Sensitivity of friction coefficients, material constitutive laws and yield functions on the accuracy of springback prediction for an automotive part

Abstract

High Strength Steel is more used in automotive industry to combine safety improvements and weight reductions of the vehicles. However, these steels raise new challenges for manufacturer, especially for springback and its prediction. Finite Element Analysis is widely used for the prediction of springback but accuracy is strongly depending on numerical as well as physical parameters. Such parameters affecting prediction are for example frictional behavior, constitutive model, loci function. In this paper the sensitivity of friction coefficients and the influence of some material laws on the springback prediction were studied and analyzed in details. Several numerical models are investigated in this work in order to allow the different aspects of materials mechanical behavior to be better described: isotropic work-hardening (Swift, Hockett-Sherby, SHS), isotropic-kinematic hardening, yield loci (Hill 48, Hill 90 and Barlat 91). In this study, we use an industrial example of the deep drawing process: the side sill closing with the material steel TRIP800. All the simulations are carried out with the commercial code Pam-Stamp 2G and comparison with the experimental results are presented.