• State
  • Locked Locked
  • Replies 9 replies
  • Subscribers 83 subscribers
  • Views 2086 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.

Instaspin Motion 28069F and Hall sensor

Aj Thomas
Prodigy 130 points
Other Parts Discussed in Thread: MOTORWARE

I based my changes on the earlier post http://e2e.ti.com/support/microcontrollers/c2000/f/902/t/321420.aspx?pi303962=1
and i have modified the files to make use of these new apis.
The Hw is drv8301kit_revD. The Hall sensor is attached on Jumper J10 on the kit.
This motor is identified correctly and is working in Lab5c.

I am not able to spin my motor with the Hall changes. I am sure the Hall sensors work fine and i am able to
deduce the angle in the HALL_ReadTheta function. On manual rotation the angle i do see all the values seen in
the switch ladder.  

I am able to see that there is small displacement of few degrees when i execute the Lab 12.
But that is all the motion i see. The current rises first time from 0.08A to about 0.21 A at first jerk.
Later on the current rises to max of 0.81A and there is no motion at all. I changed the reference
speed but this does not have any change in outcome. I even checked using different values of theta offset
and the Bandwidth but this does not alter the behaviour.

I am attaching my modified files for lab12 and my user.h
I am trying to improve the low speed response using some feedback loop. i would like to go to lower speed of 1 rpm.
I am using Hall since these should provide approximate commutation sequence from power off unlike the encoder. There is also mention of
IPD in the thread but i think this is not provided in latest version of motorware.
I will verify the response with a quadrature encoder once i get it. Do suggest what can be wrong in current setup.

Best regards
Aj
Motorware version:motorware_1_01_00_12
Tools version:c2000_6.2.6

  • 0
    Prodigy 130 points

    Attached user.h

    I am using a PMSM motor for evaluation that has integrated Hall sensors.

    user.h
  • 0 in reply to Aj Thomas
    TI__Guru** 119340 points

    Aj,

    1. can you confirm your chip is actually F28069M and not F?  The Motion portions only run on the M devices.

    2. Adam from LineStream is bringing up the Hall based projects and I believe will have them in MotorWare _13.  I think he also had an issue with the version I posted on the previous thread. There are some updates from our FAE who worked on it and Adam is now evaluating.

     

  • 0 in reply to ChrisClearman
    Prodigy 130 points

    Hello Chris,

    I am using the 69M Motion control developer kit. I incorrectly mentioned it as 69F.

    Can you provide any pointers to debugging the output from Observer module.
    Any document would help to understand the implementation basics.
    If possible can you share your working setup.

    Hello Adam,
    Were you able to get to root cause the failure on your setup. Let me know if you progressed.
    Best regards

    Aj

  • 0 in reply to Aj Thomas
    TI__Guru** 119340 points

    Aj Thomas said:
    Can you provide any pointers to debugging the output from Observer module.
    Any document would help to understand the implementation basics.
    If possible can you share your working setup.

    I don't understand what you are asking for.

    The output of the Hall observer?  As mentioned on the previous thread this is not supported by TI until it is in MotorWare. I posted a version I had (limited testing) to help some people out. 

    I think if Adam pulls this into MotorWare as an example it will have some further details on tuning the Hall based observer.

     

  • 0 in reply to ChrisClearman
    Prodigy 130 points

    Chris,
    I wanted to understand the Math of Observer to debug further. There was implementation for Sensored BLDC motor for Hall in control suite,  but that is different than the one you had posted. This looks simpler to me and it will serve purpose of comparison.
    When is the next motorware release scheduled for.

    Thanks for replies and support.

    Aj

  • 0 in reply to Aj Thomas
    TI__Guru** 119340 points

    Aj,

    Aj Thomas said:
    I wanted to understand the Math of Observer to debug further.

    Which observer?

    The FAST observer: we do not release any of the math, it is embedded and protected in on-chip ROM.

    The Hall Observer: all of the math is in source code that I provided.

    Aj Thomas said:
    There was implementation for Sensored BLDC motor for Hall in control suite,  but that is different than the one you had posted. This looks simpler to me and it will serve purpose of compariso

    Sensored BLDC is just using the Hall signals for commutation.

    What we showed in the InstaSPIN-MOTION example is using the Hall signals for angle interpolation, then using that estitmated angle for FOC.  They are quite different and one is certainly more complex.

    What are you trying to solve in our motor control application?

    Aj Thomas said:
    When is the next motorware release scheduled for.

    2nd half of June.

  • 0 in reply to ChrisClearman
    Prodigy 130 points

    Chris,
    My application is to move a pulley at fixed angle. This will be supported with
    gears but will need finer control. With instaspin-FOC i was not able to go lower than
    60RPM without loss of performance. I thought i can better that with sensored instaspin-Motion chip.
    My motor has Hall sensors so wanted data with it for comparision sake.
    I have planned for quadrature encoder. If i find such issues i will get back to you since it is
    supported by TI in present release of motorware.
    Thanks for the date for Motorware. i will check the solution later.
    Best regards
    Aj

  • 0 in reply to Aj Thomas
    Guru 12800 points

    Aj,

    Have you tried running InstaSPIN-MOTION sensorlessly in your application?  InstaSPIN-MOTION has a more advanced velocity controller and will be able to run at a slower speed than with InstaSPIN-FOC alone.  

  • 0 in reply to Aj Thomas
    TI__Guru** 119340 points

    Since you are planning for an encoder I wouldn't waste any time on the halls. You will get excellent performance using MOTION and proj_lab12 or 13