Post Go back to editing

LT3791 Buck mode question

Hello

      Customer has and issue about LT3791EFE-1 and that is when the input voltage > output voltage, the buck-boost converter should be worked at buck mode, and the M1 & M2 should be work as buck converter, and M4 should be 100% on and M3 should be OFF, however when they measured the waveform at boost side of inductor, it is not a flat waveform like Vout, it is also a switching waveform, that violate customer understanding about H bridge buck-boost converter. 

      Why we see switching waveform, on boost side of inductor?  Is it normal? Thank you. 

Patrick

  • Hello,

    are you sure the Vin is large enough to operate in buck mode? if Vin is only slightly higher than Vout the Dutycycle might have a value where the part operates in BuckBoost Region:

    First measure is to increase Vin and check if something changes at the switch nodes of the converter

    It would be helpful to show measured waveforms for both Switch nodes and give more details on the real circuit.

  • Hello

            Customer use 24Vin and 12Vout. while at light load, 3791 looks not turn on M4 and the current flow to body diode of MOS as below waveform on DCM mode. If we increase load current to exit DCM mode to CCM mode, the vout waveform is flat 12V waveform. 

            Could we know M4( boost side MOSFET) will not turn off on DCM actually? 

             Customer thinks has risk due to current flow via MOS body diode, so they have two solution for below. 

    1. Pull high CCM pin, let converter run at CCM mode, and this is our solution now.
    2. It is recommend that CCM pin must be connected to C/10 for low current operation. Ex: 5A output current, 0.5A will be enter DCM mode. And also add schottky diode parallel with MOSFET to prevent current flow through the body diode.

    CH1: buck side phase node waveform
    CH2: boost side phase node waveform
    Switching frequency is correct as setting 200kHz

    BR

    Patrick