Two-dimensional analysis for implementing nondestructive crack detection system in automotive production line

Abstract

According to the automotive market trend, the vehicle machine needs to the electronic components. In the automotive body panel, appearance of cracks is one of the most serious challenge. Sometimes surface cracking will cause an unexpired expenses in production process. The most common method used to detect the crack in automotive press line is a visual scanning of objects, parts or components which is the oldest and reliable non-destructive testing method. This test method is applied to almost every automotive product as a quality assurance.However, the specific optimization method with great accuracy the time and effort is required for high speed throughput in an automated press line system. An acoustic emission is a technique which centered on the concept of utilizing the transducer action of a flaw in a stressfield. This technique was used to investigate fatigue crack characteristics such as initiation closure and propagation on smooth specimens. It is shown that acoustic emission from unflawedtensile specimens can be treated from a dislocation dynamics approach. The choice of analytical method is extremely important and should not only focus on high-accuracy crack detection, but should also low-cost with high-efficiency in this system. In cases where crack is expected or necessary, the analytical method should detect the acoustic emission signal while the accuracy of the resulting measurements should fall within an acceptable range. The purpose of the detection system is acquisition correctly and to verify accuracy of the measurements. When the accuracy of crack detection falls out of predetermined acceptability criteria, usually within 20% accuracy, the measured data should be reanalyzed by using other methods, if necessary. The system consists of two parts: the hardware and the DSP (digital signal processing) part which includes AE parameter analyzer, based on the LabVIEW program. The crack acquisition system is set to sampling rate of 300 KHz with 20dB pre–amplification. As a result, maximum received frequency range is 150 kHz according to the field test. Operating temperature is -40˚C ~ +85˚C considering the severe press factory environment with 7 seconds to analyze the data. The proposed system was tested and successfully demonstrated crack detection in an actual automotive production line.