STEVAL-SPIN3201. Incorrect signal waveforms to drive N-channel power MOSFETs

[Shevtsov]

I designed PCB with STSPIN32F0 and my MOSFETs sometimes breaks down. I think because of break-in. I started to look at oscillograms and i was surprised. Then I took the STEVAL-SPIN3201 and repeated research. It's the same.

I explored waveforms for gate(MOSFET) of high- and low-side. You can see that MOSFET of high-side has stair. And there (eval1&eval3) has small time when exists short curcuit because low-side and high-side mosfet are opened simultaneously. That situation is not always. Approximately 50% of wafeforms. Correct wafeform is shown in "eval4".

How to fix it?

Thank you in advance.

[Shevtsov]

I made an oscillogram still.

Red-HSV (here you can see part of waveform that shouldn't)

Blue-LSV

LightBlue - SENSE2P(see the picture above)- in my case resistor=5mOhm

You can see that there that value is above 600mV.

I=600mV/5mOhm=120A

It means that there is short-circuit. High-side MOSFET is not closed when low-side MOSFET is opening. High-side MOSFET has parasitic value between high- and low-signal.

[Cristiana SCARAMEL]

Hi @Shevtsov​ sorry for the late reply.

I have some questions to better understand your application behavior.

First of all the high side gate signal (HSx) must be referred to respective OUTx.

If you do not have a differential probe you can monitors both signal versus ground and show them (if a Math function of oscilloscope if available you can display 'HS - OUT' voltage).

From the scopes it seems you are supplying the board with VM = 8 V, could you confirm it?

How you are driving the device?

In particular in which way are you set the Advanced-control timer (TIM1) used to generate the PWM signal for the three half-bridge gate drivers and relative deadtimes?