Detection of flip chip solder joint cracks using correlation coefficient and auto-comparison analyses of laser ultrasound signals

Abstract

The detection of cracks in solder joints has been a challenge for the current nondestructive inspection techniques. This paper investigates the capability of a laser ultrasound and vibrometer inspection system to detect thermal cycle induced cracks in flip chip solder joints. Correlation coefficient analysis of ultrasound vibration signals is presented and compared with the previous error ratio method. The correlation coefficient method improves the system's signal-to-noise ratio. Optimum detection locations on the chip surface are investigated and an auto-comparison method is introduced to inspect a chip with fewer detection points and without the dependence on a reference chip. The results show that the system is able to identify thermally cracked solder joints from well-connected joints. When fully developed, the system can be put in-line and serve as a low-cost go/no-go inspection tool to screen out defects in solder joints at an early stage in the manufacturing process. It can also be used to evaluate the quality of many other types of electronic packages and other devices.