Please login to view abstract download link
In this study, we investigate the dynamics of Selective Laser Melting (SLM) in powder-based additive manufacturing using a recently developed solver in the CFDEM framework [1]. SLM is a popular method for producing complex metal parts with high precision, however, the process can be affected by various phenomena such as powder spreading, melting, and solidification. The semi-resolved approach applied in this study incorporates the Discrete Element Method (DEM) and Volume of Fluid (VOF) to simulate the interactions between powder particles and molten bed. The solver applies LIGGGHTS implementation of the DEM for particles coupled with OpenFOAM for Computational Fluid Dynamics (CFD) simulation of the continuum phase, i.e., inert gas. A kernel approximation facilitates the semi-resolved solutions as opposed to unresolved solutions which is the bottleneck of the traditional solvers in CFDEM. Furthermore, the interface between the phases is reconstructed using the isoAdvector [2] method based on VOF. The melting and solidification process strongly depends on the laser power and scan speed. As the laser power increases, the melting zone becomes wider and deeper, resulting in a smoother surface finish. However, excessive laser power may cause excessive evaporation and a spattering of the molten metal, leading to defects and reduced mechanical properties. Similarly, the scan speed affects the cooling rate, and a lower scan speed can result in better bonding between layers and reduced defects. We perform a series of simulations to investigate the effects of laser power, scan speed, and layer thickness on the powder melting, solidification, and entrainment processes. In conclusion, our study demonstrates the importance of numerical simulations in understanding the dynamics of the SLM process and optimizing the process parameters. The developed solver in the CFDEM framework provides a powerful tool to investigate the complex interactions between the laser beam and powder particles and can be used to further improve the SLM process for industrial applications.