Predicting distortion in vehicle body assemblies Iain Masters & Xinmin Fan EuroPAM 5th October 2005 © 2004 IARC Outline PARD Programme Background Causes of Deformation Software Requirements Initial results with Steel assemblies Future work © 2005 IARC EuroPAM – 5th October 2005 2 PARD Programme Collaboration between Advantage West Midlands, Jaguar Land Rover, Warwick Manufacturing Group and the automotive supplier base Objective is to improve competitiveness of OEM’s and their suppliers, through partnership in the programme, to develop leading edge technologies and best practices © 2005 IARC EuroPAM – 5th October 2005 3 Background Dimensional Variation in body assemblies associated with: Component variation DVA type software assumes rigid parts Fixture / Jig Wear addressed by regular maintenance Joining method Mechanical, thermal distortions not addressed © 2005 IARC EuroPAM – 5th October 2005 4 Need for Simulation tool More accurate control of BiW assembly variation Optimising manufacturing cycle times and quality Reduced jig/fixture development time Introduction of ‘new’ materials Aid to decision making © 2005 IARC EuroPAM – 5th October 2005 5 Joining Methods Resistance Spot Weld (RSW) Self Piercing Rivet © 2005 IARC EuroPAM – 5th October 2005 6 Observed Distortions with RSW and SPR in Al Top-Hat Assemblies Aluminium Top-hat / Top-hat assemblies showing deformation caused by Resistance Spot Welding and Self Pierce Riveting © 2005 IARC EuroPAM – 5th October 2005 7 Observed Distortions with RSW and SPR in Top-Hat Assemblies 3.0 Displacement (mm) SPR 2.5 2.0 RSW 1.5 RSW 1.0 0.5 SPR 0 2 mm gauge 1.5 mm gauge Aluminium Mild steel Maximum distortion across the flanges of two top hats joined flange to flange © 2005 IARC EuroPAM – 5th October 2005 8 Resistance Spot Welding Distortion around RSW caused by sheet separation Sheet separation 5mm Sheet separation arises from expansion and contraction in the fusion zone Electrode penetration Nugget penetration HAZ dia. Nugget dia. © 2005 IARC EuroPAM – 5th October 2005 9 New Automotive materials Reduce environmental impact of vehicles Issues for RSW Joining High Strength Low Alloy steels • higher energy input than conventional steels • short hold time to avoid quenching Aluminium • High welding currents • High electrode forces • Short weld times © 2005 IARC EuroPAM – 5th October 2005 10 Software requirements Ability to model local distortions around a spot weld Electro / Thermal / Mechanical interactions Modelling of assembly process Welding sequence Clamping conditions © 2005 IARC EuroPAM – 5th October 2005 11 Local - Global Modelling Technique LO 2d model of individual weld CA L 3d model of individual weld M OD EL Application of 1st weld Application of final weld GL OB AL M OD EL Release from fixture © 2005 IARC EuroPAM – 5th October 2005 12 Validation of Local Weld Model (a) Nugget size (b) Sheet separation (Welding current=10.5 KA, Electrode force = 2.5 KN, Welding time = 15 cycles) Nugget size and Sheet separation © 2005 IARC EuroPAM – 5th October 2005 13 Validation of local weld model Target Actual Simulation >4.9 6.0 6.1 Nugget penetration (mm) - 1.1 1.3 HAZ dia. (mm) - 7.2 7.6 Sheet separation 5mm from centre (mm) <0.15 0.1 0.12 Electrode penetration (mm) <0.15 0.006 0.16 Nugget dia. (mm) © 2005 IARC EuroPAM – 5th October 2005 14 Distortion Prediction – Effect of Weld Sequence (a) Inline joining sequence z Weld sequence path x 11 10 9 8 7 6 5 4 3 2 1 z x y y (b) Spiral joining sequence z Weld sequence path y x 12 13 14 15 16 17 18 19 20 21 22 21 17 13 9 5 1 4 8 z 12 16 x 20 y 22 18 14 10 6 2 3 7 11 15 19 Steel Top-hat / Flat plate assembly © 2005 IARC EuroPAM – 5th October 2005 15 Physical Validation Welding Top Hat assembly CMM measurement of assembly © 2005 IARC EuroPAM – 5th October 2005 16 Comparison of Test and Simulation 0.171 -0.389 -0.259 CMM measurement © 2005 IARC -0.18 0.187 Simulation result DC01 + Zintec Top-hat / Top-hat assembly – in-line welding sequence EuroPAM – 5th October 2005 17 Comparison of Test and Simulation STEEL TOP-HAT / TOP-HAT INLINE WELDING SECTION 1 0.5 0.4 CLAMP - Meas Displacement (mm) 0.3 0.2 0.1 CLAMP - Meas UNCLAMP - Meas 0 0 50 100 150 200 250 -0.1 300 350 UNCLAMP - Meas -0.2 -0.3 -0.4 -0.5 Distance from Left end of test piece(mm) © 2005 IARC EuroPAM – 5th October 2005 18 Comparison of Test and Simulation STEEL TOP-HAT / TOP_HAT INLINE WELDING SECTION 5 0.5 0.4 0.3 Displacement(mm) CLAMPED - Meas 0.2 0.1 UNCLAMPED - Sim 0 CLAMPED - Sim 0 50 100 150 200 250 300 350 -0.1 -0.2 -0.3 -0.4 UNCLAMPED - Meas -0.5 Distance from Left end of test piece(mm) © 2005 IARC EuroPAM – 5th October 2005 19 Comparison of Test and Simulation STEEL TOP-HAT / TOP-HAT INLINE WELDING SECTION 7 0.5 CLAMPED - Meas 0.4 Displacemnet (mm) 0.3 0.2 0.1 UNCLAMPED - Sim 0 -0.1 0 UNCLAMPED - Meas 50 100 150 200 250 CLAMPED - Sim 300 350 -0.2 -0.3 -0.4 -0.5 Distance from Left end of test piece(mm) © 2005 IARC EuroPAM – 5th October 2005 20 Conclusions Successful prediction of direction of deformation Underestimate of magnitude of distortion Component & Fixture accuracy? Mesh size? Projection of welds on to global model and model size could be extremely large Extensive material data is required © 2005 IARC EuroPAM – 5th October 2005 21 Future Work Case study with actual part Validate model for RSW of aluminium Use Pam Stamp Simulations as the starting point to incorporate distortion and residual stress information into SYSWELD © 2005 IARC EuroPAM – 5th October 2005 22 Acknowledgements © 2005 IARC EuroPAM – 5th October 2005 23 Predicting distortion in vehicle body assemblies Iain Masters & Xinmin Fan Warwick manufacturing Group International Automotive Research Centre The University of Warwick Coventry CV4 7AL UNITED KINGDOM +44 (0)24 7657 5380/5408 © 2004 IARC