In situ defect detection in selective laser melting via full-field infrared thermography

Selective laser melting (SLM) has become one of the most commonly utilized processes in metal additive manufacturing (AM). Despite its widespread use and capabilities, SLM parts are still being produced with excessive volumetric defects and flaws. The complex dependence of defect formation on process parameters, geometry, and material properties has inhibited effective quality assurance in SLM production. Exacerbating these issues are the difficulties thus far in accurately detecting and identifying defects in-process so that parts may be qualified without destructive testing. Some of the most detrimental defects produced during SLM processing are lack of fusion (LoF) defects, which are frequently found to be in excess of 100 mu m in size, thus these defects are of critical importance to detect and remove. In this work, we have developed and demonstrated the capabilities of a novel in situ monitoring system using full-field infrared (IR) thermography to monitor AlSi10Mg specimens during SLM production. Using layerwise relative surface temperature measurements, subsurface defects were identified via their retained thermal signature at the surface; transient thermal modeling was performed, which supported these observations. Parts were characterized using ex situ scanning electron microscopy (SEM) to validate data identified defects and, critically, to estimate detection success. The IR defect detection method was highly effective in identifying defects, with an 82% total success rate for LoF defects; detection success improved with increasing defect size. The method was also used statistically to analyze the presence of systematic process errors during SLM production, expanding the capabilities of IR monitoring methods. This unique analysis method and simple integration for in situ IR monitoring can immediately improve non-destructive qualification methods in SLM processing.