Accelerating LPBF Qualification Through Multi-sensor Data 'Safe Range' Specification

Laser Powder Bed Fusion (LPBF) has established itself as a pivotal additive manufacturing technique, offering the ability to fabricate intricate geometries with remarkable precision. However, qualifying machine parameters and ensuring part quality remain time-intensive due to the potential occurrence of defects such as porosity, cracks, and incomplete fusion. To address these challenges, this study introduces the concept of a "Safe Range" for accelerating the qualification of both machine and part parameters, combined with a multi-sensor analysis approach.

The "Safe Range" specification defines an optimal window of machine settings and process parameters that consistently produce high-quality parts without defects. This range is determined through a combination of multi-modal sensors—such as optical imaging, thermal sensor data, and optical tomography—and data-driven models. These sensors capture comprehensive process data in situ, allowing for a detailed understanding of the LPBF process and the identification of key parameters that influence process stability and defect formation.

Artificial intelligence and machine learning (AI/ML) models, including convolutional neural networks (CNNs), are used to analyze the data from the multi-sensor system, helping to rapidly detect deviations from the "Safe Range." By focusing on maintaining parameter settings within this range, manufacturers can significantly reduce the number of trial-and-error builds required for qualification, thus speeding up the process.

This integrated approach not only ensures in situ defect detection but also supports faster qualification by identifying and maintaining critical machine parameters. By leveraging the "Safe Range" concept, LPBF technology can achieve more efficient and reliable production, contributing to its broader adoption in industrial applications.