Improved SiC Power MOSFET Model Considering Nonlinear Junction Capacitances

Silicon carbide (SiC) power metal-oxide-semiconductor field-effect transistors (MOSFETS) have been applied in high-power and high-frequency converters recently. To effectively predict characteristics of SiC power MOSFETS in the design phase, a simple and valid model is needed. In this paper, a simpl...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on power electronics 2018-03, Vol.33 (3), p.2509-2517
Hauptverfasser: Duan, Zhuolin, Fan, Tao, Wen, Xuhui, Zhang, Dong
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Silicon carbide (SiC) power metal-oxide-semiconductor field-effect transistors (MOSFETS) have been applied in high-power and high-frequency converters recently. To effectively predict characteristics of SiC power MOSFETS in the design phase, a simple and valid model is needed. In this paper, a simple improved SiC power MOSFET behavioral model is proposed using SPICE language. Key parameters in the model are analyzed and determined in detail, including parasitic parameters of the power module, steady-state characteristic parameters, and nonlinear parasitic capacitances. The effect of negative turn-off gate drive voltage is considered and a continuously differentiable function is proposed to describe the gate-source capacitance. Experimental validation is performed under a double pulse circuit employing an N-channel power MOSFET half-bridge module CAS300M12BM2 (Cree Inc.) rated at 300 A/1200 V. The main switching dynamic characteristic parameters of the model have been compared with those of the measured results. The results show that taking gate-source capacitance as a linear value as most previous models do will cause significant turn-on deviations between experiment and simulation results, while the improved model is more accurate compared with the measured results.
ISSN:0885-8993
1941-0107
DOI:10.1109/TPEL.2017.2692274