24/7 emergencies 814 777 6921812 230 6079WhatsApp 811 786 0768
Electrónica MonterreyElectrónica MonterreyLaboratorios Creativeit Talk to an engineer
Menu

How do you measure dead time in a power inverter?

The dead time of an inverter is the interval during which both switches in the same leg are off; it is measured by comparing the gate signals of both with probes that tolerate the floating reference of the high-side switch.

Why dead time exists

In an inverter leg, the high-side and low-side switches must never conduct at the same time, because that short circuit across the supply would destroy the devices. Since real switches take time to turn off, the controller adds an interval during which both commands are off. That dead time protects the leg, but it also distorts the output voltage, so it is worth knowing it accurately.

What is measured

Dead time is the interval between the instant the first switch stops conducting and the instant the second begins. It can be measured on the controller’s command signal or, more usefully in practice, on the actual gate-source (or gate-emitter) voltage of each device, which includes the driver delays. The difference between the two readings is useful information in itself.

Setup

  1. Connect one channel to the gate of the low-side switch and another to the gate of the high-side switch. The source of the high-side switch is not at the oscilloscope’s reference but swings with the output voltage, so a probe common with the instrument’s ground would cause a short circuit or an invalid measurement.
  2. Use differential or isolated probes, or optically isolated measurement systems, with enough bandwidth for the edges and good common-mode rejection.
  3. Trigger on the edge of one of the gates and expand the time base at the transition.
  4. Use cursors to measure between the crossing of a defined level on the turn-off edge and the crossing of the same level on the turn-on edge of the other switch. Choose a threshold with physical meaning, such as the device’s turn-on threshold.

Sources of error

  • Different delays between channels or probes: compensate for them with the deskew function, because the dead time can be on the same order as the delay differences in the measurement.
  • Insufficient common-mode rejection, which injects noise from the switching voltage into the gate signal.
  • Ringing on the gate that crosses the threshold several times and makes the reading ambiguous.

An example

If the controller declares a dead time and the measurement at the gates is shorter, the driver or the device is introducing a difference between the turn-on and turn-off delays. Measurements on energized circuits or circuits with hazardous voltages must be made by qualified people, following the applicable safety regulations and using accessories with the appropriate category and voltage rating.

We do real tests, on video! Watch them on our YouTube channel →

Do you have a problem nobody has been able to measure?

Talk directly to the engineer in charge, with no middlemen.

Talk to an engineer