Diese Website verwendet Cookies. Mit der Nutzung der Website stimmen Sie deren Verwendung zu. Weitere Informationen erhalten Sie in unserer Datenschutzerklärung.

Calibration and Appliance of the Wilkins Damage Model for Aluminium Cast Alloys

Aluminium cast alloys gain more relevance in structural applications. Therefore, numerical Simulations and appropriate material models are necessary. The material modelling of cast alloys without consideration of material damage leads to significant overestimated simulation results. Hence, this study covers damage modelling of an AlSi cast alloy by the Wilkins Damage Model. A strategy for the characterization of the damage behaviour is developed and applied on a die cast alloy. Therefore, flat tensile tests, e.g. Merklein and Notched Tensile Test, with various geometries are performed. These geometries represent different triaxiality regimes, which deliver data for the description of damage behaviour. The calibration of the Wilkins Model is done by inverse simulations of these test geometries. A Material Degradation Parameter allows the modification of the slope for the modelling of damage induced material softening effects. The approach of Wilkins considers both, the influence of hydrostatic stress and deviatoric stress for the damage evolution. This is performed by a separated damage formulation in application of the hydrostatic stress part and the deviatoric principal stresses. Hence, a quasi-separated consideration in the calibration process may be applied to ensure well accordance in inverse simulation of the test coupons. This means, shear dominated test coupons leads to data for the phenomenological parameters of the first damage evolution consideration. The remaining parameters may be calibrated by the second damage evolution, which includes the hydrostatic stress part. A validation of the material model is carried out by a crushing test of hat profiles. The numerical simulation of the experiment is done by LS-Dyna R8 explicit. The comparison of the energy absorption shows a low divergence of 3 %.