Prediction of lack of fusion porosity in selective laser melting based on melt pool monitoring data

Selective laser melting is already established as a commercial production technique. Some high-end users are, however, struggling with the complexity, consistency and cost associated with the qualification of high-end products built using the technique. In-situ process monitoring is a promising means to accommodate this issue, but quantitative correlations between monitoring signals and actual part defects have been lacking. In this paper, results are presented that have been obtained with an off-axis melt pool monitoring system on a 3D Systems ProX DMP 320 using Ti-6Al-4 V ELI. The focus is on the development of a method for predicting the presence and location of lack of fusion porosities as they can have a large impact on part quality and are not always easily detected post-build. The processed signals from the monitoring system are shown to have a high degree of correlation with the presence of lack of fusion porosities as measured by CT scans. A prediction sensitivity of 90% for lack of fusion events in the range of pores having a volume greater than 0.001 mm3, roughly equivalent to 160 μm in diameter, was obtained.