Experimentally Extracting Multiple Spatial Thermal Models that Accurately Capture Slow and Fast Properties of Assembled Power Semiconductor Converter Systems

Models describing transient heat transfer within power electronics are commonly formed using step response curves and numerical models. Unfortunately, data sheet transient thermal impedance curves only characterize the junction-to-case spatial domain segment. Also, computationally intense transient numerical models require a detailed specification procedure. This paper presents an experimental methodology to capture thermal response properties of assembled power semiconductor converter systems, including fast (die-level) and slow (sink-level) modes. The method utilizes semiconductor devices as modulators of dynamic loss (heat) injections, and discrete temperature measurements, to form multiple thermal impedance models using the frequency response function (FRF) metric. Complementary investigation of numerical and analytical models yields a general, compact thermal model topology including a physics-based, spatiotemporal delay term, critical for correctly interpreting FRFs. The evaluation utilizes a power converter having integrated temperature sensing. Ambient-referred and relative FRF results are obtained across wide frequency ranges and are parameterized using the compact thermal model.

[1]  J.D. van Wyk,et al.  Void induced thermal impedance in power semiconductor modules: some transient temperature effects , 2001, Conference Record of the 2001 IEEE Industry Applications Conference. 36th IAS Annual Meeting (Cat. No.01CH37248).

[2]  Kexin Wei,et al.  Online Calculation of the Increase in Thermal Resistance Caused by Solder Fatigue for IGBT Modules , 2017, IEEE Transactions on Device and Materials Reliability.

[3]  P. Welch The use of fast Fourier transform for the estimation of power spectra: A method based on time averaging over short, modified periodograms , 1967 .

[4]  Robert D. Lorenz,et al.  Control of Junction Temperature and Its Rate of Change at Thermal Boundaries via Precise Loss Manipulation , 2017, IEEE Transactions on Industry Applications.

[5]  Bongtae Han,et al.  Physics-of-Failure, Condition Monitoring, and Prognostics of Insulated Gate Bipolar Transistor Modules: A Review , 2015, IEEE Transactions on Power Electronics.

[6]  Arne Hinz,et al.  Spatial electro-thermal modeling and simulation of power electronic modules , 2016, 2016 IEEE Energy Conversion Congress and Exposition (ECCE).

[7]  P. W. Webb,et al.  Transient thermal impedance measurement in power semiconductor devices , 1996 .

[8]  Mounira Berkani,et al.  Ageing and Failure Modes of IGBT Modules in High-Temperature Power Cycling , 2011, IEEE Transactions on Industrial Electronics.

[9]  F. Blaabjerg,et al.  Frequency-Domain Thermal Modeling and Characterization of Power Semiconductor Devices , 2016, IEEE Transactions on Power Electronics.

[10]  William A. Sethares,et al.  Thermal parameter estimation using recursive identification , 1991 .

[11]  Marco Liserre,et al.  Computational light junction temperature estimator for active thermal control , 2016, 2016 IEEE Energy Conversion Congress and Exposition (ECCE).

[12]  Dawei Xiang,et al.  An Industry-Based Survey of Reliability in Power Electronic Converters , 2011, IEEE Transactions on Industry Applications.