Case Study
Leinfelden-Echterdingen,
Reliable weld seam testing
Inline thermography for complex welds in miniature format.
Automated weld seam inspection using thermography testing methods is also possible for small and complex parts. In a high-pressure injector for a diesel engine, two individual parts are joined together with three laser beam welds. In the event of a fault, the welds may crack. Defects and leaks would be the consequences. Together with the manufacturer, an inline crack test stand based on induction thermography was developed.
Project Description
Motivation
The customer is a German automotive supplier with extensive know-how in the area of injection components. The component is part of a magnetic valve injector or injection valve, which is exposed to an injection pressure of up to 2,700 bar. However, the application is also suitable for many other laser beam weld tests on complex or miniaturized components.
Challenge
The task was to develop a non-destructive testing method for inline testing using induction thermography and the corresponding testing software. The evaluation should be automated through image processing. In a cycle time of just under 5 seconds, 3 laser beam welds could be reliably tested for cracks, holes, pores or puncture defects. To visualize all welds, the component had to be repositioned twice.
Benefits
The previous testing process consisted of visual crack testing under a microscope: not fast, very expensive and error-intensive because, in contrast to thermography testing, depends on human factors. This process has now been replaced by a fully automated 100% inline inspection — with rapid payback and completely non-destructive.
Solution
For the small components with a height of 30 mm and a width of 15 mm, the welds are less than 10 mm short and the cracks are microscopically small. In addition, the laser beam weld of the two components is made in a groove of 3 to 4 mm deep. All of this places increased demands on the performance of the hardware and application. The particular challenge is to concentrate the induction field precisely on the area of the groove in which the test part is welded without losing energy at the edges. edevis works with a patented process that is able to test even such hard-to-reach welds. In addition, the relatively inhomogeneous surfaces of laser welds place high demands on non-destructive testing technology. The camera optics and image processing used for testing must make it possible to reliably distinguish normal irregularities and defects such as cracks or pores. Especially when parts have no weld at all due to faults.
Machine learning
If necessary, as in the present case, edevis uses sophisticated “machine learning” as a precursor to artificial intelligence behind the application. The effects and inhomogeneities of the weld seam that occur here were examined in order to reduce the interference effects for the software's image processing and thus make the inspection more robust.
Implementation
It began with the customer sending components with the requirement to find known faults. In this case, this is the usual procedure. Customers send faulty parts and edevis proves that they can be discovered using the appropriate thermographic methods. On this basis, a test idea was developed and the feasibility was validated with a study. This was followed by the offer phase. edevis supported the project from application development and later on to the acceptance of the test system in an advisory capacity. The test system was implemented by the customer himself using edevis components. The highlight was successfully achieving a high level of sensitivity during application, even for very small cracks.
FAQ
Our frequently asked questions — answered quickly and easily.
