SBIR-STTR Award

Reducing the waste by recovering more powder will greatly drive the cost of parts. In most cases, only a small amount of the powder that flows into the chamber during the build process is actually fused into a component. This leaves a significant amount of powder available either to be reused or to scrap. The process of recycling the material for future projects, manufacturers can better manage powder stock, requiring fewer orders, which will eventually reduce overall cost per part. The feedstock utilization can be maximized and drives the production towards increased sustainability.

The proposed innovation delivers a product quality assurance approach that integrates an automated particle shape, size and color measurement tool with a machine learning architecture, that enables use of recycled powder without compromising fatigue life. The machine learning algorithm consists of a two-fold module – first, imaging the particle size, shape and color and correlate to the recycled powder characteristics and secondly, enabling a high throughput testing tool to establish the correlation of recycled powder impact on component fatigue life. This integrated approach can be applied for any additive manufacturing process, machine, or material, thus enabling a broad applicability to the overall emerging AM industry