Resolving the three-dimensional flow field within commercial metal additive manufacturing machines: Application of experimental Magnetic Resonance Velocimetry

In laser-based powder bed fusion of metals (PBF-LB/M) systems, the shielding gas flow is important for controlling the accumulation of melting byproducts such as soot and ejecta near the melt pool as well as sweeping ejecta, agglomerates, and “flying” powder away from the build region. It is necessary to understand the flow field throughout the build chamber in order to sufficiently control these contaminants to prevent their interference with delivered laser power and maintain part quality. A measurement technique called Magnetic Resonance Velocimetry (MRV) was applied to measure the entire three-dimensional three-component mean velocity field in a scaled flow model of a commercial EOS M290 PBF-LB/M build chamber. Flow conditions in the model were at a Reynolds number of 36000 (based on the average bulk velocity in the supply duct and its diameter) comparable to conditions in the full size M290. MRV measurements reveal five large jets issuing from the upper vent of the M290: the top jet supplies fresh gas flow across the optics keeping them clean, three jets issue horizontally, and the last jet travels downward impinging on the center of the build plate potentially sweeping contaminants toward the exit or delivering contaminants to the surface from the upper portion of the chamber. The lower vent supplies flow across the build plate, but this flow issues from above the build plate creating a backward facing step flow in which a region of separation with backwards flow exists. The backwards flow extends several centimeters over the edge of the build plate nearest to the vent before the inlet flow “attaches” to the build plate and proceeds to the exit. Single layer laser scans on individual titanium plates spaced uniformly around the build surface are performed, and their melt pool depths are measured as an indication of delivered laser power. Measured depths for plates in the separated flow region are measured to be 23 % shallower than in the central region of the build plate where the flow is unidirectional toward the exit. The influence of the shielding gas flow on melt pool depth is demonstrated using an extra vent placed below the lower vent to supply unidirectional flow across the build plate in the separated flow region. With the flow improvement, the melt pool depths increased by 15 %. These experiments support arguments that shielding gas flow is important for building consistent high quality parts and demonstrate the utility of MRV measurements fo