Back to Search Start Over

A physical lifetime prediction methodology for IGBT module by explicit emulation of solder layer degradation.

Authors :
Wu, Xinlong
Yang, Xin
Dai, Xingyu
Tu, Chunming
Liu, Guoyou
Source :
Microelectronics Reliability. Dec2021, Vol. 127, pN.PAG-N.PAG. 1p.
Publication Year :
2021

Abstract

Physical lifetime prediction methods based on inelastic strain energy density (energy-based lifetime prediction) have been proven to be effective in producing reliable results on small solder joints (ball-grid arrays or chip-scale package) due to their capabilities to represent physical characteristics of soldering materials and loading history. However, in the existing energy-based lifetime models for solder layer failure the inelastic strain energy density is usually assumed to unchanged during aging, which is untrue particularly for large-area solder joints when crack propagation is considered. In this paper, an energy-based lifetime prediction method is proposed for failure of IGBT modules by explicit emulation of solder layer degradation. Inelastic strain energy density variances with respect to aging are meticulously modeled. The large-area solder layer is evenly divided into meshed elements, birth and death technique in finite element analysis (FEA) is used to successively emulate crack propagation. In this way, the increment of thermal resistance due to the solder layer degradation can be predicted. The lifetime prediction results by our proposed method show nice agreement with the experiment under different aging conditions. • An energy-based lifetime prediction method is proposed for failure of IGBT modules • Birth and death technique in FEA is used to successively emulate crack propagation • The increment of thermal resistance due to the solder layer degradation can be predicted • ΔW will change with different aging stage [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
00262714
Volume :
127
Database :
Academic Search Index
Journal :
Microelectronics Reliability
Publication Type :
Academic Journal
Accession number :
153681010
Full Text :
https://doi.org/10.1016/j.microrel.2021.114384