In-process quality assurance (QA) systems play a critical role in ensuring the quality and consistency of parts produced using L-PBF technology. These systems involve a set of procedures and protocols that are applied during the manufacturing process to ensure that the final product meets the required specifications and quality standards. Some of the failure types that in-process QA systems can identify are delamination, overheating, uneven powder distribution or wrong machine set-up.
The analysis and interpretation of the QA data is often still a manual process requiring deep expertise. More and more intelligent software is coming up, however, that promise to automate the QA process and potentially even perform corrective actions.
On a global level, sensors are used to monitor the overall state of the L-PBF process. This includes the O2 content and pressure in the atmosphere and the laser current.Â
Compared to other monitoring systems, they have a low sampling rate and low resolution and thus lead to relatively small data packages.Â
Powder layer control
Identification of missing powder and part defects
Schematic illustration of the local L-PBF monitoring process
Schematic illustration of the powder layer control monitoring
Powder layer control is an effective way to monitor the recoating during the L-PBF process that has become standard for most machine manufacturers. A picture of the powder bed is taken before and after the application of a new layer of powder and the applied powder is depicted as grey scale picture.Â
This way it can be analysed if powder is missing on a layer, e.g. because of poor powder quality or false machine settings, or if parts are bending up and are thus defective. Since such a system generates over 1 000 pictures for a typical build job, the analysis must be guided by intelligent software.Â
Melt pool monitoring
Melt pool control for in-depth quality assurance
Melt pool monitoring provides the most in-depth analysis of the AM process. The systems include high speed CCD cameras that translate the electrical signals into digital values to measure the melt pool dimensions as well as diodes and a pyrometer to measure the melt pool intensity. Due to the high complexity, 2D or 3D mapping of the data is required for making the most out of the systems. The identification and correlation of abnormalities to defects remains extremely difficult, especially when building a part for the first time. Melt pool monitoring systems have a very high sampling rate and resolution and thus lead to very large data packages.Â
Schematic illustration of the melt pool monitoring process
Comparison of sensor data between identical parts on a CONCEPT LASER system
Source: CONCEPT LASER
The picture above shows how melt pool monitoring can be used for quality assurance in the LPBF process. 30 identical parts are produced in the same build job, each using the exact same process parameters. The signals measuring the melt pool dimensions and intensity should thus be identical for all 30 parts. Part 1 shows a deviating signal at a certain point, indicating a process abnormality that could lead to a certain defect. This part should thus be either be directly rejected or analyzed in more detail, e.g. using a CT scan.