6 Schottky Diode Method

Another method for addressing bootstrap overcharge is to place a diode parallel to the lower GaN FET. In this configuration, the diode behaves like the body diode of a MOSFET. The diode limits the negative HS voltage to the VF of the diode, which is often less than 1 V. Reducing the negative voltage on HS reduces bootstrap overcharging and dead time losses in the low-side GaN FET. Schottky diodes have better reverse recovery and VF performance compared P-N diodes, and are favored in this application.

Figure 6-1shows that the GaN FET reaches a higher dv/dt and switches faster than the GaN FET with a parallel diode. The GaN FET reaches approximately –1.8 V on HS, while the GaN FET with the diode only reaches approximately –0.6 V. Therefore, the Schottky diode can effectively limit the negative HS voltage and prevent overcharge. The downside to using this Schottky diode is the addition of extra capacitance to the HS node, which results in increased switching time and losses.

Schottky diodes are less feasible in higher-voltage systems. Silicon Schottky diodes (SBDs) are available with blocking voltages of 100–200 V but start to lose their advantages over P-N diodes at higher voltages. SiC SBDs are becoming popular for higher voltages and can reach blocking voltages past 1200 V. However, these diodes have a higher forward voltage (greater than 1.3 V), which can be too high to prevent overcharge.

Systems with very high loads or load transients require special considerations. For example, in a 3-kW, 48-V to 12-V DC/DC converter, the load current is approximately 250 A, requiring a very large (and expensive) Schottky diode. Interleaving reduces the current requirement on each diode but requires more diodes. Adding Schottky diodes is very expensive in terms of cost and board size.