 0.1 and

0.01mm corner radii (b) Experimental load-displacement diagram at corner radius 0.1mm and

0.01mm [Picas et al, 2010] ...... 12

Figure 3.7: (left) Dynamic behavior of stripper at 1200SPM; (right) Mechanical model of newly designed stripper for I-shaped frame [Sekine, 2005] ...... 15

Figure 3.8: Characteristics of tool materials by Böhler Uddeholm [Bell, 2006]...... 18

xiv

Figure 3.9:(a)Part formed (b) tooling used in the experiment (c) Burr height measured for various tool materials/treatments/coatings[Straffelini 2010] ...... 21

Figure 3.10: Schematic illustration of flank and face wear in punch [Hernández et al, 2006] ...... 23

Figure 3.11: Chipping wear seen on punch cutting edge [Luo, 1999], [Uddeholm and SSAB 2008]

...... 24

Figure 3.12: Cracking as seen on punch surface [Luo, 1999] ...... 24

Figure 3.13: Gross Fracture on punch [Luo, 1999] ...... 25

Figure 3.14: Galling on punch face [Uddeholm and SSAB 2008] ...... 25

Figure 3.15: Increase in burr volume with strokes because of tool wear [Makich et al, 2008] ...... 29

Figure 3.16: Effect of sheet material on burr volume [Makich et al, 2008] ...... 30

Figure 3.17: Effect of tool wear and blanking clearance on part edge quality as predicted by simulations on Cu alloy, 0.58mm thick[Husson et al, 2008] ...... 31

Figure 3.18: Tooling commonly used in high speed blanking [Hirsch et al, 2011]...... 32

Figure 3.19: Forces in the blanking process for stroke rates 100, 300, 500 and 1000 strokes per minute [Hirsch et al, 2011] ...... 33

Figure 3.20: Forces during snap-thru [Miles, 2004] ...... 34

Figure 3.21: (top) micrographs of cross sectional area in punched (a), drilled (b) and wire cut (c) hole of DP800; (bottom) Influence of the hole edge condition on HER ...... 36

Figure 3.22: Smoothed surfaces and relationships between percentage of depths of rollover, burnished surface, fracture surface and smoothed surface on sheared edge and smoothing load:

(a) P = 16.2 kN; (b) P = 21.8 kN; (c) depth percentage [Mori, 2010] ...... 37

Figure 3.23: Relationship between limiting expansion ratio and smoothing load [Mori, 2010] ...... 37

Figure 4.1: Experimental setup for the punch force measurement ...... 43

Figure 4.2: Punch force and ram displacement (recorded continuously over a number of strokes)

...... 46

Figure 4.3: Forces identified on the punch during blanking for Case A at blanking velocity

465mm/sec ...... 48

Figure 4.4: Forces identified on the punch during blanking for Case B at blanking velocity

465mm/sec ...... 49

Figure 4.5: Forces identified on the punch during blanking for Case C at blanking velocity 930 mm/sec ...... 50

Figure 4.6: Influence of blanking velocity on maximum blanking force ...... 51

Figure 4.7: Influence of blanking velocity on maximum reverse loading on punch ...... 52

Figure 4.8: Influence of stripper pinning on maximum blanking force ...... 53

Figure 4.9: Influence of velocity on loading on the punch during stripper plate vibration during pinning ...... 54

Figure 4.10: Influence of velocity on loading on the punch during stripper plate vibration during unpinning ...... 55

Figure 4.11: (left) Blanking load curve at 60 SPM; (right) blanked edge profile ...... 56

Figure 4.12: Blanked edge obtained for Case A at different blanking velocities (left to right) (i)

187mm/sec (ii) 500mm/sec (iii) 875mm/sec and (iv) 1250mm/sec...... 57

Figure 4.13: Influence of blanking velocity on shear length ...... 57

Figure 4.14: Flow Stress data of C51100 used in simulations (experimental – obtained using

Viscous Pressure Bulge test) ...... 59

Figure 4.15: Schematic of the blanking process ...... 61

Figure 4.16: (top) Velocity of the ram as a function of stroke (sinusoidal waveform); (bottom) approximated to a straight line in the 0.2mm length of the contact with the sheet ...... 63

Figure 4.17: (top) Simulation setup for blanking; (bottom) enlarged view of the sheet mesh at the blanking interface ...... 64

Figure 4.18: Different stages of punch penetration into the sheet ...... 65

xvi

Figure 4.19: Comparison of experimental and simulated blanked part edge (Zb – burr, Zf – fracture zone, Zs – shear zone, Zr – rollover zone) for blanking 0.2mm thick C51100 material,

1.5mm diameter hole ...... 66

Figure 4.20:Comparison of blanking load-stroke curve obtained from experiments and FE simulations ...... 67

Figure 4.21: Flow chart to obtain flow stress curve using blanking tests for low strain rate ...... 70

Figure 4.22: Points taken in the deformation zone to average stress, strain and temperature values ...... 70

Figure 4.23: Flow chart to develop flow stress curve using blanking tests for higher strain rate .. 71

Figure 4.24: Flow stress curve for C51100 obtained using bulge test (extrapolated by fitting

Hollomon’s equation  842.4 0.1355) and used in FE simulations ...... 73

Figure 4.25: (From top left clockwise) FE model setup of blanking C51100 sheet, 0.2mm thick; strain distribution; strain rate distribution; temperature distribution in the deformation zone of the sheet (at 0.1mm punch stroke) ...... 74

Figure 4.26: Comparison of experimental and simulated force-stroke curves (at 20mm/sec) ...... 75

Figure 4.27: Influence of blanking speed on maximum blanking force ...... 76

Figure 4.28: Flow stress curves for higher strain rates obtained using the methodology ...... 77

Figure 4.29: Temperature dependent true stress-strain curve for OFHC copper [Nemat-Nasser et al, 1998] ...... 78

Figure 4.30: Flow stress curves for higher temperatures obtained using the ratios calculated for

OFHC ...... 79

Figure 4.31: Comparison of experimentally obtained and simulated force-stroke curve for high strain rates (left) έ = 2 x 104 s-1, (right) έ = 3 x 104 s-1 ...... 80

Figure 5.1: Effect of punch-die clearance on part edge quality (obtained using FE analysis) [sheet

- 0.2mm thick, C51100 material] ...... 84

xvii

Figure 5.2: Maximum punch loads at different punch-die clearances (obtained using FE analysis)

[sheet - 0.2mm thick C51100] ...... 84

Figure 5.3: Maximum Punch Stress at different punch-die clearances (obtained using FE analysis) [sheet - 0.2mm thick C51100] ...... 85

Figure 5.4: Influence of punch corner radius on maximum punch stress (obtained using FE analysis) [sheet - 0.2mm thick C51100] ...... 86

Figure 5.5: Influence of punch corner radius on part edge zones (obtained using FE analysis)

[sheet - 0.2mm thick C51100] ...... 87

Figure 5.6: Punch tip geometries studied for their effect on punch load ...... 88

Figure 5.7: Shape of the slug obtained using the different punch tip geometries (obtained using

FE analysis) [sheet - 0.2mm thick C51100] ...... 89

Figure 5.8: Punch load during blanking using the different punch tip geometries (obtained using

FE analysis) [sheet - 0.2mm thick C51100] ...... 90

Figure 5.9: Effect of coefficient of friction on part edge quality (obtained using FE analysis) [sheet

- 0.2mm thick C51100] ...... 91

Figure 5.10: Effect of coefficient of friction on maximum punch load (obtained using FE analysis)

[sheet - 0.2mm thick C51100] ...... 91

Figure 5.11: Effect of coefficient of friction on maximum punch stress (obtained using FE analysis) [sheet - 0.2mm thick C51100] ...... 92

Figure 5.12: Effect of contact pressure on blanked edge quality (obtained using FE analysis)

[sheet - 0.2mm thick C51100, punch diameter – 0.2mm] ...... 93

Figure 5.13: Effect of contact pressure between stripper plate and sheet material on part edge quality (left) low contact pressure <10 MPa (right) high contact pressure > ½ YS of sheet material

...... 94

Figure 5.14: Misaligned punch (punch diameter – 1.5mm); maximum misalignment – α = 0.015⁰

...... 95

xviii

Figure 5.15: Schematic of (a) Plane strain condition with no misalignment (b) Tilted punch and (c)

Vertical punch with unequal clearances on either side used in FE simulations ...... 96

Figure 6.1: Worn punch showing more wear along the radius than straight edges ...... 99

Figure 6.2: Curvilinear length (CL) along which stresses are plotted on the punch ...... 103

Figure 6.3: Effect of geometry on contact pressure – blanking AISI 1010 sheet, 0.25mm thick

(when the punch has penetrated 0.25t into the sheet, corresponding to maximum stress) ...... 104

Figure 6.4: Influence of radius on average contact stress on the punch (sheet material AISI 1010,

0.25mm thick) ...... 105

Figure 6.5: Influence of radius on average contact stress for different materials (at ~0.25t penetration, which gave maximum stress)...... 106

Figure 6.6: Effect of geometry on blanked edge zones obtained using FEA (sheet material – AISI

1010 – 0.25mm thick) ...... 107

Figure 6.7: Effect of geometry on blanked edge zones obtained using experiments (sheet material

– SS301 – 0.25mm thick) ...... 108

Figure 6.8: Blanked edge cross section (a) straight edge (b) 0.25mm radius (c) 0.15mm radius

(sheet material – SS301 – 0.25mm thick) ...... 108

Figure 6.9: Influence of material strength on average contact stress during blanking of straight edge) (at ~0.25t penetration, which gave maximum stress) ...... 110

Figure 6.10: Flow stress data of the materials showing the stresses at ε=0.9 for different materials

(these stress values are plotted along the x-axis in Figure 6.9) ...... 111

Figure 6.11: Effect of sheet thickness on average contact pressure during blanking C51100 material (at ~0.25t penetration, which gave maximum stress) ...... 112

Figure 6.12: Effect of punch-die clearance on average contact stress on the punch (sheet material – AISI 1010 – 0.8mm thick) ...... 113

Figure 6.13: Relationship between tool wear and punch-die clearance obtained experimentally when blanking Docol 1400 DP, 1mm thick by [Hogman, 2004] ...... 114

xix

Figure 6.14: FE models used in the simulations (from top left clockwise) (a) round, (b) straight edge, (c) corner of rectangle, (d) oblong...... 116

Figure 6.15: Effect of punch-die clearance on contact stress during blanking of a straight edge

(sheet material – AISI1010 0.8mm thick) ...... 117

Figure 6.16: Effect of punch-die clearance on average contact stress on the punch (sheet material – AISI 1010 – 0.8mm thick) ...... 118

Figure 6.17: Effect of punch-die clearance on contact stress during blanking of 1.27mm radius hole (sheet material – AISI1010 0.8mm thick) ...... 119

Figure 6.18: Effect of clearance on average contact stress on the punch for 1.27mm radius hole

(sheet material – AISI1010 0.8mm thick) ...... 119

Figure 6.19: Effect of clearance on stress (on the bottom surface of the punch) during blanking of an oblong shaped geometry (sheet material – AISI1010 0.8mm thick) ...... 120

Figure 6.20: Effect of clearance on stress (on the bottom surface of the punch) during blanking of a rectangle shaped geometry ...... 121

Figure 6.21: Optimal clearance selected for different geometries for different sheet materials and thickness ...... 123

Figure 7.1: Schematic top view of punch and die for a) uniform clearance and b) higher clearance

[Högman 2004] ...... 126

Figure 7.2: Schematic top view of the punch showing the clearance % for (a) uniform clearance and (b) larger clearance at corner ...... 126

Figure 7.3: Schematic of the view of the punch used in results ...... 127

Figure 7.4: (left) Punch with radius 0.2 mm at the corner of the rectangle chipped after 40000 strokes (right) punch with radius 0.5 mm at the corner of the rectangle did not chip after 200000 strokes [Högman 2004] ...... 128

Figure 7.5: Yield data for Docol 800DP [Söderberg 2006] ...... 129

Figure 7.6: Stress distribution at the corner of the rectangular punch ...... 130

Figure 7.7: Normal pressure [MPa] on the punch surface for 0.2 mm radius (uniform 10% clearance) and 0.5mm radius (larger 20% clearance in the corner); arrows are pointing to the highest contact pressure ...... 131

Figure 7.8: (right) Wear starts around the corners; (left) and then progresses to the straight edges of the punch used in production ...... 133

Figure 7.9: Punch shapes used in the study ...... 133

Figure 7.10: Examples showing the variable punch-die clearance design ...... 136

Figure 7.11: Comparison of punch wear between regular clearance (after 126000 hits) and variable clearance (after 350000 hits); also shown are pictures of the sharp punch ...... 138

Figure A.1: Arbitrarily extrapolated flow stress curve using straight line for C51100 (Low global strain rate έ ~ 10s-1) ……………………………………..…………………………………………….152

Figure A.2: Comparison of experimental and simulated blanking force (flow stress curve used in

Figure A.1 is used in simulation) …………………………………….……...……….………………..153

Figure A.3: Flow Stress obtained using the methodology ………………...………………………..154

Figure A.4: Comparison of experimental and simulated blanking force (flow stress curve used in

Figure A.3 is used in simulation).………………………………………..…………………………...155

xxi

CHAPTER 1: Introduction

Sheet metal forming processes like blanking, stamping and bending are very commonly used in the manufacture of sheet metal parts and it takes a combination of different processes to manufacture sheet metal parts. Blanking and piercing are metal shearing processes in which the incoming sheet material is sheared to a desired shape. In blanking, the removed piece of material is the product and while in piercing, the material that is removed is scrap while the remaining part of the strip is the product, as shown in the schematic in Figure 1.1. In this study, blanking and piercing are used interchangeably since they mean the same physical process. Blanking is one of the processes in which the sheet undergoes severe deformation since the sheet metal is sheared or separated to have the slug and part.

Figure 1.1: Different blank and scrap for blanking and piercing

Figure 1.2 shows the schematic of the blanking process. During blanking, the sheet material is held by the stripper plate (which is also the blank holder) by applying a certain stripper force, fb.

The sheet material in between the punch and die undergoes very high deformation and is

1 sheared as the punch penetrates the sheet material with velocity vp. The stripper plate strips the slug off the punch during the upwards motion of the punch.

Figure 1.2: Schematic of the blanking tool setup

1.1. Blanking Process

The blanking process can be considered to include a series of phases in which the sheet metal undergoes deformation and separation, as seen in Figure 1.3.

Contact of the punch: The punch first touches the fixed sheet. At impact, a compressive stress rapidly builds on the punch and sends a shock wave through it.

Elastic and plastic deformation: The punch penetrates into the sheet, first causing an elastic and then plastic deformation.

Shearing and crack formation: When the stresses increase, shearing occurs followed by fracture.

Fracture begins from both the punch end and die end of the sheet. They usually meet and complete fracture of the material takes place.

Breakthrough: If the sheet material has a high strength or is thick, a large force is required for the blanking process. During shear and fracture, compressive forces are stored in the tool. When

2 complete fracture occurs, there is an instant release of these compressive forces. These generate shock which can lead to breakage of the punch in some cases.

Stripping: The punch moves down to the bottom dead center and ejects the part / slug. At the bottom dead center, the direction of punch motion is reversed. There is friction between the stock and the surface of the punch, which causes the sheet to lift along with the punch. A stripper or blank holder strips the blank from the punch.

Figure 1.3: Phases of the blanking process [Schuler Handbook, 1998]

1.2. Part Edge Characteristics

The sheared/blanked edge is made of different zones based on material deformation that has occurred. The zones and deformation modes of a blanked part edge are given below and in

Figure 1.4.

 Rollover zone (Zr): caused by plastic material deformation.

 Shear zone (Zs): smooth and shiny area, created during material shearing.

 Fracture/Rupture zone (Zf): rough surface, results after the material cracks.

 Burr zone (Zb): caused by plastic deformation.

 Depth of crack penetration (Dcp): angle of fracture zone, depends mainly on clearance.

 Secondary shear: created if cracks do not run towards each other and material is

sheared again.

The ratio of the different zones is influenced by parameters like the punch-die clearance, punch corner radius and sheet material properties to name a few. A large shear zone and small rollover and burr are generally preferred in the blanked part for post-processes like assembly and flanging.

1.3. Punch/Die Wear

Punch and die wear influences the blanked edge quality because it changes the punch-die clearance and punch/die corner radius. The wear on punch and die is affected by the sheet material blanked, punch/die material and coating, geometry of the punch and die and punch/die corner radius to name a few.

Figure 1.4: Different zones of blanked part edge

CHAPTER 2: Research objectives and approach

The objectives of this study are to

1. Improve punch/die life in high speed and high volume blanking operations

2. Improve blanked edge quality of parts produced at high speeds

There are two major directions from which the objectives can be approached (i) material (ii) design. The punch and die material play a very important role in affecting wear. Hence, the most suitable punch and die material/coating for a given sheet material can be used to minimize the wear. The other approach is making small changes to the design of the punch and dies to improve the life of the tool. This study focuses on design-driven minimization of tool wear.

The objectives of this study are achieved by the following approach.

1. Understand punch-sheet material and punch-stripper plate interaction in high speed

blanking through experiments and FE simulations.

2. Develop a methodology to obtain flow stress data of materials at high strains and strain

rates by a combination of experiments and simulations to be able to conduct more

accurate FE simulations.

3. Identify the most important parameters that affect punch/die life and blanked edge quality

by the studying the influence of various parameters like punch material and shape,

punch/die clearance, stripper pressure and coefficient of friction.

4. Understand the effect of punch geometry, sheet material and thickness and punch-die

clearance on punch stress.

5. Establish a relationship between punch-die clearance and geometry of the punch for

achieving uniform punch stress and hence uniform wear.

6. Verify the concept of variable punch-die clearance using experiments.

CHAPTER 3: State-of-the-art review

3.1. Fundamentals in blanking – effect of various process parameters on punch life and

blanked edge quality

3.1.1. Effect of punch-die clearance

In general, as the clearance between punch and die increases, the roll over zone, fracture zone, fracture angle and burr increase while the shear zone decreases. Insufficient clearance produces secondary shear i.e the cracks originating at the punch and die do not meet; hence a ring of material is further stressed to its shear limit, expending more energy. Excessive clearance causes large plastic deformation, large burr and high fracture angle. Furthermore, tool life is also lowered by improper clearance. There are studies which have studied the effect of punch-die clearance on part edge quality and punch life, some of which are mentioned below.

[Husson et al, 2008] conducted blanking simulations of a copper alloy of sheet thickness 0.58mm and compared it with experimental results. The effect of punch-die clearance on the part edge quality was studied in the range of 15 µm (2.5%) to 110 µm (19%) during blanking of 3.5mm diameter holes. From the FE simulations, it was found that rollover and shear edge increase and fractured edge decreases with increase in blanking clearance as shown in Figure 3.1. The fracture angle increases significantly with clearance.

o zone lengths / t / lengths zone

Figure 3.1: Effect of blanking clearance on part edge as predicted with finite element simulations on 0.58mm Cu alloy [Husson et al, 2008]

Figure 3.2: Effect of blanking clearance on part edge in DP590 steel of 1.4mm thickness [Wiedenmann et al, 2009]

[Wiedenmann et al, 2009] conducted blanking studies on DP590, 1.4mm thick. The effect of clearances on part edge quality in the range of 5% -20% was studied during blanking of 10mm diameter holes. Roll over and fracture increase with clearance while the shear zone decreases with clearance as shown in Figure 3.2.

[Grünbaum et al, 1996] also observed the same effects as [Weidenmann et al, 2009] when blanking low carbon steels, high carbon steels, copper, aluminum and brass at clearances ranging from 5% to 20% sheet thickness. 12.7mm diameter holes were blanked. Experiments were conducted at blanking velocities ranging from 0.15m/sec to 3.6m/sec. The length of the shear zone decreases with increase with clearance for all materials in this study, as seen in

Figure 3.3. However, some materials are more sensitive than others to punch-die clearance. The tests are conducted at different blanking velocities. The shear length increases for all materials, more significantly in copper, increases with blanking velocity.

Influence of punch-die clearance on shear zone at (left) 0.15m/sec and (right) 3.6 m/sec

100 100 low carbon steel low carbon steel high strength steel 80 high strength steel80 Al 2008 Al 2008 Copper 110 Copper 110

60 60 % shear % 40 shear % 40

20 20

0 0 0 5 10 15 20 250 5 10 15 20 25 punch-die clearance [%] punch-die clearance [%]

Figure 3.3: Influence of punch-die clearance on shear edge length for different blanking velocities [Grünbaum et al, 1996]

[Bell, 2006] conducted a study in which 1400 MPa grade sheet material, 1mm thick was blanked with PM 4% V, 60 HRC punch at three different clearances, 6%, 10% and 14% of sheet thickness. Punch wear was measured after 200 thousand strokes. It was found that lower

9 clearance caused galling while higher clearance caused high bending stresses in the cutting edge increasing the risks of edge chipping. There is an optimum clearance in between the two which will have lower wear on the tool (Figure 3.4).

Figure 3.4: Effect of punch-die clearance on tool wear [Bell 2006]

[Högman, 2004] studied the influence of punch-die clearance on the punch life in a rectangular part. Experiments were conducted on 1mm thick Docol 800DP sheet material. Two radii were used in the punch for the corner of the rectangle for the experiments, 0.2mm and 0.5mm. A constant punch-die clearance of 10% is maintained in the case of 0.2mm radii. A punch-die clearance of 10% is maintained along the straight edge but increases to a maximum value of 20% in the corner in the case of 0.5mm punch corner radius as shown in Figure 3.5. The tool material used was Vanadis 4 with a hardness of 60 HRC.

20%

Figure 3.5: Influence of punch-die clearance on punch wear at the corner of the rectangle (left) with 10% and (right) with 20% punch-die clearance [Högman, 2004]

The punch with a punch-die clearance of 10% at the corner of the rectangle lasted 45000 strokes before chipping. The punch with a punch-die clearance of 20% at the corner of the rectangle lasted 200000 strokes without chipping. Figure 3.5 shows the worn tools with the different corner radii on the rectangular punch.

All the studies have shown consistently similar results for the effect of punch-die clearance on blanked edge quality and punch wear. It should be noted that all studies except [Högman, 2004] are conducted for blanking of round parts, may be because of the simplicity involved. Hence, the effect of sheet material and thickness, punch-die clearance and blanking velocity on blanked edge quality and punch wear are studied. The effect of geometry of the pierced shape on punch wear is not studied.

3.1.2. Effect of punch corner radius [Picas, 2010]

The effect of punch corner radius and punch-die clearance on punch stress and punch wear was studied by [Picas et al, 2010] experimentally. Blank material used was DP1000, 2mm thick.

Punch material was a cast steel (D2) tempered and hardened to 60-62 HRC. The simulations indicated that the effect of punch corner radius on Von Mises stresses generated in the punch is more dominant than the effect of clearance. The maximum stresses in the punch show a decrease of 500 MPa when the corner radius is increased from 0.01mm to 0.1mm, while there is a decrease of only about 50 MPa when the clearance is increased from 10% to 20% (Figure 3.6 a). Figure 3.6 b shows that the punching load does not change significantly with change in corner radius. Experiments reveal that the punch ran for only a few hundred strokes before showing the first signs of wear and 3000 strokes before fracture with the 0.01mm corner radius. The punch with 0.1mm corner radius ran for 15000 strokes without showing any signs of damage. 30000 strokes showed some wear and edge chipping.

Figure 3.6: (a) Maximum Von Mises stress shown for 10%, 15% and 20% clearances at 0.1 and 0.01mm corner radii (b) Experimental load-displacement diagram at corner radius 0.1mm and 0.01mm [Picas et al, 2010]

The effect of having a very large punch corner radius on the blanked edge quality is not discussed.

3.1.3. Effect of stripper pressure on blanked edge quality of thin parts

Very little data is available in the literature on the effect of stripper pressure on blanked part quality. Stripper plate, which is primarily used to help the punch withdraw from the sheet, also plays an important role in holding thin sheets in place while blanking and also in changing the stress patterns and the path of fracture.

The effect of using a spring stripper in sheet metal cutting was experimentally studied by [Bing &

Wallbank, 2008]. They found that:

1. Using a sprung stripper reduces the height of burr formed on a cut edge; in some places it is reduced by a factor of two. It also reduces secondary shear on the exterior part of the edge.

2. The process is more controlled with a sprung stripper holding the strip down and preventing bending— results show under half the amount of variation as without.

3. On the return stroke, horizontal burr can be formed when a sprung stripper is used. This is from a combination of the strip being pulled upwards by friction from the withdrawing punch and the bulk being held flat by the stripper, and is present where punch to die misalignment causes unbalanced lateral forces.

4. The mean of the burr height varies around the hole, as does the variation in burr height readings. This is independent of the type of stripper used.

[Jimma et al, 1990] studied the various factors that affect the dimensional accuracy of lead frames. Some of them are (i) mechanical properties of the strip (ii) static and dynamic accuracy of press and blanking tools (iii) accuracy of feed (iv) slenderness of punch (v) tool wear (vi) lubricant

(vii) blanking speed (viii) strip-holding force (ix) order of progression (x) shape of the curved lead frames (xi) ratio of length and width to thickness of leads.

[Sekine, 2005] has used varying stripper pressures and changed the shut height of stripper stopper to improve accuracy of lead frames dimensions and reduce vibrations at higher blanking

13 speeds. When the lead width is as narrow as the thickness of the sheet, a slight unbalance of the clearance causes lead frame deflection. Influence of stripper force on dimensional accuracy of the lead frame is studied. At high speeds, stripper vibrations also cause unsteadiness during blanking. I-shaped and L-shaped models are blanked in a progressive die in the study. In this study, 42% Ni steel strips, 0.25mm thick of tensile strength 618 MPa are blanked. 2% punch-die clearance is used in this study. Both punch and die are made of cemented carbide. Clearance between punch and stripper is 6µ. Strip-holding pressure is 0.8MPa (116 psi) and strip-holding force is 3500N. Blanking is carried out at 100 to 1200 SPM using a 60tonf straight sided crank press.

Increasing blanking speed makes the vibration of the stripper plate more remarkable. At higher speeds, strip holding force cannot be effectively applied to the blanked material because of the vibration of the stripper. Hence, the blanked lead deflects. As soon as the stripper hits the sheet, there is bending vibrations which changes to seesaw vibration after 0.1ms. To avoid the vibration, the stripper force was increased from 2600N to 5500N and gap between stripper and stopper was reduced from 140µm to 100µm (almost equal to the upper surface of the material). The dynamic vibration of the stripper at 1200 SPM was recorded for different stripper forces and different gaps

(Figure 3.7).

For the I-shaped lead frame, a new stripper with wider pressing area and a structure protecting the lead’s falling sideways was developed (Figure 3.7). Sheet material is 0.15mm thick and lead widths are 0.12, 0.15 and 0.20mm. When thickness to width ratios is bigger than one, the new design of stripper performed well in comparison with conventional one, but for ratios lesser than one, they performed the same. Hence, the new design was successful for thinner lead widths.

Stripper Holding Force (N) Gap between stopper and stripper (µ) 1) 2600 140 2) 2600 100 3) 5500 140 4) 5500 100

Figure 3.7: (left) Dynamic behavior of stripper at 1200SPM; (right) Mechanical model of newly designed stripper for I-shaped frame [Sekine, 2005]

According to [Wang, 2010], a strong stripper-force increases the range of hydrostatic compressive stress and delays the formation and development of crack, giving a longer and better shear zone.

3.2. Punch materials and coatings

Punch failure can be lowered by using newer and better materials, surface treatments and coatings. WC is the most commonly used punch material in high speed blanking of miniature parts.

3.2.1. Different Punch Materials and Coatings Used in Blanking

The punch materials and coatings used in blanking depend largely on the sheet material and expected tool life.

Steels

Some of the tool steels by Uddeholm are [Sandberg, 2004]:

AISI D2 – high carbon/chromium tool steel with high amount of chromium carbides, 13%, and a hardness potential of 59-61 HRC after secondary hardening

AISI S7 - shock steel with exceptional impact properties. It is used widely for medium-run cold- work tools.

Carmo/Calmax – the first generation of car body die steels with a pure martensitic matrix structure. Optimized properties profile with regard to wear, ductility, weldability and induction hardening. 58 HRC is the potential hardness after low temperature tempering.

Diemax – new developed matrix steel with an optimal combination of ductility, hardenability and temper resistance at a hardness up to 57-58 HRC. The hardness is achieved by high temperature tempering, which facilitate surface coating.

Caldie – new car body die steel developed from the old grade, Carmo, but with high temperature tempering facilities giving a hardness of 60-62 HRC and still maintaining a very good ductility, hardenabilty and temper resistance. A matrix is steel also well suited for surface coating.

Sleipner – a development of AISI A2 and D2. Improved ductility compared to D2 due to less amount of chromium carbides, 6% vs.13% (D2) and higher hardness potential, 64 HRC vs. 61

HRC (A2/D2) after secondary hardening.

Roltec – a spray formed grade in between PM and conventional cold work steels. A steel with a good combination of abrasive wear resistance and ductility with a hardness potential up to 65

HRC.

Some of the other commonly used tool steels for blanking are:

AISI M2 - possesses an admirable combination of wear resistance, toughness and compressive strength. This combination of properties makes it superior to many of high alloyed cold work steel.

CPM M4 - high-vanadium special purpose HSS exhibiting better wear resistance and toughness than M2 and M3 in cold work punches, die inserts and cutting applications involving high speed and light cuts. [Crucible Industries]

Powder metallurgy (PM), Cermets and Ceramics

PM [Sandberg, 2004]

Vanadis 4 - PM steel offering an extremely good combination of wear resistance and ductility for high performance tools.

Vanadis 4 Extra – An optimal combination of ductility and wear resistance, mixed abrasive and adhesive, of all PM cold work steels.

Vanadis 6 – PM steel with better abrasive wear resistance than Vanadis 4 Extra.

Vanadis 10 – a high vanadium alloyed PM grade with the best combination of high abrasive wear resistance and ductility.

Vancron 40 – a nitrogen alloyed PM steel with very good low-friction properties for excellent galling and adhesive wear resistance.

Cermets / Cemented Carbides

WC – the most widely used composites in different wear applications owing to their excellent combination of high wear resistance and strength, as well as toughness [Klassen 2011]

TiC - TiC-based cermets (with Ni-alloy or steel binder) have proven successful in some applications because of their high specific strength (low density), high adhesive wear resistance, good weldability and improved resistance to oxidation [Klassen 2011].

When machining copper, nickel and pure iron, the wear of cemented carbide may be fast. The reason for this is that the affinity between the cobalt present in cemented carbide and copper or nickel causes the wearing of the cemented carbide to progress faster. [Misumi Tech Central]

Ceramics

Advanced ceramics are characterized by high strength, high fracture toughness, fine grain size and little or no porosity. They often are used in high-wear or corrosive environments.

Y-TZP (yttria-tetragonal zirconia polycrystal) – A zirconia-alloyed ceramic is used in some punching applications. Although it has great wear resistance, it is not very commonly used in metal forming applications because it is very unforgiving with bending.

A comparison of wear characteristics of the tools by Uddeholm are shown in Figure 3.8.

Figure 3.8: Characteristics of tool materials by Böhler Uddeholm [Bell, 2006]

Surface Treatments and Coatings

The common surface treatment processes are:

(i) Nitriding

(ii) Physical Vapor Deposition (PVD)

(iii) Chemical Vapor Deposition (CVD)

(iv) Thermal Diffusion (TD)

CVD and TD work best in forming applications that do not require high levels of precision.

Because of the high processing temperatures involved, distortion and size changes occur. PVD has lower processing temperatures; hence it can accommodate more precision in the tooling.

Some of the coatings that are commonly used in metal forming tools are TiN, TiC, TiCN, TiAlN,

TiCrN, AlCrN, CrN. MoST™ is a PVD solid lubricant coating, composed of sulfur and molybdenum [Dayton Progress].

It should be understood that not all coatings go well with all tool materials in terms of coating adhesion. Similarly, surface treatments are not recommended for all tool materials.

3.2.2. Comparison of WC punches with Ceramics (Y-TZP) [DiRuggiero 2000]

A case study evaluating punch life with two different materials, WC and Y-TZP is presented. The experimental conditions are given in

Table 3.1

Table 3.1 Case Study: Comparison of WC and ceramic punches in precision stamping of copper lead frames

Item Existing Stamping Parameters Z-mat Ceramic Tooling

Punch Type Pierce punch Pierce punch

Punch Dimensions 0.008in – 0.012in cross sections 0.008in – 0.012in cross sections

WC: 10%-15% Co, sub-micron Z-Mat Ceramic, sub-micron Punch material particles particles

Stripper clearances .0001in max per side .0001in max per side

Die manufacturing EDM Split and ground process

.008in - .012in thick .008in - .012in thick Material stamped Copper alloy 194 Copper alloy 194

Press speed 450 SPM 450 SPM

2.5 million strokes between 11+ million strokes between RESULTS resharpenings resharpenings

According to the author, a two phase material, tungsten carbide for example, is susceptible to metal pick-up, even with a high surface finish, for two reasons: 1) the metallic binder phase

(typically cobalt) has an affinity for other metals like copper and nickel and 2) cobalt depletion occurs during EDM. These are overcome while using ceramic punches. It must be borne in mind

19 that ceramic punches cannot take bending loads like WC punches. This must be taken into account while designing dies and other tooling with ceramic punches.

3.2.3. Environmentally Benign Tribo-systems for Metal Forming [Bay, 2010]

Various punch materials and coatings for dry blanking of layered electrical steel sheet have been investigated, where the intermediate, insulating glass coating causes excessive wear. Testing of punches in AISI D2 and PM High Speed Steel (HSS) with PVD coatings of DLC (diamond-like carbon), Me-DLC (metal containing DLC) and EDC (Electro Discharge Coating) with TiC were carried out. All the coatings extended punch life by about a factor of three. Combining the EDC with cryogenic process treatment of AISI D2 resulted in an extension of tool life by a factor of nine. A similar investigation on blanking of layered electrical steel sheet was conducted, testing punch materials of PM HSS and cemented carbide with coatings of AlCrN, TiCN and TiAlN +

WC/C. The best results were obtained by combining PM HSS with AlCrN. Good results have been obtained in coating of punches with TiCN for Al sheet perforation, increasing the tool life by a factor 10 due to reduced pick-up and increased wear resistance. Oerlikon Balzers reports that coating of punches with AlCrN gives three times longer tool life than TiN and TiCN.

3.2.4. Improving the wear resistance of tools for stamping [Straffelini 2010]

Various tool materials and coatings were evaluated. A correlation between burr height and tool wear is first established in this study. Various tool materials and treatments/coatings are evaluated based on the burr height. The blank has a thickness of 2.2mm and is made of a cold rolled strip subjected to spheroidisation annealing. Its chemical composition is: 0.7% C, 0.7% Mn and 0.2% Si. Figure 3.8 (a,b) shows the shape of the part and the tooling used in this study.

Figure 3.8(c) shows the burr height for the various tool materials and hence their performances.

S390 is a PM HSS from Uddeholm, QTC and QCT are two different cryogenic treatments on the

S390 steel. Hard metal is carbide, Ceratizit H40S. AlCrN is PVD coated on S390 steel by

Oerlikon Balzers Italy. From this graph, it can be seen than S390 with an AlCrN PVD coating is the best performer, followed by the ceramic.

(a) (b) (c)

Figure 3.9:(a)Part formed (b) tooling used in the experiment (c) Burr height measured for various tool materials/treatments/coatings[Straffelini 2010]

3.3. Material model and fracture model used in simulations

3.3.1. Material Model

High speed blanking simulations require flow stress data of materials at very high strains (~ 3) and strain rates (up to 105). Currently, materials are not tested to very high strains and strain rates at the same time. Tensile tests can be conducted for very high strain rates, however necking starts to occur at strains much lower than 1, especially at high strain rates. Torsion tests are available, but they do not reach high strains of ~3 either. Hence, material models with extrapolated data are used in modeling the material for simulations. The accuracy of material data and hence the accuracy of simulations at higher strains and strain rates is unknown.

3.3.2. Fracture Model

Ductile fracture criteria works under the empirical hypothesis that the ductile fracture occurs when the maximum damage value of the work piece exceeds a critical damage value. Many efforts have been made by various researchers to formulate a ductile fracture criterion that can accurately predict fracture.

 [Yu, 2011] suggests that it is difficult to determine the criteria and the corresponding critical

damage value to be selected in complex situation because the criterion is related to not only

material property but also stress state and deformation process.

 [Bao & Weirzbicki, 2004] suggests that it is impossible to capture all features of ductile crack

formation in different stress states with a single criterion. Different functions are necessary to

predict crack formation for different ranges of stress triaxiality.

 [Goijaerts et al, 2001] conducted experiments, blanking X30Cr13 stainless steel with varying

punch-die clearance. Clearance was varied from 1% to 15 % of sheet thickness. Simulations

were conducted with different damage criteria incorporated into them; (i) plastic work (ii)

Cockroft & Latham (iii) Rice & Tracy and (iv) Oyane et al. Rice & Tracy model was adapted to

represent the triaxiality in blanking. Adapted Rice & Tracy and Oyane were found to

represent the ductile fracture occurring in blanking with reasonable accuracy.

 [Sartkulvanich et al, 2010] conducted a study on blanking of AHSS DP 590 with four punch-

die clearances. Different fracture criteria were evaluated for FE modeling of blanking and the

results were compared with experimental data. The adapted Rice & Tracy criterion gave the

closest results, when comparing the part edge measurements from experiments and

simulations.

3.4. Punch failure – simulation modeling and experiments

3.4.1. Punch Failure Mechanisms

Punch failure can be broadly classified into the following types.

Wear

Wear is damage to a solid surface, involving loss or displacement of material. Wear is caused by sliding contact between the work piece and tool. Mainly, there are two types of wear [Uddeholm and SSAB 2008; ASTM G40 2005]:

1. Abrasive wear is caused by hard particles forced against and moving along a solid surface.

2. Adhesive wear is defined as wear due to localized bonding between contacting solid surfaces leading to material transfer between the two surfaces or loss from either surface.

Side / flank wear [Luo, 1999]

This is the side surfaces of the punch during shearing to produce worn conditions. Punch side wear is a result of the combination of adhesive wear, fatigue wear, and abrasive wear. Such side wear of the punch would cause the internal diameter of the punched holes to become small, and it increases the clearance between the punch and the die. Furthermore, the deformation (edge draw-in) of the work piece during shearing is also increased.

Face wear [Luo, 1999]

The face of the punch in the shearing process produces wear, which then causes the punch edges to become round and obtuse. It may be a result of mechanical attrition and micro-chipping.

This would reduce the sharpness of the punch during shearing, and increase the deformation of the punched work piece. Moreover, the burrs of parts become larger, and the noise level in the press also becomes very high. Figure 3.10 shows face and flank wear in a punch schematic.

Figure 3.10: Schematic illustration of flank and face wear in punch [Hernández et al, 2006]

Chipping

This punch appearance would show some micro crushes, fragments and breakage on the cutting edges. The reason for this may be the repeated impact loads or thermal shocks. If the surface of the punch is too rough, then chipping of the punch cutting edges would occur easily [Luo, 1999].

Chipping is a result of high stresses, exceeding the fatigue strength of the tool material

[[Uddeholm and SSAB 2008]. Figure 3.11 shows images of chipping on the punch cutting edge.

Figure 3.11: Chipping wear seen on punch cutting edge [Luo, 1999], [Uddeholm and SSAB 2008]

Cracking [Luo, 1999]

This appearance displays many irregular micro-cracks on the face of the punch edges (Figure

3.12). It may be the result of mechanical fatigue or thermal fatigue during shearing. When these cracks propagate further during shearing, the punch edges would produce chipping or, in an extreme case, macro-fracture.

Figure 3.12: Cracking as seen on punch surface [Luo, 1999]

Gross fracture [Luo, 1999]

This is a macro-fracture appearance on the punch surfaces as shown in Figure 3.13. It can be seen that the beach markings on the cyclically grown portions of the fracture are a result of fatigue failure. Furthermore, it is also possibly additional damage of chipping and cracks on the cutting edges during shearing.

Figure 3.13: Gross Fracture on punch [Luo, 1999]

Galling [Uddeholm and SSAB 2008]

Galling (pick-up), which is a result of heavy friction forces due to the sliding contact and the adhesive nature of the work material (Figure 3.14). The galling mechanism is closely related to adhesive wear.

Figure 3.14: Galling on punch face [Uddeholm and SSAB 2008]

3.4.2. Factors affecting tool wear and galling [Billur, 2009]

Contact Pressure

All failure types can be avoided by reducing contact pressure. The contact pressure between punch and sheet during blanking depends on (i) sheet material (ii) punch-die clearance, (iii) punch corner radii and (iv) profile being blanked.

Surface Quality

Although the surface of the tool is much smoother than the surface of the sheet, galling is affected by tool’s surface quality. Polishing the tool surfaces before and after coating helps to reduce galling. Roughness of sheet has little influence [Rooij and Schipper 2001].

Tool Material and Coating

Selection of proper tool material and coating is crucial to avoid galling and tool wear. Tool materials with high toughness and compressive yield strength must be selected according to the application. Coatings that may have lower coefficient of friction should be selected [Podgornik et al 2006].

Lubrication

Unless there is a solid film lubricant coating, lubrication is needed to reduce wear and galling with traditional coatings (CrN or TD). Lubricants help in better performance in processes that involve high contact pressure and temperature. Extreme pressure (EP) additives may be required [Rooij and Schipper 2001; Janoss 2008; Kim et al 2006].

3.4.3. Wear Models

There are numerous empirical-based relationships in the literature, which describe both abrasive and adhesive surface wear in sliding contact as a function of the contact conditions experienced.

These include some of the well-known equations presented by [Rhee, 1970], [Bayer, 1993] and

[Archard, 1953], in which wear rate W is commonly expressed as a function of normal load L, sliding distance S, and wear coefficient K, in the following form: W = K Lm Sn, where m and n are

26 empirical constants [Yan 2012].The wear coefficient K is often determined experimentally. Some of the wear models commonly used in numerical FE simulations are shown in Table 3.2.

Table 3.2: Wear Models used in numerical simulations

WEAR MODEL COMMENTS

[Archard, 1953] W= wear volume W K  F s= sliding distance  n s H K=wear coefficient

Fn=normal load

H=yield stress or hardness of the softer

material

[Archard, 1953[: V= wear volume K q  s K= wear coefficient V  H q= normal pressure

s= sliding distance

H= hardness of the worn material

[Painter et al, 1995]: Zab= abrasive wear depth a b K=experimental coefficient Kl ( p V t) Zab  p= local pressure H c d V=local sliding velocity

Hd=tool hardness

a,b,c= experimental constants

Table 3.2 Continued

[Kang et al, 1999] dfin= allowable amount of wear of the dies

H(T,t,winitial)=function of hardness softening

considering tempering parameter n fin kPL H d fin    H= hardness of the die at steady-state 1 3H H (T , t, winitial ) temperature

k= experimental coefficient

P= local pressure

L= sliding distance

w= amount of wear

[Behrens, 2005] σN= normal pressure n H= Hardness dependent on temperature, θ,   N   w    vrel tinc and time, t H(,t) inc1  Vrel= sliding velocity

Δt= interval of time

n= number of cycles

[Choi et al, 2009] w = wear depth

a K - the abrasive wear coefficient c  P  w(c)  K  vdt   H( j)  P - normal pressure on the contact surface, j1   H(j) - die hardness at the blanking stroke(j),

v - sliding velocity at the contact surface,

c - total number of blanking strokes,

a - experimental constant

From the wear equations given in Table 3.2, it can be observed that certain specific parameters define the wear. They are normal load/pressure, hardness of the die, temperature, sliding distance/velocity at the contact surface. The hardness of the die can be varied by varying the die material, temperature by improving lubrication conditions and normal load/pressure by making small design changes.

3.4.4. Effect of tool wear on part edge quality

The effect of tool wear on part edge quality is significant. Tool wear leads to the formation of burrs and increase in burr length. Burr length is generally an important criterion in the industry to judge part quality. Burr length indicates when the tool should be reground to get the sharp die and punch radius. It has also been observed that the effect of tool wear is more pronounced at higher blanking clearances.

A method to quantify the volume of burr using optical methods is presented by [Makich et al,

2008]. This method is used to measure the effect of tool wear on the volume of burr generated. It can be seen from Figure 3.15 that the volume of burr generated increases rapidly with number of strokes initially. This is followed by a constant burr with increase in stroke, followed by an increase in burr again. This pattern may also be dependent on the sheet material.

Figure 3.15: Increase in burr volume with strokes because of tool wear [Makich et al, 2008]

Figure 3.16: Effect of sheet material on burr volume [Makich et al, 2008]

The volume of burr is also dependent on the sheet material blanked. Figure 3.16 shows that some sheet materials generate much less burr compared to others. For example, Mat 3 shows

~30% reduction in burr compared to Mat A even though the tool wear has well set in by the time

Mat 3 is blanked.

Tool wear can be estimated in simple ways using FE simulations by increasing the punch and die corner radius. The effect of tool wear on part edge quality is significant. Tool wear leads to the formation of burrs and increase in burr length as shown in Figure 3.17 in the work done by

[Husson et al, 2008]. It has also been observed that the effect of tool wear is more pronounced at higher blanking clearances (Figure 3.17).

Figure 3.17: Effect of tool wear and blanking clearance on part edge quality as predicted by simulations on Cu alloy, 0.58mm thick[Husson et al, 2008]

3.5. High speed blanking

Figure 3.18 shows the tooling generally used in high speed blanking. A high speed blanking cycle in the manufacturing of small electronic components consists of the following steps:

(i) Ram moves downwards bringing the stripper plate to come in contact with the blank

(ii) Further downward motion of the ram enables the stripper springs to apply pressure

on the stripper plate and in turn on the blank

(iii) Blanking operation occurs

(iv) Punch moves further down until BDC

(v) Upward motion of the ram, punch rubs sheet in opposite direction

(vi) Punch is stripped off the sheet by the stripper plate

(vii) Stripper springs are released and stripper plate bounces off the sheet

Figure 3.18: Tooling commonly used in high speed blanking [Hirsch et al, 2011]

[Hirsch et al, 2011] conducted experiments to measure punch loads in high speed blanking up to

1000 SPM. Nine holes 0.6mm each were blanked on Cu alloy sheets 0.29mm thick. The various forces recorded on the punch are shown in Figure 3.19. The vibrations due to stripper plate pinning are first recorded which is followed by a spike in force, which is due to the blanking operation. This is followed by the frictional force due to the lateral face of the punch rubbing the sheet. Then, the stripper springs are released and vibrations are recorded. The velocity at which the sheet was blanked is not known. The blanking force seems to increase with speed a bit. But the vibrations due to unpinning of the stripper plate increase rapidly with speed.

Figure 3.19: Forces in the blanking process for stroke rates 100, 300, 500 and 1000 strokes per minute [Hirsch et al, 2011]

This is probably the only study where a wide range of speeds (SPM) are studied for their effect on the change in the various forces acting during the blanking cycle. As high speed presses that can operate at speeds of up to 4000 SPM are available, it becomes extremely important to understand the different forces that are generated during the blanking cycle and understand which of them are influenced by speed.

3.6. Press stability and reverse loading

Stability of the press, tooling and punches plays a very important role in having a long tool life and good part quality. There are several reasons that can lead to press and tooling deflections, some of which are (i) uneven blanking loads (ii) inaccurate guiding systems (iii) reverse loading during blanking. [Jimma et al, 1990] has looked into the relation between press rigidity and part quality in high speed blanking. [Behrens et al, 2010] has considered machine properties in sheet forming simulations to couple the press performance with sheet forming process. [Groche et al, 2007] has described a procedure for the generation of numerical model for forming machines and also showed methods for experimental acquisition of dynamically relevant machine parameters to be used in the numerical simulations.

Snap-thru or reverse loading is another important factor that contributes to the life of the press, tools and punches. Snap-thru forces are evident while blanking thick or very strong materials.

During blanking, some portion of the material is sheared and the rest is fractured. The fracturing happens much faster than shearing and there is a significant reverse tonnage at the end of fracture on the press. The press components reach their maximum deflection just before fracture.

After fracture, they spring back to their original shape and beyond unopposed at a very high velocity and there is a sudden release of energy. The ram is accelerated to a high speed. The drive train of the press has clearances around every moving component. When the clearances have reversed, the ram stops suddenly expending all the energy and sending a shockwave. The press deflects in the opposite direction as when the forward tonnage was developed and the snap-thru forces are also called reverse tonnage. The reverse tonnage can go as high as 50% or more in case of very strong materials, as shown in Figure 3.20.

Figure 3.20: Forces during snap-thru [Miles, 2004]

Punch staggering, nitrogen cushion cylinders and hydraulic dampers are some of the commonly used solutions used to address the problem of snap-thru. However, nitrogen cylinders and hydraulic dampers are not viable solutions in high speed blanking, since they cannot operate at such high speeds.

3.7. Flanging

There are several factors that affect the flangability of a hole like (i) the sheet material (ii) punch- die clearance used in blanking (iii) quality of the sheared hole (iv) shape of the punch used in flanging (v) position of the burr during flanging. The influence of blanked edge quality upon Hole

Expansion Ratio (HER) has been studied by a few researchers in the past, by which a better understanding of the subject has been obtained. However, more needs to be done in this area since a good understanding of how to characterize a blanked edge for flangability is not very clear yet.

[Karelove et al, 2007] studied the influence of cutting methods on the hole flangability of DP800 and CP800 steels. Figure 3.21shows that wire cutting causes only minor deviations from the desired hole geometry. The sharp edges created by wire cutting can have a detrimental effect, as they are the possible stress concentrators during the hole expansion. This fact can also explain that in spite of the evidently higher accuracy of the wire cutting procedure, the increase HER is not much higher than that measured for drilled holes. This study shows the strong influence of the hole edge and hole surface quality on the hole flangability of the material.

Figure 3.21: (top) micrographs of cross sectional area in punched (a), drilled (b) and wire cut (c) hole of DP800; (bottom) Influence of the hole edge condition on HER

[Mori et a, 2010] conducted similar studies where a secondary smoothing operation was employed to improve the hole flangability. DP980 grade material was used in this study. The smoothing operation was conducted with a punch that had a conical angle (30°) lesser than the angle on the flanging punch (60°). The smoothing load was varied to see its effect on HER.

Figure 3.22 shows the blanked edge after smoothing with different loads where the fractured zone is smoothed. The difference in smoothing load causes a difference in the edge quality before flanging. This in turn makes a difference in the flangability, which is shown in Figure 3.23.

These experimental studies have shown that the blanked edge quality plays an important role in hole flangability of the sheet material.

Figure 3.22: Smoothed surfaces and relationships between percentage of depths of rollover, burnished surface, fracture surface and smoothed surface on sheared edge and smoothing load: (a) P = 16.2 kN; (b) P = 21.8 kN; (c) depth percentage [Mori, 2010]

Figure 3.23: Relationship between limiting expansion ratio and smoothing load [Mori, 2010]

3.8. Research Needs

Based on the review of literature, some areas have been identified for further research for improving part edge quality and tool life in blanking of small parts.

1. Understanding tool- sheet material interaction in high speed blanking

Although there have been several studies in high velocity blanking, none but one study by Hirsch et al. have looked at the interaction between the tools and sheet material during the entire blanking cycle. It was a good first step, however data is available only for up to 1000 SPM, while today’s presses can go way faster than that. A good understanding of tool - sheet material is very important to design robust tooling. Hence, as a first step, a good understanding between tool and sheet material is sought after in this research.

2. Methodology to develop flow stress data using blanking tests

In order to conduct accurate FE simulations of blanking, a good flow stress model which is obtained under the same deformation conditions as blanking is required. Currently, high strain rates are achievable by using the Split-Hopkinson’s bar (SHB) test and high strain rates up to 1 are achievable by using shear tests. However, SHB can be very expensive and strain rates up to

3 are reached in blanking. Since blanking is one of the very few forming operations in which deformation reaches high strains and strain rates, it needs to be investigated as a method to obtain flow stress data of materials at high strains and strain rates.

3. Improving punch life by having variable punch-die clearance

The influence of punch-die clearance on punch/die life in blanking has been studied by a number of researchers. However, all but one study has been conducted on blanking of round parts, which is not the majority of blanked shapes. The effect of geometry of the punch (shape of the cut) on punch/die wear is not available in literature. The relationship between geometry of cut, punch-die clearance and punch wear is also not available in literature. Since improving tool life (largely by using better and newer tool materials and coatings) is always an area that has been focused,

38 having design guidelines like optimal geometry dependent punch-die clearance has not been looked into in the past by researchers. Hence, geometry dependent punch-die clearance needs to be studied as a potential way to improve punch/die life. If the concept proves to improve punch/die life, a relationship can be established between the punch-die clearance and geometry being cut.

4. Improved stripper designs to provide better part edge quality

Strippers have traditionally been used only to hold the sheet material during blanking and strip the sheet material off the punch when the punch is returning. However, if the stripper plate holds the sheet material with enough pressure, it may generate compressive stress on the sheet which can help reduce rollover and delay onset of fracture. The influence of stripper pressure on the blanked edge quality is not found in literature, although it is a well-known fact that a better blanked edge is obtained when the stripper pressure applies pressure than when it does not. The change in blanked edge quality with change is stripper pressure is not found in literature. Hence, an understanding of the relationship between applied stripper pressure and quality of blanked edge is required.

5. Optimized punch corner design to improve punch/die life

Punch/die corner radii are usually associated with corner radii. The effect of punch corner radius by using a constant radius at the intersection of the face and flank is found in literature. In reality, the punches are first machined with a certain punch corner radius, but in later sharpening, the bottom surface of the punch is ground to achieve the sharpness. Instead, if a punch corner design which gives an improved tool life is designed, where in the tool life improves ten folds or so, the resharpenings can be avoided and ‘one time use’ punch inserts can be manufactured.

6. Guidelines to design punches to prevent punch breakage

Punch length and cross-section design can be crucial in determining punch breakage. In the manufacture of small electronic components, the punch passes through a slot in the stripper plate. Since the clearance between the punch and stripper plate is in the order of a few microns,

39 the interaction between punch and stripper plate may lead to punch wear and breakage in the lateral area of the punch which is within the stripper plate. A correlation between the forces acting on the lateral area of the punch, punch geometry and breakage is an important relationship to be understood in high speed blanking of electronic parts.

CHAPTER 4: Interaction between punch, sheet material and stripper plate

in blanking

The first step in this study is to obtain a good understanding of the fundamentals involved in high speed blanking. In order to be able to design robust tooling for high speed blanking, it is important to understand the interaction between the tooling and sheet material at various speeds of operation of the press. The interaction between punch, stripper plate and sheet material is studied for different blanking velocities. Since the clearance between punch and stripper plate is generally very small, in the order of a few microns, in blanking of small electronic components, it is important to study the punch-stripper plate interaction. The effect of blanking velocity on blanking load and reverse loading are also important factors of which a good understanding is required.

4.1. Approach

The technical approach used in this section of the study is as follows:

 Conduct blanking experiments at blanking velocities ranging from 20mm/sec to 1600mm/sec.

 Determine the effect of blanking velocity on the blanking force, reverse loading and stripper-

punch interaction.

 Perform FEA of blanking using DEFORM 2D at quasi-static conditions and compare the

simulated results with experimental results.

 Using a combination of experimental results and FEA, develop a methodology to obtain high

strain rate dependent flow stress data for materials.

4.2. Experimental studies on punch loading during blanking at various speeds

Blanking experiments are conducted at various speeds to understand punch loading and punch- stripper interaction.

4.2.1. Experimental Setup

The experimental setup is explained in the following sub-sections.

4.2.1.1. Press and Tooling

Experiments were conducted with a super high-speed mechanical precision press which has a capacity of 30 metric tons and stroke length of 13mm.

The tooling used for the experiments is shown in Figure 4.1.

A punch of round cross section, 1.5mm diameter is used in this study. It is made of Tungsten

Carbide (WC). Two punches of lengths 32.74mm and 37.26mm are used in this study to obtain different blanking velocities for the same ram speed. The punch is screw mounted to the piezoelectric sensor which is screwed to the punch block.

The stripper plates are spring loaded, see Figure 4.1. The springs exert force on the stripper plate during the bottom motion of the ram through the depressor rods. There is initially a small gap between the stripper plate and the stock. The springs first move the stripper plate downwards to initiate contact with the stock and then exert the pressure during further downward motion of the ram. During the upward motion of the ram, the pressure is released from the stripper plate and two return springs (from the die side) raise the stripper plate to its original position. This would enable feed of the stock.

4.2.1.2. Stock Material

The stock is a phosphor bronze alloy, C51100 of 0.2mm thickness. The flow stress data of this material is obtained from bulge test and is shown in Figure 4.14.

Stripper springs

Stripper depressor rods

Piezoelectric sensor

Stripper plate Punch Sheet

Die block

Figure 4.1: Experimental setup for the punch force measurement

4.2.1.3. Sensors

Laser Sensor

Keyence LK-H057 laser displacement sensor is sued to measure the ram displacement.

Piezoelectric force sensor

The piezoelectric sensor is used to measure the punch force. As shown in Figure 4.1, the punch is threaded to the sensor which is mounted on to the punch block. The sensor is from PCB

Piezotronics, model - 221B03.

The tooling parameters are summarized in Table 4.1. The punch-die clearance is 6.5% sheet thickness at 13µ. It can be seen that the punch-stripper clearance is only 3µ, which is a very tight clearance especially at high ram speeds. The stripper pressure of 1.3MPa is obtained by calculating the spring force at the instant the punch touches the sheet material and dividing it by the area of the stripper plate.

Table 4.1: Tooling Parameters Used in Experiments

Parameters Value

Punch

- Material Tungsten Carbide (WC) - Diameter 1.5mm - Lengths (i)32.74mm (ii) 37.26mm - Tip Flat

Sheet C51100, 0.2mm thick

Stripper Pressure ~1.3MPa

Punch-Die Clearance 13µ (6.5% sheet thickness)

Punch-Stripper Clearance 3µ

4.2.2. Experimental Procedure

Blanking experiments were conducted to measure the forces on the punch during the entire blanking cycle for various blanking velocities. Three cases (A, B, C) were studied. Experimental parameters used in each of the cases are shown in Table 4.2. A minimum of 20 readings were recorded for each velocity of each case. The data acquisition rate ranged from 50 – 250 KHz depending on the press speed. Punch force is measured using the piezoelectric sensor and the ram displacement is measured using the laser sensor.

Table 4.2: Experimental parameters used in measuring forces during blanking

Case A Case B Case C

Distance from BDC at which punch touches 0.86/32.74 0.86/32.74 5.38/37.26

sheet (mm) / Punch Length (mm)

Range of punch velocity tested (mm/sec) 20-800 20-465 40-1600

Stripper pinning* the sheet during punching Yes No No**

(~1.3MPa)

*Pinning refers to holding the sheet down by applying force

**Although the stripper plate was pinning the sheet, the punch was too long that pinning occurred after punching through the sheet.

4.2.3. Method used to analyze experimental results

Ram displacement and punch force is recorded for different speeds. It is cleaned using a low pass Butterworth filter in-built in MATLAB when the data is noisy at higher blanking velocities.

The ram displacement and punch force for every speed is recorded over a number of strokes continuously as shown in Figure 4.2.

20 400

15 300

10 200

Punch Force (N) Ram Motion Ram (mm)

5 100 (N) Force Punch Ram Displacement (mm) Displacement Ram

0 0

-5 -100 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Data Point 5 Data point (105) x 10

Figure 4.2: Punch force and ram displacement (recorded continuously over a number of strokes)

This continuous curve is split into the different strokes by considering the maximum and minimum ram displacement for top and bottom dead centers.

The maximum blanking force and the maximum force on the punch during loading and unloading of the stripper are averaged across all the strokes for which measurements are taken and plotted with respect to speed. However, to show all the forces measured in a blanking cycle, one of the strokes is selected (at random) to represent the various forces acting on the punch for every speed. This method was found to be better than any averaging technique used to average the forces, since the magnitude of vibrations and the maximum blanking force is lost during averaging.

The punch force and ram displacements are plotted against ‘normalized time’ where

Normalized time = instantaneous time / total time for a stroke. Hence, the normalized time would vary from 0 to 1 for every stroke, independent of the press speed.

4.2.4. Results

The forces and ram displacements measured for various speeds are analyzed. The graphs in

Figure 4.3, Figure 4.4 and Figure 4.5 explain the different forces experienced by the punch for

Cases A, B and C respectively.

In Figure 4.3 corresponding to Case A, the stripper plate then pins the material at ~ 0.3 normalized time. The stripper plate is pinned to the material using stripper springs and depressor rods. When the ram moves further down towards the BDC, the stripper springs compress, applying pressure on the stripper plate. This causes some vibration to be induced into the stripper springs and hence the stripper plate. This vibration of the stripper plate in turn applies lateral loading on the punch because of the very small clearance between the punch and stripper plate.

The spike seen at about 0.43 normalized time is the blanking load, when the punch pierces through the material. When the punch touches the sheet material and forces itself through the sheet material, it elastically compresses. When the sheet material undergoes fracture, the blanking load drops almost instantaneously releasing the elastic energy stored in the punch. The punch returns to its original state and overshoots a little in tension which is represented by the snap-thru or reverse loading in Figure 4.3. During further downward motion of the ram, frictional forces are exerted on the punch because of the rubbing of the lateral sides of the punch on the sheet. This frictional force changes direction at BDC. At about 0.68 normalized time, the stripper plate is unpinned i.e the springs are relaxed and the stripper plate is pushed back to its original position by the action of the retracting springs on the die block. This causes some vibrations to be induced into the system, similar to that seen at ~0.3 normalized time.

Punch ForcePunch Case ForceA – Blanking Case A Velocity -1400 SPM 465mm/sec 700 14 600 Blanking load 13

500 Stripper plate-punch 12 Frictional force 400 interaction when it 11 between punch pins the sheet Stripper plate-punch 300 and sheet material 10 material interaction when it (changes direction 200 unpins the sheet 9 at BDC) 100 material 8 0 7

-100 6 Punch Force(N) -200 Snap thru force/ 5 Motion(mm) Ram -300 reverse loading 4 -400 3 -500 2 -600 1 -700 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Normalized Time

Figure 4.3: Forces identified on the punch during blanking for Case A at blanking velocity 465mm/sec

Figure 4.4 shows the punch force measured for Case B, where the stripper plate was not used to hold the sheet material. Hence, no stripper plate-punch interactions are observed in this case.

The blanking load, reverse load and friction forces on the punch are recorded. Although stripper plates are used in reality in production, it was removed for experimental purposes to understand the influence of pinning the sheet material on blanking load, all other conditions being the same.

Punch Force Case B -1400 SPM 700 Punch ForcePunch Case Force B -CaseBlanking B -1400 velocity SPM 465mm/sec 14 600700 1314 500600 1213 400500 Blanking load 1112 300400 1011 200300 910 100200 89 1000 78 -1000 67

-200-100 56 Punch Force(N)

Figure 4.4: Forces identified onPunch the punch Force during Case blanking C -1400for Case SPM B at blanking velocity 465mm/sec 700 Punch Force Case C -1400 SPM 14 600700 1314 Figure500600 4.5 showsBlanking the punch load force measured for Case C, where a longer punch was used. 1213Hence, Stripper plate the punch400500 pierces through the sheet material first before the stripper plate pins the material.1112 Stripper plate pins unpins the 300400 1011 Although this is not a common practicethe material in reality, a longer punchmaterial was used to have a higher 200300 910 blanking100200 speed. Hence, the effect of blanking speed on force was recorded for even8 9higher speeds100 0than achievable in Case A and B. 78 -1000 67

-200-100 56 Punch Force(N)

Punch Force Case B -1400 SPM 700 Punch Force Case B -1400 SPM 14 600700 1314 500600 1213 400500 Blanking load 1112 300400 1011 200300 910 100200 89 1000 78 -1000 67

-200-100 56 Punch Force(N)

Punch ForcePunch Case Force C -CaseBlanking C -1400 velocity SPM 930mm/sec 700 Punch Force Case C -1400 SPM 14 600700 1314 500600 Blanking load 1213 Stripper plate 400500 1112 Stripper plate pins unpins the 300400 1011 the material material 200300 910 100200 89 1000 78 -1000 67

-200-100 56 Punch Force(N)

-300-200 45 Motion(mm) Ram Punch Force(N) -400-300 34 Motion(mm) Ram -500-400 23 -600-500 12 -700-600 01 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 -700 0 0 0.1 0.2 0.3 Normalized0.4 0.5 Time0.6 0.7 0.8 0.9 1 Normalized Time Figure 4.5: Forces identified on the punch during blanking for Case C at blanking velocity 930 mm/sec

4.2.4.1. Influence of velocity on blanking load and reverse load

Velocity (and hence strain rate) is an important parameter influencing the forces generated in blanking. A wide range of blanking velocities is studied for their influence on blanking load, as shown in Figure 4.6. Force required for blanking the material is 378N at 40mm/sec while it is

522N at 1616mm/sec. There is a 38% increase in load for the range of velocities studied. C51100 being an alloy of copper has a very high thermal conductivity compared to steel; therefore the heat generated during blanking is dissipated in the sheet quickly. This prevents temperature build up in the sheet. Hence, the blanking load is influenced more by strain rate than temperature in

C51100.

Influence of Velocity on Maximum Blanking Force 700

600

500

400

300

200 Maximum Blanking Force(N)Blanking Maximum

100 Case A - Punch contacts sheet @ 0.86mm above BDC Case C - Punch contacts sheet @ 5.45mm above BDC 0 0 200 400 600 800 1000 1200 1400 1600 1800 Punch velocity at the instant of contact with sheet (mm/sec)

Figure 4.6: Influence of blanking velocity on maximum blanking force

Interestingly, the reverse loading on the punch also increases with velocity, as seen in Figure 4.7.

The reverse loading is 13.5% of blanking load at 20mm/sec blanking velocity while it increases to

40% of blanking load at 808mm/sec. This is because the elastic forces stored in the punch during blanking are released more suddenly at higher speeds. This emphasizes that the punch and other tooling need to be designed to absorb the reverse loading and not transfer it to the members of the press. A damper on the punch head can be used to lower the reverse loading. A relationship between the unloading curve angle and stiffness of the press and die set is shown in

[Guo et al, 1998]. This can be used to lower the reverse loading on the tooling.

Maximum Reverse Loading on Punch 100

-100

-200

-300

Maximum ReverseLooading (N) Maximum -400 Case A - with stripper pinning Case B - without stripper pinning -500 0 100 200 300 400 500 600 700 800 900 1000 Punch velocity at the instant of contact with sheet (mm/sec)

Figure 4.7: Influence of blanking velocity on maximum reverse loading on punch

4.2.4.2. Influence of stripper plate pinning on blanking load

There is a very small difference in the punch load between Case A, where the stripper plate held the sheet material and Case B where it did not. Case A shows higher load because of either (i) experimental deviation when the tooling was taken out and put back in for the two cases, or (ii) when the stripper pressure pins the material in Case A, the sheet material is under compression and slightly strain hardened. Since the sheet material is more hardened in Case A than in Case

B, the blanking load increases in Case A.

Influence of Stripper Pinning on Maximum Blanking Force 600

500

400

300

200 Maximum Blanking Force(N)Blanking Maximum

100

Case A - Stripper plate pins material during blanking Case B - No stripper plate pinning during blanking 0 0 100 200 300 400 500 Speed (mm/sec)

Figure 4.8: Influence of stripper pinning on maximum blanking force

4.2.4.3. Influence of velocity on stripper plate vibration

The amplitude of vibration during stripper plate pinning is mostly influenced by the characteristics of the spring. This can be seen in Figure 4.9 where the maximum amplitude occurs at blanking velocity of ~ 933mm/sec. The amplitude of vibration during pinning was significantly lower than that during unpinning. Figure 4.10 shows that the amplitude of vibration during unpinning is largely dependent on the ram velocity. The amplitude of punch force at higher speeds reaches

1300N. These vibrations are experienced by the lateral face of the punch. For long slender punches, this is an issue that may be of concern. Hence, it is important that the vibrations of the stripper plate be reduced and the stripper-plate punch interaction which currently exists be minimized.

Maximum Force on Punch During Stripper Plate Pinning 400 Mean amplitude 300 deviation

200

100

-100

-200

-300

-400 Maximum Force on Punch During Stripper PlateMaximumStripper Pinning(N) Punch Force on During

-500 0 200 400 600 800 1000 1200 1400 1600 1800 Ram Velocity at the instant of stripper plate unpinning (mm/sec)

Figure 4.9: Influence of velocity on loading on the punch during stripper plate vibration during pinning

Figure 4.10: Influence of velocity on loading on the punch during stripper plate vibration during unpinning

4.2.4.4. Analysis of the blanking curve and correlating with part edge quality

Blanking load curve can be divided into various segments, as shown in Figure 4.11 for Case A.

The initial portion of the curve (a-b) corresponds to the elastic deflection of the punch, dies and sheet. The next portion of the curve (b-c) corresponds to rollover and shear, with the initial linear section of 0.02mm corresponding to rollover and the remaining to shear. The shape of the curve in the shear section is determined by mechanical properties of the sheet material. If the sheet material strain hardens, the curve tends to remain almost flat. This is because the decrease in the thickness of sheet material to be cut is compensated by the strain hardening in the region of

55 sheet deformation. The shear is followed by fracture of the material, shown by region ‘c-d’ in the curve. This sudden drop in force leads to reverse loading of the punch as seen in region ‘d-e’.

Experimentally obtained blanking load at 60 SPM 400 Case A Case B 350 Case C

300

250

200

150 Punch Force (N) PunchForce 100 Initial elastic Reverse region Rollover and shear loading 50 Fracture 0 a b c d e

-50 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2 Punch Stroke (mm)

Figure 4.11: (left) Blanking load curve at 60 SPM; (right) blanked edge profile

4.2.4.5. Influence of velocity on blanked edge quality

Figure 4.12 shows the blanked edge for Case A where a flat punch is used. The blanked edge quality does not vary much with speed. The sheared to fractured zones maintain the same ratio at different speeds. Overall, it can be concluded that the blanked edge does not change very much with speed for speeds from 20mm/sec to 1616mm/sec during blanking of C5100 material.

Figure 4.12: Blanked edge obtained for Case A at different blanking velocities (left to right) (i) 187mm/sec (ii) 500mm/sec (iii) 875mm/sec and (iv) 1250mm/sec

Figure 4.13 shows the very small (10%) increase in sheared edge length obtained at different blanking speeds from 40mm/sec to 1250mm/sec. Although blanking velocity does affect the length of the sheared zone, it is not very significant in the range of velocity investigated.

Influence of% blanking shear at velocity different on shearedspeeds edge length

% shear % ) as % of sheet thickness as sheet of % )

Zs 20

0 Shear zone ( Shear 40 500 1250 60 800 2400 Blankingspeed velocity (SPM) (mm/sec)

Figure 4.13: Influence of blanking velocity on shear length

4.3. FE simulations of blanking

Although simulating the blanking process using FE methods is fairly common, accuracy of the model depends on the input data such as flow stress curve of the material, thermal properties, damage calculation used to predict crack to name a few. Blanking is a very high deformation process where fracture has to occur for the operation to be complete. The stresses and strains in the deformation region of the sheet reach very high levels.

4.3.1. Simulation setup

The sheet properties and tooling parameters used in FE simulations of blanking are explained in this section.

4.3.1.1. Sheet Material Properties

C51100 (phosphor-bronze alloy) of 0.2mm thickness is the sheet material used in this study.

Material property was obtained using viscous pressure bulge tests (conducted at Center for

Precision Forming, shown in Figure 4.14) and tensile test data [Ling, 1996]. Sheet material undergoes high strain deformations. At these high strains, the stress-strain relationship is unknown. Hence, a constant value is maintained during extrapolation.

Figure 4.14: Flow Stress data of C51100 used in simulations (experimental – obtained using Viscous Pressure Bulge test)

Fracture criteria and damage value

The initiation point of fracture during FEM simulation is indicated by the ductile fracture criterion applied to the simulation. Equation 4-1 shows the general form used to express most fracture criteria to estimate a damage value C. Depending on the fracture criteria, the function ‘f’ is modeled as a function of maximum principal stress, effective stress or other deformation parameters.

 f (deformation)d  C  (Equation 4-1) 0 where  is the effective strain and C is the damage value

In modeling of ductile fracture initiation, the deformation history is required. In most cases, ductile fracture criteria that incorporate the stress and strain history are applied for the prediction of fracture initiation.

If the integral on the left-hand side of Equation 4-1 reaches the critical damage value (CDV) during the simulation, ductile fracture is initiated in the simulation. In the blanking process, this initiation determines the height of the shear and fracture zone. A comparison between experimental and simulation results of the blanked edge is used to determine the critical damage value for each material (the length of the shear and fracture zones in simulation has to be equal to the length of the shear and fracture zones in experiments).

Various ductile fracture criteria have been proposed. Studies conducted by [Weidenmann et al,

2009] show that Rice & Tracy damage criterion is best suited for simulating fracture in blanking.

Hence, this criterion is used in the current study. The critical damage value is determined by comparing blanked edge (length of shear and fracture zone) obtained experimentally with simulated results.

Rice & Tracy criterion shown in Equation 4-2 used in the current study

  (Equation 4-2) exp( m )d  C  0 

With α = 0.3 (Rice & Tracy constant)

C = 17 (critical damage value)

4.3.1.2. Geometry Setup

The schematic of the blanking process is shown in Figure 4.15. The notations shown in Figure

4.15 are used in Table 4.3 to give the dimensions of the tool and sheet and parameters used in simulations.

Figure 4.15: Schematic of the blanking process

Table 4.3 Parameters and values used in simulations

Parameter Value

Sheet material and thickness ‘t’ C51100, 0.2 mm thick

Punch diameter ‘dp’ 1.5mm

Punch corner radius ‘rp’ 0.0127mm (0.005in)

Stripper diameter ‘db’ 1.506mm

Die diameter ‘dd’ 1.526mm

Die corner radius ‘rd’ 0.0127mm (0.005in)

Stripper pressure (‘fb’/area) 4 MPa (580 psi)

Punch and die material Tungsten carbide (WC)

Punch velocity Function of stroke (Figure 4.16)

Coefficient of friction 0.1

Additional simulation parameters and boundary conditions are shown in Table 4.4. Figure 4.16 shows the approximation done on the blanking velocity used in simulations. Since it is a

61 mechanical press, the velocity at other SPMs can be calculated from if the velocity at any one

SPM is known.

Table 4.4: Simulation Parameters and Boundary Conditions

Object Simulation Parameters & Boundary Condition

- Rigid body Punch - If stresses are to be calculated, elastic material with E = 650GPa (WC)

- Plastic body Sheet material - Axisymmetric

Stripper - Force corresponding to 4 MPa pressure

Ram Velocity 4 3.75 3.5 3.25 3 2.75 2.5 2.25 2 1.75 1.5

Distance above BDC (mm) above Distance BDC 1.25 1 0.75 0.5 0.25 Aida Crank Motion 0 0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500

Velocity (mm / second)

Figure 4.16: (top) Velocity of the ram as a function of stroke (sinusoidal waveform); (bottom) approximated to a straight line in the 0.2mm length of the contact with the sheet

The mesh in the deformation zone of the sheet is very important to predict (i) stresses on the punch and (ii) fracture initiation and propagation in the sheet. For a sheet 0.2mm thick, there are approximately 100 elements along the thickness of the sheet in the deformation zone. Figure 4.17 shows the mesh distribution in the sheet in the deformation zone.

Figure 4.17: (top) Simulation setup for blanking; (bottom) enlarged view of the sheet mesh at the blanking interface

Figure 4.18 shows the different stages on punch penetration into the sheet. The punch first touches the sheet after which it causes rollover and shear in the sheet. After some shear, fracture begins in the sheet. At the end of fracture, the blank separates from the slug. This process, which happens in reality during blanking, is replicated in simulations.

Figure 4.18: Different stages of punch penetration into the sheet

4.3.2. Comparison of experimental and simulated results

The zones of the blanked edge and load-stroke curve obtained from simulations were compared with those obtained from the experiments for the same parameters like punch-die clearance, punch and die dimensions, stripper pressure and blank material and thickness.

4.3.2.1. Comparison of blanked edge zones to validate damage model used for the sheet material

Figure 4.19 shows how the different zones are measured in the simulated and experimental blanked edge zones. The lengths of the different zones are compared for experimental and simulated cases in Table 4.5. The simulation results are in fair agreement with experimental measurements. The variation in the shear and fracture zones may be due to the fracture model used in the simulation. The critical damage value would decide the instant of crack initiation in the sheet, which would in turn affect the shear and fracture zone lengths in the sheet. Hence, the

65 fracture (damage) model and critical damage value used in the simulations seem to be accurate enough to be used in future simulations.

Roll over zone (Zr)

Shear zone (Zr) Zs

Fracture zone (Zf)

Burr zone (Z ) b

Figure 4.19: Comparison of experimental and simulated blanked part edge (Zb – burr, Zf – fracture zone, Zs – shear zone, Zr – rollover zone) for blanking 0.2mm thick C51100 material, 1.5mm diameter hole

Table 4.5: Comparison of experimental and simulated blanked part edge

Zone Simulation (µm) Experiment* (µm)

Roll over (Zr) 22 10

Shear(Z ) 105 100 s Fracture(Z ) 71 84 f

* Shear and fracture were measured from a picture taken using digital microscope. Roll over zone estimated from the picture.

4.3.2.2. Comparison of load-stroke curve to validate the flow stress model used for sheet material

A comparison between the simulated and experimental load-stroke curve is shown in Figure 4.20.

Although there is less than 10% difference in the maximum blanking load obtained in simulations and experiments, the shape of the curves are very different from each other. The experimental

66 load-stroke curve reaches a maximum at about 0.05mm stroke and remains almost the same until the beginning of fracture after which there is a sudden drop in load. However, in simulations, the load reaches a maximum and drops visibly with stroke before fracture begins. During fracture, a sudden drop in load in observed. The curves differ from each other because of the flow stress curve used for the sheet material in simulations. The strains in the sheet reach much higher than

1, while the flow stress curve has data only until strains <1. The stresses at higher strains are extrapolated and the accuracy of the extrapolation is not known. Therefore, there is a discrepancy in the load-stroke curve obtained with simulations and experiments. This difference can be resolved by remodeling the flow stress curve of the sheet material for higher strains. The next section describes a method to obtain flow stress data of materials at higher strains and different strain rates.

Comparison of blanking load-stroke curve obtained using simulations and experiments 450 400 350 300 Simulated 250 Experimental 200

150 Blanking Load Load (N) Blanking 100 50 0 0 0.05 0.1 0.15 0.2 Penetration into the sheet (mm)

Figure 4.20: Comparison of blanking load-stroke curve obtained from experiments and FE simulations

4.4. Development of high strain rate material model

Flow stress data must be available for up to high strains (up to 4) at high strain rates (up to 104 s−1) and temperatures (200 °C and even higher) to accurately simulate the blanking process. This data is difficult if not impossible to obtain reliably with conventional tensile and compression tests.

Often the high-speed Hopkinson’s bar compression tests are used but these tests are: (a) difficult and expensive to conduct and (b) can provide data only for limited strain (about 1.5). Torsion tests are also available although they are not as widely accepted and used as the tensile and compression tests. There are no test standards for the torsion test to obtain flow stress data of materials. In addition, the specimens need to be in the form of solid or hollow cylinders.

[Sartkulvanich et al, 2004] and others have proposed a method to obtain the Johnson-Cook parameters using metal cutting (machining) using a hybrid method of combining experimental and

FEA.

Since the above methods either provide insufficient data to conduct FEA of blanking or involve a cumbersome test procedure, a method to obtain flow stress data of materials using blanking tests is proposed in this study. In addition, it is always more accurate to obtain the flow stress data through tests that emulate (and have similar stress state as) the actual deformation process

The following approach is used in obtaining the flow stress data using blanking.

 Blanking test and FEA are conducted at low speeds (20mm/sec cutting speed) corresponding

to strain rate of ~500sec-1 in the deformation zone.

 Flow stress data for the material at low strain rate is obtained using the method explained in

4.4.1.1

 Using the flow stress data obtained for low strain rate and using the method explained in

4.4.1.2, flow stress curves for higher strain rates are obtained.

This approach also considers the influence of temperature on the flow stress data of materials.

The following assumptions/approximations are made in this methodology.

 The strain distribution in the sheet does not vary with blanking speed (strain rate).

 Strain rate is approximated to be a constant throughout the blanking process and changes

only with blanking speed since the variation with speed is much greater than within the

process.

 Friction between the sheet and punch/die is found to have a negligible effect on the blanking

force. There is a 2% difference in blanking force for CoF values of 0 and 0.2. Hence, a CoF of

0.1 is used in the FE simulations.

4.4.1. Description of the methodology

4.4.1.1. Method to obtain stress-strain data at low strain rate (~100 sec-1)

The flow chart in Figure 4.21 shows the proposed methodology to obtain flow stress curves for quasi-static blanking condition. Temperature effects are not taken into account since there is only a very small increase in temperature in quasi-static blanking. Flow stress curve obtained from tensile test or biaxial test and extrapolated to higher strains are used in this step in order to conduct the FEA of blanking. (An example is shown in section 4.4.2 to show how the methodology is applied to a specific material.)

The overall idea is to compare the force-stroke curve from experiments and simulations at small increments of stroke. At every step, if the forces obtained from experiment and simulations do not match, the flow stress curve is modified for the particular average effective strain corresponding to that point in stroke.

Figure 4.21: Flow chart to obtain flow stress curve using blanking tests for low strain rate

Figure 4.22: Points taken in the deformation zone to average stress, strain and temperature values

4.4.1.2. Method to obtain stress-strain data at high strain rate (in the order of 103 – 105 sec-1)

High strain rate flow stress data can be calculated by using the combination of flow stress data obtained from Figure 4.21 and experimental blanking force curves at higher speeds. A simple procedure for determining strain rate dependent flow stress data for high strains is shown in

Figure 4.23. The effect of temperature is not considered in this methodology because the maximum blanking force was found to occur at the very beginning of stroke, at which time, the temperature effects are small enough to be neglected. This also helps in separating the strain rate effect from temperature effect on the flow stress of the material. (An example showing the force-stroke curve at higher strain rates for C51100 is shown in section 4.4.4.)

Here, the flow stress curve is scaled by a factor equal to the ratio of the maximum blanking force at high strain rate to the maximum blanking force at quasi-static conditions.

Figure 4.23: Flow chart to develop flow stress curve using blanking tests for higher strain rate

4.4.2. Evaluation of the methodology for C51100 (using simulation and experimental results)

The methodology explained in section 4.4.1 is applied to obtain the flow stress curve of C51100 at both quasi-static and high strain rate conditions. Experiments were conducted corresponding to

71 strain rates of ~ 102 - 104sec-1 in the deformation zone. Corresponding FEA was also conducted in order to compare the two.

4.4.2.1. Experimental setup and procedure

Blanking experiments were conducted using a 300 kN high speed mechanical press. The details of the tooling used in this study are shown in Table 4.1 and the schematic is shown in Figure 4.1.

The stripper plate was spring loaded. Punch force was measured using a piezoelectric sensor

(221B03 from PCB Piezotronics) and ram displacement was measured using a Keyence LK-

H057 laser displacement sensor. The data acquisition rate ranged from 50 – 250 KHz depending on the press speed.

The experimental load-stroke curve is obtained for blanking speed of 20mm/sec to 1600mm/sec.

A minimum of 20 readings were recorded for each velocity tested. Further details of the experimental setup, conditions and results are presented in 4.2. The blanking velocities from experiments used in the methodology to obtain the flow stress curves are shown in Figure 4.27.

4.4.3. Flow stress curve for C5100 at quasi-static conditions

The method explained in the flow chart of Figure 4.21 is used to obtain the flow stress curve at low strain rate. The following sections give the details of the procedure.

4.4.3.1. Simulation set up

In this study, DEFORM 2D is the software used to simulate the blanking process. Since the process takes place in a very short period of time (in the order of a few milliseconds), heat transfer to the dies is negligible and hence not considered in the model. Temperature rise and heat conduction within the sheet material are taken into account. FEA of the blanking process corresponding to the experimental parameters shown in Table 4.1 and blanking velocity of

20mm/sec is conducted. Additional parameters used in the simulation are shown in Table 4.6.

Table 4.6: Parameters used in simulations – methodology to obtain high strain rate flow stress data using blanking tests

Simulation Parameters Value

Punch

- Material model Elastic with Young’s Modulus = 650 GPa (WC)

Sheet

- Material model Plastic (flow stress shown in Figure 4.24)

- Thermal conductivity 84 W/mK

Fracture

- Damage model Adapted Rice and Tracy

- Critical damage value 3*

* Critical damage value was chosen by matching the shear and fracture lengths of the blanked edge from experiments and simulations.

Flow Stress Curve for C51100 used in FE simulations 1000

900

800

700

600

500

400 True Stress True

300

200

100 experimentally determined extrapolated 0 0 0.5 1 1.5 2 2.5 3 True Strain

Figure 4.24: Flow stress curve for C51100 obtained using bulge test (extrapolated by fitting Hollomon’s equation  842.4 0.1355) and used in FE simulations

Figure 4.25 shows the strain, strain rate and temperature distribution in the sheet obtained from

the simulations. Strain value averages at about 2, while the strain rate averages at about 375/s.

There is only 20°C increase in temperature in the sheet, the effect of which on flow stress can be

neglected.

Punch Punch Punch Punch Punch Stripper plate Punch Punch Stripper plate Stripper plate Punch Stripper plate Stripper plate Stripper plate Stripper plate Stripper plate DieDie DieDie DieDie DieDie

4.4.3.2. Comparison of experimental and simulated force curves

Experimental and simulated force-stroke curves are compared and plotted in Figure 4.26. It can

be seen that the two curves almost overlay on each other implying that the flow stress

extrapolation used in the simulations is a good approximation of the material stress-strain

Experimentally obtained blanking force at 60 SPM 400 Case A 350 Case B Case C 300

250

200

150 relationship. However, this need not be the case for other materials. Hence, another extrapolation

100 Punch Force (N) PunchForce of the same flow stress curve is tested using this methodology in Appendix A where the 50 extrapolated flow0 stress curve is used in simulations and then the flow stress curve is ‘adjusted’ to

-50 match the experimental0 0.02 and0.04 simulated0.06 0.08 force0.1-strok0.12e curve.0.14 0.16 0.18 0.2 Punch Stroke (mm)

Comparison of experimental and simulated blanking force at 6020mm/sec SPM for C51100 400 Experimental 350 Case A Simulated 300

250

200

150 Punch Force (N) PunchForce 100

0 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2 Punch Stroke (mm)

Figure 4.26: Comparison of experimental and simulated force-stroke curves (at 20mm/sec)

4.4.4. Flow stress curve for C5100 at higher strain rates

Now that the flow stress curve for C51100 at quasi-static strain rate is available, it can be used to obtain the flow stress curve for higher strain rates using the methodology stated in Figure 4.21.

The flow stress curve of the material is scaled by a factor equal to the ratio of the maximum blanking force at higher strain rates to the maximum blanking force at strain rate of 375s-1. The maximum blanking force at different blanking speeds (obtained from experiments) is shown in

Figure 4.27.

The average strain rate in the deformation zone for blanking velocity of 20mm/sec is found to be

375 sec-1 from FE simulations in section 4.4.3.1. Since the average strain rate increases linearly

75 with speed with everything else being the same, strain rates are labeled for some of the velocities in Figure 4.27.

Influence of Velocity on Maximum Blanking Force 700

600

έ1610 3x104 sec-1 500

400 έ έ 1060 400 2x104 sec-1 7500 sec-1 300

200 Maximum Blanking Force(N)Blanking Maximum

100

Case C - Punch contacts sheet @ 5.45mm above BDC 0 0 200 400 600 800 1000 1200 1400 1600 1800 Speed (mm/sec)

Figure 4.27: Influence of blanking speed on maximum blanking force

Based on the maximum blanking force and strain rates, the factor by which the flow stress needs to be scaled is calculated in Table 4.7.

Table 4.7: Factor used to scale the flow stress curve for higher strain rate

Average Strain Rate in the Maximum blanking force (N) Ratio of maximum blanking

-1 deformation zone in s forces Fhigh strain rate : F375

(blanking speed in mm/sec)

375 (20) 378 1

7500 (400) 394 1.04

2 x 104 (1060) 447 1.18

3 x 104 (1600) 522 1.38

The flow stress curve obtained for quasi-static strain rate (showed in Figure 4.24) is scaled by the ratios obtained in Table 4.7 to obtain the flow stress curves at higher strain rates. The flow stress obtained for different strain rates is shown in Figure 4.28.

Strain Rate Dependent Flow Stress Curve for C51100 Obtained using the methodolody 1600

1400

1200

1000 True Stress (MPa) Stress True 800

strain rate 375 600 3 strain rate 7.5*10 4 strain rate 2*10

4 strain rate 3*10 400 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 True Strain

Figure 4.28: Flow stress curves for higher strain rates obtained using the methodology

4.4.5. Effect of temperature on flow stress of C51100

Since temperature dependent flow stress data for C51100 was not available in literature, the relationship between flow stress and temperature of pure copper was used from [Nemat-Nasser et al, 1998]. An example of the temperature dependent true stress-strain curves of OFHC copper is shown in Figure 4.29. True stress-strain curves for other temperatures can be obtained from

[Nemat-Nasser et al, 1998].

The ratio of true stress at 296K to the true stress at higher temperature is calculated for different values of strains and averaged (the red vertical lines in Figure 4.29 show the strains for which the stress values were averaged). This gives the factor by the true stress is scaled down due to the effect of temperature. From preliminary simulations, it was found that the maximum temperature in the sheet is ~300°C. Hence, the temperature dependent stress-strain curves were obtained for up to 300°C.

Figure 4.29: Temperature dependent true stress-strain curve for OFHC copper [Nemat-Nasser et al, 1998]

Table 4.8: Factor used to scale the flow stress curve for higher temperatures

Temperature (T) Ratio - flow stress T / flow stress22°C

22 1

222 0.8048

322 0.6272

For each strain rate, temperature dependent flow stress curves are calculated (similar to that shown in Figure 4.30).Hence, there is a total of 12 curves (4 strain rate dependent x 3 temperature dependent) are input as material property in DEFORM 2D for the simulations.

Temperature dependent flow stress data for C51100 at strain rate 375s-1 (obtained using the temperature-flow stress relation from 1200 OFHC)

1000

800 600 T = 22 C 400 Stress (MPa) Stress T = 222 C 200 T = 322 C 0 0 1 2 3 4 5 6

Strain

Figure 4.30: Flow stress curves for higher temperatures obtained using the ratios calculated for OFHC

4.4.6. Compare experimental and simulated force curves at high strain rate

Simulations are conducted at blanking speeds of 1060mm/sec and 1600mm/sec with the same simulations parameters as shown in Table 4.6 except the flow stress curve, which is obtained from sections 4.4.4 and 4.4.5. The force-stroke curve from experiments and simulations is compared in Figure 4.31. There is a good correlation between the maximum force predicted using

79 simulations and that obtained experimentally. Moreover, the shape of the load-stroke curve obtained using simulations is in good agreement with that obtained using experiments. This good correlation throughout the stroke of the curve shows that the flow stress data for higher strains and strain rates has been predicted with reasonable accuracy.

BlankingBlanking Force force at 1600 at 1060mm/sec SPM (1.06 m/sec) BlankingBlanking Force force at at2400 1600mm/sec SPM (1.6 m/sec) 600 600 experimental experimental simulated simulated 500 500

400 400

300 300

Force (N) Force (N) Force

200 200

100 100

0 0 0 0.05 0.1 0.15 0.2 0 0.05 0.1 0.15 0.2 Stroke (mm) Stroke (mm)

Figure 4.31: Comparison of experimentally obtained and simulated force-stroke curve for high strain rates (left) έ = 2 x 104 s-1, (right) έ = 3 x 104 s-1

4.5. Summary and conclusions

1. Blanking tests were conducted for different velocities ranging from 20mm/sec to

1600mm/sec.

a. There was a 38% increase in the force required to blank the part at 1600mm/sec

compared to 20mm/sec.

b. The reverse loading is 13.5% of blanking load at 20mm/sec blanking velocity while it

increases to 40% of blanking load at 808mm/sec.

80 c. The vibrations during stripper plate unpinning apply forces on the lateral areas of the

punch. This may lead to punch breakage. Depending on the area of cross-section and

length of the punch, this is an area that needs further investigation.

A methodology to obtain flow stress data of sheet materials at high strains and high strains

is proposed. The flow stress of C51100 material is obtained using this methodology.

CHAPTER 5: Influence of various parameters on blanked edge quality and

punch load/stress

There are various process parameters that can influence the quality of blanked edge, blanking load and stress on the punch. Punch-die clearance, punch and die corner radius, punch tip geometry, stripper pressure and friction between the punch are the factors that are studied in this section. The objective of this section is to identify the parameters that have the most influence on blanked edge quality and punch stress using FE simulations.

Table 5.1: Parameters Used in Simulations to study the influence of various parameters

Parameter Value

Sheet material / thickness C51100 / 0.2mm

Punch

Material / diameter WC / 1.5mm

Stripper

- diameter ‘db’ 1.506mm

- pressure (‘fb’/area) 4 MPa (580 psi)

Punch-die clearance 6.5% sheet thickness

Punch and die corner radius 0.0127mm

Punch velocity Function of stroke (Figure 4.16)

Coefficient of friction 0.1

A punch of diameter 1.5mm blanking C51100 material 0.2mm thick is used in the following simulations to study the influence of various parameters. The other parameters used in this study are listed in Table 5.1. These parameters correspond to the conditions for which the blanked edge quality and blanking load-stroke curve were compared for experimentally obtained and simulated results. This is the baseline simulation for which various parameters are varied in the following sections.

5.1. Effect of punch-die clearance on part quality and punch stress

Simulations were carried out with different punch-die clearances varying from 1% to 12% sheet thickness to investigate its effect on the punch and part edge. Punch-die clearance is one of the most important parameters that significantly affect the length of shear zone of the blanked part.

Figure 5.1 shows the effect of punch-die clearance on blanked edge quality. It can be seen from

Figure 5.1 that punch-die clearance of 1% - 3% gives a shear edge of 75% sheet thickness, while it drops to 55% at 5% clearance and remains a constant of about 50% up to 12% clearance.

Lower punch-die clearance is known to cause higher tool wear and hence an optimum value of clearance should be chosen based on the part quality requirement and tool wear allowance.

Figure 5.2 shows its effect on punch load. The punch load varies negligibly (3%) from 384 N at

1% clearance to 372 N at 12% clearance. This effect can be neglected.

Effect of punch-die clearance on part edge quality 180 160 140 120 100 80 60 40

Zone Length (µ) Length Zone 20 0 0 2 4 6 8 10 12 Clearance (% sheet thickness) Roll over zone Shear Zone Fracture zone

Figure 5.1: Effect of punch-die clearance on part edge quality (obtained using FE analysis) [sheet - 0.2mm thick, C51100 material]

Maximum Punch load at different punch-die clearances

386 384 382 380 378 376 374 372

Maximum PunchLoad (N) Maximum 370 0 2 4 6 8 10 12 14 Punch-die clearance (% of sheet thickness)

Figure 5.2: Maximum punch loads at different punch-die clearances (obtained using FE analysis) [sheet - 0.2mm thick C51100]

Figure 5.3 shows its effect on maximum punch stress. There is a decrease of punch stress with punch-die clearance until the clearance reaches 6.5% after which it increases with clearance.

There is a 25% decrease in the punch stress when the punch-die clearance increases from 1% to

6.5%. The small punch-die clearance causes severe shearing of the material, causing high strains in the sheet material which in turn leads to higher punch stress. As the punch-die clearance increases, the strain levels reached in the sheet decreases causing the punch stress to lower. There is an optimum range of punch-die clearance which causes the least punch stress. If the punch-die clearance is increased beyond this optimum range, the punch stress increases because the sheet material undergoes sever bending inducing bending stresses in the punch.

This optimum range of punch-die clearance is dependent on the sheet material and thickness and geometry of the punch.

Figure 5.3: Maximum Punch Stress at different punch-die clearances (obtained using FE analysis) [sheet - 0.2mm thick C51100]

5.2. Effect of corner radius on part quality and punch stress

The influence of punch corner radius on punch stress and part edge quality is studied. Three punch corner radii, 0.0127mm, 0.02mm and 0.03mm are used. There is a dramatic (50%) reduction in punch stress when 0.02mm / 0.03mm corner radii are used instead of 0.0127mm corner radius as seen in Figure 5.4. A very small increase in corner radius reduces the punch stress by half, which may in turn increase the punch life by many folds. However, a practical difficulty of achieving the exact corner radius during resharpenings is an issue that would need further investigation.

Figure 5.4: Influence of punch corner radius on maximum punch stress (obtained using FE analysis) [sheet - 0.2mm thick C51100]

The influence of punch corner radius on part edge quality is seen in Figure 5.5. There is a marginal change in the lengths of shear and fracture zones of ~7% which can be neglected.

However, the burr length increases considerably, about 80% when the punch corner radius increases from 0.0127mm to 0.03mm. This can be an issue since the allowable burr length in production generally adheres to very low numbers, which can be below 10µ in the case of

86 electronic components. For parts where the part specification does not call for tight burr tolerances, a higher punch corner radius can be useful in improving the tool life dramatically.

Figure 5.5: Influence of punch corner radius on part edge zones (obtained using FE analysis) [sheet - 0.2mm thick C51100]

5.3. Effect of punch tip geometry on punch load

Depending on the strength and thickness of the blanked material, the punch load during the blanking process can be estimated. If the punch cannot sustain such high loads or there is a possibility of occurrence of snap-thru forces, punch load has to be lowered. Load on the punch can be lowered by altering the punch tip geometry by having a shear angle or by having a conical punch. The advantages and disadvantages of different punch tip geometries are discussed in

Table 5.2. In this study, the shear angles used in conical, single shear and double shear is 10⁰.

Simulations were conducted in Deform-3D to obtain the punch forces while using the geometries shown in Figure 5.6. Part edge quality is not compared for these geometries since fracture cannot be visualized using Deform-3D. The shape of the deformed slug for various tip geometries can be seen in Figure 5.7.

Figure 5.6: Punch tip geometries studied for their effect on punch load

Table 5.2: Advantages and Disadvantages of different punch tip geometries

Punch End Advantages Disadvantages Shape

Flat • Easy to manufacture • Large punch force generated

• Lesser wear

Single shear • Lower punch force because shear • Unbalanced force on punch; locally instead of entire tool punch tends to deflect periphery • Reduce reverse tonnage

Double shear, • Lower punch force • Slug formed is deformed conical • No unbalanced forces on punch • Chipping can occur in conical

Flat conical

single shear double shear

Figure 5.7: Shape of the slug obtained using the different punch tip geometries (obtained using FE analysis) [sheet - 0.2mm thick C51100]

Figure 5.8 shows the loading on the punch while using different punch tip geometries. The two important characteristics to be observed are (i) the punch load is reduced to almost half when single shear is used and is reduced slightly when a double shear is used(ii) except for the flat punch, the punch load never rises immediately, but gradually increases with stroke which can greatly reduce shock and instant loading of the punch. It should also be noted that the shear angle has an influence on the punch load. A different shear angle for double shear can result in a reduced punch force.

Single shear gives the least load on the punch and can be used in applications where a high strength or thick gauge sheet material is cut. Since it is not symmetric, the punch has the possibility of deflecting depending on the punch geometry and load.

Punch Load Distribution for Different Punch Shapes 500 450 400 conical 350 300 double shear 250 200

PunchLoad(N) 150 single shear 100 50 flat 0 0 50 100 150 200 250 Stroke (penetration into sheet) (%) flat punch conical single shear double shear

Figure 5.8: Punch load during blanking using the different punch tip geometries (obtained using FE analysis) [sheet - 0.2mm thick C51100]

5.4. Effect of friction on punch and part

The effect of friction on part edge quality, punch load and punch stresses is studied. The coefficient of friction is generally in the range of about 0.1-0.3 in blanking and it can be observed from Figure 5.9 that there is not a significant change in the part edge quality in this range of CoF.

However, very low values of friction like 0.01 show a small increase in the shear edge quality.

The maximum punch load also does not change much (~ 3%) in the range of 0.1-0.3 CoF (Figure

5.10). The maximum punch stress is the lowest at 0.2 CoF at 950 MPa. There is not a significant difference in the maximum punch stress until the CoF is 0.3 after which an increase of about 38% is seen for CoF 0.5 (Figure 5.11). CoF of 0.5 is high for blanking, especially with lubrication. hence, it can concluded that under common operating conditions, a variation in CoF between 0.1 and 0.3 does not cause a significant difference in blanked edge, punch load and punch stress.

Effect of CoF on part edge zones

120

100 80 60 40

20 Edge Length (µm) Length Edge 0 0 0.1 0.2 0.3 0.4 0.5 0.6 CoF roll over shear fracture

Figure 5.9: Effect of coefficient of friction on part edge quality (obtained using FE analysis) [sheet - 0.2mm thick C51100]

Maximum Punch Load for different CoFs 400

395 390 385 380 375

Maximum punch load (N) load punch Maximum 370 365 0 0.1 0.2 0.3 0.4 0.5 0.6 CoF

Figure 5.10: Effect of coefficient of friction on maximum punch load (obtained using FE analysis) [sheet - 0.2mm thick C51100]

Maximum Punch Stress for different CoFs

1600 1400 1200 1000 800 600 400 200

Maximum punch stress (MPa) stress punch Maximum 0 0 0.1 0.2 0.3 0.4 0.5 0.6 CoF

Figure 5.11: Effect of coefficient of friction on maximum punch stress (obtained using FE analysis) [sheet - 0.2mm thick C51100]

5.5. Effect of stripper pressure on part edge quality

The stripper plate not only helps in stripping the sheet material off the punch but also applying compressive force on the sheet when it is being blanked. The sheet material is stressed under compression when it undergoes shear and is stressed under tension during fracture. The more the onset of tensile stress is delayed, the more shear is obtained on the blanked edge. In addition, the application of compressive stress also affects the rollover on the sheet. The relation between stripper pressure and part edge quality is studied in this section.

A higher compressive stress is induced in the sheet material by increasing the contact pressure between the sheet material and stripper plate. The contact pressure between the strip and stripper plate can be varied by varying either the stripper spring configuration or geometry of the stripper plate. For FE simulation purposes, the contact pressure between the stripper plate and strip is increased by increasing the force on the stripper plate. In reality, this can be brought about by design changes. The contact pressure is increased from 5MPa to 350 MPa (2/3rds of yield strength of the material) in FE analysis to understand the effect on part edge quality. Although

92 there is not a significant difference with small increase in contact pressure, higher contact pressures of ½ to 2/3rds of yield strength of the sheet material give better edge quality, as shown in Figure 5.12. Increasing the contact pressure to 350MPa reduces the roll-over zone to half and

15% increase in shear zone compared to contact pressure of 10 MPa or less.

Effect of contact pressure on part edge zones 140

120 100 80 60 40 Zone Length (µ) Length Zone 20 0 0 100 200 300 400 Contact Pressure (MPa) roll over Zr shear zone Zs fracture zone Zf

Figure 5.12: Effect of contact pressure on blanked edge quality (obtained using FE analysis) [sheet - 0.2mm thick C51100, punch diameter – 0.2mm]

The effect of contact pressure between the stripper plate and sheet material on blanked edge quality is verified experimentally. A hole of diameter 0.2mm is blanked using two contact pressures (i) corresponding to a low contact pressure of < 10 MPa and (ii) high contact pressure greater than ½ YS of sheet material. This is achieved by having small design changes incorporated in the tool design. The exact contact pressure could not be measured experimentally, but is approximately obtained using simulations and calculations for both cases.

Figure 5.13 shows the difference in the part edge quality obtained using two very different contact pressures. It can be seen that there is a very large difference in the quality of blanked edge

93 obtained in the two cases. As seen in the FE analysis, the rollover zone shows a significant decrease when a higher contact pressure between the stripper plate and sheet material is used.

In addition, there is a clean shear-fracture when a higher contact pressure is used.

Roll over Shear

Transition from shear to fracture Fracture Burr

Figure 5.13: Effect of contact pressure between stripper plate and sheet material on part edge quality (left) low contact pressure <10 MPa (right) high contact pressure > ½ YS of sheet material

Hence, the compressive stress can be induced in the sheet material by increasing the pressure applied by the stripper plate for applications that require a very high part edge quality.

5.6. Effect of punch misalignment on part edge quality

The punch is not rigidly fixed to the punch holder and can hence be misaligned during its course of travel. The punch is guided by the stripper plate. There is a small clearance between the inside wall of the stripper plate and outer surface of the punch. The punch can get misaligned in this small clearance. This study is conducted to determine the effect of this misalignment on the punch stress and blanked edge quality.

The maximum misalignment that can occur for a 1.5mm diameter punch with in a 1.506mm diameter stripper hole is 0.015⁰. This is illustrated in Figure 5.14. The length of the stripper plate in this case is taken to be 21mm.

1.506mm

Stripper plate α 21 mm Punch

Sheet

Figure 5.14: Misaligned punch (punch diameter – 1.5mm); maximum misalignment – α = 0.015⁰

Three conditions are simulated for plane strain condition. Figure 5.15 gives the schematic of the three cases. Figure 5.15 (a) represents the perfectly straight punch (b) represents the titled punch and (c) represents a straight misaligned punch.

Table 5.3 shows the difference in the blanked edge quality and maximum punch stress between ideal condition of plane strain and misaligned punch condition. The punch stress does not vary more than 3% for the different conditions studied. There is not much variation in the maximum punch stress which is of importance. The blanked edge quality also does not show a significant difference. There is less than 10% difference in the length of the shear and fracture zone and even less in the roll over zone. Hence, the misalignment of the punch does not affect the punch stress and part quality to an extent of concern.

(a)

A B

(b)

(c)

Figure 5.15: Schematic of (a) Plane strain condition with no misalignment (b) Tilted punch and (c) Vertical punch with unequal clearances on either side used in FE simulations

Table 5.3: Blanked edge quality and maximum punch stress determined with various misalignments of punch (simulations)

Max Punch

Condition Edge Roll over Shear zone Fracture zone stress during

(µ) (µ) (µ) blanking

(MPa)

(a) Plane strain with no A 33 98 68 795 punch misalignment B 29 98 68 795

(b) Tilted punch A 35 89 75 785

B 34 97 69 783

5.7. Summary and Conclusions

1. Design and process parameters are studied for their effect of blanked edge quality, punch

load and punch stress using FE analysis.

2. Punch-die clearance affects the punch stress and blanked edge quality. There is an optimum

range of punch-die clearance which gives lowest punch stress depending on the sheet

material and thickness. Punch-die clearance less than 5% gives a longer shear and smaller

roll-over after which it remains almost the same.

3. Punch corner radius affects the punch stress significantly. A change in the punch corner

radius from 0.0127mm to 0.02mm reduces the punch stress to half, but has negligible effect

on rollover, shear and fracture in the blanked edge. The burr length increases 50% when the

punch corner radius is 0.02mm and 80% when the punch corner radius is 0.03mm compared

to the burr length with 0.0127mm corner radius. If the burr length obtained using the larger

corner radii is acceptable, the punch life can be improved by using larger punch corner radii.

4. Punch tip geometry affects the punch load, with a single shear angle on the tip reducing the

maximum punch load to 50% of the load obtained with a flat punch. Having a shear angle can

(i) reduce the peak blanking load and (ii) apply the load on the punch gradually avoiding

shock.

5. CoF and punch misalignment have negligible effect on the punch stress and part edge quality

according to the results obtained using FE analysis.

CHAPTER 6: Relation between punch-die clearance, punch stress and part

geometry in blanking

There are several factors in the punch and die that affect punch and die failure in blanking including the contact pressure between punch and sheet, sliding velocity, surface quality of the punch, lubrication and friction conditions, tool material and coating. Of these factors, the ones that can be modified by design changes are contact pressure and may be sliding velocity. Contact stress on the punch can be reduced by selecting the optimum punch-die clearance and/or punch corner radii.

In the production environment, punches are sharpened by grinding the bottom surface of the punch by a few microns depending on the amount of wear. In this case, the punch corner radius is generally not controlled to have a certain dimension. From the standpoint of implementation in an industry environment, a study of the influence of punch-die clearance on stresses and wear may be more practically applicable. Hence, this study will focus on studying the relation between punch-die clearance, punch stress and punch geometry in blanking of different sheet materials.

Punches seldom have round cross-sections; instead they have corners and sharp edges or profiles. While the stresses are mostly uniform all around a round punch, the stresses for punches with other shapes are not. Stresses are higher in corners or other areas with high curvature. For example, in rectangular punch geometry, the corners of the rectangles are expected to wear before the straight edges. These localized stress concentrations lead to chipping and higher punch wear. This study is undertaken to understand the effect of punch-die clearance on punch stress and in turn, punch failure for different geometries. This would help in designing punch-die clearance to prevent premature wear around corners or sharp edges.

There are many studies like those by [Bell, 2006], [Hernàndez, 2006], [Hambli 2002], [Ship-Peng

Lo, 2007] which have looked into the effect of punch-die clearance on blanked edge quality and tool wear. However, all the above studies have only used a punch with a round cross-section for blanking in order to find the optimal clearance. The influence of geometry of the punch on wear has not been found in any of the previous studies in literature. [Högman, 2004] showed through experiments that increasing the radius around the corner of a rectangle in the punch eliminated chipping and excessive wear on the punch. However, the study did not go into details of determining how much would be the optimal punch-die clearance for different radii and also the effect on blanked part dimensions. Furthermore, the study was purely experimental and there was no established guideline or methodology to select an optimal punch-die clearance for other geometries.

Observation of worn punches and dies, such as in Figure 6.1 shows that wear and chipping initiates at locations on the punch which have a very sharp radius or sudden changes in geometry. Sheet deformation and punch-sheet interaction are the important parameters that change locally at different locations of the cutting surface (face-flank interface) for a given punch- sheet material combination. Hence, these parameters are studied for different punch geometries to understand their influence on punch/die stresses and hence punch/die wear.

Figure 6.1: Worn punch showing more wear along the radius than straight edges

The objective of this study is to improve the punch/die life by selecting the optimal punch-die clearance depending on the localized geometry of cut/punch to obtain a more uniform wear and reduce chipping. ‘Optimal’ punch-die clearance refers to the value of clearance that gives least stress on the punch while at the same time not deviating from the required part dimensions. The hypothesis used this study is that the variation in wear on a non-round punch/die is related to the variation in normal/contact stress in the punch/die. Hence, obtaining a more uniform stress distribution on the punch by varying the punch-die clearance along the perimeter of the shape blanked can give rise to a more uniform wear pattern. Hence, the wear along sharp radii can be reduced. This, in turn, can improve the punch/die life.

Terminologies: Punch corner radius – radius between the face and flank of the punch

Punch radius / punch geometry / geometry of cut – radius of a round punch/ geometry on the face of a flat punch / the shape that is blanked out

The following approach is taken to select the optimal punch-die clearance depending on the geometry of the blanked shape.

 The influence of punch geometry (different radii and straight edge), sheet material properties

and punch-die clearance on punch stress is studied using FEA.

 The relationship between punch radius and punch-die clearance to obtain the least stress on

punch given everything else to be a constant is identified for two different sheet materials and

thickness.

 The wear pattern on the punch is compared experimentally for the two cases (i) uniform

punch-die clearance and (ii) variable punch-die clearance.

6.1. Simulation Setup

The influence of four parameters on normal/contact stress on the punch is studied. They are (i) punch geometry (different radii and straight edge), (ii) sheet material strength (iii) sheet material thickness and (iv) punch-die clearance.

100

Numerical simulations are conducted using DEFORM 2D/3D to study the effect of various parameters on punch stress and blanked edge quality. The simulation parameters common to all the cases in the study are shown in Table 6.1. Parameters specific to each case are shown in the respective sections.

Table 6.1: Parameters used in FEA of blanking in studying the effect of various parameters on punch stress

Parameters Used in Simulations Value

Sheet Material

- material / flow stress model AISI 1010 / σ = 698.63ε0.2057

SS301 / σ = 1929.9ε0.0804

C51100 / σ = 842.42ε0.1355

- material model plastic

Punch / Die

- Material model Elastic

- Material AISI D2

- Corner radius 0.0127mm

- Clearance 8%

Coefficient of Friction 0.1 Shear Friction

Blank Holder Force 10 MPa

6.2. Effect of punch geometry

Any given complicated punch geometry can be broken down into straight edge and radii. Hence, in this study, blanking of a straight edge and round geometry with different radii are studied. The effect of punch geometry on normal stress and blanked edge quality are studied.

101

The dimension of radius in the pierced shape affects the stress in the punch. Different radii are used with different sheet materials to study the influence of radius on punch stress and also to study if it is consistent across materials. Table 6.2 shows the sheet materials and geometries studied in this section to understand the effect of punch geometry on punch stress and blanked edge quality.

Table 6.2: Simulation matrix used to study the effect of geometry of the pierced shape

Parameters Used in Simulations Value

Sheet Material - AISI 1010

- SS301

- C51100

Sheet Thickness 0.25mm

Geometries - Radii – (i) 0.15mm,(ii) 0.25mm, (iii) 0.5mm, (iv) 1mm

- Straight edge

Punch/die clearance 8%

6.2.1. Influence on contact stress on the punch

The contact pressure is measured along the most critical region of the punch, i.e along the punch corner radius (at the face-flank interface of the punch). The line AC along which it is measured is shown in Figure 6.2. The contact stress is measured at the position in stroke showing maximum stress. In all the cases tested, the punch penetrates about 0.25t – 0.3t into the sheet material where ‘t’ is the thickness of sheet material. This also corresponds approximately to the position in stroke showing maximum blanking load. Figure 6.3 shows the contact stress along the curvilinear length AC for different punch radii when AISI1010 sheet material, 0.25mm thick is used. Smaller the radius, higher the contact pressure on the punch. The highest contact pressure is between A

102 and B, closer to B for all punch geometries studied. It is also observed that when the radius of the punch is 0.6t, very high stresses are observed on the punch. But as the radius approaches 4t, the stress levels are similar to that on the straight edge of the punch (which is mathematically ∞).

Similarly, the influence of geometry on contact stress is studied for other materials too, i.e SS301 and C51100. The contact pressure curves for the two materials look very similar to that shown in

Figure 6.3 for AISI 1010.

ACL = 0mm

C = 0.05mm B = 0.02mm CL CL

Figure 6.2: Curvilinear length (CL) along which stresses are plotted on the punch

103

Influence of geometry on contact pressure (sheet - AISI 1010 0.25mm thick)

1800

1600

1400

1200

1000

800

600

400 Normal/Contact Pressure (MPa) Pressure Normal/Contact

200

0 0.005 0.015 B 0.025 0.035 0.045 C 0.055

Curvilinear length (mm)

axi 0.15mm R axi 0.25mm R axi 0.5mm R axi 1mm R straight edge

Figure 6.3: Effect of geometry on contact pressure – blanking AISI 1010 sheet, 0.25mm thick (when the punch has penetrated 0.25t into the sheet, corresponding to maximum stress)

The contact stresses along the curvilinear length are averaged and plotted against radius in

Figure 6.4 for AISI 1010 sheet material. There is an exponential decrease in stress with increase in radius up to 1mm, after which there is not a very significant change in stress.

In this section and the following sections, average contact stress corresponds to the average of the contact stresses along the curvilinear length AC of the punch.

104

Influence of Radius on Average Contact Stress on Punch (Sheet material - AISI 1010 / 0.25mm thick)

1250 1200 1150 1100 1050 1000 950

900 Contact stress (MPa) (MPa) stress Contact 850 800 750 straight edge 0 0.5 1 1.5 2 2.5 (radius = ∞)

Radius (mm)

Figure 6.4: Influence of radius on average contact stress on the punch (sheet material AISI 1010, 0.25mm thick)

In Figure 6.5, the influence of radius on contact stress for different materials is plotted. The normalized radius plotted along the x-axis is the radius/sheet thickness. Normalized stress plotted along the y-axis is contact stressradius/ contact stressstraight edge for every sheet material. Hence, normalized stressstraight edge = 1. Figure 6.5 shows that the stress on the punch during blanking a hole the radius of which is 0.6 times sheet thickness is ~50% higher than the stress during blanking of a straight edge. It can also be observed that the relation between contact stress and radius of the punch is independent of the sheet material (at least for the materials studied). The absolute values of the contact stress changes with material. However, the percentage increase in contact stress at different radii with respect to the contact stress for straight edge is the same for all materials investigated in this study.

105

Influence of Radius on Contact Stress on Punch 1.6 (for different materials) SS301 0.25mm thick 1.5

edge AISI 1010 0.25mm thick

1.4 straight

Stress C51100 - 0.25mm thick

1.3 stress

/ / 1.2 radius

Normalized Normalized 1.1 stress 1

0.9 straight edge 0 2 4 6 (radius8 = ∞) 10

Normalized Radius (radius/sheet thickness)

Figure 6.5: Influence of radius on average contact stress for different materials (at ~0.25t penetration, which gave maximum stress)

6.2.2. Influence on part edge quality

The geometry of the punch has a significant influence on the blanked edge zones, as shown in

Figure 6.6. The blanked edge zones are obtained through simulations for different geometries for

AISI 1010 sheet material. There is a steady increase of rollover and fracture and decrease in shear zone with increase in radius up to normalized radius of 4. The relation between shear zone and radius is very similar to the relation between contact stress on the punch and radius. The shear zone length decreases exponentially with increase in radius. This is because smaller the radius of the punch, higher the compressive stresses in the sheet along the deformation zone.

106

This high compressive stress delays the fracture and hence causes an increase in shear length.

This is also the reason for high contact stresses on the punch shown in section 6.2.1.

Effect of geometry on blanked edge zones (AISI 1010 - 0.25mm thick) 60

10 Zone Lengths (% of sheet thickness ) thickness sheet of (% Lengths Zone 0 0 2 4 6 8 Straight edge10 (radius = ∞) Normalized Radius (radius/ sheet thickness)

rollover shear fracture

Figure 6.6: Effect of geometry on blanked edge zones obtained using FEA (sheet material – AISI 1010 – 0.25mm thick)

Experimentally obtained blanked edge measurements for SS301, 0.25mm thick support the results obtained from simulations. Figure 6.7 shows the measured shear edge lengths for different radii and straight edge for SS201 material. The relation between shear edge measurements and geometry obtained from simulations and experiments (Figure 6.6 and Figure

6.7) correlate well with each other. Although they are two different sheet materials, the effect of geometry on the blanked edge is similar.

107

Effect of radius on shear zone length

10 Shear zone (% (% thickness)ofsheet Shearzone

0 straight edge 0 2 4 6 8 (radius10 = ∞)

Normalized radius (radius/sheet thickness)

Figure 6.7: Effect of geometry on blanked edge zones obtained using experiments (sheet material – SS301 – 0.25mm thick / only 3 experimental points were obtained)

(a) (b) (c)

Figure 6.8: Blanked edge cross section (a) straight edge (b) 0.25mm radius (c) 0.15mm radius (sheet material – SS301 – 0.25mm thick)

In addition, Figure 6.8 shows the blanked edge for the three radii plotted in Figure 6.7. The difference in the blanked edges for different radii is clearly visible. Figure 6.8(c) shows the sharp radius of 0.15mm and the straight edges that it connects. It can be clearly seen that the radii

108 alone has a large shear zone, but the straight edge has a smaller shear zone which is similar to that shown in Figure 6.8(a).

Hence, it can be shown both experimentally and through FEA that the geometry of the punch does affect the blanked edge zones.

6.3. Effect of material properties on punch stress

Strength and thickness of the sheet material may affect the deformation of the sheet, thereby influencing the stress on the punch. Hence, the effect of sheet material on punch stress is studied using three sheet materials.

6.3.1. Effect of sheet material (strength)

In the previous section, it was found that the normalized stress for different radii varies the same way for different materials. The influence of sheet material on the absolute contact stress of the punch for blanking of a straight edge is studied in this section.

Three materials, AISI 1010, SS301 and C51100 are studied for their effect of contact stress while blanking straight edge. The contact stress along the curvilinear length of the punch corner (AC) is studied for the three materials.

109

Influence of Material Strength on Contact Stress Flow Stress Curves of Sheet Materials Used in This Study 3000 2500 SS301 2500

2000

) )

2000 SS301 MPa

MPa 1500 (

1500 AISI 1010 Stress ( Stress C51100 1000 1000 True Stress Stress True C51100 AISI 1010 500

500 Average Contact Contact Average 0 0 0 500 1000 1500 2000 2500 0 0.5 1 1.5 2 2.5 3 3.5 True Strain Strength of Material at ε =0.9 (MPa)

Figure 6.9: Influence of material strength on average contact stress during blanking of straight edge) (at ~0.25t penetration, which gave maximum stress)

Average contact stress along the curvilinear length AC is plotted in Figure 6.9. The x-axis is plotted as the strength of material at ε = 0.9 since the average strain in the deformation region of the sheet material at 0.25t punch was found to be ~0.9. Figure 6.10 shows the flow stress data of the sheet materials used in this study and also indicates the stresses at strain ε = 0.9. A linear relation between the contact stress on the punch and strength of the material is seen in Figure

6.9, which tends to be intuitive.

110

Influence of Material Strength on Contact Stress Flow Stress Curves of Sheet Materials Used in This Study 3000 2500 SS301 2500

2000

) )

2000 SS301 MPa

MPa 1500 (

1500 AISI 1010 Stress ( Stress C51100 1000 1000 True Stress Stress True C51100 AISI 1010 500

500 Average Contact Contact Average 0 0 0 500 1000 1500 2000 2500 0 0.5 1 1.5 2 2.5 3 3.5 Strength of Material at ε =0.9 (MPa) True Strain

Figure 6.10: Flow stress data of the materials showing the stresses at ε=0.9 for different materials (these stress values are plotted along the x-axis in Figure 6.9)

6.3.2. Effect of sheet thickness

One would expect that the contact stress has a linear relationship with sheet thickness (similar to

blanking force having a linear relationship with sheet thickness).

Blanking simulations were conducted with C51100 sheet material with different thicknesses

varying from 0.2mm to 1.8mm. The contact stress on the punch is found to have a linear

relationship with thickness, as seen in Figure 6.11. This shows that the sheet thickness does not

alter the strain distribution and deformation in the sheet significantly in the range studied.

111

Effect of Sheet Thickness on Contact Pressure

1400 )

1300 MPa

1200 y = 251.91x + 883.32

Pressure Pressure 1100

1000

900

Average Contact Contact Average 800 0 0.5 1 1.5 2 Sheet Thickness (mm)

Figure 6.11: Effect of sheet thickness on average contact pressure during blanking C51100 material (at ~0.25t penetration, which gave maximum stress)

6.4. Effect of clearance on punch stress

Punch-die clearance is the only parameter among the ones studied here that can be varied in a real production environment to influence punch/die stress.

The effect of punch-die clearance on normal/contact stress is studied for blanking of a straight edge. DEFORM 3D was the software used for the simulations. Average contact stress of the nodes along the curvilinear length AC is plotted in Figure 6.12. When there is a very low clearance, the sheet material shears severely applying higher stress on the punch. As the clearance increases, the shearing decreases and bending of the sheet material increases. There is a range of clearance where an optimal balance between the two can be struck. As the clearance is further increased, the sheet material undergoes too much bending, applying bending stress on the punch. This explains the ‘V’ or ‘U’ shaped relationship between the punch-die clearance and contact stress on the punch. Although the actual optimal punch-die clearance may

112 vary with the geometry blanked and the sheet material and thickness, the relationship between punch-die clearance and contact stress is assumed to be similar for all cases studied.

Effect of Punch/Die Clearance on Contact Stress for Straight Edge Condition (AISI 1010 - 0.8mm thick) 1200 1150 1100 1050 1000 950 900

Average Contact Stress (MPa) Stress Contact Average 850 800 4 5 6 7 8 9 10 11 12 Clearance %

Figure 6.12: Effect of punch-die clearance on average contact stress on the punch (sheet material – AISI 1010 – 0.8mm thick)

The V-shaped curve seen in Figure 6.12 is very similar to the curve obtained by [Hogman, 2004] in which the relationship between tool wear and punch-die clearance was studied, shown in

Figure 6.13. This further shows that contact stress obtained from simulations is a good indicator of tool wear and that a correlation can be drawn between contact stress and tool wear.

113

Figure 6.13: Relationship between tool wear and punch-die clearance obtained experimentally when blanking Docol 1400 DP, 1mm thick by [Hogman, 2004]

6.5. Punch-die clearance to obtain least punch stresses for different geometries

FE simulations were conducted in DEFORM 3D to optimize the punch-die clearance for each geometry such that minimum punch stresses are generated during the blanking process.

Blanking simulations were conducted for only 30% of stroke since the maximum punch stress

(which is of interest to us) occurs within the initial 30% of stroke. Since fracture of the sheet does not occur in the first 30% of punch penetration into the sheet, damage criterion and critical damage value are not important in this study. The parameters used in the simulations in this case study are shown in Table 7.3.

114

Table 6.3: Parameters Used in FE Simulations to determine the optimal punch-die clearance for different geometries (sheet material AISI 1010, 0.8128mm thick)

Parameter Shape Value

Sheet

material AISI 1010

thickness 0.8128mm

Punch & Die

Material ,Coating AISI D2, TiAlN

Material model Elastic (E = 200GPa)

Punch and Die cutting edge radius 0.0127mm

CoF 0.1 shear friction

Straight edge 5,7,9,11

1.27mm radius (on a 5,7,9,11

round punch)* Punch-die clearance (as % of sheet 0.5mm radius (on the 5,12,18 thickness) corners of an oblong)*

0.127mm radius (corner 5,18,25

of rectangle)*

* commonly found dimensions and parameters in the progressive die studied in section 7.2

The FE models used in simulations are shown in Figure 6.14. One-eighth model (45°) is used in the case of round. Half symmetry is used in the case of straight edge. Quarter model is used in simulating the corner of rectangle and oblong shapes.

115

Punch Punch Stripper plate

Stripper plate

Sheet Sheet

Die (insert) Die (insert)

(a) (b)

Punch Punch

Sheet Sheet Die (insert) Die (insert)

Figure 6.14: FE models used in the simulations (from top left clockwise) (a) round, (b) straight edge, (c) corner of rectangle, (d) oblong

Effective stress on the punch and die is used as the critical parameter to select the best suited clearance for a given geometry. The maximum stress on the punch and die is compared for each of the cases. This maximum stress on punch generally occurs at the instant when the blanking load reaches its peak.

6.5.1. Straight Edge

The influence of clearance on the stresses on the bottom surface of the punch is shown in Figure

6.15. The maximum stress on the punch is seen at the face-flank interface, corresponding to the curvilinear length AC mentioned in section 6.2 . Since it is not very clear from Figure 6.15 which

116 of the clearances gives the least stress, the stress in the face-flank interface region is averaged to find the average contact stress for different clearances. The average contact stress is shown in

Figure 6.16. 7% clearance shows the least contact stress among the clearances studied. A clearance between 7% and 9% will give the least stress on the punch.

4.6% 7% 9% 11% Contact Stress

(MPa)

Bottom face of the punch shown in the picture

Figure 6.15: Effect of punch-die clearance on contact stress during blanking of a straight edge (sheet material – AISI1010 0.8mm thick)

117

Effect of Punch/Die Clearance on Contact Stress for Straight Edge Condition (AISI 1010 - 0.8mm thick) 1200 1150 1100 1050 1000 950 900

Average Contact Stress (MPa) Stress Contact Average 850 800 4 5 6 7 8 9 10 11 12 Clearance %

Figure 6.16: Effect of punch-die clearance on average contact stress on the punch (sheet material – AISI 1010 – 0.8mm thick)

6.5.2. Round

The influence of clearance on the stresses on the bottom surface of the punch is shown in Figure

6.17. Since it is not very clear from Figure 6.17 which of the clearances gives the least stress, the stress in the face-flank interface region is averaged to find the average contact stress for different clearances. The average contact stress is shown in Figure 6.18.

118

5% 7% 9% 11% Contact Stress (MPa)

Figure 6.17: Effect of punch-die clearance on contact stress during blanking of 1.27mm radius hole (sheet material – AISI1010 0.8mm thick)

9% punch-die clearance gives the least contact stress on the punch for blanking a 1.27mm radius hole in AISI1010 0.8mm thick sheet material, as seen in Figure 6.18.

Effect of Punch-Die Clearance on contact stress for 1.27mm radius diameter punch (AISI 1010 - 0.8mm thick) 1200

1000

800

600

400

200 Average Contact Stress (MPa) Stress Contact Average 0 0% 2% 4% 6% 8% 10% 12%

Clearance (%)

Figure 6.18: Effect of clearance on average contact stress on the punch for 1.27mm radius hole (sheet material – AISI1010 0.8mm thick)

119

6.5.3. Oblong

Figure 6.19 shows the influence of clearance on the stresses plotted on the bottom surface of the oblong punch. It can be seen that both 12% and 18% show a reduction in stress when compared to 5% clearance. However, there is not a significant difference in stress between 12% and 18% clearance. Hence, 12% clearance could be used in the experiments since a higher clearance could cause slug retention problems.

Contact stress (MPa)

punch 5%

5% die

punch 12%

7.8% 12% die

18% punch

7.8% 12% die

Figure 6.19: Effect of clearance on stress (on the bottom surface of the punch) during blanking of an oblong shaped geometry (sheet material – AISI1010 0.8mm thick)

6.5.4. Corner of rectangle

Figure 6.20 shows the effect of clearance on the stresses in the bottom surface of the punch for the corner of a rectangle. Around the sharp corner of the rectangle (which has a radius of

0.127mm, refer to Table 6.2), a decrease in the contact stress with increase in clearance can be observed. Clearances of 18% and above were tried based on the results of section 6.5.3, where up to 18% clearance was simulated. Again, an increase in clearance shows decrease in contact

120 stress. 25% punch-die clearance showing slightly lower contact stress distribution compared to

18%. However, 18% would be used in the experiments since the quality of the blanked edge

(which cannot be predicted accurately by DEFORM 3D) is not known at such high clearances.

enlarged Contact Stress

5% clearance 18% clearance 25% clearance (MPa)

Figure 6.20: Effect of clearance on stress (on the bottom surface of the punch) during blanking of a rectangle shaped geometry

6.5.5. Variable punch-die clearances for different geometries

Based on the results obtained from FE simulations, geometry-specific punch-die clearance is defined in Table 6.4.

121

Table 6.4: Geometry-specific variable punch-die clearance for blanking sheet material AISI 1010, 0.8128mm thick

Geometry Selected Clearance (Cv)

Straight edge ~ 8%

1.27mm radius (on a round punch) ~ 8%

0.5mm radius (on the corners of an ~ 12.5%

oblong)

0.127mm radius (corner of rectangle) ~ 18%

6.6. Empirical relation to determine variable clearance as a function of radius of curvature,

material strength and thickness

In section 6.5, optimal punch-die clearance for blanking different geometries in 0.8128mm thick

AISI 1010 sheet material was selected based on the contact stress distribution obtained using

FEA. Similarly, optimal punch-die clearances for blanking different geometries in SS301 sheet material, 0.25mm thick are selected using the same procedure. The optimal clearance selected for both the materials (with different thicknesses) is plotted in Figure 6.21. In addition to the optimal clearance for the two materials, another series (which is the optimal clearance for AISI

1010 material multiplied by a factor equal to clearanceSS301 / clearanceAISI 1010 at 0.6 normalized radius) is plotted. Since this curve overlaps very well on the SS 301 curve, it can be concluded that the radius - optimal punch-die clearance follows a definite relationship. Furthermore, the curves in Figure 6.21 are very similar to the curve plotted in Figure 6.5 (which shows the influence of radius on contact stress). There is a negative exponential relationship between optimal punch-die clearance and normalized radius until the radius of punch is about 1.5 - 2 times the sheet thickness, after which the optimal clearance remains a constant.

122

Optimal Clearance For Different Radii (Based On Simulations) 25%

AISI 1010 - 0.8128mm thick 20% SS 301 - 0.25mm thick AISI 1010 - multiplied by a factor* 15%

10% optimal clearance optimal

0% 0 0.5 1 1.5 2 2.5 Straight3 edge (r3.5 = ∞) 4 normalized radius (radius/sheet thickness)

Figure 6.21: Optimal clearance selected for different geometries for different sheet materials and thickness

*factor is equal to clearanceSS301 / clearanceAISI 1010 at 0.6 normalized radius

Hence, it may be concluded that the optimal punch-die clearance for blanking straight edge depends on the sheet material and thickness. The optimal punch-die clearance for the different radii depends only on the dimension of the radius (once the clearance for the straight edge is determined) and can be calculated using the curve/relationship shown in Figure 6.21.

6.7. Summary and Conclusions

The influence of various parameters on punch stress is studied. It was found that

I. Sheet material and thickness have a linear relationship with punch stress.

II. The radius of the punch has a negative exponential relationship with punch stress up to

radius = 4t where ‘t’ is the thickness of sheet material. When radius > 4t, the stress on

punch is almost the same as that for a straight edge.

123

III. Punch-die clearance vs. punch stress has a ‘V’ or ‘U’ shaped curve for blanking of

different radii and straight edge.

IV. Based on the geometry of the punch, the optimum punch-die clearance for different

sheet materials and thickness is obtained.

V. It is found that the relation between optimum punch-die clearance for least stress and

geometry is the same for different sheet materials and thickness.

VI. Hence, a geometry dependent variable clearance can be designed based on the

relationship established if the optimum clearance for blanking a straight edge is known.

124

CHAPTER 7: Case Studies showing the effect of variable punch-die

clearance on tool wear

This chapter presents two case studies / applications in which the performance of the punch using uniform clearance and geometry-dependent variable clearance along the perimeter of the punch are compared. The results obtained in Chapter 6 are implemented and proved experimentally in this chapter.

7.1. Influence of variable punch-die clearance on a rectangular punch

In this section, a comparison is made between experimentally observed punch wear and simulated punch stress. Using the experimental data and results are presented in [Högman

2004], FEA was conducted using the experimental conditions to determine if a correlation could be drawn between experimentally obtained punch failure and simulated punch stress. Blanking tests were conducted by Högman with two square punches with different corner radii as shown in

Figure 7.1. The different corner radii give different punch-die clearances. These conditions are used in Finite Element Analysis (FEA) and the punch stress distribution for the two conditions is obtained. The results obtained using experiments and simulations are then compared. The idea is to be able to relate failure of the punch to the stresses generated in the punch.

7.1.1. Experimental Setup [Högman 2004]

[Högman 2004] conducted experiments to determine the influence of punch-die clearance on punch failure. Since sharp corners are the first locations to fail in a punch, punch-die clearance was varied in the corner. To increase the clearance, the corner radius of the punch was increased. A schematic of the two different geometries is given in Figure 7.1.

125

Figure 7.1: Schematic top view of punch and die for a) uniform clearance and b) higher clearance [Högman 2004]

In this study, the work piece is 1mm thick Docol 800DP. Two corner radii, 0.2mm and 0.5mm were used in the experiments. A constant punch-die clearance of 10% is maintained in the case of 0.2mm radii. A punch-die clearance of 10% is maintained along the straight edge but increases to a maximum value of 22% in the corner in the case of 0.5mm punch corner radius (Figure 7.2).

The tool material used was Vanadis 4 with a hardness of 60 HRC. The dimensions for different punches are given in Table 7.1.

22%

Figure 7.2: Schematic top view of the punch showing the clearance % for (a) uniform clearance and (b) larger clearance at corner

126

Table 7.1: Tool geometries used by [Högman 2004]

Tool Clearance Corner Overall Maximum

along radius dimensions clearance at the

straight corner

edge

Die 0.2 mm 12.2 mmx 12.2 mm

Punch 0.2mm radius 10 % 0.2 mm 12 mm x 12 mm 10%

Punch 0.5mm radius 10 % 0.5 mm 12 mm x 12 mm 22%

7.1.2. Experimental Results [Högman, 2004]

The tests with 0.2mm punch corner radius were repeated three times. The corner radius was chipped out in all three tests, after 12000, 40000 and 45000 strokes respectively. The punch with a corner radius of 0.5 mm lasted 200000 strokes without chipping.

Figure 7.3 indicates the view of the punch used to show the simulation and the experimental results. The arrow shows the viewing direction that was used for the pictures of the experimental results as well as for the simulation results.

Comparison of the punch corners at the end of the experiments is shown in Figure 7.4.

Figure 7.3: Schematic of the view of the punch used in results

127

Table 7.2: Parameters used in simulation (using Högman’s experimental data)

Parameter Value

Sheet material - Docol 800 DP

- Flow Stress Data [Söderberg 2006] (Figure 7.5)

- Plastic

Sheet thickness 1 mm

Punch material - Vanadis 4 hardened to 60 HRC

- Elastic with E=225 GPa (Assab Tool Steels) Symmetry Quarter of Punch simulated

Punch-Die Clearance Refer Table 7.1

Punch Geometry Refer Table 7.1

Die Geometry Refer Table 7.1 Table 7.1 Punch rectangular dimensions 12x12 mm Table 7.1 Punch and Die corner radii 0.05 mm (assumed)

Friction 0.1 Shear friction

Heat transfer Isothermal

Blankholder Pressure 8.5 MPa

128

7.1.3. Simulation Setup

With the punch and die geometries used by [Högman 2004], FE simulations of blanking are conducted to obtain stress distribution on the punch. Table 7.2 gives a summary of the parameters used in simulations. Since the punch and die corner radii was not specified in

[Högman 2004], a value of 0.05 mm was assumed as the corner radius for the punch and die.

Another effect in blanking of high strength steels is the temperature generated in the material by the deformation. As a blanking speed is not known, the real strain rates in the material are not known. In summary, it can be said that only an approximate flow stress for the material Docol

800DP is available.

Figure 7.5: Yield data for Docol 800DP [Söderberg 2006]

7.1.4. FEA Results

Non-uniform stress on rectangular punches could be shown by a die stress analysis of the punch.

The stress on the straight sides is much lower than the stress in the corner region.

It was assumed that the highest stresses in the punch appear at the instant/step of maximum blanking load. Therefore a die stress analysis was performed for the step having maximum

129 blanking load. In blanking of rectangular parts, stress in the punch are much higher in the corners of the punch which can be related to significantly more wear in the corners of the punch.

The maximum effective stress during blanking occurred in the corner of the rectangle, as expected. The maximum effective stress in the case of 0.2mm corner radius is 2270 MPa while that in the case of 0.5mm corner radius is 2010 MPa. This is a reduction in the maximum stress by 8%, while the maximum clearance was increased from 10% to 22%. In order to validate the stresses obtained by the simulation, the stresses were compared to the ultimate compressive strength of the tool material. The tool material used in the experiments is Vanadis 4 with a hardness of 60 HRC. Since material data for this material is not available, a similar material

Vanadis 6 was used for comparison. The ultimate compressive strength (UCS) for Vanadium 6 of hardness 60 HRC is 2290 MPa (Uddeholm 2006). This comparison with the UCS of the material can also suggest that although the stress reduces only 8% while using a larger corner radius, it makes a significant difference because the stress levels are very close to the UCS of the material.

Figure 7.6: Stress distribution at the corner of the rectangular punch Left: radius 0.2 mm (uniform 10% clearance); Right: radius 0.5 mm (larger 20% clearance at the corner)

Normal pressure on the face of the punch is another important parameter influencing punch failure. The normal pressure on the punch face is evaluated for the two geometries. The

130 simulation results show high normal pressures in the corner. The maximum normal pressure for

0.2mm corner radius is 2890 MPa, while this value was reduced to 2630 MPa for the punch with

0.5 mm corner radius. This equals a pressure reduction of 9%, which is about the same as the reduction of the effective stresses. The maximum normal pressure on the punch occurs in the corner of the rectangular punch. This is also the region of the punch, where chipping occurs.

Decreasing this value can possibly help to prevent the tool from chipping.

Radius on Punch 0.2 mm Radius on Punch 0.5 mm Max. normal pressure: 2890 MPa Max. normal pressure: 2630 MPa

7.1.5. Comparison of experimental and FE results

Comparing the experimental and simulated results for 0.2mm corner radius (from Figure 7.4 and

Figure 7.6), the tool has chipped 0.4mm in the experiments while the maximum punch stress is

2270 MPa for 0.47mm in simulations. Increasing the punch corner radius to 0.5mm leads to no

131 chipping in experiments and the punch stress is 2010 MPa in simulations. The stresses were compared to the ultimate compressive strength of the tool material. The tool material used in the experiments is Vanadis 4 with a hardness of 60 HRC. Since material data for this material is not available, a similar material Vanadis 6 was used for comparison. The UCS for Vanadium 6 of hardness 60 HRC is 2290 MPa (Uddeholm 2006). If the UCS of the tool material is known, a more accurate correlation could have been drawn.

Based on the experimental and simulated results, it can be concluded simulated punch stress is correlated to punch failure in experiments.

7.2. Influence of variable punch-die clearance on complex geometries

A case study is conducted to evaluate the influence of variable punch/die clearance on the punch and die life. A die with several stations used to progressively blank a nickel coated AISI 1010 steel sheet, 0.8128mm thick in a production environment is taken as an example for the case study. The punches and die inserts are made of powder metallurgy (PM) steel. Punches are coated with TiAlN coating. When using a uniform punch-die clearance, the punches and die inserts last for 126,000 parts before the punch coating wears thereby deteriorating both the blanked edge quality (increasing burr) and punch/die tip geometry. It was observed that the punch coating wears away at the corners first before progressing to straight edges as shown in

Figure 7.8, which further indicates that the variable clearance in blanking non-round parts maybe a good approach to alleviate the issue of non-uniform punch wear.

132

Figure 7.8: (right) Wear starts around the corners; (left) and then progresses to the straight edges of the punch used in production

The punch shapes that are used in this study are shown in Figure 7.9.

Figure 7.9: Punch shapes used in the study

The following approach is used to optimize the clearance for this case study:

1. Simplify / break down the punch shapes to geometries that take significantly lesser

computational time for FE simulations (straight edge and different radii).

133

2. Simulate the blanking process for all the identified geometries using different punch-die

clearances. Find the clearance that gives minimum punch stress for each of the

geometries.

3. Incorporate the clearances determined from 2 in the respective die*.

4. Conduct tryouts in production to compare the punch wear in the two cases (i) uniform

clearance and (ii) variable punch-die clearance.

*dimensions on the punch will not be altered since it is a piercing operation; dimensions on the final part can be maintained this way.

Based on the blanked shapes identified in Figure 7.9, four simplified geometries were identified to select the optimal punch-die clearance. These geometries along with the clearance used in the original design are shown in Table 7.3.

Table 7.3: Geometries used in FE simulation to determine punch stress

Geometry Dimension Original Clearance (Co)

Straight edge Length of the straight edge ~5% does not influence required clearance Round R 1.27mm ~3%

Oblong 0.5mm ~5%

1.27mm Corner for rectangle 0.127mm ~5%

134

7.2.1. Selecting variable punch-die clearances for different geometries

FE simulations were conducted for AISI 1010, 0.8128mm thick sheet material to select the geometry dependent variable clearance for the simplified geometries shown in Table 7.3. The details of the simulations are shown in section 6.5. In addition, section 6.6 also shows the geometry dependent punch-die clearance in Figure 6.21.

Based on the results shown in Figure 6.21, geometry-specific punch-die clearances for the different simplified shapes are defined in Table 7.4.

Table 7.4: Geometry-specific variable punch-die clearance

Geometry Original Clearance (Co) Selected Clearance (Cv)

Straight edge ~5% ~ 8%

Round (1.27mm radius) ~3% ~ 8%

Radius in oblong (0.5mm radius) ~5% ~ 12.5%

Corner of rectangle (0.127mm radius) ~5% ~ 18%

The dies corresponding to all the punches shown in Figure 7.9 were redesigned to implement variable punch-die clearance, Cv. Some examples are shown in Figure 7.10.

135

Punch Die with original clearance of constant 5%

Die with variable clearance

punch Die – original clearance 5% Die – variable clearance

12.5% clearance at radius of oblong

18% clearance at corners

8% clearance along straight edges

Figure 7.10: Examples showing the variable punch-die clearance design

136

7.2.2. Experimental Procedure

Experiments were conducted with two clearance designs (i) original clearance corresponding to a uniform punch-die clearance Co and (ii) variable punch-die clearance Cv corresponding to Table

7.4. The following procedure was used to conduct the experiments and evaluate the results.

 Since the experiments were conducted under production conditions, the blanked part had to

meet tolerance requirements specified for that product. One of the tolerance requirements

was a maximum allowable burr height of 5µ. Based on experience with the production die, it

was known that the uniform clearance of 5% can blank 126000 parts before the coating on

the punch shows excessive wear and part shows burr > 5µ. Hence, tests with uniform

clearance were run until this limit of 126000 parts was met.

 The variable-clearance tests were to be run until the punch coating showed wear and/or burr

height reached the maximum acceptable level of 5µ.

 Pictures of the worn punches from both cases were taken to compare the performance of the

two different clearance conditions.

 Since it is a challenge to obtain accurate and reliable burr height measurements, it was not

used as a measure to evaluate the performance of the different clearances. However, it was

made sure that the burr height does not exceed 5µ.

7.2.3. Experimental Results

The punches with variable clearance were run for 350,000 hits. The blanked edge at 350,000 hits still met all the tolerances specifications for the part. The burr height also remained below 5µ. The punches did not see excessive wear on them either. However, since the production order was completed, further parts were not run using the punches. Moreover, the punches with variable clearance already showed significant improvements over the uniform clearance since it was able run almost three times longer.

137

Regular tooling Variable clearance Sharp after 126 K hits after 350K hits

Figure 7.11: Comparison of punch wear between regular clearance (after 126000 hits) and variable clearance (after 350000 hits); also shown are pictures of the sharp punch

Pictures of worn punches are used as the parameter to evaluate the effect of variable clearance on wear. Figure 7.11 shows the wear on three punches using regular clearance and variable clearance. It can be seen that although the variable clearance tooling ran almost three times longer than the uniform clearance tooling, the punches have more uniform and significantly less wear than the uniform clearance punches. This result verifies the hypothesis that variable punch-die clearance depending on the geometry of the blanked shape gives a more uniform wear

138 on the punch. The results also prove that correlating punch stresses to punch wear is also a good approach to select the optimum punch/die clearance.

139

CHAPTER 8: Summary and Conclusions

8.1. Summary

The geometries investigated in this study include a) straight edge b) round c) oblong d) corner of rectangle and are shown in Table 7.3.

The objective of this study was to improve tool life and part edge quality in high speed blanking.

The objective was met by following the approach given below.

1. Forces on the punch during the entire blanking cycle were recorded for different blanking

velocities from 20mm/sec to 1600mm/sec using a round punch. Influence of blanking

velocity on the different forces that are generated in during the blanking cycle is studied.

2. A combination of blanking simulations and experiments were used to obtain high strain

and strain rate dependent flow stress data of sheet materials. Using this methodology,

the flow stress curve for C51100 material was obtained at high strains and strain rates.

3. FE simulations were conducted to study the effect of various parameters such as punch-

die clearance, punch tip geometry, stripper pressure, punch corner radius and coefficient

of friction on punch stress, part edge quality and punch load on a round punch.

4. The effect of punch geometry on normal stress on the punch was studied for different

sheet materials and thicknesses on four geometries, a) straight edge b) round c) oblong

and d) corner of rectangle.

5. Geometry-dependent (straight edge and different radii) optimal punch-die clearance was

designed using FE analysis, for a specific case study.

140

6. The performance of geometry-dependent variable punch-die clearance was compared

with the performance of uniform punch-die clearance experimentally.

8.2. Conclusions

The following conclusions were obtained from this study.

1. The punch force required for blanking C51100 sheet material using a round punch

increased 38% when the blanking speed increased from 20mm/sec to 1600mm/sec. The

stripper springs cause large vibrations at high blanking speeds which exerts force of

amplitude over 1000N on the lateral area of the punch, which may be or is of concern for

long slender punches.

2. Using experimental force-stroke curve at different blanking speeds and FE simulations, a

methodology was developed to obtain flow stress data of materials at high strains and

strain rates. The methodology was verified by applying it to C51100 material and a round

punch.

3. Of the various parameters that were studied for their influence on part edge quality using

FEA, punch-die clearance and stripper pressure were found to have an influence on part

edge quality. Simulations were conducted on C51100 sheet material, 0.2mm thick with a

round punch.

a. Punch-die clearance <4% gives a shear zone of 80% while punch-die clearance

>6.5% gives a shear zone of 60%.

b. Stripper pressure ~ 2/3 yield strength of sheet material reduces the roll-over

length to 50% of the length obtained using the widely used stripper pressure of

<10 MPa.

4. Of the various parameters that were investigated for their influence on punch stress using

FEA, punch-die clearance and punch corner radii were found to have the most influence

on punch stress. Simulations were conducted on C51100 sheet material, 0.2mm thick

with a round punch.

141

a. Punch stress first decreases and then increases with an increase in punch-die

clearance in the range studied (5% - 20%)

b. Punch corner radius of 0.02mm gives 50% of the punch stress obtained with

0.0127mm punch corner radius. Increasing the punch corner radius can help in

improving the tool life many folds, if the burr length with the higher punch corner

radii is within acceptable limits in production conditions.

5. The punch tip geometry is found to be the most significant factor in lowering punch loads

in FE simulations conducted using C51100 sheet material and a round punch. The single

shear geometry (end surface of the punch at an angle) can reduce the punch load by

50% compared to flat geometry on the tip. Deflection and punch breakage are factors

that should be taken into account while designing single shear punch tips.

6. The geometry of the punch (shape of cut) affects the localized punch stress at the face-

flank interface. In punching of a rectangular shape, the punch with radius 0.6t (at the

corner of the rectangle in plan view) gives 50% higher punch stress compared to a punch

with radius 4t, where t is the thickness of the material. In addition, the stress on a round

punch of radius 4t is the same as the stress on the punch with a straight edge.

7. Punch stress increases linearly with increasing sheet thickness or sheet material

strength.

8. A relationship between punch shape (straight edge and different radii) and optimal

punch-die clearance is obtained (in Figure 6.21). Geometry dependent variable punch-die

clearance is developed to achieve more uniform wear on the punch thereby reducing

localized excessive punch wear and chipping.

9. Geometry-dependent variable punch-die clearance gives three times more life than

uniform punch-die clearance experimentally.

142

8.3. Research contributions

The following research contributions came out of this study.

1. The study provides an understanding of the punch, stripper plate and sheet interaction at

high speed blanking. Areas that need attention to reduce vibrations and improve

robustness of tooling are identified.

2. A new methodology to obtain flow stress data of materials at high strains and strain rates

using blanking tests is developed. The data is useful not only in high deformation metal

forming operations but also in other application that require data at high strains and

different strain rates like FE modeling for crash and impact analyses.

3. Stripper pressure (when it is as much as 2/3rds of Yield Strength of sheet material) is

identified to reduce the rollover in the blanked edge, which is a very important part

specification in certain applications.

4. Guidelines for selecting geometry dependent punch-die clearance are provided in this

study. The guidelines can be applied to improve punch life across all applications and

sectors in the industry that use blanking as a manufacturing process. This method to

improve punch and die life comes at no additional cost except selecting clearances

during the design stage of the tooling.

8.4. Future work

The following areas are identified for future work of this study.

1. Areas that need attention to reduce vibrations and improve robustness of tooling are

identified. Methods to reduce the vibration and interaction between stripper plate and punch

need to be developed.

2. A higher stripper-plate – sheet material contact pressure is found to give a better edge

quality. Designs to locally achieve high contact pressure between the sheet and stripper plate

need to be developed.

143

3. A larger punch corner radius is found to reduce punch stress in this study. The punch corner

can have a specific profile instead of a simple radius. The effect of punch corner profile on

punch life and wear needs to be studied.

4. Very slender punches (with dimensions as small as 0.1mm radius) are used in blanking sheet

material 0.2mm thick. The punch design including the design of punch head, body and tip

and punch holder designs needs to be developed for better dynamic stability of the punch

during high speed blanking.

5. The effect of punch end shape, conical, single flat inclined, double flat inclined, etc. on

blanked edge quality and punch life for different sheet materials needs to be investigated.

144

REFERENCES

[Archard, 1953] Archard, J., “Contact and Rubbing of Flat Surfaces”, Journal of

Applied Physics, Vol. 24, 1953, pp. 981-988

[Bao & Weirzbicki, 2004] A Comparative Study on Various Ductile Crack Formation

Criteria. Journal of Manufacturing Technology. Volume 126,

Issue 3, 314-325

[Bay, 2010] Bay, N., Azushima, A., Groche, P., Ishibashi, I., Merklein, M.,

Morishita, M., Nakamura, T., Schmid, S., & Yoshida, M. (2010).

Environmentally benign tribo-systems for metal forming. CIRP

Annals - Manufacturing Technology, 59(2), 760-780

[Bayer, 1993] Bayer, R. G. (1993). A general model for sliding wear in electrical

contacts. Wear, 162–164, 913-918.

[Behrens et al, 2010] Behrens, B.-A., Bouguecha, A., Czora, M., Krimm, R., Matthias,

T., & Salfeld, V. Consideration of machine properties in FE

analysis of sheet metal forming processes. 2nd

INTERNATIONAL CONFERENCE Process Machine

Interactions, June 2010, Vancouver.

[Behrens, 2005] Behrens, B. -., & Schaefer, F. (2005). Prediction of wear in hot

forging tools by means of finite-element-analysis. Journal of

Materials Processing Technology, 167(2–3), 309-315.

[Bell, 2006] Bell,T. Böhler Uddeholm Autosteel Partnership Presentation,

2006

145

[Billur, 2009] Billur,E., Kardes,N., Altan,T. (2009). Tool Wear in Stamping of

Advanced High Strength Steels (AHSS) – A Review. CPF 2.3-

09-02. Center for Precision Forming, The Ohio State University,

Internal Report.

[Bing & Wallbank, 2008] Bing, G. J. A., & Wallbank, J. (2008). The effect of using a

sprung stripper in sheet metal cutting. Journal of Materials

Processing Technology, 200(1-3), 176-184.

[Breitling et al, 1996] Breitling, J., Wallace, D., & Altan, T. (1996). Investigations of

different loading conditions in a high speed mechanical press.

Journal of Materials Processing Technology, 59(1-2), 18-23.

[Choi et al, 2009] Choi, C., Groseclose, A., & Altan, T. (2012). Estimation of plastic

deformation and abrasive wear in warm forging dies. Journal of

Materials Processing Technology, 212(8), 1742-1752.

[Crucible Industries] http://www.crucibleservice.com

[Dayton Progress] www.daytonprogress.com/pdf/form993.pdf

[DiRuggiero 2000] DiRuggiero,C. (2000). Advanced Ceramics Excel in High-Speed

Metalforming Tools. Metalforming Magazine, February 2000, 48-

[Goijaerts et al, 2001] Goijaerts, A. M., Govaert, L. E., & Baaijens, F. P. T. (2001).

Evaluation of ductile fracture models for different metals in

blanking. Journal of Materials Processing Technology, 110(3),

312-323.

[Gou et al, 1998] Guo, B., Chen, W. M., & Wang, Z. R. (1998). Analysis of

blanking vibration with consideration of the break-through state.

Journal of Materials Processing Technology, 75(1-3), 117-121.

146

[Groche et al, 2007] Groche, P., Großmann, K., Hofmann, T., & Wiemer, H. (2007).

Advanced experimental and numerical methods for the analysis

of the dynamic forming press behavior. Production Engineering,

1(3)

[Grünbaum et al, 1996] Influence of high cutting speeds on the quality of blanked parts.

Report No. ERC/NSM - S-96-19, Center for Precision Forming,

Columbus, OH, USA

[Hambli 2002] Prediction of burr height formation in blanking processes using

neural network. International Journal of Mechanical Sciences,

44(10), 2089-2102.

[Hernández et al, 2006] Hernández, J. J., Franco, P., Estrems, M., & Faura, F. (2006).

Modelling and experimental analysis of the effects of tool wear

on form errors in stainless steel blanking. Journal of Materials

Processing Technology, 180(1–3), 143-150.

[Hirsch et al, 2011] Hirsch,M., Demmel,P.,, Golle,R., and Hoffmann,H. (2011) Light

Metal in High-Speed Stamping Tools. Key Engineering Materials,

473, 259-266.

[Husson et al, 2008] Husson, C., Correia, J. P. M., Daridon, L., & Ahzi, S. (2008).

Finite elements simulations of thin copper sheets blanking: Study

of blanking parameters on sheared edge quality. Journal of

Materials Processing Technology, 199(1-3), 74-83.

[Janoss, 2008] Selecting, Using Tool Coatings to Stamp AHSS, The Fabricator

July 2008(2008)

[Jimma et al, 1990] Jimma, T., & Sekine, F. (1990). On high speed precision

blanking of IC lead-frames using a progressive die. Journal of

Materials Processing Technology, 22(3), 291-305.

147

[Kang et al, 1999] Kang, J. H., Park, I. W., Jae, J. S., & Kang, S. S. (1999). A study

on a die wear model considering thermal softening: (I)

construction of the wear model. Journal of Materials Processing

Technology, 96(1–3), 53-58.

[Karelove et al, 2007] Influence of the Edge Conditions on the Hole Expansion

Property of Dual-Phase and Complex-Phase Steels by A.

Karelova, C. Krempaszky, E. Werner, T. Hebesberger, and A.

Pichler

[Kim et al 2006] Kim, H., H. Palaniswamy & T. Altan. Investigation of Tribological

Conditions in Forming Uncoated and Galvanized Advanced/Ultra

High Strength Steels, CPF - 2.3/06/01(2006)

[Klassen 2011] Klaasen, H., Kübarsepp, J., Tšinjan, A., & Sergejev, F. (2011).

Performance of carbide composites in cyclic loading wear

conditions. Wear, 271(5–6), 837-841.

[Luo, 1999] Luo,S.Y (1999). Effect of the geometry and the surface treatment

of punching tools on the tool life and wear conditions in the

piercing of thick steel plate. Journal of Materials Processing

Technology, 88(1–3), 122-133.

[Makich et al, 2008] Makich.H., Carpentier.L., Monteil,G., Roizard,X., Chambert,J.,

Picart,P.(2008). Metrology of the burr amount - correlation with

blanking operation parameters (blanked material – wear of the

punch). International Journal of Material Forming April 2008,

Volume 1, Issue 1 Supplement, pp 1243-1246

[Miles, 2004] Miles, R. (2004). Combating Snapthrough. Metal Forming

Magazine, March,2004.

[Misumi Tech Central] http://www.misumi-techcentral.com/tt/en/

148

[Mori et al, 2010] Mori, K., Abe, Y., & Suzui, Y. (2010). Improvement of stretch

flangeability of ultra high strength steel sheet by smoothing of

sheared edge. Journal of Materials Processing Technology,

210(4), 653-659.

[Nemat-Nasser et al, 1998] Nemat-Nasser,S.,Li,Y.(1998). Flow stress of F.C.C polycrystals

with application to OFHC Cu. Acta Mater,46, 2, 565-577.

[Painter et al, 1995] Painter, B., Shivpuri, R., & Altan, T. (1996). Prediction of die

wear during hot-extrusion of engine valves. Journal of Materials

Processing Technology, 59(1–2), 132-143.

[Picas, 2010] Picas,I., Hernández,R., Casellas,D., Valls,I. (2010). Strategies to

increase the tool performance in punching operations of UHSS.

IDDRG 2010. 325-334

[Podgornik et al 2006] Podgornik, B., S. Hogmark & O. Sandberg. Proper coating

selection for improved galling performance of forming tool steel,

Wear 261, 15-21 (2006).

[Rhee, 1970] Rhee, S. K. (1970). Wear equation for polymers sliding against

metal surfaces. Wear, 16(6), 431-445.

[Rooij and Schipper 2001] Rooij, M. B. d. & D.J. Schipper. Analysis of Material Transfer

From a Soft Workpiece to a Hard Tool: Part II---Experimental

Verification of the Proposed Lump Growth Model, Journal of

Tribology 123, 474-478 (2001).

[Sandberg, 2004] Sandberg.O., Johansson,B. (2004). Tool steels for blanking and

forming – new developments. Recent Advances in

Manufacturing & Use of Tools & Dies and Stamping of Steel

Sheets. Pp 53-70

149

[Sartkulvanich et al, 2004] Sartkulvanich, P., Koppka, F., & Altan, T. (2004). Determination

of flow stress for metal cutting simulation—a progress report.

Journal of Materials Processing Technology, 146(1), 61-71.

[Sartkulvanich et al, 2010] Sartkulvanich,P., Kroenauer,B., Golle,R., Konieczny,A. and

Altan,T. (2010). CIRP Annals -Manufacturing Technology, 59,

1,279-282

[Schuler Handbook, 1998] Schuler – Metal Forming Handbook, Springer, 1998

[Sekine, 2005] Sekine, F. (2005). Effective stripper-holding on high precision

blanking of electronic machine parts. Advanced Materials

Research, 6-8, 805-808.

[Ship-Peng Lo, 2007] Lo, S., Chang, D., & Lin, Y. (2007). Quality prediction model of

the sheet blanking process for thin phosphorous bronze. Journal

of Materials Processing Technology, 194(1–3), 126-133.

[Söderberg 2006] Söderberg, M.: Finite Element Simulation of Punching (Master's

Thesis): Department of Applied Physics and Mechanical

Engineering. Lulea University of Technology. Lulea (2006).

[Straffelini 2010] Straffelini, G., Bizzotto, G., & Zanon, V. (2010). Improving the

wear resistance of tools for stamping. Wear, 269(9–10), 693-

697.

[Uddeholm and SSAB 2008] Uddeholm & SSAB. Tooling Solutions For Advanced High

Strength Steels, Selection Guidelines, (2008)

[Wang, 2010] Wang,B. (2010). Finite element simulation of blanking

mechanism for precision progressive die. 2010 2nd International

Conference on Computer Engineering and Technology (ICCET),

V4, 82 -84.

150

[Wiedenmann et al, 2009] Wiedenmann,R., Sartkulvanich,P., & Altan,T.(2009). Finite

element analysis on the effect of sheared edge quality in

blanking upon hole expansion of advanced high strength steel.

IDDRG 2009 International Conference, June 2009.

[Yan 2012] Yan,W., Pereira,M.P., & Rolfe,B.F.(2012) Tool Wear in Sheet

Metal Stamping. Advanced Materials Research. Vol. 421, 750-

753

[Yu, 2011] Yu,S., and Feng,W. (2011) Experimental research on ductile

fracture criterion in metal forming. Frontiers of Mechanical

Engineering Volume 6, Number 3, 308-311

[Ling, 1996] Ling.Y., (1996) Uniaxial True Stress-Strain after Necking”, AMP

Journal of Technology, vol.5, 37-48

151

APPENDIX A : Alternate Flow Stress to Demonstrate the Methodology to

Obtain Flow Stress Data

Since the flow stress of C51100 used in section 4.4.3 resulted in a very good comparison between experimental and simulated force-stroke curve, the flow stress did not have to be

‘adjusted’ to match the experimental data. Hence, another extrapolation of the flow stress curve is used here in order to demonstrate the methodology.

Artifically extrapolated flow stress curve using straight line for C51100 1200

1000

800

600 bulge test power law extrapolation(same as Fig 400

True Stress (MPa) Stress True 4.24) artificially extrapolated 200

0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 True Strain

Figure A.1: Arbitrarily extrapolated flow stress curve using straight line for C51100 (Low global strain rate έ ~

10s-1)

152

Comparison of experimental and simulated blanking force 450 400

350 simulated using arbitrarily extrapolated flow stress 300 250 experimental 200 150

Blanking Force (N) ForceBlanking 100 50 0 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 Stroke (mm)

Figure A.2: Comparison of experimental and simulated blanking force (flow stress curve used in Figure A.1 is used in simulation)

At each stroke, the experimental force, average strain and average stress are obtained from simulations. The experimental force is used to find the ‘factor’. The average stress is then multiplied by the ‘factor’ to obtain the new stress for the respective strains. Table A.1 shows the simple calculations involved in obtaining the scaling factor and the modified flow stress. The modified flow stress is plotted in Figure A.3.

153

Table A.1: Calculating the factor to ‘modify’ the flow stress

Simulations Experiment Factor Modified Stress

(experimental force/ (Average Stress simulation Stroke Force (N) Average Strain Average Stress Stroke (mm) Force (N) simulated force) * Factor) 0 0.0 0 0 0.000 301.0 0.0000 0.0 0.002 288.8 0.0262 559 0.002 316.6 1.0961 612.7 0.004 298.6 0.0535 575 0.004 328.4 1.0996 632.3 0.006 307.6 0.0782 596.4282201 0.006 336.9 1.0954 653.3 0.008 316.3 0.1021 618.3746207 0.008 343.2 1.0850 671.0 0.010 324.0 0.1254 635.8402045 0.010 348.2 1.0746 683.3 0.012 332.2 0.1413 646.2088952 0.012 352.3 1.0605 685.3 0.014 337.4 0.1662 660.5781008 0.014 355.4 1.0534 695.8

Flow stress curve obtained using the methodology 1000 900

800

700 600 500 400 300 truestress (MPa) modified flow stress from methodology 200 bulge test extrapolated by fitting Holloman's equation 100 0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 true strain

Figure A.3: Flow Stress obtained using the methodology

Simulations are run with the modified flow stress shown in Figure A.3 and the force-stroke curves are compared with experimental results in Figure A.4. The simulated force-stroke curve correlates very well with the experimental curve. Hence, the methodology seems to be a good approach to obtain flow stress for materials up to high strains.

154

Comparison of experimental and simulated blanking force 450 simulated - using modified flow stress 400 experimental

350 300 250 200 150 Blanking Force (N) Force Blanking 100 50 0 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 stroke (mm)

Figure A.4: Comparison of experimental and simulated blanking force (flow stress curve used in Figure A.3 is used in simulation)

155