Forming simulation is a widely used tool to estimate production forming capabilities. During the last three to four years the prediction of process robustness by sensitivity analysis has been developed for industrial applications. The change of material parameters is one key figure and has a large impact on the final findings. Thus the user has to ensure that the variations done are in correlation with real material behaviour and the selected numerical model is reliable. Up to now the permitted changes in modelling are seldom secured by real measurements, especially when more advanced material models and hardening options are applied.
Various materials chosen out of the production process for mild steel grades have been investigated for their mechanical properties using different tests. The limits of failure are defined by a reduced number of Nakajima tests to predict the forming limit curve (FLC). All data is prepared in the same manner to meet simulation program needs. The calibration of yield locus by solely tensile tests and in combination with hydraulic bulge tests gives a first estimation of possible compromises in data preparation. More advanced material calibration strategies allow a better fit of the models to the measured values (Hill ´90, Banabic 2005). Furthermore some principles can be noticed for yield locus calibration in biaxial stress space. The experimentally measured forming limits are compared to the TKS-in-house regression formula, Keeler´s approach and other models.
Finally simplified state of the art simulations (input tensile test) are compared to simulation results gained with more accurate material data input parameters for yield locus and hardening for different complex parts. The resulting difference in the feasibility findings are then discussed with respect to a virtual process design chain.
