Electromigration-Aware Reliability Optimization of MCPM Layouts Using PowerSynth

2022 IEEE Energy Conversion Congress and Exposition (ECCE)(2022)

引用 0|浏览3
暂无评分
摘要
Modern power electronics are experiencing significant demand for ultra-high-power density in the grid, data center, automotive industries, and aerospace applications. To satisfy the increasing demand, innovative packaging technologies for multi-chip power modules (MCPMs) are proposed by leveraging the advantages from wide bandgap (WBG) devices (i.e., SiC, GaN). Since higher density modules are more vulnerable to electromigration (EM) risk, the MCPM layout optimization requires EM consideration along with electro-thermal aspects. In this paper, an EM-aware reliability optimization methodology is proposed and implemented in PowerSynth. Mean time to failure (MTTF) is used as the reliability metric for EM risk assessment. MTTF estimation needs a detailed current density and temperature distribution in a relatively fast method that can be used in an optimization loop. Our approach has shown 1,100 times speedup in current density extraction compared to the FEM simulation. MCPM with wire bonding and solder bump arrays are optimized using the proposed workflow. Experimental results are used to predict MTTF for wire-bonded module case.
更多
查看译文
关键词
mcpm layouts,reliability,electromigration-aware
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要