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.

Motor Control on CLA



A pleasure to watch CLA executing (core) motor control code under 6 µs using  under 1KW PGM-RAM & 1KW DATA-RAM :)

Good for withdrawal symptoms of numerical kind...

 

 

  • So you moved motor control loop over to the CLA?  Can you share the specs on your system?

    We are actually in process of releasing a CLA DMC Library. We have an internal release and have a sensorless FOC 10KHz loop running on the Dual Axis LV Kit.  We are also obviously playing with PFC in the CLA.

    I'm interested to know what the interest level is for this, and if there are any other requests.

     

  • This is excellent !! Will it be possible to share more information on your system? The type of motor, type of algorithm, which C2000 device, application etc. If you can't share detail information then any overview, high level description of your system will also be highly appreciated.

  • Ibrahim Kolbas said:

    A pleasure to watch CLA executing (core) motor control code under 6 µs using  under 1KW PGM-RAM & 1KW DATA-RAM :)

    Good for withdrawal symptoms of numerical kind...

    Very cool!  I'd like to hear more about your application as well.

    -Lori

     

  • More about the system for Lori and Arefeen:

    • Main application is -but not limited to- sensored and sensorless ACIM/PMSM drives (already in production).
    • My primary focus was on benchmarking available processors to find a low cost & high performing one (to replace my existing  µC with moderate recoding effort ).
    • F28035 CLA code runs on the stick and not on the real drive yet; but, this is of no relevance at the moment with regard to performance evaluation.
    • 6µs includes all essential modules needed for sensored ACIM/PMSM plus some non-essential transformations (ca. 0.5µs).
    • On a real drive, additional 5µs..9µs overhead is estimated.
    • Part of the same code was benchmarked on F28335 in tandem with my existing controller on a real drive.
    • To get back to my primary focus, ample headroom seems to be present for sensorless operation @20kHz.

    About the good news (also for me being in the process of porting the sensorless part  to the CLA) from Chris:

    • A CLA DMC Library including modules  for sensorless operation is absolutely necessary. 
    • This library would save precious engineering time all over the world.
    • It also would accelerate the widespread use of the CLA-based family.
    • With SW retrofitting, PFC+1PM on CLA would make a good kit.

    ..."CLAmathLib_type0.inc" is as important as the price&performance of the F28035 itself...

     

  • Thanks for sharing the information,  ... one more question .. what is the power range of these drives?

  • The power range is 0.1kW up to 355kW.

  • I am assuming you will have even more bandwidth with higher power drives as the switching frequency may not be as high as 20kHz for those drives. This will allow to implement other system related functions besides core motor control algorithms.

  • Yes; as you go up with the power, switching frequency is reduced and you theoretically have more time. But, high power motors tend to be more problematic than their low power counterparts; so, more real-time processing capability is always welcome....

  • Thanks for the info - your feedback has been a nice end of year present for the C2000 development team.  I hope the CLA continues to serve you well!

    -Lori

  • Ibrahim,

    just for clarification: what type of "stick" did you use for  your evaluation? All I know of is a 28027 USB stick (without CLA). Am I right that you used a F28035 DIMM100 control card?

    Regards

     

  • I think you are right. It reads "PICCOLO B EXPERIMENTER KIT" TMDXDOCK28035 on the box.