In-situ detection of stochastic spatter-driven lack of fusion: Application of optical tomography and validation via ex-situ X-ray computed tomography

Artikel i vetenskaplig tidskrift, 2023

The presence of random defects in laser powder bed fusion (LPBF) parts is an issue that challenges the reliability of this manufacturing process and hinders its employment in structural, defect-sensitive components. A potential solution to increase the reliability of LPBF is employing in-process monitoring targeting defect detection. This study aims to detect stochastic defects driven by spatter particles via in-situ monitoring and validate the detection method ex-situ via X-ray computed tomography (XCT). By means of in-situ optical tomography (OT), monitoring images were registered layerwise during the manufacturing of Hastelloy X specimens. The images were analyzed to detect spatters landing within specimen boundaries, and the spatial coordinates of the detections were obtained. The specimens were also measured ex-situ by means of XCT, from which key features and coordinates of internal defects were obtained. The in-situ spatter detection method was then compared to the XCT measurements. It was found that 79 % of lack of fusion defects were detected in OT images. The detection was particularly successful for large defects. Spatter-induced lack of fusion defects were present in the specimens manufactured with optimized processing parameters in different degrees, depending on the robustness of the processing conditions to spatters. This study demonstrates the applicability of optical tomography in-situ monitoring for indirect detection of stochastic lack of fusion, whose presence is inferred from spatter redeposits on the powder bed.