The future of 3D printing has led researchers to jump dramatically with the success of developing new methods for detecting and predicting defects in 3D printing materials in real time. This groundbreaking approach, combining advanced diagnostic tools with machine learning, allows for the conversion of additive manufacturing by ensuring higher quality and reliability of printed components.
Scientists from the US Department of Energy and the University of Virginia have created a system that can detect pores of a common type of metal 3D printing, with near perfect accuracy on a sub-millisecond timescale. This development addresses one of the most important barriers to the widespread adoption of 3D printing for critical applications.
Promoting 3D printing defect detection
The researchers studied the 3D printing process at the microscope level using a combination of high-speed synchrotron x-ray imaging and thermal imaging. Their focus is a common 3D printing technique called laser powder bed fusion, in which high power lasers melt metal powders and create three-dimensional objects per layer.
During this process, one of the major defects that can occur is the formation of keyhole pores. These structural defects form when the unstable vapor depression zone, known as the keyhole, instantaneously collapses. When this occurs, the gas bubbles can be pinched from the tip of the keyhole and trapped in the solidified metal, which can impair the performance of the printed part.
Researchers discovered two different types of keyhole vibrations in titanium alloy (TI-6AL-4V) during the printing process. The first type, known as natural vibration, occurs in both stable and unstable keyholes, but does not lead to defects. The second type, known as perturbation oscillations, occurs only under unstable conditions and leads to the creation of porous keyholes.
“The clear frequency of these vibrations allows for the detection of faithful pore-generating events with high fidelity and high resolution,” explained one of the lead investigators. “We can now accurately identify when pores form during the printing process on a sub-millisecond timescale.”
Machine learning enables real-time detection
The breakthrough came when researchers correlated X-ray images inside samples with thermal images of the melt pool. They discovered that the formation of keyhole pores creates clear signals on the surface of material that can be detected by thermal cameras.
Using this knowledge, they trained machine learning models using x-ray images and predicted pore formation using only thermal imaging. Once tested, the approach can detect accurate moments where holes have been formed with 100% accuracy during the printing process.
This method provides a practical way to monitor the 3D printing process in real time without the need for expensive and complex X-ray imaging equipment. Many 3D printing machines already have thermal imaging sensors that monitor what is being built, but in many cases they miss the formation of pores. The new approach can be implemented using these existing sensors, allowing access to industrial applications.
Extending applications for 3D printing
The development of this detection method will expand the use of additive manufacturing in aerospace and other industries that rely on high-performance metal parts. Industry currently uses 3D printing to create everything from rocket engine nozzles to pistons for high-performance vehicles to custom-made orthopedic implants.
“New and advanced diagnostic tools for detecting and potentially repairing defects will expand the use of additive manufacturing in industries where quality is critical,” said researchers involved in the project. “The goal is to create a system that not only detects defects, but also allows for repairs during the printing process.”
Researchers have demonstrated that their approach can be implemented in two practical ways: by adjusting the time stamp to improve prediction accuracy, or by training machine learning models using well-tuned simulations. Both methods showed high prediction accuracy when tested with experimental data.
About Manufactur3D Magazine: Manufacturer3D is an online magazine about 3D printing. Please visit us Technology News For more information about 3D Printing Technology News, please see the page. Just like us to stay up to date with the latest events in the 3D printing world Facebook Or follow us LinkedIn and Twitter.
