• State
  • Locked Locked
  • Replies 7 replies
  • Subscribers 31 subscribers
  • Views 1091 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.

PWM Phase on DRV8432

John Cole
Intellectual 340 points
Other Parts Discussed in Thread: DRV8412, DRV8332, DRV8312

Folks,

First, everyone on the forum has been awesome about answering problems.  Thank you very much!

We have a DRV8432 setup as 2 full bridges, CBC mode.  For each full bridge, our PWM comes out slightly out of phase when running at 100% (the logic comes out slightly out of phase from the microcontroller.  We have been chasing bugs in the motor-controller the past week, and when we saw this, we thought it might be affecting the motor controller.  We don't know how exactly it would affect the motor controller, but we wanted to see if this could be related to anything we've seen.  

When sending a 50% PWM, we get what we think is ideal:

When sending out 100% PWM there both signals seem out of phase.  Could this be doing damage or harming the DRV8432 in 2 full bridge, CBC mode?

  • 0
    TI__Mastermind 21320 points

    Hi John,

    I do not believe this would cause any damage to the DRV8432. This just appears to be due to how your controller generates 100% and 0% duty cycle waveforms. What issues were you experiencing?

    -Nick

  • 0 in reply to Nicholas Oborny
    Intellectual 340 points

    I've been back through the microcontroller code and can't figure out why that's happening.  Good to know that that won't damage the DRV8432 though.

    We don't start seeing problems until we try to decrease the PWM frequency.  We're trying to lower the heating characteristics by lowering the PWM frequency.  When we drop below 1.25 kHz, and we limit the CBC current limiting, we start seeing funny things happen:  the motors get jerky, and in some cases won't start in a specific direction.  For example, when we start trying to run a 500 hz PWM, with a 47k resistor in the CBC current limit, we start to see the motor shake and start and stop, and sometimes not start at all in a specific direction (and mysteriously run just fine in the opposite direction).

    We have increased capacitance on teh bootstrap circuits from 0.1 to 1.0uF.  

  • 0 in reply to John Cole
    Intellectual 340 points

    Sorry, just to followup on that, we thought there was an off-chance it had to do with the out of phase PWM cycling you see in the pictures above.

  • 0 in reply to John Cole
    TI__Mastermind 21320 points

    Hi John,

    It is possible that even with the increased capacitor that this frequency is simply to low. This does not explain why the motor runs fine in the other direction though. Have you tried with a different motor and gotten the same results?

    -Nick

  • 0 in reply to Nicholas Oborny
    Intellectual 340 points

    Thanks Nick.  We will keep working on it.  We have tried this on the same part number, different motors.  When we first saw it, we thought the motor itself was failing.  After switchign a few motors out, we didn't see any different in performance.

  • 0 in reply to John Cole
    Prodigy 110 points

    Hello John,

     

    In my personal opinion, you are facing natral feature of DRV8412/DRV8432/DRV8312/DRV8332 silicon, because

    1)  your PWM frequency is tooo low. 20kHz is recommended in DS, down to 10kHz should be with 1uF and 5 Ohm serial resistor.

    Too low frequency means, that Boost capacitor gets decharged during ON time, and DRV8XX2 chip will report FAILURE - loosing connectivity.

    2) you shold never go above 99percent of PWM, because Cboost will not get enoug time to be charged.

     

    Thanks.

     

  • 0 in reply to John Cole
    Prodigy 110 points

    Hello John,

    you are simply facing simple feature of DRV8412/8432/8332/8312 silicon, that:

    100percent on time (forewer on for half/full bridge) is possible only for LOW (bottom) side. It clearly explains, why you have no problem in one motor direction.

    To fix it for HIGH(upper) side, you must:

    - fix PWM to be bellow 99percent/20kHz.

    - be watching FAULT pin.