Quality control is one of the biggest challenges in production scale additive manufacturing. Main sources of variable quality could be from machine set up, process control, and raw materials. However, qualifying machine set up and print parameter can be challenging, time consuming and expensive. In-order to industrialize additive, we need a quick and cost effective way to evaluate multiple variables to map out the process control boundaries.
In this study, we present a novel solution for rapid qualification using Deep learning-based CT imaging and automated image processing. This solution can help with rapid evaluation of process control space to optimizing the print parameters, track variabilities in a multi-laser system as well as location in the build envelope as well as evaluate equivalency between printers.
We will present three case studies each covering various challenging aspects of AM industrialization. (1) rapid development of print parameter for novel high temperature Aluminum alloys, (2) Reduction of cost and time of the final part by adopting print recipe with thicker layers, and (3) Quality machine set- up to study the variability between the lasers and location in a large format multi-laser system.
Learning Objectives:
- Learn about X-ray CT-based automated defect detection for full 3D defect mapping
- Minimize the time and cost of their process parameter development and enable printing with novel alloys
- Learn about a novel method to evaluate printer equivalency
