This study introduces an effective and efficient dynamic electro-thermal coupling analysis (ETCA) approach to explore the electro-thermal behavior of a three-phase power metal–oxide–semiconductor field-effect transistor (MOSFET) inverter for brushless direct current motor drive under natural and forced convection during a six-step operation. This coupling analysis integrates three-dimensional electromagnetic simulation for parasitic parameter extraction, simplified equivalent circuit simulation for power loss calculation, and a compact Foster thermal network model for junction temperature prediction, constructed through parametric transient computational fluid dynamics (CFD) thermal analysis. In the proposed ETCA approach, the interactions between the junction temperature and the power losses (conduction and switching losses) and between the parasitics and the switching transients and power losses are all accounted for. The proposed Foster thermal network model and ETCA approach are validated with the CFD thermal analysis and the standard ETCA approach, respectively. The analysis results demonstrate how the proposed models can be used as an effective and efficient means of analysis to characterize the system-level electro-thermal performance of a three-phase bridge inverter.

Temperature resulting from the joule heating power and the turn-on and turn-off dissipation of high-power, high-frequency applications is the root cause of their thermal instability, electrical performance degradation, and even thermal-fatigue failure. Thus, the study presents thermal and electrical characterizations of the power MOSFET module packaged in SOT-227 under natural convection and forced convection through three-dimensional (3D) thermal-electric (TE) coupled field analysis. In addition, the influences of some key parameters like electric loads, ambient conditions, thermal management considerations (heat sink, heat spreader) and operation conditions (duty cycle and switching frequency) on the power loss and thermal performance of the power module are addressed. The study starts from a suitable estimation of the power losses, where the conduction losses are calculated using the temperature- and gate-voltage-dependent on-state resistance and drain current through the device, and the switching losses are predicted based on the ideal switching waveforms of the power MOSFETs applied. The effectiveness of the theoretical predictions in terms of device’s power losses and temperatures is demonstrated through comparison with the results of circuit simulation and thermal experiment.