Monitoring of Resistance Spot Welding Process.

Resistance spot welding (RSW) is one of the common assembling process of thin metals. Due to the inaccessibility of nugget in inspection process, usually destructive tests are used to assess weld quality which is time consuming and costly. Finding a practical solution for monitoring of RSW would reduce necessity of these tests for quality control. In this study, a new designed algorithm is used for processing of a video recorded by a high speed camera from resistance spot welding process, to determine the displacement of the electrode (DE), which can be easily applied to any similar process. It was found that the diagram of DE can provide appropriate information about the ongoing events during the resistance spot welding process and gives a suitable tool for monitoring and controlling RSW. The effect of environmental conditions on the DE diagram was also investigated. Impact of expanded heat, melting and expulsion on displacement diagram during the process was discussed. The results showed that current and time have similar trend in DE, higher pressure will result in diagram's compaction. It was also concluded that shunt and contamination have considerable effects on displacement diagram and strengths of weld. Using holders as external constraint can improve weld strength for about 12%. An artificial neural network (ANN)-based program was developed to introduce a relation between displacement diagram and weld quality. In most cases, the predicted values are closed to the experiments. 36 different input modes for welding were investigated in this regard. It is shown that by applying the developed code on DE graph the weld quality can be predicted without need to any destructive test. [ABSTRACT FROM AUTHOR]