Generally, the switching speed of the power switch during turnon and turnoff should be as fast as possible to minimize switching power losses. The gate driver device must be able to provide the required peak current for achieving the targeted switching speeds for the targeted power MOSFET.
Using the example of a power MOSFET, the system requirement for the switching speed is typically described in terms of the slew rate of the drain-to-source voltage of the power MOSFET (such as dVDS/dt). For example, the system requirement might state that a SPP20N60C3 power MOSFET must be turned on with a dVDS/dt of 20 V/ns or higher under a DC bus voltage of 400 V in a continuous-conduction-mode (CCM) boost PFC-converter application. This type of application is an inductive hard-switching application and reducing switching power losses is critical. This requirement means that the entire drain-to-source voltage swing during power MOSFET turn-on event (from 400 V in the OFF state to VDS(on) in on state) must be completed in approximately 20 ns or less. When the drain-to-source voltage swing occurs, the Miller charge of the power MOSFET (QGD parameter in SPP20N60C3 power MOSFET data sheet = 33 nC typical) is supplied by the peak current of gate driver. According to power MOSFET inductive switching mechanism, the gate-to-source voltage of the power MOSFET at this time is the Miller plateau voltage, which is typically a few volts higher than the threshold voltage of the power MOSFET, VGS(TH).
To achieve the targeted dVDS/dt, the gate driver must be capable of providing the QGD charge in 20 ns or less. In other words a peak current of 1.65 A (= 33 nC / 20 ns) or higher must be provided by the gate driver. The UCC2753x series of gate drivers can provide 2.5-A peak sourcing current, and 5A peak sinking current which clearly exceeds the design requirement and has the capability to meet the switching speed needed. The 1.5x sourcing, and 3x sinking overdrive capability provides an extra margin against part-to-part variations in the QGD parameter of the power MOSFET along with additional flexibility to insert external gate resistors and fine tune the switching speed for efficiency versus EMI optimizations. However, in practical designs the parasitic trace inductance in the gate drive circuit of the PCB will have a definitive role to play on the power MOSFET switching speed. The effect of this trace inductance is to limit the dI/dt of the output current pulse of the gate driver. To illustrate this, consider output current pulse waveform from the gate driver to be approximated to a triangular profile, where the area under the triangle (½ ×IPEAK × time) would equal the total gate charge of the power MOSFET (QG parameter in SPP20N60C3 power MOSFET datasheet = 87 nC typical). If the parasitic trace inductance limits the dI/dt then a situation may occur in which the full peak current capability of the gate driver is not fully achieved in the time required to deliver the QG required for the power MOSFET switching. In other words, the time parameter in the equation would dominate and the IPEAK value of the current pulse would be much less than the true peak current capability of the device, while the required QG is still delivered. Because of this, the desired switching speed may not be realized, even when theoretical calculations indicate the gate driver can achieve the targeted switching speed. Thus, placing the gate driver device very close to the power MOSFET and designing a tight gate drive-loop with minimal PCB trace inductance is important to realize the full peak-current capability of the gate driver.