No abstract

Acoustic microimaging (AMI) has been widely used to nondestructively evaluate microelectronic packages for the presence of internal defects. To detect defects in small devices such as BGA, flip-chip, and chip-scale packages, high acoustic frequencies are required for the conventional AMI systems. The acoustic frequency used in practice, however, is limited by its penetration through materials. In this paper, a novel acoustic microimaging technique, which utilizes nonlinear signal processing techniques to improve the resolution and robustness of conventional AMI, is proposed and investigated. The technique is based on the concept of sparse signal representations in overcomplete time-frequency dictionaries. Simulation and experimental results show its super resolution and high robustness.Index Terms-Acoustic microimaging (AMI), microelectronic packages, overcomplete dictionaries, sparse signal representations.

PurposeThe purpose of this paper is to evaluate the application of an acoustic micro‐imaging (AMI) inspection technique in monitoring solder joints through lifetime performance and demonstrate the robustness of the monitoring through analysis of AMI data.Design/methodology/approachAccelerated thermal cycling (ATC) test data on a flip chip test board were collected through AMI imaging. Subsequently, informative features and parameters of solder joints in acoustic images were measured and analysed. Through analysing histogram distance, mean intensity and grey area of the solder joints in acoustic images, cracks between the solder bump and chip interface were tracked and monitored. The results are in accord with associated Finite Element (FE) prediction.FindingsAt defective bumps, the formation of a crack causes a larger acoustic impedance mismatch which provides a stronger ultrasound reflection. The intensity of solder joints in the acoustic image increase according to the level of damage during the ATC tests. By analysing the variation of intensity and area, solder joint fatigue failure was monitored. A failure distribution plot shows a normal distribution pattern, where corner joints have the lowest reliability and are more likely to fail first. A strong agreement between AMI monitoring test data and FE prediction was observed, demonstrating the feasibility of through lifetime monitoring of solder joints using AMI.Originality/valueThe paper indicates the feasibility of the novel application of AMI inspection to monitor solder joint through lifetime performance non‐destructively. Solder joints' real life conditions can be tracked by an AMI technique, hence solder joint fatigue failure cycles during the ATC tests can be monitored and analysed non‐destructively.

Sparse signal representations from overcomplete dictionaries are the most recent technique in the signal processing community. Applications of this technique extend into many fields. In this paper, this technique is utilized to cope with ultrasonic flaw detection and noise suppression problem. In particular, a noisy ultrasonic signal is decomposed into sparse representations using a sparse Bayesian learning algorithm and an overcomplete dictionary customized from a Gabor dictionary by incorporating some a priori information of the transducer used. Nonlinear postprocessing including thresholding and pruning is then applied to the decomposed coefficients to reduce the noise contribution and extract the flaw information. Because of the high compact essence of sparse representations, flaw echoes are packed into a few significant coefficients, and noise energy is likely scattered all over the dictionary atoms, generating insignificant coefficients. This property greatly increases the efficiency of the pruning and thresholding operations and is extremely useful for detecting flaw echoes embedded in background noise. The performance of the proposed approach is verified experimentally and compared with the wavelet transform signal processor. Experimental results to detect ultrasonic flaw echoes contaminated by white Gaussian additive noise or correlated noise are presented in the paper.

No abstract