• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 83 subscribers
  • Views 726 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.

Phase Current feedback instability for 28069F InstaSpin-FOC

Kuo Brad
Prodigy 65 points

Other Parts Discussed in Thread: CONTROLSUITE, DRV8312, DRV8301, MOTORWARE

Hi all,
     Recently we found Phase current feedback is unstable when we monitor the ADC value through watch window under this condition (Flag_enableSys=1, Flag_Run_Identify=0). The current feedback shows abt 40 unit vibration at idle (avg. ADC ~2020, vibration around 2000~2040 at idle, ADC total for 4096 unit),
This vibration behavior will be resulted in ADC_offset inaccuracy for Lab3a.
In addition, this ADC_offset inaccuracy will be also made some side effect as below: 

 
1.  3 phase current with unequalization at idle (not zero phase current at idle as make a few current in 3 phase at idle.)
2. Make Motor-ID(Lab2a) fail as Rs inaccuracy due to start current is not at zero current.
3. After motor driving, that is difficult to make motor stop (iq_Ref_A =0 in Lab5a) at no loading condition, suppose that should be CloseLoop system could not convergence due to Current feedback vibration slightly at iq_Ref_A =0.

After our circuit hardware verification, we know root cause should be related to our DC-DC system vibration & Layout noise slightly. But we have to admit this slightly power vibration & Layout noise due to controller size & cost limitation.

In fact we had tried to implement average algorithm for ADC acquisition of 3 phase current feedback, but that still have no quite good result. So, we would like to know whether TI has any possibility to filter those noise from firmware point of view ?!

As we know, TI Controlsuite of F28069 has example code (SineFIRFFT32) is for filter purpose, could we use this example code to apply into Lab5a to filter ADC noise ?! ,or have any other solution to overcome this ADC instability issue from firmware point of view?!  

Thanks in advance !

 

  • 0
    TI__Expert 5180 points

    Let me first replicate the issue. I ran the DRV8312 kit and I see a variation of +/- 2 counts out of 4095 when Flag_enableSys=1 and Flag_Run_Identify=0. What board are you using? Is it possible that the board is damaged? can you try another one?

    -Jorge

  • 0 in reply to Jorge Zambada
    Expert 1620 points

    Hi,Jorge.

         He said"After our circuit hardware verification",so he might do the test with his own hardware,not TI's board .

    I have the same question.I design my own board by reference to TI's board.If phase current feedback is unstable,How to use digital filtering? 

    The average algorithm? 

    A low-pass first order filtering algorithm? 

    Recursion average filtering algorithm?

    Disappears shakes algorithm?

    FIR or IIR ?

    Do the filtering algorithms affect the efficiency of CPU seriously ?How to achieve with C language?or have any other solution to overcome this ADC instability issue?

    I think anyone will encounter ADC interference issue more or less.

     

  • 0 in reply to Jorge Zambada
    Prodigy 65 points

    Hi Jorge,
         Thanks for your prompt reply.
    That is my apology to be unable make clear you.
    Regarding to ADC instability issue, that circuit hardward had been designed by ourself, not TI EVM board.
    In fact we also had verified TI EVM board (DRV8301), indeed the ADC vibration is around 1~2 unit.
    But that is difficult to have same ADC vibration performance in our circuit hardware design due to controller size & cost limitation. So, that is why to ask any TI solution (e.g. FIR or IIR...) for filter the slight ripple of DC-DC system and Layout noise issue .      

    by the way, May I know which latest Motorware & CCS version is suitable for instaSpin-FOC project develop so far?!

    Thanks in advance!

  • 0 in reply to Kuo Brad
    TI__Expert 5180 points

    It is really not desirable to have a filter in the currents unless it is absolutely needed. Is it possible for you to share the current feedback circuit you have? We might need to fine tune your circuit to avoid using filters unnecessarily.

    -Jorge

  • 0 in reply to Jorge Zambada
    Prodigy 65 points

    Hi Jorge,
          Thanks for your comment.
          We can send you the current feedback circuit if you can provide your official E-mail due to our business confidential concern.

    Sorry for that we can not post our current feedback circuit in TI Forum directly.

    That is appreciation to provide your E-mail to my E-mail address: bradtom911@yahoo.com.tw
    We will use our official E-mail to send you current feedback circuit.

    Thanks a lot !