Because of the Thanksgiving holiday in the U.S., TI E2E™ design support forum responses may be delayed the week of November 23.

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.

TIDM-02002: PWM generation: phase shift between primary and secondary

Part Number: TIDM-02002

Dear TI team,

I´m checking the design of the TIDM-02002: CLLLC reference design and while comming to the PWM generation, I got some doubts:
I see that the primary PWM carrier and the secondary PWM carrier are shifted by 90° (see picture 25, also attached below).
I´m wondering why is done in this way. What about using both triangular carrier sync and generating the PWM with the same COMPA for prim and sec? Or am I missing something?  

Another question, the COMPA of the synchronous rectification would be 50% but triggered with the current comparator, right? 

Thanks for your support.

image.png


  • Hi,

    TIDM-02002: PWM Phase Shift and Synchronous Rectification

    Why the ~90° (50% of period) phase shift between primary and secondary carriers?

    The secondary PWM is phase-shifted by approximately 50% of the period (which appears as 90° in the up-down counting triangular carrier) for two key reasons:

    1. Blanking window alignment: The blanking window for the CMPSS (used in synchronous rectification) is only available at the time-base counter zero and period events. By phase-shifting the secondary PWM by ~50%, the secondary's zero/period events align with the midpoints of the primary switching — exactly where the blanking needs to suppress noise during the secondary-side switching transitions 1. 

    2. Isolator propagation delay compensation: The PWM pulse to the secondary side passes through an isolator, adding propagation delay. The phase shift implements a small advance of the secondary PWM relative to the primary's falling edge to compensate for this delay 2. The phase shift is calculated in CLLLC_calculatePWMDutyPeriodPhaseShiftTicks_primToSecPowerFlow() (ISR2) and applied via CLLLC_HAL_updatePWMDutyPeriodPhaseShift() (ISR1) 2.

    3. F28004x hardware constraint: The PWM sync chain on the TMS320F28004x is fixed (PWM1→2→3→4), so the secondary PWMs must always be slaves to the primary PWMs. You cannot sync in the reverse direction on this device 3.

    I´m wondering why is done in this way. What about using both triangular carrier sync and generating the PWM with the same COMPA for prim and sec? Or am I missing something?  

    Why not use both carriers in sync with the same COMPA? If you synchronized both carriers and used the same COMPA, you would lose the blanking window alignment and the ability to compensate for isolator delay. The trip actions (xA tripped when time base counts UP, xB tripped when counting DOWN) rely on the phase-shifted relationship to correctly generate the synchronous rectification turn-off at zero-current crossing 4.

    Another question, the COMPA of the synchronous rectification would be 50% but triggered with the current comparator, right?

    The synchronous rectification does not use a fixed 50% COMPA duty in the traditional sense. Instead, the CMPSS (Comparator Subsystem) with integrated DACs detects when the secondary tank current reaches zero, and a trip action turns off the secondary-side switches at that point 5. The blanking window (generated by EPWM3 at both TBCTR zero and period events via EPWM_DC_WINDOW_START_TBCTR_ZERO_PERIOD) suppresses false triggers during noisy switching instants 6. The trip is latched until the next zero or period event to prevent spurious turn-on from noise 6.

    So the rising edge of the SR drive is controlled by the blanking window release (at zero/period), and the falling edge is controlled by the current comparator trip — it is not a simple 50% COMPA compare value.

    1. TIDM-02002: ISR1 Phase Shift Discussion
    2. TIDM-02002: PWM Synchronous Rectification
    3. TIDM-02002: ISR1 Phase Shift Discussion - Reply #4
    4. TIDM-02002 Design Guide - Figure 25
    5. TIDM-02002 Design Guide - CMPSS Features
    6. TIDM-02013: Blanking Window Discussion (Srikanth Yerra)

    Thanks

    Srikanth Yerra