Menü des aktuellen Bereichs:



Die Filling

Simulation of the die filling process

Die filling process deals with a prohibitive number of particles to be simulated in most of the cases. Coarse graining methodology replaces individual particles by representative parcels, substantially reducing the required number of particles to represent a process. Analytical considerations and numerical simulations were performed to ensure that the physics is captured correctly in the different regimes the material may be inserted. A shear box using Lees-Edwards boundary conditions in different shear rates was simulated with different solid fractions to investigate different flow regimes (Fig. 1).

Fig. 1: Effect of coarse graining in the particle stress for flow regimes (left). Stress box with boundary conditions to mimic different regimes (right).

A characteristic of many metallic powders used in the die filling process is their complex dynamic and static behaviour. We use cohesion models like JKR and capillary in addition to standard DEM contact models to depict this behaviour (Fig. 2). The calibration process of cohesive material is not yet well known. Most promising calibration process we conduct involve a shear cell like device where material can reach a unique consolidation stage and Coulomb cohesion can also be captured (Fig. 3).

Fig. 2: Metallic powders used in die filling process and tests. On the left a non cohesive powder. On the right highly cohesive powder.

Fig. 3: Shear cell depicted in the left. In the right top result of calibration without cohesion for artificially high friction parameters. Right bottom result of calibration process incorporating cohesion and using intermediate realistic friction parameters.

The usage of cohesion models combined with coarse graining has impacts in the void fraction distribution and have to be considered in numerical simulations. CFDEM captures void fraction distribution in the die region while Monte Carlo method is preferable to capture these effects near to the walls (Fig. 4).

Fig. 4: Void fraction distribution in the die region (left) and very near to the wall ~ particle diameter / 100 - for different particle sizes (graph right).

(Daniel Schiochet Nasato)