UCC28950: Shim inductor overheating problem at light load.

hi soongil,
kindly please load data as below,
1, schematic
2, shim inductor detail data.
3,saturation waveform.

thanks

Hi Bliss,
Thank you for your reply.
We want our development materials not to be exposed.
So, I'd like to share schematics and transformer by e-mail.

1. The waveform overheating shim inductor.

    C1 : QD2 drain(QB in yours block diagram), C4 : shim inductor current.

2. The waveform overheating shim inductor. 

    C1 : QD2 drain(QB in yours block diagram), C2 : QD2 gate, C3 : QD1 gate, C4 : shim inductor current.

    It seems to be overheated when excessive ringing on the drain is observed.

3. modify RD821 and RD822(RdelAB and RdelCD in yours block diagram) , 22kohm -> 56kohm.

   Ringing is disappear, but regulation is going to poor.

Others are attached.

Shim inductor_SER1512-103ME.pdf

3225.UCC28950 Design Tool_SLUC222C_1U-CONVERTER.xlsx

Hi bliss, 

I have one more question. Is it possible to disable burst operation at light load?

Best Regards.

dear,

you can try to reduce 1 or 2 turns of transformer primary side  what you can get better regulation.

thanks

Hi,
I haven’t heard back from you, I’m assuming you were able to resolve your issue. If not, just post a reply below (or create a new thread if the thread has locked due to time-out) .

Hi Bliss

I haven't resolved yet, and it is assuming to be a malfunction due to the noise increase related to the Trr of the switching FET at light load.

Therefore, if the dead time is greatly increased, the overheat is improved, but the efficient and regulation is to be worse.

Also, Since it takes 14 weeks to adjust the number of transformer input turns, changing transformer must be determined by checking more data.

Is it hard to design PSFB to guarantee ZVS operation with whole load area?

I think the parasitic diode of the MOSFET will not be able to withstand in the high frequency operatng if the ZVS fails.

P.S
I have not received much of the answers to many of the questions I have asked.

Best regards.

hi soongil,

It is difficulty to disable burst mode function. according to datasheet of UCC28950, If the converter is commanding a duty cycle lower than TMIN, then the controller will go into Burst Mode. you can try to adjust the Rtmin to change the burst off threshold.
But please do not short TMIN to GND. If TMIN was shorted to GND, this is unpredictable based on data sheet equation. Entry point into Burst Mode will be unpredictable. This should not immediately damage the PSU.

Hi soongil,
1, overheat
You need to confirm if the circuit is working in ZVS. in order to the circuit work in ZVS state, you can try below items:
a, testing variety of Coss mosFET
b, testing Higher inductance Ls
c, try to increase dead time.
2, It is difficulty to match ZVS at whole load conditions. If you don't want to change transformer,You need to balance between the ZVS and regulation.
And I agree your idea about parasitic diode of the MOSFET, you can try other mosFET(better diode).

Hi Bliss

Thank you so much for your answer.

I has been confirmed that the overheat of the inductor is a problem with ZVS.

Now, our circuit is working in ZVS from 50w to 1kw without any malfunction.

 1. Increased Shim inductance(10uH -> 20uH)

 2. Increased dead time

     concerning this item, I have a one more question.

      Do I have to equal the resistance value of DELAB and DELCD?

      I has changed DELAB=16k and DELCD=47K to work in ZVS, and the result waveform is below(Po=1KW).

     (ch1 : Vds of QD, ch2 : Vds of QB, ch3 : Gate of QD, ch4 : drain current of QD)

  (ch1 : Vds of QD, ch2 : Vds of QB, ch3 : Gate of QD, ch4 : Current of transformer)

To work in ZVS down to zero load, I think it's good to reduce the MOSFET's coss, but what do you think about the increase?