Laser additive manufacturing: monitoring during defect occurrence
Researchers at the Federal Institute of Technology in Lausanne have resolved the long-standing debate surrounding laser additive manufacturing processes through a groundbreaking defect detection method.
The development of laser additive manufacturing is often hindered by unexpected defects. Traditional monitoring methods, such as thermal imaging and machine learning algorithms, have shown significant limitations. They often overlook or misunderstand defects, making precision manufacturing elusive and excluding this technology from important industries such as aviation and automotive manufacturing.
But what if defects can be detected in real-time based on the sound differences and irregular sounds emitted by the printer during the perfect printing process? So far, the prospect of detecting these defects in this way is considered unreliable. However, researchers from the Thermomechanical Metallurgy Laboratory at the Federal Institute of Technology in Lausanne have successfully challenged this hypothesis.
Professor Roland Log é, the head of the laboratory, said, "There has been controversy over the feasibility and effectiveness of acoustic monitoring in laser based additive manufacturing. Our research not only confirms its relevance, but also emphasizes its advantages over traditional methods.".
This study is crucial for the industrial sector as it introduces a breakthrough and cost-effective solution for monitoring and improving the quality of products manufactured through laser powder bed melting.
Dr. Milad Hamidi Nasab, Chief Researcher, stated that the synergistic effect of synchrotron X-ray imaging and acoustic recording provides real-time insights into the LPBF process, helping to detect defects that may endanger product integrity. In an era of constant pursuit of efficiency, accuracy, and waste reduction in various industries, these innovations not only save a lot of costs, but also improve the reliability and safety of manufactured products.
How does LPBF manufacturing work?
LPBF is a cutting-edge method for reshaping metal manufacturing. Essentially, it uses high-intensity lasers to carefully melt tiny metal powders, layer by layer creating detailed 3D metal structures. Treating LPBF as a metallic version of traditional 3D printers adds a certain degree of complexity.
It is not melted plastic, but uses a layer of small microscopic metal powder, whose size can range from the thickness of human hair to fine salt particles. The laser moves on this layer, melting specific patterns according to the digital blueprint. This technology can produce customized complex parts with minimal excess, such as lattice structures or unique geometric shapes. However, this promising approach is not without challenges.
When laser interacts with metal powder to form a so-called molten pool, it will fluctuate between the liquid phase, gas phase, and solid phase. Sometimes, the process may fluctuate due to variables such as the angle of the laser or specific geometric properties of the powder or part. These situations, known as "inter regime instability," sometimes lead to a shift between two melting methods, known as "conduction" and "lockhole" systems.
In an unstable lockhole state, when the molten powder pool is drilled deeper than expected, it will generate pores, ultimately leading to structural defects in the final product. In order to facilitate the measurement of the width and depth of the melt pool in X-ray images, the Image Analysis Center of the Imaging Center at the Federal Institute of Technology in Lausanne has developed a method that makes it easier to visualize small changes related to liquid metals, as well as a tool for annotating the geometry of the melt pool.
Use sound to detect these defects
In a joint venture with the Paul Scherrer Institute and the Swiss Federal Laboratory for Materials Science and Technology, the EPFL team has developed an experimental design that combines operational X-ray imaging experiments with acoustic emission measurements.
The experiment was conducted on the TOMCAT beam line of PSI Swiss Light Source, using a small LPBF printer developed by Dr. Steven Van Petegem's team. The combination with the ultra sensitive microphone located in the printing room can accurately locate significant changes in acoustic signals during state transitions, thereby directly identifying defects in the manufacturing process.
A crucial moment in this study was the introduction of adaptive filtering technology by Empa's signal processing expert Giulio Masinelli. "This filtering method," Masinelli emphasized, "enables us to distinguish the relationship between defects and accompanying acoustic features with unparalleled clarity.".
Unlike typical machine learning algorithms, machine learning algorithms excel at extracting patterns from statistical data, but are typically customized for specific scenarios. This approach provides a broader understanding of the physics of melting states while providing excellent temporal and spatial accuracy.
Through this study, the Federal Institute of Technology in Lausanne has contributed valuable insights to the field of laser additive manufacturing. These findings have significant implications for potential industrial applications, particularly in fields such as aerospace and precision engineering. This study consolidates Switzerland's reputation in meticulous craftsmanship and manufacturing accuracy, emphasizing the need for consistent manufacturing technology.
In addition, it also has the potential for early detection and correction of defects, thereby improving product quality. Professor Log é concluded, "This study paves the way for a better understanding and improvement of manufacturing processes, and in the long run, it will ultimately lead to higher product reliability.".
The research results are published in the journal Nature Communications.
Source: Laser Net
