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.

Piccolo ePWM Dead-Band Submodule is inverting!?



Hi all,

I have a Problem with understanding of the Piccolo ePWM Dead-Band Submodule (described in spruge9a.pdf):

From Figure 2-28 in spruge9a.pdf, I would say that when enabling the deadband submodule (setting EPwm1Regs.DBCTL.bit.OUT_MODE from 0 to 3), the polarity of the ePWM signals should not change when keeping POLSEL=0.

This is not what my scope says. I have to set POLSEL=2 for a standard PWM (PWMA high active, PWMB low active) to stay the same with or without deadband module (OUT_MODE=0 or 3).

So is my understanding wrong or is Figure 2-28 wrong?

  • Stephan,

    I don't see any problems with Figure 2-28.

    At the moment when you switch OUT_MODE from 0 to 3 (S1=S0=1) you enable the deadband unit.  Let's assume you keep the switches S2,S3 (POLSEL = 0) and S5,S4 (IN_MODE=0) at their default levels. As a result EPWM1A_IN will be routed both to EPWM1A_OUT and EPWM1B_OUT - they will show an identical pattern, because EPWM1B_IN is no longer used!

    Now, if you set POLSEL = 2 you add an inverter into the path to EPWM1B_OUT. As a result, the pattern of EPWM1A_OUT and EPWM1B_OUT is back to active high / active low.  Please note that in this setup EPWM1B_IN is not used. 

    If you set IN_MODE = 2 , POLSEL = 0 and OUT_MODE = 3 you will route EPWM1A_IN directly to EPWM1A_OUT and EPWM1B_IN directly to EPWM1B_OUT. I guess this was your intention.

    The deadband unit allows a lot of different operating modes. For motor control I usually use ePWM1A_IN as source both for rising and falling edge delay (IN_MODE = 0) and an inverter for EPWM1B_OUT (POLSEL = 2) combined with OUT_MODE = 3. Both power lines ( ePWM1A_OUT and ePWM1B_OUT)  are controlled by ePWM1A_IN only. This allows me to optionally use high - resolution PWM for both output signals.

    Regards

     

     

     

     

        

     

  • Frank,

    Thanks for the info and clarification.

    The point is that depending on my configuration, only EPWM1A_IN or EPWM1B_IN was routed to the output.

    Now this is clear - the Figure 2-28 is correct.