APPLICATION-AWARE ESTIMATION OF THE JUNCTION TEMPERATURE SWING UNDER ACTIVE CYCLING

This paper presents a methodology for application-aware estimation of the junction temperature swing of smart power devices under active cycling, based on electro-thermal simulation. According to the most known lifetime model, i.e. the Coffin-Manson law, the maximum junction temperature swing within...

Full description

Saved in:
Bibliographic Details
Published inScientific Bulletin. Series C, Electrical Engineering and Computer Science no. 1; p. 131
Main Authors Pop, Ciprian V, Buzo, Andi, Diaconu, Cristian, Pelz, Georg, Cucu, Horia, Burileanu, Corneliu
Format Journal Article
LanguageEnglish
Published Bucharest University Polytechnica of Bucharest 01.01.2020
Subjects
Coffin-Manson law
Computer simulation
Cycles
Electronic devices
Service life assessment
Thermal simulation
Online AccessGet full text

Cover

More Information
Summary:This paper presents a methodology for application-aware estimation of the junction temperature swing of smart power devices under active cycling, based on electro-thermal simulation. According to the most known lifetime model, i.e. the Coffin-Manson law, the maximum junction temperature swing within a fast thermal cycle (∆T) represents the main stressfactor of the lifetime. Consequently, for application-aware lifetime estimation of power devices an important step is to estimate ∆T in the space of application operating conditions. The proposed application-aware ∆T model is fitted from data on a grid of electro-thermal simulations based on estimated power profiles over a wide space of operating conditions. The model is validated on electro-thermal simulation data based on the power profiles corresponding to the active cycling experimental scenarios, showing a maximum unsigned relative prediction error of 6.56%. With the proposed model, the dependencies of ∆T with the operating parameters can be observed and predictions on different application operating conditions can be done.
ISSN:2286-3540