The most important uses of diodes are not in DC conduction, they are actually in AC circuits. The fact that AC circuits are time-dependent also means that diodes have time-dependent characteristics that affect operation, but these are not always covered in a semiconductor devices class. For systems designers, the time-dependent characteristics of a diode are also important from the EMC perspective as the component’s time-dependent behavior can create radiated and conducted EMI.
One of the main specifications that defines time-dependent behavior of a diode is the reverse recovery time. The specification has a simple definition and can be thought of as a time constant required to switch between forward and reverse bias states. To see what you need to consider in diodes for AC circuits, keep reading below to learn about reverse recovery time.
How Diodes React Under AC Voltages
Diodes have particular characteristics that can be observed under AC driving due to the internal structure of the device and their semiconductor properties. Diodes have a property called the junction capacitance, which is related to the material properties of the semiconductor and the geometry of the diode. Together with the average charge carrier recombination time, there will be some delay or lag in the response of the diode when driven with an AC voltage.
In AC driving, the important parameter determining diode response is the reverse recovery time. This parameter defines how the diode switches between forward and reverse bias based on the amount of time required to sweep carriers out from the junction (or the depletion region in a p-n diode). After an appropriate amount of time, excess charge will leave the junction and the conduction state will be fully switched.
Reverse Recovery Time Defined
The reverse recovery time has a simple definition:
- If a diode is initially driven in forward bias, and the polarity suddenly switches to reverse bias, the diode will still remain conducting for some time.
- The time required for conduction to settle into the reverse bias state is the diode’s reverse recovery time.
The reverse recovery time appears to behave somewhat like a discharging capacitor, where the conducted current eventually decays to zero. The behavior can be most clearly seen on a graph of diode forward bias current vs. time.
Driving voltage and diode current vs. time
Because the diode is still conducting after the polarity switch, the current momentarily switches from positive to negative, and it remains negative for some time before decaying to zero. Technically, the time required to decay to zero remnant current is infinite, but it will eventually become so small as to be indistinguishable from inherent fluctuations in background noise.
The above graph implies there is a single diode reverse recovery time, but this is not true. The reverse recovery time will depend on several factors:
- The initial forward bias voltage value
- The initial forward current value
- Any load in the diode’s current loop
- Junction capacitance
Forward bias current measurements or specifications will normally be given with values for these parameters, rather than as a universal recovery time. If the recovery time is known, then a switching limit or frequency for the circuit can be determined.
Frequency or Switching Limit
Due to the delay in the switch from forward to reverse bias, there is some limit on the switching rate or frequency that should be used with a diode. The major diode manufacturers do an excellent job of calculating and measuring the reverse recovery time of their diodes, and they will specify this information in their datasheets. Once you have found the recovery time of your component, you can compare this to the frequency or edge rate you plan to use in your system.
Suppose for a moment that a diode has a reverse recovery of 50 ns under specific DC driving and loading conditions. If the diode needs to switch the driving voltage polarity from positive (forward bias) to negative (reverse bias), then the polarity switch should occur in longer than 50 ns.
Example datasheet entry showing reverse recovery time specifications.
If the switch is too fast (or equivalently, if the dI/dt rate for the diode current is too high), then the leftover current pulse in the reverse direction could excite a transient in the circuit. The result is ringing; this creates a high conducted EMI problem in switching power systems running at high currents. To prevent this, you would need to:
- Reduce the edge rate during switching
- Reduce inductance along the current path
- Use a different diode, such as a fast recovery diode
Whenever you need to simulate diode reverse recovery in your electronic circuits, use the comprehensive set of simulation tools in PSpice from Cadence. PSpice users can access a powerful SPICE simulator as well as specialty design capabilities like model creation, graphing and analysis tools, and much more.