Predicting density distributions in die filling

A computational tool which predicts density distributions after die filling is a key component in establishing a virtual process chain for the die compaction and sintering processing route. In order to simulate die filling a discrete element method (DEM) code has been developed regarding the particulate nature of flowing powder. The code is parallelized enabling detailed simulations of three-dimensional parts. Spherical grain representations and geometrically complex models were employed in order to distinguish their predictive capabilities. Mechanical model parameters have been calibrated within reference and validation experiments using ASC100.29 iron powder. While all models are able to predict flow rates for simple geometries they differ substantially when tested in their static properties like angles of repose. The most realistic grain model has then been used for filling simulations of simple and stepped ring cavities. Density inhomogeneities within these parts are discussed.