• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 29 subscribers
  • Views 587 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.

DRV8889-Q1: stall detection does not work

Lenio Cacula
Genius 16280 points
Part Number: DRV8889-Q1

I have a motor that is difficult to detect a stall condition. Register TRQ_COUNT (CTRL7) is zero even at zero load - what does that mean? Is there a specific parameter in the motor, inductance or otherwise, that dictates whether stall detection is going to work? I'm afraid the back EMF is too little for this particular stepper, but I am not sure how to precisely measure it - can you help? Are there any configurations that help mitigate the problem with low back EMF?

Thank you!

  • 0
    TI__Guru 56650 points

    Lenio,

    Would you take the winding current measurement with a scope when the motor is stuck?

    Also, when the motor is in stall condition, does the motor vibrate?

  • 0 in reply to Wang5577
    TI__Genius 16280 points

    Hi Wang,

    yes, the motor vibrates when stalled.

    I could not notice much difference in the motor curves between running and stalling. They are below:

    First, with the motor running normally:

    The second is with the motor stalled - I can only see occasional transients in the current curve that are hard to capture in the scope:

    Thank you!

  • 0 in reply to Lenio Cacula
    TI__Guru 56650 points

    Lenio,

    When the motor vibrates, I can see a subharmonic current waveform and stall detection condition cannot meet.

    The vibrating is related to micro-step, pps, torque setting and system resistance. Please increase the motor speed and adjust micro-step, pps and torque setting to see if it work or not. 

  • 0 in reply to Wang5577
    TI__Genius 16280 points

    Unfortunately I will need a bit more of details on how to systematically achieve the ideal configuration. I have too many parameters to adjust and so far the results have been inconsistent. I do see a TRQ_COUNT at higher speeds, but the stall detection is still very erratic, even when I see the count crossing the threshold up and down. How should I adjust the decay mode? What is the relation between step size and speed? Sometimes the motor stalls on its own depending on the combinations I choose.

    Thank you.

  • 0 in reply to Lenio Cacula
    TI__Guru 56650 points

    Lenio,

    The stall detection requires the ripple control decay. A good step size and speed setting is for avoid a strong mechanical vibration which makes DRV8889 stall detection function not work. I don't have a good equation for it. It also related to the system load/resistance. I know it is hard work due to many micro step setting and speed option. To me, the quick way could be

    a. set winding current higher enough to make sure the torque higher enough to overcome the system and load resistance

    b. pick a reasonable high speed to avoid strong mechanical vibration during the following test.

    c. Check 1/2 micro step to 1/16 micro step to find out a good setting to have a stall detection.

  • 0 in reply to Wang5577
    TI__Genius 16280 points

    Wang,

    with a higher speed and a very steep acceleration I can get the motor to more consistently (yet not 100%) to detect stall and reverse direction. I do not see however the chart showing the threshold being crossed every time -  I assume GUI Composer has a delay in reporting the data. More importantly, the stall flag does not get set, even though the motor is reversing. If I disable the reverse feature, the stall detection is very poor, the motor vibrates, and the flag never gets set. 

    If you have any suggestions let me know.

    Thank you!

  • 0 in reply to Lenio Cacula
    TI__Guru 56650 points

    Lenio,

    Since we change the conversation to email, let me close this post.