• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 83 subscribers
  • Views 592 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.

Original question:

CCS/TIDM-1000: TIDM-1000 sample project, boot from flash

CCS/TIDM-1000: Operation with light loads

Lorenzo Lutti
Intellectual 665 points
Part Number: TIDM-1000


Tool/software: Code Composer Studio

Hello, I have a question regarding the TIDM-1000 Vienna Rectifier sample project pfc3phvienna_F2837x. I've noticed that the control algorithm does not work very well with light or no loads. In this condition, the output of the voltage loop (VIENNA_iLRef_pu) becomes negative. Being the Vienna rectifier a unidirectional converter, I suppose that a negative current does not bring good results. In fact, when the current reference becomes negative, the voltage regulation does not work anymore and the voltage goes above the setpoint value (I'm currently using a 350V setpoint). With no load, the voltage output goes above 600V.

Maybe part of the problem could be due to the unavoidable offsets in the current sensor reading; obviously the power transfer is not supposed to change direction with light loads. But since the voltage control loop is of PI type (and therefore it also has an integral part) it should compensate for this offset. Perhaps since the voltage loop works on the powers (i.e. on the currents squared) it does not notice the change of sign of the reference current?

Of course I will need to study the control model in more detail, but I would really appreciate some hints and opinions directly from the authors.

Thank you!

L.

  • 0
    TI__Genius 15935 points

    Hi Larry,

    We have not considered the very light load and no load condition for this design. Below is a thread that might help you to some extent.

    Regards,

    Chen

  • 0 in reply to Chen Jiang ti
    Intellectual 665 points

    Hello, the other thread gave me some ideas, thanks.

    I think I've found the culprit. As I was saying previously, the voltage loop uses a PI control, and therefore it should compensate for this negative offset on the output current reference. In fact it does, but the problem is that the integral part is saturated to a very small negative value:

    #define VIENNA_GV_PI_MAX ((float32_t)0.7)
    #define VIENNA_GV_PI_MIN ((float32_t)-0.02)

    By changing the constant VIENNA_GV_PI_MIN to -0.7, it is possible to regulate the voltage even with very light loads.

    For no load operation you would need to go even lower (below -1) but this does not concern me much. I can add a small "dummy load" if I find out that the regulation is too noisy at no load.

    I think that this thread can be closed. I will post a related question if I have further problems.

    L.