• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 82 subscribers
  • Views 971 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

ePWM interrupt not occuring on both ZERO and PRD

Michael He
Expert 2170 points

I am working with the F28M35H52C1 RevB controlCard.

I have an ePWM interrupt configured to occur both when CTR = ZERO and CTR = PRD, as follows:

EPwm2Regs.ETSEL.bit.INTEN = 1; // Enable PWM Interrupt
EPwm2Regs.ETSEL.bit.INTSEL = 0x011; // Choose events to be both Zero and PRD
EPwm2Regs.ETPS.bit.INTPSSEL = 0x0; // Use INTPRD to determine interrupt prescale
EPwm2Regs.ETPS.bit.INTPRD = 0x01; // Interrupt on each event

The only thing the interrupt does is toggle a GPIO pin.

If I look at the GPIO pin and my PWM signals, the GPIO signal is not toggling at the same frequency as the PWM. It appears that the toggle only happens when CTR = ZERO, not when CTR = PRD. In other words, the high-side PWM output is center-aligned with the GPIO square wave.

Am I missing anything with my configuration?

EDIT: I put a break point in the interrupt handler and watched EPwm2Regs.TBCTR. When the breakpoint is hit, this register ranges between 24 and 31. The full period is 18750. This seems to indicate that it is indeed only occuring on the zero of the counter.

For reference, here is my PWM configuration:

    EPwm2Regs.TBCTL.bit.CLKDIV = 000;  // No Clock prescale - use full SYSCLK
    EPwm2Regs.TBCTL.bit.HSPCLKDIV = 000; // No clock division
    EPwm2Regs.TBPRD = PWM_HALF_PERIOD;  // 2x this period gives # of cycles for PWM period
    EPwm2Regs.TBPHS.half.TBPHS = 0;  // Set Phase register to zero
    EPwm2Regs.TBCTL.bit.CTRMODE = TB_COUNT_UPDOWN;  // Symmetrical mode
    EPwm2Regs.TBCTL.bit.PHSEN = TB_DISABLE;  // Master module
    EPwm2Regs.TBCTL.bit.PRDLD = TB_SHADOW;
    EPwm2Regs.TBCTL.bit.SYNCOSEL = TB_CTR_ZERO;  // Sync down-stream module
    EPwm2Regs.CMPCTL.bit.SHDWAMODE = CC_SHADOW;
    EPwm2Regs.CMPCTL.bit.SHDWBMODE = CC_SHADOW;
    EPwm2Regs.CMPCTL.bit.LOADAMODE = CC_CTR_ZERO;  // load on CTR=Zero
    EPwm2Regs.CMPCTL.bit.LOADBMODE = CC_CTR_ZERO;  // load on CTR=Zero
    EPwm2Regs.AQCTLA.bit.CAU = AQ_SET;  // set actions for EPWM1A
    EPwm2Regs.AQCTLA.bit.CAD = AQ_CLEAR;
    EPwm2Regs.DBCTL.bit.OUT_MODE = DB_FULL_ENABLE;  // enable Dead-band module
    EPwm2Regs.DBCTL.bit.POLSEL = DB_ACTV_HIC;  // Active Hi complementary
    EPwm2Regs.DBFED = DEADBANDCOUNT;  // FED = 50 TBCLKs
    EPwm2Regs.DBRED = DEADBANDCOUNT;  // RED = 50 TBCLKs
    EPwm2Regs.AQSFRC.bit.RLDCSF = FORCE_IMMEDIATE;  // Software Force loads immediate

  • 0
    Expert 2170 points

    Any help on this issue?

    I have an additional question: when I configure the PWM interrupt from the main() function, it works. However, if I configure it within a different file, it does not. For example, I have the following code in a separate file to group together PWM related functions:

        EALLOW;
        PieVectTable.EPWM2_INT = (PINT) (&PWMToggleGPIO);
        PieCtrlRegs.PIEIER3.bit.INTx2 = 1;  // ePWM2 Int

        IER |= M_INT3;
        EDIS;

    When I run tis version of the code, the PWM interrupt never occurs. This function is called from the main() function! If I write the exact same lines within the main() function itself, the PWM interrupt works properly.


        PieVectTable.EPWM2_INT = (PINT) (&PWMToggleGPIO);
        .....
        PieCtrlRegs.PIEIER3.bit.INTx2 = 1;  // ePWM2 Int
        IER |= M_INT1;
        IER |= M_INT3;
        EINT;
        ERTM;


    Is there some reason for this?

  • 0 in reply to Michael He
    Expert 2090 points

    The PWM functions that enable the interrupt, do they have access to PieVect/PieCtrl registers?  They need to have the right .h files included.

    Also, are interrupts enabled by the main() before these PWM functions are run?

  • 0 in reply to Fulano de Tal
    Expert 2170 points

    Your second point is the answer. I was being stupid and called the interrupt configuration function in the other file BEFORE the whole table was wiped by the InitPieCtrl() and InitPieVectTable() functions. Thanks for pointing that out!

    This issue is independent of the original post though. Regardless of where the interrupt is configured. It does not seem to occur when the PWM TBCTR = PRD. Do you have any ideas why this might be?

  • 0 in reply to Michael He
    Expert 2090 points

    are you clearing the INT flag in the ISR?

    in EPWM2_INT_ISR you need to do

        PieCtrlRegs.PIEACK.all = PIEACK_GROUP3;     /* Must acknowledge the PIE group                                     */
    EPwm2Regs.ETCLR.bit.INT = 1;                /* Clear INT flag

  • 0 in reply to Fulano de Tal
    Expert 2170 points

    Yes I am:

        PieCtrlRegs.PIEACK.all = PIEACK_GROUP3;
        EPwm2Regs.ETCLR.bit.INT = 1;

    The issue here is that the interrupt clearly is working - it triggers off the TBCTR = ZERO just fine, just not when TBCTR = PRD.

  • 0 in reply to Michael He
    Expert 2090 points

    You can check a few things to diagnose this.

    Set a breakpoint while running and make sure that INTSEL is in fact 011 in the Registers window of CCS.

    Also, are you ever modifying TBPRD at all in your code?  Maybe you can check TBSTS.bit.CTRMAX to see if it's overflowing somewhow.

    Do you have any other interrupts going off on CTR=PRD that may be higher priority (PIE groups 1 & 2) and take precedence?

  • 0 in reply to Fulano de Tal
    Expert 2170 points

    I'm seeing that INTSEL is 1 when running. I'm not sure why the value I give it isn't sticking, I don't use that register anywhere else in my code (I did a search). I will investigate further.

  • 0 in reply to Michael He
    Expert 2170 points

    I've figured out my mistake, I was writing INTSEL = 0x11, which is not 3. That's in hex, so is actually 17. Silly mistake.

    Thanks for the help.