[SOLVED] IGBT Overshoot and Weird Slope

Akanimo said:

Thermal paste conducts heat. They do not conduct electricity. In fact, it acts like a dielectric with the IGBT and the heatsink acting like the two parallel plates of a capacitor, the thermal paste sandwiched between them. If the heatsink is connected to earth, then the setup produces common-mode noise when switched rapidly.

Single-ended oscilloscopes like the ones we find everywhere can pick up this noise.

Click to expand...

I used metal screws so there is electrical conduction. I tested short circuits with middle pin and yes they are electrically connected. I'm not sure if it is the problem. I think I should get a differential probe to clear the risks.

Thermal paste conducts heat. They do not conduct electricity.

Click to expand...

It depends. Some thermal pastes with high thermal conductivity are metal filled and also electrically conductive. But as far as I can see from the photo you have separate heat sinks for each IGBT. So there should be no problem with heat sink connection.

My interpretation of your probe connection is shown below, it's drawn according to the waveforms and given information. Probe ground is shorting the SAAR diode and load current is flowing through probe cable, causing undefined voltage drops. Hard to predict the exact waveform.

FvM said:

My interpretation of your probe connection is shown below, it's drawn according to the waveforms and given information. Probe ground is shorting the SAAR diode and load current is flowing through probe cable, causing undefined voltage drops. Hard to predict the exact waveform.

Click to expand...

I've removed the blue channel, measuring the yellow only. Result is still the same. Still there is a big overshoot.
Soon I'm going to add a 4700uF 63V capacitor to output rail and test it again.

Replace the gate resistor with a higher value one. Try 75 ohm

Thanks

FvM said:

FvM

Click to expand...

,

Akanimo said:

Akanimo

Click to expand...

and others for help.

Found the problem. I'm not commutating, just switching the DC positive ON and OFF. I mean;
when gate signal high Load = Vdc
when gate signal low Load = Nothing, open circuit.
Assuming there is a little inductance on rheostat, overshoot is not abnormal. I guess I will not have an overshoot problem when I'm working phase to phase with real load like ACIM.

I'll start coding the MCU after solder rest of the measurement, switching and sources. See you with my next problem.

Sincerely.

a really good look around your circuit would have saved you a lot of time I'm thinking ...

I've learned that I need differential probes and probes can be calibrated (except way too cheap probes)