| Abstract Scope |
In additive manufacturing, ensuring product quality by inspecting printed parts is critical prior to shipment. Detecting internal defects using X-ray computed tomography imaging is indispensable for quality assurance. However, this approach has two drawbacks. One is the challenge in inspecting defects in heavy metals such as tungsten and pure copper. Due to the limited penetration of X-rays, defect analysis becomes difficult. Another issue is that analysis and 3D data reconstruction require significant time and resources. Delays occur if unacceptable defects are found after printing, resulting in time and financial losses.
To address this, JEOL has developed defect detection using Backscatter Electron monitoring. This application integrates machine learning to identify defects, allowing users to detect unacceptable defect sizes and porosity. Furthermore, the application can halt printing or isolate problematic parts, ensuring efficiency.
We will introduce the features and principles of this application, emphasizing its role in enhancing productivity in AM processes. |