The Research Of Online Detecting On Can Body Weld Defects Based On Eddy Current Principle

Resistance welding machine is commonly used for welding work on can body. With improper operation, it will cause the weld defects such as rosin joints or flaws. Currently, there is still a lack of reliable online nondestructive detecting methods for can body weld defects. The defective products are easy to mix to the qualified products, which will lead to the difficulty of subsequent process. Therefore,the research of this subject is very important.The weld defects on the surface of the can body will change its conductivity and magnetic permeability. Eddy current sensor is the device that can detect the defects. According to the results of measurement, we can estimate whether there is defect and the size of the defect. So this paper uses eddy current sensor to detect defects.The main work of this paper includes four aspects. The first aspect is to model the sensor in the ANSYS software and study the influence of structure parameters change on the sensor linearity and sensitivity, so as to complete the optimization design of the sensor which is applicable to the system. The second aspect is to design the system’s signal processing circuit design, which includes the oscillation circuit, amplifying circuit, filter circuit and the other functional circuit modules. The third aspect is to build the system test platform and the data acquisition and processing system, complete the design of the upper computer and the lower computer. The forth aspect is to do the defect analysis and the defect recognition research, realize the judgement to the location and number of the defects, and then analyze the influence of the defect size change on the output signal. Based on the above four aspects, this paper complete the research on the main problems of the subject.