| MS TSA Meeting

Name of the Speaker: Mr. SATWIK KOMMA (EE21S090)
Guide: Dr. Lakshminarasamma N
Online meeting link: http://meet.google.com/ewh-ebbe-ecd
Date/Time: 19th June 2025 (Thursday), 11 AM
Title: Dead time optimisation and turn off loss modelling of a SiC MOSFET based Series Resonant converter.

Abstract :

Series resonant converters are widely used for high power applications with wide load variations. In resonant converters, changes in operating conditions such as input voltage and load current will significantly impact the turn off behaviour and turn off switching losses induced in the Silicon Carbide Metal Oxide Semiconductor Field Effect Transistor (SiC MOSFET). Hence, a fixed dead time would not ensure complete zero voltage switching (ZVS) and minimum body diode conduction losses, thereby reducing the efficiency of the DC-DC converter.

This thesis proposes a method to compute dead time adaptively for a typical half bridge circuit in a series resonant converter. The proposed model uses forward euler method to iteratively compute the drain to source voltage of the top and bottom MOSFETs in a typical half bridge circuit, the gate to source voltage, and the drain current through the top and bottom SiC MOSFETs and uses them as state variables to determine the turn off time, the optimal dead time, and the turn off losses. The proposed model is used to determine the turn-off characteristics, also considering external circuit parasitics, device parasitics, and gate driver parameters to accurately estimate the turn off time and optimal dead time. The proposed analytical model has been validated with the turn off losses and turn off times obtained from MATLAB Simulink model of a half bridge configuration for two different SiC MOSFETs with voltage and current ratings of 1200 V, 36 A and 1200 V, 17.7 A, respectively.

The proposed analytical model used to compute the turn off time, optimal dead time, and turn off losses is verified experimentally through a double pulse test setup under varying load and line conditions, and the measured turn off characteristics are in close agreement with turn off characteristics obtained from the analytical model, thus verifying the accuracy of the proposed analytical model. The proposed analytical model is also verified with the turn off characteristics obtained from a 600 V, 1.2 kW series resonant DC-DC converter for varied load conditions to verify the zero voltage switching obtained across the SiC MOSFET switches in a half bridge configuration by placing optimal dead time computed from the analytical model and by adaptively computing the dead time, the body diode conduction losses are reduced, and hard switching is avoided.