DRV8424: nFAULT behavior

Zeming Kong

Expert 3515 points

Part Number: DRV8424

Hi team,

We are testing DRV8424 with auto-retry mode, output pins are connected GND directly to confirm OCP protection.

And we see nFAULT is latched to L, after we turn EN from H to L.

Can I consider that nFAULT will be latched when EN is turned to L during OCP, even if the device is in auto-retry mode?

Best,
Zeming

over 2 years ago

0 Ryan Kehr over 2 years ago

TI__Guru 55615 points

Zeming,

If ENABLE pin is brought back to the HIGH state, does the device auto-recover is short is removed? It should.

Is nSLEEP pin held HIGH this whole time?

Regards,

Ryan

0 Zeming Kong over 2 years ago in reply to Ryan Kehr

TI__Expert 3515 points

Hi Ryan,

If ENA is brought back to H, it will return to auto-retry, as you mentioned.

nSLEEP is always H.

Best,
Zeming

0 Ryan Kehr over 2 years ago in reply to Zeming Kong

TI__Guru 55615 points

Zeming,

Then this is expected operation of the device. It is still in a FAULT state until ENABLE is taken back HIGH and a retry attempt is made.

Regards,

Ryan

0 Zeming Kong over 2 years ago in reply to Ryan Kehr

TI__Expert 3515 points

Hi Ryan,

Thanks for your reply, understood this behavior is expected.

The customer connected all nFault outputs to the same GPIO port of CPU. When fault happens, all motor drivers will be shutdown, and then enabled again one by one to find out the abnormal device. But if the fault is latched, it will be hard to find out the abnormal one.

Also this behavior is not descripted in the datasheet, the customer requests us to add the description.

Best,
Zeming

+1 Ryan Kehr over 2 years ago in reply to Zeming Kong

TI__Guru 55615 points

Zeming,

I see. Let me check with team on update.

Regards,

Ryan

0 Zeming Kong over 2 years ago in reply to Ryan Kehr

TI__Expert 3515 points

Hi Ryan,

Sorry to open this thread again. Please let me confirm more about this behavior:

1. DRV842x/3x series all have this kind of behavior?

2. Any method to keep auto-retry when EN=L? If not, any method to clear fault when EN=L?

Best,
Zeming

+1 Ryan Kehr over 2 years ago in reply to Zeming Kong

TI__Guru 55615 points

Zeming,

1) Yes, behavior is the same across the device family.

2) This doesn't make much sense. When EN=L, the bridges are disabled, so there is no possibility to "retry" when the bridges can't be enabled because EN=L is holding off the bridges. nSLEEP can be taken low and back high again to clear the fault.

Regards,

Ryan

Motor drivers

Motor drivers forum

DRV8424: nFAULT behavior