• State
  • Locked Locked
  • Replies 7 replies
  • Subscribers 30 subscribers
  • Views 615 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.

MCT8316Z-Q1: MCT8316: Noisy coil waveform; takes 5x the current I think it should (compared to DRV10970)

Robert Stoddard16925
Prodigy 190 points
Part Number: MCT8316Z-Q1
Other Parts Discussed in Thread: DRV10970, MCT8316Z, MCT8316ZTEVM, MCT8316A

That being said, your document about tuning with this motor controller needs to be revised, almost needs to be totally rewritten.   It seems to apply to a very very old version of the silicon.

Forgive me, I have yet to "tune" this motor.  I intend to do that right now, or at least try.  Considering the waveform I'm getting, I don't think it'll pan so well.


Anyway, where I'm going with this is that I've gotten my motor up and running on the MCT8316.   It's spinning the right direction, has decent strength even though I haven't tuned it yet, however, the waveform is very noisy, and that noise translates to mechanical/audio noise from the motor.   Even unloaded, the motor gets rather warm, I can't imagine it being good for the motor.   More alarming, however, is that the motor I have attached to this controller takes half an amp to spin even though it's totally not loaded.  The identical motor in-situ (with a little load, basically it's spinning a free wheel on a bearing) running the DRV10970 takes 1/10th of an amp.  I have taken pictures of both the DRV10970 waveform and the MCT8316 waveform running 7.5v.   The PWM frequency I'm using is about 100 kHz, just a little less than that actually.   The PWM is generated with a 555 circuit.

Also, as a side note, the motor, if provided sufficient load, will suddenly switch direction, get considerably weaker and go about half speed.   I suspect tuning may alter this behavior.

MCT8316 trace taken from the W coil wire:



DRV10970 trace taken from the W coil wire:

Thank you for any input you may provide.

  • 0
    Prodigy 190 points

    Addendum:  I'm using the MCT8316ZT   so no SPI control.

  • 0 in reply to Robert Stoddard16925
    TI__Expert 5578 points

    Hi Robert,

    Thanks for your question on the Motor Drives E2E Forum. 

    I will aim to provide an answer on this by the end of the week!

    Best,

    Robert

  • 0 in reply to Robert Wedan
    TI__Expert 5578 points

    Hi Robert, 

    I have some initial suggestions and further questions to help you debug this issue:

    Quesitons:

    - Is the board you are using for both motor drivers similar or identical? 

    - Just to confirm, you are not using our EVM correct?

    - Which tuning guide are you referring to for that device?

    Suggestions:

    - Even though the motor is identical, can you test the MCT8316Z on the motor that you have working with the DRV10970?

    - Have you tried operating this motor with our MCT8316ZTEVM and testing this performance?

    - Please try tuning the motor, even if you feel that it may not provide many advantages. It should resolve some more of your issues. 

     

    I also want to caution that MCT8316Z (sensored trap) and MCT8316A (sensorless trap) represent two different variants and have different application requirements and considerations.

    Please update me on your answers and results and let me know if you have any more additional questions.

    Best,

    Robert W.

  • 0 in reply to Robert Wedan
    Prodigy 190 points

    I am not using the EVM.

    The board with the DRV10970 has a buck converter on it that the board for the MCT8316 does not.   The buck converter converts up to 60v down to 12v, however I am running both boards at 7.5v through the same power supply.

    The tuning guide is SSLVUCC2.  Ah, this is a tuning guide for a sensorless device.  That might be the issue.

    I do not currently own the MCT8316 EVM.   Seems like every source is backordered into at least late summer if not next year.

    I have now tuned the motor.   The same noise persists, however, I was able to "shave off" 100mA of current by using an advance of 4 degrees; the motor (at the desired speed) runs about 430mA.

    I will try putting the DRV motor on the MCT, see if that changes anything.   I will also try the reverse; I'll put the DRV on the MCT motor.

  • 0 in reply to Robert Stoddard16925
    Prodigy 190 points

    I have soldered the motor from the DRV10970 onto the MCT8316Z, and inspected the waveform of the coils.   The waveform looks the same on all coils as the MCT8316 image I posted above.

    Is this chip trying to do some PWM-like DAC of the trapezoidal rise or something?

    The motor I am using is the Electrocraft RPX-22-042V24-100-S  and the details about it can be found here   www.electrocraft.com/.../

  • 0 in reply to Robert Stoddard16925
    TI__Expert 5578 points

    Hi Robert,

    Thanks for trying that out and relaying the results, I will try to provide you will additional guidance on this in the next few days.

    Best,

    Robert

  • 0 in reply to Robert Wedan
    TI__Expert 5578 points

    Hi Robert,

    From what you have told me, may I ask how you are configuring the lead angle settings on the MCT8316Z?

    DRV10970 has Adaptive Drive Angle Adjust, see below:

    "The drive angle can be adjusted to achieve better efficiency by measuring the phase current and comparing it with the Hall signals. But, because of variation in parameters such as motor resistance and inductance with operating point, the BEMF voltage and phase current are not aligned for fixed drive angle when an operating condition is changed."

    "The drive angle must be adjusted when an operating condition, such as speed and load, is changed to operate the motor at optimal efficiency. This dynamic change of the drive angle requires continuous monitoring of phase current and additional processing of data. To overcome this problem of manually adjustment the drive angle, Texas Instruments' new adaptive drive-angle adjust (ADAA) algorithm can be used."  (https://www.ti.com/lit/an/slaa699/slaa699.pdf) 

    MCT8316Z does not have this feature so if you are not properly configuring the lead angle settings it may lead to inefficiencies and a higher current draw.

    To learn more about Lead Angle Adjustment, please refer to this video (https://training.ti.com/lead-angle-adjustment?context=1139747-1138777-1139742-1148348) and application note (https://www.ti.com/lit/an/slaa561/slaa561.pdf). 

    Best,

    Robert