Digital Data Management to Certify am Parts with Reduced Inspection: System Health Monitoring and Final Part Quality

Conference Abstract: Process qualification typically requires substantial non-recurring engineering (NRE) frequently requiring more than six months and $500,000 in cost. Traditional approaches rely on performing redundant, statistically substantiated testing even for seemingly basic process modifications such as the qualification of a new AM system with identical specification or a change in feedstock provider. Advanced process qualification approaches, such as integrated computational materials engineering (ICME) and data-driven machine learning have been studied to reduce NRE. These techniques have shown promise; however, certifying bodies have yet to approve these methods as a means of process qualification.

LIFT, America Makes, and MxD, three DoD manufacturing innovation institutes, have executed a collaborative DoD-funded program to reduce cost and time required for AM qualification.

This presentation summarizes conclusions against the following steps:
1. Implement AM equipment health monitoring, i.e., direct data measurement of equipment key health parameters such as gas flow and laser quality to establish acceptable system performance limits that reduce the cost of scaling across a fleet of identical AM systems;
2. Test assertions and assumptions by manufacturing a statistically significant quantity of parts to assess how AM process and post-process variables affect final quality.

The effects of variation in equipment health on quality was quantified across two thermal histories using thermal simulation to design coupons with different support features. Repeatable methods of varying equipment performance were established, a set of experiments was designed, and builds were performed. Coupons were thermally post-processed, machined, and mechanical testing was performed. Regression models were developed to link equipment health to part properties and thresholds for machine performance were defined.