What Determines Reverse Recovery Time in MOSFETs

Just like all other transistors, MOSFETs are used as electrically controlled switches, but their switching speed and low ON-state resistance make them very useful in advanced power electronics. MOSFETs have a built-in body diode that prevents them from blocking current in the reverse direction. When the gate is toggled with a square wave signal, the blocking nature is repeatedly toggled.

As the gate driving switches the MOSFET’s operation between reverse and forward operation, there is a delay known as reverse recovery time. What determines this time lag and how does it affect operation? We’ll examine the losses associated with this transition and how to balance this specification when choosing MOSFETs.

Watch For the Reverse Recovery Time

All diodes exhibit losses during switching, and in particular during the period known as reverse recovery, i.e., while it switches between forward bias and reverse bias. The same applies to the body diode in an MOSFET. Reverse recovery losses result from the excitation and movement of charge carriers in the channel; as the MOSFET is switched to the OFF state and forward current decreases, remnant charge carriers are swept out in the reverse direction and flow in reverse over the body diode.

Because the charge carriers are essentially flowing in reverse over the body diode, they create significant losses during the reverse switching time. To estimate the average reverse recovery loss during switching, we can use the total reverse recovery charge (Q), peak drain-source voltage (V), and switching frequency (f):

P = ½QVf

Typical reverse switching times for power MOSFETs can be as high as 100 ns, during which time the reverse recovery charge will be swept out of the body diode. This value is the total charge that moves along the body diode during reverse recovery and it will be specified for rate of change in the drain-source current (di/dt).

How Much Loss is Too Much?

The amount of acceptable reverse recovery loss depends on the thermal resistance of the package as this will determine how hot the junction becomes due to repeated switching. For example, suppose we have the following parameters:

• Drain-source voltage V = 30 V

• Reverse recovery charge Q = 900 nC

• Switching frequency f = 100 kHz

Plugging these numbers into the above formula gives switching losses of 1.35 W. This might not sound like very much loss, especially if you consider the amount of power that might be delivered with this switcher. However, if you look at typical package thermal resistance values (see below), this much loss can cause a FET to overheat. This is why it is important to balance package thermal resistance against reverse recovery characteristics.

MOSFET Selection Based on Reverse Recovery

Although MOSFET reverse recovery does create resistive losses (and thus heat), MOSFET selection is not just about selecting the lowest reverse recovery time or charge. Given two MOSFETs with all other specifications being identical, you should naturally select the MOSFET with the lower reverse recovery charge. Unfortunately, it is almost never that simple. Instead, try to answer the following questions:

• Will losses due to reverse recovery create excessive heating?

• Will an alternative package reduce heating despite long reverse recovery time?

To better understand how this works, see the specifications listed below for the IRF640N MOSFET. These two tables outline the reverse recovery time, charge, and the junction-to-ambient thermal resistance values.

Notes 4 and 5 in the Thermal Resistance table indicate that the thermal resistance values apply to different packages for this component. In total, the reverse recovery time depends only on the circuit construction on the semiconductor die, while the thermal resistance depends on the package construction. Cost also depends on package construction, and certain packages may be smaller and thus give some advantage in a PCB layout.

From the above table, we can see that if the reverse recovery losses in the TO package (Note 4) are too large, then the D2PAK package (Note 5) would be preferable. Because the thermal resistance is lower, the expected heating during operation would be lower. This could be very important when the device is being used to generate pulses with high current, which is often the case for MOSFETs used in power systems.

No matter what FETs you decide to use in your power electronics, you can use the Smoke Analysis feature in PSpice from Cadence to evaluate losses and efficiency in your systems. PSpice users can access a powerful SPICE simulator as well as specialty design capabilities like model creation, graphing and analysis tools, and much more.

Subscribe to our newsletter for the latest updates. If you’re looking to learn more about how Cadence has the solution for you, talk to our team of experts.