Deep Drawing Simulation Of High And Ultrahigh Strength Steels Under Consideration Of Anisotropic Hardening

Abstract

In today’s sheet metal forming simulation, most attention is paid to yield loci functions, which describe the anisotropy of the material in yielding. The coefficients, defining the shape of the yield locus in these functions are usually fitted at a certain level of plastic work and are then valid for the whole range of plastic deformation. Modern high and ultrahigh strength steels, especially those with induced plasticity, may often exhibit only a very small anisotropy in yielding, but a severe anisotropy in work hardening for different loading conditions. This behavior can not be described by fitting the yield locus at a specific value of plastic deformation. An approach to take into account the anisotropic hardening of sheet metals is to provide different yield curves for several loading conditions and expand the yield locus dependent on the current form of load. By doing this, one can use a comparatively simple yield locus, like that of Hill from 1948, because all anisotropy is given by the different hardening curves. For the commercial FEM code LS DYNA the material model MATFEM Generalized Yield is available as a user subroutine, which supports this approach. In this paper, forming simulation results of different yield loci are compared with experimental results. The simulations were carried out in LS‐DYNA with the Barlat 89 and 2000 yield loci and isotropic hardening and with the GenYld model combining a Hill 48 yield locus and anisotropic hardening. The deep drawing experiments were conducted on a hydraulic press, measuring binder and punch forces. The deformation of the sheet was measured by optical grid analysis. A comparison of the simulated and measured plastic strains shows that using a model including anisotropic hardening can produce better results than the usage of a complex yield locus but isotropic hardening for the examined materials. This might be interesting for e.g. spring back simulations. By combining a simple yield locus with anisotropic hardening an effective way for industrial part simulation seems to be available, which does not require the consideration of thermal effects or crystal plasticity.