Technical Information Site of Power Supply Design

2021.09.29 Simulation

# PFC Circuits: Examination of Optimum Dead Time Values

Optimization of PFC Circuits

PFC Circuits: Examination of Optimum Dead Time Values

Here we examine how best to estimate the optimum dead time in a bridge circuit.

Example Circuit

As an example circuit, the simulation circuit“A-6. PFC CCM Synchro Vin=200V Iin=2.5A” in the Power Device Solution Circuit/AC-DC PFC list is used (see Fig.16). Details of the circuit diagram can also be seen here.

This circuit performs synchronous rectification, and so simulations were used to study optimum values for the dead times of the SCT2450KE SiC MOSFETs on the high side (HS) and low side (LS), that is, the shortest times for which shoot-through currents do not occur. The respective dead times are set using the parameters TD1 (HS) and TD2 (LS) of the PWM controllers in the simulator.

Fig. 16: PFC simulation circuit” A-6. PFC CCM Synchro Vin=200V Iin=2.5A”

Fig. 17 shows the flow of current during a dead time period. In a circuit with a bridge configuration, dead times must be kept long enough to ensure that shoot-through currents are prevented, but if they are set unnecessarily long, losses increase. This is because during a dead time period a SiC MOSFET is in the off state, so that current flows through the body diode. Body diodes generally have large conduction losses, and if a body diode is conducting for a long time period, losses increase.

Fig. 17: Current flowing during a dead time period

Fig. 18 shows the relation between the dead time length and the inductor current IL. If the dead time is too long, operation in the low-voltage region becomes discontinuous, the inductor current waveform is distorted, and the power factor is worsened. Hence setting the dead time to be unnecessarily long is disadvantageous from the standpoint of the power factor as well.

Fig. 18: Relation between dead time length and inductor current IL

Fig. 19 shows loss simulation results for a SiC MOSFET when the dead time is changed.

Fig. 19: Loss simulation results for a SiC MOSFET in which the dead time is changed

When the dead time is 50 ns or less, the flowing of a shoot-through current causes losses to increase sharply. Conversely, if the dead time is too long, the conduction time of the body diode in the HS SiC MOSFET is lengthened, and under these conditions as well, losses increase. SiC MOSFET losses are minimized when the dead time is the shortest possible time without the occurrence of a shoot-through current; in this example, we see that this optimum dead time is 100 ns. However, the switching speed varies depending on the temperature, variation in the manufacturing lot, and other factors, and so in general a margin of about 100 ns must be included. Hence in this case, 200 ns should be regarded as the optimum dead time.

#### Key Points:

・The dead time settings in a bridge circuit are related to losses and safety, and so must be studied thoroughly.

・The optimum value for the dead time is the shortest time at which a shoot-through current does not occur.

・The switching speed of a switching element varies depending on such factors as the temperature and variation in the manufacturing lot, and so a margin must be added to this shortest time in the design stage.