An Analytical Switching Process Model of Low-Voltage eGaN HEMTs for Loss Calculation

Abstract

This paper proposes an improved analytical switching process model to calculate the switching loss of low-voltage enhancement-mode Gallium Nitride high-electron mobility transistors (eGaN HEMTs). The presented eGaN HEMTs models are more or less derived from silicon MOSFETs models, whereas eGaN HEMTs are different from three aspects: higher switching speed, much more reduced parasitic inductance in switching loop, and absence of reverse recovery. Applying the traditional model to eGaN HEMTs results in inaccurate prediction of switching waveforms and losses. The proposed model considers the effect of low-parasitic inductances, nonlinearity of junction capacitances, and nonlinearity of transconductance. The turn-on and turn-off switching processes are described in detail and the resulting equations can be easily solved. The accuracy of the proposed model is validated by comparing the predicted switching waveforms and converter's efficiency with the experimental results, respectively. Based on the analytical model, the effects of gate resistance, gate supply voltage, and parasitic inductances on switching losses are investigated. Meanwhile, a novel current measuring method based on magnetic coupling is proposed to measure the switching current waveform with improved accuracy.