About this Abstract |
Meeting |
2024 TMS Annual Meeting & Exhibition
|
Symposium
|
Additive Manufacturing Modeling, Simulation and Machine Learning
|
Presentation Title |
Analyzing Debinding and Carbide Pickup for Quality Control of Binder Jet Printed SS 316L Using Computer Vision |
Author(s) |
Pooja Maurya, P.Chris Pistorius, Alex Gaudio, Asim Smailagic |
On-Site Speaker (Planned) |
Pooja Maurya |
Abstract Scope |
A potential issue that arises during the post-processing of binder-jet printed samples is incomplete binder removal, leading to carbon pick-up by the material during sintering. In this project, samples of wrought stainless steel with different carbon concentrations were heat treated to produce different extents of carbide precipitation. These samples were metallographically polished and etched to generate a set of images that will be used to test a computer-vision approach to classifying the extent of carbide precipitation, to be applied to sintered parts. These samples were subjected to electrochemical measurements to identify the carbide precipitation, which would serve as the ground truth for the computer vision approach. In addition, the thermal response of cubes of different sizes will be measured during debinding treatment and the resulting thermal history will be used to fit the thermal diffusivity – since the heating rate can affect the extent of binder removal. |
Proceedings Inclusion? |
Planned: |
Keywords |
Additive Manufacturing, Machine Learning, Characterization |