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.

TMS320F28379D: f28379

Part Number: TMS320F28379D
Other Parts Discussed in Thread: TIDA-01606

Tool/software:

Hi,

Is there a way to generate complementary signals between ePWM1, 2 and so on; a complementary signal between ePWM1A and ePWM2A ?

Generating complementary signals in one ePWM (1A and 1B ) is straightforward. But, ePWM1A and ePWM2A and so on, not so.

I was trying to modify TIDA-01606 3L-2L. The first step is to shut down switches Q3  and Q4 (A,B,C). Then Q1 and Q2  (A,B,C) need to be in complementary mode. This means ePWM1A and ePWM2A as complimentary signals.

This 3L to 2L could have been easy if ePWMs could be redirected arbitrarily to any GPIO. Going through the Diagrams of XBARS, Input, output and ePWM XBARs,I think this is not be possible as GPIOs remain limited to the mux options described in the datasheet. XBAR helps take any GPIO of the device into any place in the processor, however, XBAR does not send ePWMs to any arbitrary GPIO.

Another way around 3L---> 2L PWM Driver modification may be if complementary signals can be generated between ePWMs (1A and 2A and so on). This also does not appear to be a possibility,

The next step then is the intervention in HW by generating ePWM A,B complementaries and disabling (in HW) Q3 (and Q4) and using the signal from Q3 to drive Q2. A SW only solution would be preferred. Can you please provide a code snippet if you think this is possible.

Thanks a lot,

en

  • Hi Erol,

    You can sync ePWM1 and ePWM2 timers. Generate same duty cycle from both PWMs but ensure that the ePWM2 is inverse of ePWM1.

    If you are designing new hardware, you can update the PCB to correct PWM pins.

    Also, would recommend looking at our latest device the F28P65x, this has more memory and updated peripheral set.

    Best regards, Pawan

  • Hi Pawan,
    Thank you !
    I can generate ePWM1A/2A static complementary. I had tried this before. However, using sync out from ePWM1 to ePWM2A and activating the complementary mode will not work when the duty cycle changes (not fixed CMPA value but sinusoidally variable value compare value). Please try it and let me know !
    Thanks a lot,
    Erol
  • Erol, What do you mean by will not work? Can you share some waveforms on what you are observing.

    Best regards, Pawan

  • Hi Pawan,
    Thank you !
    I mean it will not generate complementary signals on ePWM1A/2A.; ePWM1A can be set up to generate any duty cycle and sync out signals ; ePWM2A can be made to be the inverse of ePWM1A by setting Action qualifiers to generate a complementary signal with the opposite action compared to ePWM1A and the RED and FED for the deadtime. However, when ePWM1A changes the duty cycle dynamically (due to a sinusoidal signal passed on to its compare), ePWM2A will not track it in a complementary fashion. So, ePWM1A and ePWM2A will not be complementary signals for the sinusoidal modulation. Setting up ePWM1A and ePWM1B to generate complementary signals is straightforward. Then, for a 3PH System, ePWM1 (1A and 1B set up as complimentary) Module is set up to sync out into ePWM2 and  ePWM3 (slaves) and this way a 3PH drive is generated TRM page 1984/2971-Fig.15-67. So, I think that the PWM Drive on TIDA 1606 cannot be turned from 3L---->2L without the modification on HW as well, like I initially stated at the start of this correspondence and was wondering if ePWM1A and ePWM2A can be made complementary (which would be needed for TIDA1606 conversion to 2L PWM).
    Thanks a lot,
    Erol
  • Hi Erol,

    Can you share the PWM configurations you have done. A diagram of counter with compare and action qualifiers set would be helpful to start with. Configuration code would also help.

    Best regards, Pawan