The development of a method and apparatus to detect lap and butt welds in a moving steel strip

Abstract

The first objective of this investigation was to determine the physical differences between the weld and the base metal. The physical differences were used to create amplitude differences in the retained fields of the magnetized weld and the. magnetized base metal. The difference in mechanical hardness between the weld and base metal was the distinguishing characteristic of the butt welded samples. This property was related to the retained fields and experimentally verified. The difference in thiclmess between the weld and base metal was the outstanding property of the lap welded samples. This property was also related to the retained fields and was experimentally verified. The second objective of this investigation was to test several different sensing devices. Some of the devices were specially constructed, while others were purchased. All sensors were tested in the laboratory on a rotating disk with a butt weld through the center. All the devices were evaluated and the sensor that had the best characteristics was selected to be used for the experimental weld detector. The third objective was to design and construct an experimental weld detector. The circuit for the detector was designed and constructed to compensate automatically for different velocities of the welded strip. The compensation was for frequency and amplitude changes of the sensor signal due to different velocities. The last objective was to test the detector and to evaluate the basic concepts and experimental results of this investigation. The fields of the magnetized welds of all the samples tested were related to the hardness and thickness of the welds, All the welds tested exhibited the predicted characteristics needed for detection. Before this method could be considered reliable enough for a permanent mill installation many more welds should be tested in the laboratory. The test results of the experimental weld detector were very encouraging. The detector compensated very well for amplitude and frequency changes of the sensor signal due to velocity changes. In conclusion, if after testing many more weld samples, it is found that all of them exhibit the required characteristics, then a weld detector, based on the findings of this investigation, is definitely feasible.