Defect detection of flip-chip solder joints using modal analysis

Abstract

The reliability of solder joints is a critical element for the successful application of flip chip technology. This paper presents a nondestructive inspection method using modal analysis combined with image processing for the defect detection of flip-chip solder joints. Simulations are implemented to explore the feasibility of applying the chip modal shapes and resonance frequencies for defect detection. In experimental investigations, the signal generator and power amplifier are applied to drive the capacitive air-coupled ultrasonic transducer to produce continuous ultrasonic waves for exciting the test chips. The vibration responses of the chips are measured by the laser scanning vibrometer in order to reconstruct modal shapes and extract resonance frequencies. The chip modal shapes are analyzed by virtue of image processing technique, and the resonance frequencies are investigated through frequency spectra. The results reveal that the defective chips can be distinguished from the good chips by the modal shape differences represented by the characteristic angles, and the chip resonance frequencies can also serve as an indicator for defect detection, which prove the feasibility of this method.