In-situ monitoring of geometric accuracy in laser powder bed fusion processes

The increasing interest for metal additive manufacturing (AM) technologies in highly regulated sectors like aerospace and healthcare imposes the need to achieve challenging part and process qualification requirements. Powder bed fusion processes represent the family of AM processes that provides the highest accuracy and resolution performances. However, several source of instability may yield geometrical distortions that shall be detected as soon as possible. This paper presents a study for in-situ identification of geometrical defects via high-resolution imaging of each layer. It allows comparing the geometry observed via online image segmentation with the nominal one for a quick detection of major departures. The calibration of the sensing equipment together with the repeatability and reproducibility (R&R) analysis of image sequences gathered in-situ is discussed, to highlight the possibility of using this tool for reliable in-situ measurements and to reduce the need for expensive and complicated post-process inpections.