Mobile-Menu

Don't go all out—take it easy Digital Measurement of Welds Saves a Tremendous Amount of Steel

Source: Fraunhofer IWM | Translated by AI 3 min Reading Time

Related Vendors

FAULHABER Drive Systems
dataTec AG
SIMTOS (Seoul International Machine Tool Show)

As part of the “ORka3D” collaborative project funded by the BMWE, researchers have developed a system that uses 3D scanning and artificial intelligence to capture the actual geometry of individual welds and evaluate them on a case-by-case basis ...

When it comes to ensuring that a welded joint lasts as long as possible, this has traditionally been assessed on a blanket basis using so-called notch fracture classes. However, this usually meant accepting that the joint was oversized. Researchers have therefore set out to change that ...(Image: AI-generated)
When it comes to ensuring that a welded joint lasts as long as possible, this has traditionally been assessed on a blanket basis using so-called notch fracture classes. However, this usually meant accepting that the joint was oversized. Researchers have therefore set out to change that ...
(Image: AI-generated)

Every weld is unique and has a geometric notch at the interface with the base material, from which most fatigue cracks originate. Because this notch varies depending on the welding process, torch guidance, and even the welder’s performance on a given day, regulations have so far relied on a blanket safety margin—they generally assume the most unfavorable permissible geometry to be on the safe side. However, the actual—often significantly better—quality of the finished weld is not taken into account. The result is that more welding material is consumed than necessary. According to the Fraunhofer Institute for Mechanics of Materials (IWM), changing this, however, holds enormous potential for resource savings when it comes to steel. The project mentioned above is officially titled “Optimization of Resource Use in Welded Structures through Automated In-Line Service Life Assessment Based on 3D Scans.” Under the leadership of the IWM, a method was developed that scans the actual weld bead surface and can then automatically derive a component-specific and objective service life prediction for the welded joint from the data, according to the institute. This, it is said, can usefully supplement the general assessment based on notch fracture classes (FAT classes).

How the New AI Weld Inspection Works in Practice

As mentioned earlier, a 3D scanner captures the weld surface with a pixel pitch of up to five micrometers. An algorithm automatically extracts key geometric parameters from this data—in particular, the weld transition radius, which has been neglected in standards to date, as the researchers explain. An artificial neural network uses this data to derive the individual notch shape factor and a component-specific service life prediction, rather than relying on the general FAT classes. As emphasized, the method has already been validated on approximately 400 test specimens as well as on real-world structures such as pipe joints and crane booms. With this data-driven assessment of fatigue strength, the notch class could ultimately be raised from FAT90 to as high as FAT140, which corresponds to a reduction in the standard sheet metal thickness of up to 40 percent and a halving of welding time—all while maintaining structural safety, because the verification is based on individual measurement data rather than on blanket safety margins.

Subscribe to the newsletter now

Don't Miss out on Our Best Content

By clicking on „Subscribe to Newsletter“ I agree to the processing and use of my data according to the consent form (please expand for details) and accept the Terms of Use. For more information, please see our Privacy Policy. The consent declaration relates, among other things, to the sending of editorial newsletters by email and to data matching for marketing purposes with selected advertising partners (e.g., LinkedIn, Google, Meta).

Unfold for details of your consent

Related content