AI-Enabled Board Level Vibration Testing: Unveiling The Physics of Degradation

The stringent reliability requirements of electronic packages for safety-critical automotive applications have spurred developments in real-time monitoring of electronic components. A key aspect of these advancements is the availability of physical health sensing elements and failure-predicting algorithms that can be embedded within the integrated circuit. In this paper, 4-wire resistance measurement features are embedded in Quad-flat no-leads (QFN) packages to detect physical damages at the printed circuit board (PCB)-solder interconnect interface. Additionally, several Artificial Intelligent (AI) algorithms are assessed, and the most suitable one is implemented to predict the in-situ resistance changes over time under vibration loads. The time series failure forecast from this algorithm correlates well to the experimentally determined lifetime of solder joints. This method opens avenues for investigating the physics of degradation in board level reliability.