10:00 AM - 10:20 AM
[16A-T7-15] In situ Density Prediction in Metal Binder Jetting Using Powder Bed Imaging
Keywords:metal binder jetting, green part density, powder bed imaging
Metal binder jetting promises lower costs for end-use parts, whose quality strongly depends on green part density. Various methods (e.g., Archimedes, geometric density) are available to measure this after depowdering. However, these methods require additional effort and equipment. In this paper, an in-situ density prediction tool based on process images is presented in order to minimize cost and time. For this purpose, the powder bed is photographed layer by layer with a built-in camera system. The images are then automatically analyzed using image segmentation in Python. During segmentation, the layer contour is approximated by unit cells, which are then assigned to a relative density by counting colored pixels acting as a measure for powder infiltrated with binder. It is shown that the predicted and the geometrically determined density differ on average by 1.5%. This calibration factor is then used for the in-situ density prediction of green parts.