ABICOR Innovation Award 2008 (2nd Place)
Dipl.-Ing. Konrad Willms (RWTH Aachen)
Dr.-Ing. Guido Wilms (RWTH Aachen)
Approach for the use of arc sensors for welding head guides when MIG welding aluminium alloys
Innovative light-weight construction using aluminium materials, and the associated joining tasks, are constantly gaining more significance in the development of technical products, for example, in vehicle construction. For quality assurance purposes, the automaton of welding tasks in the steel area of the arc sensor is, on the basis of the high reliability and good guide accuracy, ideal for compensation of tolerances in the joining zone. However, at the moment, the use of the arc sensor as a welding head guide when welding aluminium is not possible. This is because of the conside-rably lower electrical resistance of aluminium compared to steel and the fact that the sensor signal is reduced as a result of this.
Within the scope of the development work, a new evaluation method has been developed, specially adapted to the signal characteristics during aluminium welding, which take into account, the peculiar characteristics of aluminium materials and provides the basis for height and side guiding of the torch when welding aluminium. The prerequisite for the use of this evaluation method is, initially, the reduction of interferences. These include an oxide skin with a constant layer thickness, a stable inert gas, as well as a defined current transfer in the contact pipe due to forced contact.
To detect changes in distance, the increases in the pulse phase of the current and voltage profiles contain significant information. Thus, with both U-I and I-I modulated welding processes, the change in slope in the pulse signal can be reproduced to indicate a change in the distance of the torch. The average values of the pulse and base phase in the signal profiles are used to determine the arc length and thus provide important information for the side guide of the welding torch. Taking these relationships into account, an evaluation algorithm has been developed which can be used in conjunction with a robot to guide welding heads on a T-joint which has provided successful test results in the laboratory.