Computational analysis and experiments of spatter transport in a laser powder bed fusion machine

Abstract

This work focuses on how spatter particles are transported within a laser powder bed fusion (L-PBF) machine. The machine’s gas flow rate and salient flow features are studied with a computational fluid dynamics (CFD) model and are validated with experimental measurements of the flow near the build plate. The CFD model is also paired with a discrete phase model (DPM) to show how a spatter particle’s diameter, speed, ejection angle, material, and ejection location each affect its trajectory. The spatter model results are also validated by experiments wherein hot spatter particles are identified with an infrared camera. Overall, this work brings focus to several spatter-mitigation strategies including increasing gas flow rate, controlling spatter production through laser settings, and changing part placement. Throughout the infrared experiment’s single build cycle, over 14 million spatter particles were detected, contaminating both the parts and the powder bed. Both the spatter model and experiment suggest that spatter particles travel primarily in the downstream direction. This suggests that placing parts next to one another instead of downstream from one another could be a strategy to prevent spatter particles from traveling between parts. A low-speed zone is also discovered beneath the lower nozzle and is predicted to negatively alter the removal of spatter from that region of the build area.