• State Resolved
  • Locked Locked
  • Replies 10 replies
  • Subscribers 29 subscribers
  • Views 1044 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.

DRV8816 OCP/Thermal

Keiichi Kajino
Expert 2365 points
Other Parts Discussed in Thread: DRV8816

Our customer trying to check protection circuit.

 1. They are forcing OUT1 to HSON and OUT2 to LSON.

 2. Connecting OUT1 to OUT2 via register(Volume) to control current.

 3. Rsense connected to GND, no ITrip.

 At IOUT=1.4A following switching phenomena were produced by DRV8816.

Since IOCP is more than 3A, I guess it is not a OCP retry. The freq of retry is around 16ms.

Could you please let me know if other protection is exist?

CH1 is nFault, CH2 is OUT2

  • 0
    Expert 2365 points

    Followings are waveforms

  • 0 in reply to Keiichi Kajino
    Expert 2365 points

    I am so sorry.

    Retry to add waveforms

  • 0 in reply to Keiichi Kajino
    TI__Guru** 114905 points

    Hi Kajino-san,

    This appears to be an Over temperature shutdown event.

    nFAULT is low, indicating that overtemperature warning is already active. Once the overtemperature shutdown occurs, the device will cycle in and out of operation as the junction temperature heats and cools.

    To confirm this is overtemperature shutdown, lower the current slowly. The device should continue to operate with nFAULT low. As the current is lowered further, nFAULT should go high, indicating the junction temperature has been reduced below the warning level.

  • 0 in reply to Rick Duncan
    Expert 2365 points

    Hi, Rick

    Thank you for your reply.

    Is there any additional protection for frequent themal shutdown?

    In this cases, it seems 4 times of OTS will cause 250ms halt time for retry. 

  • 0 in reply to Keiichi Kajino
    TI__Guru** 114905 points

    Hi Kajino-san,

    There are additional protections available, but it requires one or more of the following:

    using a resistor on ISENSE to limit the current,
    reducing the duty cycle to limit the average current,
    monitoring nFAULT and perhaps VPROPI to reduce the current when nFAULT is asserted.

    Is the customer trying to check the current protection by slowing increasing the current? This can be difficult because the over temperature can enable.

  • 0 in reply to Rick Duncan
    Expert 2365 points

    Rick,

    Our customer is evaluating thermal protection behavior and want to know measured results are

    expected one or not. I guess some additonal thermanl protection senario is exist on DRV8816

    because after 4 times OCP, it skips in 250ms. Is this skip is expected behavior.

    If yes, could you explain with numbers? (xxx ms skip if xth times of OCP in yyyms... etc) 

  • 0 in reply to Keiichi Kajino
    TI__Guru** 114905 points

    Hi Kajino-san,

    The designer has looked through the circuitry and cannot find a reason for the skips. There are no circuits in the device that are intended to do this.

    He requested the following additional information:

    1) Can you confirm channel 1 is nFAULT and channel 2 is OUT2?

    2) Are the voltage scales correct? It appears that there are large excursions above and below ground in the upper waveform?

    3) Would you ask the customer for OUT1 in addition to OUT2 and nFAULT? The designer believes it will help determine the cause.

  • 0 in reply to Rick Duncan
    Expert 2365 points

    Hi, Rick

    1) These are surely correct. I have got their schematic already.

    For 2) and 3) I asked to customer but have not get reply yet.

    I will update once I get. Thanks

  • 0 in reply to Keiichi Kajino
    Expert 2365 points

    I got waveforom and also got conclusion.

    The correct waveform is below, please ignore previous one.

    In this case, it seems that 75ms of off period is recovery from OTS.

    (Customer saids that the clocking wavefrom while OTS comes from noise because signal is Hi-Z.)

    I think it is expected function. Am I correct?

    Sorry for confusing so much but appreciate your support. Please say thank you to designer also

  • 0 in reply to Keiichi Kajino
    TI__Guru** 114905 points

    Hi Kajino-san,

    No worries. Thank you for clearing this up. It makes much more sense now.

    This does appear to be the correct operation. Ignoring the noise, the DRV8816 enters thermal shutdown for approximately 75 ms. It begins driving again after the junction temperature has decreased approximately 15C (Tots_hys).

    The thermal shutdown cycling will probably change based on the current load. If the current is increased beyond 1.4A, the time between thermal shutdowns will probably decrease.