• State
  • Locked Locked
  • Replies 8 replies
  • Subscribers 29 subscribers
  • Views 1665 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

DRV8301 is not resetting the faults

Mehdi Sadaghdar
Prodigy 85 points
Other Parts Discussed in Thread: DRV8301

Using DRV8301 in our product, we are trying to intentionally trigger faults on DRV8301 and we are seeing some misbehavior from the chip:

- Under load, the supply voltage drops significantly intentionally to create fault. These should trigger two kind of faults:

- MOSFET short circuit: this is becaus eteh low supply would result in MOSFETs not turning on properly and show large VDS

- Low supply voltage

- We have set the fault to OC latch mode

- We can get these faults to happen, but although the driver stops momentarily, mostly the fault is not latched. That's the first bad behavior

- Sometimes the fault is latched, but only showed fault, and no OC condition. This could be because we continuously read the register
. This should be normal

- When the fault latches, we cannot reset the chip by toggling EN_GATE pin. This is the second issue.


Will you please tell me why we would have such behavior? Thanks.

  • 0
    TI__Mastermind 26065 points
    Hi Mehdi

    We will check this issue. Could you give more description on how you intentionally creating the fault? By reducing the power supply from what level and how many? The tests done on EVM or own board? When you say under load, you mean running the motor at normal loading, right?

    To check the fault report function, we'd better try triggling it by one factor each first. UVLO and OC may have many interaction to each other.

    Best regards,
  • 0 in reply to Wilson Zuo
    Prodigy 85 points
    Hi Wilson,

    The easy way we use to trigger faults is using a current limited supply and drive the motor hard. That way the supply voltage drops under load to below 8V, causes a fault and stops the drive.

    At that point the supply will return to normal. This action creates faults. We read the fault register to see what has gone wrong. We expect short circuit and under voltage faults. But sometimes, although faults happen (we saw short circuit momentarily triggered), they don't latch. We see motor halting momentarily, but returns back to driving normally immediately, rather than latching.

    So far what we have seen latched was just Fault bit set, or fault and under voltage. There has been no short circuit bits set, although I believe it happened. In datasheet it says reading the register a couple of times clears some faults. So to see if that makes a difference, we will read only once when fault or warning appears.

    But still the fact that it doesn't latch sometimes remains.

    So... After the fault latches, toggling EN_GATE doesn't reset the chip, and it remains in fault and doesn't respond. I wonder if the chip gets frozen in a corrupt state because of this event, although all levels are back to normal.

    While not in fault, we verified that toggling EN_GATE, the chip registers reset.

    Thanks
  • 0 in reply to Wilson Zuo
    Prodigy 85 points

    Wilson, any luck checking into my issue? Thanks.

  • 0 in reply to Mehdi Sadaghdar
    TI__Mastermind 26065 points
    Sorry for keep you waiting. I will check this and also asked help from our experts on DRV8301.

    Best regards,
  • 0 in reply to Wilson Zuo
    Prodigy 85 points

    Thanks Wilson. One more thing we are seeing is that when DRV8301 detects a short circuit (we see fault pin active and also read SPI which says a FET short circuit occurred), many times it doesn't latch the drive (we have set it to over current latch mode), but sometimes it does.

    So what condition could cause the driver to ignore the over current and not latch? I have two candidates:

    - Reading fault from SPI clears the latch?? (we are trying to see if this is the case)
    - An under voltage happening at the same time clears the latch?? (this is not indicated in datasheet)

    Let me know what your thought are. Thanks.

    [Update] The reason for the DRV8301 not latching the over current is determined to be that the supply voltage drops too low under high current and resets the chip. So latching part works fine. The only problem remains is that we are not able to clear the faults and unlatch toggling the Gate_EN pin.


    [Update-2] We tried to reset a latched chip through SPI and the chip still doesn't reset. We are unable to reset the chip with GATE_EN or SPI. Please advise.

  • 0 in reply to Mehdi Sadaghdar
    TI__Mastermind 21305 points

    Hi Mehdi,

    Addressing the reset issue. Does the device just report the FAULT bit after the lock up occurs? Is the nFAULT pin asserted or any other bits?

    Can you also provide a scope capture of the PVDD1 supply when the issue occurs (focused on the under voltage event)?

  • 0 in reply to Nicholas Oborny
    Prodigy 85 points

    Hi Nicholas,


    What we try to trigger is the FET short circuit here. The way we do it is by having a current limited supply while driving the load. What happens is that supply voltage drops and with it the gate driving voltage, causing the RDS-on of the MOSFET to rise under load. And so VDS rises, triggering the FET short circuit.

    In some cases the voltage drops so far that the driver chip resets and the faults goes away. But in some cases we trigger the fault and see nFault and FET short circuit bits, as we expect. In such cases toggling GATE_EN or reset bit through SPI will not clear faults and we are stuck in over current latch mode.

    Triggering short circuit fault due to low battery voltage is our desired function.

    So we can trigger it the way we want, just can't clear the fault, although we can talk through SPI. If you still need scope captures please let me know, but the description should be clear enough.

    Thanks,

  • 0 in reply to Mehdi Sadaghdar
    TI__Mastermind 21305 points

    Hi Mehdi,

    Let's discuss via email. I will send you a private message with my address.