Research on Current Sharing Method for SiC MOSFETs in Parallel Based on Adjustable Gate Voltage Driver

The Solid-State Power Controller (SSPC) is an intelligent switch composed of power switches, voltage-current sensors, and a microcontroller. It integrates functions such as overvoltage protection, overcurrent protection, status reporting, and remote control. It is employed in industries with high reliability demands to perform on-off control of electrical loads. Due to the high voltage withstand capability, low on-resistance, and superior temperature tolerance of SiC MOSFETs, they are preferred as power switches in solid-state power controller to enhance their reliability. However, individual SiC MOSFETs have limited current-carrying capacity, necessitating the use of multiple devices in parallel to increase their current-carrying capability. Nevertheless, due to the inherent parameter variations among devices, straightforward parallel connection of silicon carbide devices fails to achieve automatic current sharing over the entire temperature range. Consequently, prolonged operation leads to thermal imbalances, severely impacting the reliable operation of the devices. Indeed, achieving current sharing in parallel SiC MOSFETs is crucial for ensuring the safe operation of the system. In light of this, a gate voltage adjustable active current sharing method for SiC MOSFETs has been proposed. This method takes advantage of the temperature-dependent on-resistance characteristic of SiC MOSFETs and introduces a novel SiC MOSFET driver circuit. By independently adjusting the gate voltage of each SiC MOSFET within the parallel configuration, the proposed method achieves effective current sharing control. By employing the Linear Active Disturbance Rejection Control (LADRC) algorithm, control precision and response speed are enhanced, allowing for precise adjustment of the gate voltage. Consequently, this approach significantly improves the balancing and reliability of the parallel operation of SiC MOSFETs. Through simulation and experimentation, the effectiveness and feasibility of the proposed method have been validated successfully.