DRV8302: Faul in overload condition

Thomas,

Are you seeing a latching fault? Is there any activity on the nOCTW pin?

Regards,

-Adam

Hallo Adam, 

no there is no latching fault. Since whe have switched of the over current protection by removing the R107 in our schematic, there is no more activity on the nOCTW pin.

Regards, Thomas

Hi Adam, 

we have modyfide a PCB to seperate the power path from the logic side. Also with no improvement.

After this modification i measure the voltages PVDD, DVDD and GVDD again.

Ch1: nFault

Ch2: PVDD

Ch1: nFault

Ch2: GVDD

Ch1: nFault

Ch2: DVDD

As you can see there is no violation of the thresholds given in the datasheet.

OC is disabled. I have also checked the buck voltage in case of the thresholds, which is also fine.

Is there another possible source for the fault-pin that is not given in the datasheet?

Regards,

Thomas

Thomas,

Is your OC_ADJ shorted to DVDD? I think it's connected via a 10k right? Please try shorting it to DVDD. Is your VREF at 3.3V?

My guess is that OC_ADJ is in some weird state.

We see a ~800uS low followed by a ~1.3mS high on the fault line, this doesn't correspond to any protections built into the device. What does your PWM input look like?

Regards,

-Adam

Hallo Adam, 

I shorting the OC_ADJ now directly to DVDD but there is no result. 

Our VREF is 2.048V and also used for the analog measurement.

When the OC_ADJ is in some weird state we should also see an activity on nOCTW pin - right?

In my last test i put a small capacitor on the measerment pin's of the IC (SHx and SLx) - also with no Result.

Our PWM  is about 58kHz - for the high side and also invertetd for the low side (acitve free wheeeling).

At the nFault pin you can see 780us low thats korrekt. The time for the high isn't constant, it depends on the parameter of the speed control an the current control.

regards,

Thomas

Thomas,

Yes you may see activity on the nOCTW pin. 

Have you tried using a slower PWM frequency such as 10k or 20k to check if this is related?

If the fault signal frequency seems related to you control inputs then this may give us a clue.

Regards,

-Adam

Hallo Adam, 

I have checked the nOCTW pin again and there is no activity - so it seems that there is no over current event!?

Now we have tried a slower PWM frequency. We set it to 19,2KHz but the low time of the  nFault Pin is still the same - nFault is about 770µs low.

regards, 

Thomas

Thomas,

Could you try a second board, IC, or the EVM?

Regards,

-Adam

Thomas,

I have seen no update on this thread, did your issue get resolved?

I will mark this thread as closed for now but please post again if you need further assistance.

Regards,

-Adam

Hallo Adam, 

sorry for the delay. We have already checked 3 boards with the same problem but i made some measurements and tests that could give us a clue. 

I checked the bridge again and try to measure the current trough the half bridge. In this case i made some modifacation on the PCB. In special i put a short wire between every half bridge an the supply voltage to place a current probe. With this modification there is no faull anymore. It looks like an shoot through in the halfe bridge - reverse recovery time form the Body diode or something else - an in this case like a undervoltage at the PVDD for a few ns. Is it possible that the DRV Chip detects such a very short undervoltgae periode?

I have to do some more measurements and tests. I will let you know if i have a verifcation for my conclusion.

regards,

Thomas

Thomas,

I'm glad to hear we have some clues.

Yes the driver could detect this short period.

Let me know what you come up with.

Regards,

-Adam