• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 30 subscribers
  • Views 507 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.

DRV8873: Are there conditions where OLA may not detect open load?

Dmitry Marukh
Prodigy 110 points
Part Number: DRV8873

We are running DRV8873 in EN/PH mode. The OLD is active and it reliably detects disconnected motor on wakeup (OLP). However unplugging the motor while it is running (OLA) does not trigger nFAULT. 

The documentation mentions several times the possibility of erroneous fault detection. It does not mention not detecting an actual fault. Here is part of the schematic, there are 6 drivers on the board.

  • 0
    Prodigy 110 points

    An update.

    We found out that OLA works properly when PWM duty cycle is under 100%, detecting both motor unplugging and actuator reaching built-in end switches.

    Our tests before used 100% duty, which is basically DC.

    From the OLA description in datasheet it works by measuring diode voltage during recirculation. Obviously there is no recirculation when there are no pulses.

    This may help explain OLA missing open load... except no pulses actually means fault, so OLA should have detected it. Unfortunately this means it would also trigger at 100% duty cycle. Maybe the chip has some kind of OLA suppression for 100% duty, that prevents erroneous activation but also does not allow it to trigger in case of genuine fault? 

    P.S. found exactly same problem description here (posted 4 years ago, no answer yet): https://e2e.ti.com/support/motor-drivers-group/motor-drivers/f/motor-drivers-forum/908755/drv8873s-q1evm-drv8873s-open-load-detection-in-active-mode-with-100-pwm-duty

  • 0
    TI__Mastermind 19470 points

    Hi Dmitry,

    I will be happy to look into this issue. 

    Please allow until Monday for a response. 

    Best,

    David

  • 0 in reply to Dmitry Marukh
    TI__Mastermind 19470 points

    Hi Dmitry,

    Just had a few questions to better understand the layout and test conditions. 

    What is your load type, are you driving a brushed-dc motor, or solenoid? 

    How is the load connected (OUT1 to OUT2, OUTx to GND)?

    What is the resistance of the load?

    Is the EN_OLA bit enabled (write 1b) in control register prior to the tests?

    Best,

    David

  • 0 in reply to David Medis
    Prodigy 110 points

    Hi David,

    The load is an actuator with brushed dc motor, connected to OUT1 & OUT2. This is hardware version of the chip, so there are no control bits. nOL pin is activated on power up and stays active. I don't know the resistance of the motor, it draws about 4A at full speed. Please, see an update I posted earlier. The OLA works fine when duty cycle is under 100%, it does not detect open load when duty cycle is 100%. This behavior was observed by others too (see the link to the old question I provided).

  • +1 in reply to Dmitry Marukh
    TI__Mastermind 19470 points

    Hi Dmitry,

    Yes, you are correct. I was able to verify this operation on the DRV8873EVM. The OLA mode is ONLY functional when high-side recirculation occurs. When driving the load at 100% duty cycle, high-side braking won't occur and therefore no open load condition will be detected. 

    Best,

    David

  • 0 in reply to David Medis
    Prodigy 110 points

    Hi David,

    I think we are reading this note differently. My understanding is that it says "open load can be detected even if load is present", which is opposite of our case.

    Anyway, I understand now that the way OLA works creates a problem with 100% and 0% duty cycles. We will adjust our software accordingly.

    Thanks,

    Dmitry