Improvement in robustness and computational efficiency of material models for finite element analysis of metal powder compaction and experimental validation

In metal powder compaction, density nonuniformity can be a source of flaws. Material models in finite element analysis for the prediction of density distribution still lack robustness and are computationally expensive. In this work, the Drucker-Prager cap (DPC) material model is implemented into the commercial finite element (FE) software ABAQUS/Explicit using the user-subroutine VUMAT. Yield functions in this material model are pressure-dependent and the curvature of the cap yield surface is high. This can cause numerical instability. We implemented a sub-increment technique to address this instability problem in a previous paper (Kashani et al., Trans North Am Manuf Res Inst SME 38:623-631, 2010). The DPC model is also a non-smooth, multi-yield surface material model which has instability problems at the intersection of the yield surfaces; we adopted the corner region in the DPC material models for soils in order to remove instability and the results of which were presented in (Kashani et al., Trans North Am Manuf Res Inst SME 38:623-631, 2010). The computational efficiency of the DPC material model was improved using a novel technique to solve the constitutive equations analytically which was shown in a previous paper by the authors (Kashani et al., Trans North Am Manuf Res Inst SME 38:623-631, 2010). In this paper, experimental tests were conducted where cylindrically shaped parts were compacted from Distaloy AE iron-based powder to 7.0 g/cm(3) using 592 MPa of pressure. To measure local density, metallography and image processing were used to find the void area fraction of the surface. The FE results were compared to experimental results and it was shown that the FE analysis predicted local relative density within 3 % of the actual experimental measurements.