• State Resolved
  • Locked Locked
  • Replies 10 replies
  • Subscribers 31 subscribers
  • Views 1014 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DRV8350F: Fault indicator never triggers when it should

Sebastian Siegfried
Intellectual 680 points
Part Number: DRV8350F
Other Parts Discussed in Thread: DRV8320, DRV8350, DRV832X

Hi,

I'm using the DRV8350F (the F version because of a STO performance level e) and it seems, the fault indicator never triggers. I had a DRV8320

before and exactly the same fault indicator circuit, which worked fine there, so I don't think that should be a problem:

nFAULT is connected to the nFAULT pin of the DRV8350F and also a measurement with the oscilloscope shows, that nFAULT is always high, also when a VM undervoltage should be

triggered (VM is switched off by a transistor). nFAULT_B is connected to a uC Input. If VM goes to 0V, the motor controller does not work anymore (gates floating) as it should, but the nFAULT pin is still pulled up.

Is there any known problem of the fault indicator or what could be the cause of this problem?

  • 0
    TI__Mastermind 18981 points

    Hi Sebastian,

    Thanks for posting your question to the e2e forum - we are looking into this and will follow up with some feedback within the next two days 

    Thanks and Best Regards,
    Andrew

  • +1
    TI__Mastermind 18981 points

    Hi Sebastian,

    Thanks for your patience - please see the below response to your question: 

    In the case of both devices, DRV8350F and DRV8320, it does seem like the Fault indicator circuit you have should work fine.

    • Both devices have an open-drain nFAULT pin architecture, and require an external pullup to Vcc (which your circuit covers properly). 
    • both devices also have VM_UVLO protection logic that will pull the nFAULT pin down to 0V when this type of fault event occurs

    To investigate root cause, I have a few questions for you to help debug the system. 

    1. what is nFAULT_B in this case? Is this a connection between the DRV device's nFAULT pin to the microcontroller IO input or something? 
    2. when you trigger VM switching off with a transistor, are you also doing anything to the 3.3V supply rail?
      1. in order for the nFAULT diode to turn on, it needs to be forward-biased with a voltage drop & conduct current (3.3V on the supply rail, and 0V on the nFAULT pin itself) 
    3. is the nFAULT pin on the DRV8350F 3.3V logic high during regular operation?
      1. when the UVLO fault occurs, does it still read 3.3V on the scope, or is the pin observed to be floating instead?  
    4. Does nFAULT LED light up for any other fault conditions? 
      1. it may be worthwhile to check the related components on the board to make sure R17, LED1, R16 are not damaged and also have the correct values 
      2. also may need to check that there is continuity between nFAULT pin of the DRV IC and the R16, R17 resistors - just to confirm the signal path is intact
    5. If these debug items don't work, it may be worthwhile to try soldering on a fresh DRV8350F part or try replicating the issue on another copy of the same PCB (to help rule out component damage)

    Thanks and Best Regards,
    Andrew

  • 0 in reply to Andrew Liu
    Intellectual 680 points

    Hi Andrew,

    I clicked wrongly on resolved, sorry for that.

    Thanks for your answer. To your questions and recommendations:

    1.) nFAULT_B goes directly to the microcontroller input (verified, it is not an output)

    2.) No, the 3.3V supply rail is always on, it just turns off the +12V supply for VM (verified with an oscilloscope). 

    3.) Yes, the logic pin from the DRV8350F stays always high (measured with an osc), also when I disable the controller over the disable pin (I think in sleep mode, the indicator should also be low, right?) I will try out some other failure modes just to see, if they might give another result. 

    4.) I can never see the LED light up, so it seems never to get triggered. The related components are finde and have the value as in the schematics above after measuring with a DMM. Also the connection to the DRV8350F is fine.

    5.) I have the same issue on 10 boards, so this is really strange.

    Edit: What I now see is, that the LED has a forward voltage in both directions, so there might be a short over the LED and 10mA current could have been flow. Is it correct, that the nFAULT can handle a maximum of 5mA?

    Kind Regards,

    Sebastian

  • 0 in reply to Andrew Liu
    Intellectual 680 points

    Hi Andrew,

    I used now a completely new hardware, desoldered LED1, R17 and R18 and I get exactly the same issue. When I use an input voltage (Vdrain) of 5.5V, I still have a high indicator, aswell for a VM of 0V. There is nothing more connected to the nFAULT pin now, only a 10kOhm pull up:

    I really can't tell what is wrong and since it worked fine with the DRV8320 there might be another issue here. Might it be possible, that it has something to do with the splitted supply of VM and VDRAIN? VM uses directly a +12V regulator, VDRAIN is connected to the input voltage. I will anyway try to feed VM also directly from the +48V supply to see, if then the indicator works fine.

    (R9 and R8 are not assembled)

    Greetings Sebastian

  • 0 in reply to Andrew Liu
    Intellectual 680 points

    Ok, I think I see the problem. I cannot detect VM and VDRAIN undervoltage, since the logic from the DRV8350F itself needs VM to properly work. The minimum VM is 9VDC from the datasheet and a UVLO is triggered, when the voltage VM falls below 8.8VDC. When I want to trigger a VDRAIN UVLO, this is not possible, since VM is supplied over a buck converter from VDRAIN and will also decrease accordingly under 12VDC. 

    So I'm unsure what can of error I can easly trigger.

  • 0 in reply to Sebastian Siegfried
    TI__Mastermind 18981 points

    Hi Sebastian,

    Thanks for the detailed debug information -

    Let me take some time today/tomorrow to review the information and see if I can provide any new suggestions. 

    Thanks and Best Regards,
    Andrew

  • +1 in reply to Sebastian Siegfried
    TI__Mastermind 18981 points

    Hi Sebastian,

    I had a chance to talk this through with my team today, and have the below response: 

    1. You are correct that the DRV8350F device logic itself needs sufficiently high VM voltage to properly work.
      1. otherwise, the device is just completely off & has no activity - even on nFAULT circuit. 
    2. In the case of the difference between DRV8350 and DRV8320, 
      1. DRV832x has VM/Vdrain tied together and that is meant to be one supply protected by VM_UVLO 
      2. DRV835x has a split supply VM and Vdrain, which have separate VM_UVLO and VDRAIN_UVLO specs 

    In the case where you are trying to test for VM_UVLO and VDRAIN_UVLO in the DRV8350F device, 

    • it may be needed to provide the VM supply separate from Vdrain, so that you can keep VM powered on while lowering Vdrain to test the condition. 
    • secondly, since VM and VDRAIN have different UVLO thresholds, your statement sounds correct in that VM_UVLO will trigger before VDRAIN_UVLO, since the VM UVLO threshold is higher at ~9V compared to Vdrain's ~7V datasheet spec. 
    • lastly, for the VM_UVLO to trigger, it may be also important to have the ramp-down be a lot slower
      • (instead of immediately shutting off VM with a switch)
      • This way, you give a chance for the UVLO to trigger before the device becomes completely turned off. 

    Please let us know if this resolves your questions. 

    Best Regards,
    Andrew 

  • 0 in reply to Andrew Liu
    Intellectual 680 points

    Hi Andrew

    Thanks for clarification, this solved my question so far. I could now also trigger the UVLO, the main difference to the DRV8320 I see here is, that if the ENABLE pin is low, also DVDD is low on the DRV8350F and that means also the nFAULT pin does not work anymore. For my STO I have shut down the ENABLE pin faster than the VM (it is a ramp down over 400us), so it was never able to give an indicator signal.

    Kind Regards,

    Sebastian

  • 0 in reply to Sebastian Siegfried
    TI__Mastermind 21805 points

    Hi Sebastian,

    That sounds good. Let us know if you have any additional questions or if you would like us to close this thread!

    Thanks,

    Matt

  • 0 in reply to Sebastian Siegfried
    TI__Mastermind 18981 points

    Hi Sebastian,

    Closing this thread for now - if you run into any new debugs/problems, please feel free to ask a new/related question on the e2e motor drives forum

    Thanks and Best Regards,
    Andrew