Laser welding can offer significant advantages when manufacturing high value, safety critical applications - offering a low distortion and high performance means of joining.
Nevertheless, narrow process tolerances can make it a challenge to maintain weld quality. Failure to monitor and control part preparation and cleanliness, part placement and seam fit-up, and the parameters of the process itself can all result in weld defects.
The InnoSeam project is addressing this challenge by developing a system capable of real-time seam detection, inspection and tracking, multi-sensor process monitoring and, based on observed changes in those sensor signals, the automatic alerting of possible set-up, equipment or welding faults.
The project uses a new version of Permanova’s existing seam finding, inspection and tracking system, combined with in-built LED coaxial illumination and imaging of the seam and the welding zone, along with plug-and-play compatibility with sensor-embedded delivery fibres (e.g. as available from Coherent (Optoskand AB)) for additional process monitoring.
Specifically, TWI role’s in this project is to build-up knowhow in the operation and capabilities of this system, and to identify its fault-detection functions, before then assisting in the validating of this performance at GKN, and confirming its use in reducing the reliance on expensive and time-consuming offline NDT.
PERMANOVA, supported in software development by BIT ADDICT, has developed a laser welding head with in-built high intensity green LED seam/process zone illumination. A coaxially attached CMOS camera with a matching narrow bandpass filter is then used to obtain a view of the seam either prior to or during welding, for inspection and tracking purposes.
This same camera is also capable of viewing and recording (currently to 100 fps) a coaxial image of the interaction point between the incident laser beam and the materials being welded. For additional assurance, the head itself also monitors and records the temperatures of key optical components, as well as the health (scattered light and temperature) of the cover slide.
TWI is in the process of designing and executing an iterative series of welding trials using this equipment on materials and material thicknesses of industrial relevance to GKN, to ascertain the performance of this system and guide its further development.
A further development is that the system is also capable of communicating with, and recording data from (at a rate of up to 1000Hz), a sensor-enabled fibre. Using an appropriate fibre, signals such as the amount of back-scattered laser radiation and optical emissions from the process zone can also be monitored during welding.
With appropriate system training, post-weld video playback and sensor signal evaluations can then be used for weld quality and/or fault determination. Signal- and image-analysis routines also provide the means for real-time fault diagnosis.
These analyses indicate that the InnoSeam system will be capable of detecting a number of possible problems and faults during laser welding, including:
- Incorrect seam position.
- Poor joint fit-up.
- Cover slide contamination.
- Incorrect beam power being used during welding.
- Changes between partial and full penetration.
- Incorrect focus being used during welding.
- Inadequate gas shielding.
For more information, please email firstname.lastname@example.org