A Numerical Simulation Study of Deep Drawing Testing and Experimental Results of Steel Sheets, Using a Comercial Software

Present work examines the accuracy of numerical simulation results of deep drawing testing of steel sheets through employing the commercial finite element software AutoForm to reproduce the punch force versus displacement. Numerical simulation is nowadays a modern engineering practice for sheet metal product and tooling design developments, using the finite element method. Historically, sheet metal formability has been assessed by tensile testing and biaxial stretching such as the Erichsen simple test. Lately, the concept of experimental Forming Limit Curve for strains, FLC, and the numerical simulations were developed to evaluate sheet metal formability and its forming operations by predicting the onset of local necking and fracture. In the present work, five types of steel sheets and one aluminium alloy AA6022 sheet were employed in the deep drawing tests with the same flat punch and die to obtain the experimental results of force versus punch displacement which were presented by De La Cour. The steel sheets with the same blank diameter utilized in the simulation and experimental testing were DQSK, BH33, HSLA50, TRIP600 and DP600 which different thickness varied from 0.783 mm to 1.601 mm. The experimental curve of punch force versus displacement were compared with the numerical simulation results by AutoForm, using 3D shell elements, hardening law defined by the effective stress \( {\overline {\upsigma }} { = }\upsigma_{\text{o}}\left( {1 + \upbeta }\overline {\upvarepsilon } \right)^{\text{n}}\) and friction coefficient of 0.10 between the blank hold and blank and 0.075 between blank and die. The average experimental hardening law curve for tensile tests at 0°, 45° and 90° and the Hill’s 1948 yield criteria, assuming isotropic plasticity, were used to numerically simulate the plasticity behavior during the deep drawing of steel and aluminium sheets. The simulated curves are situated above the experimental points of punch load for all materials. The calculated error varied from 2% to 8.5% for points near the maximum load. The principal characteristics of reliable software for numerical simulation of deep drawing process are also discussed, considering the actual tendency to use software in the design of dies and sheet metal manufacturing process analysis.