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.

UCC28250 Half Bridge

Hi,

I have a half bridge converter with schottky diode as secondary rectification (centre tapped transformer with 2 diodes), the power rating around 400W.

we have made several boards and in some we have issues at No/Low Load. in the boards with issues I need a minimum load to achive proper operation. At no load up to say 50mA we hear a noise from controller. this current level is slightly different from board to board and increased with input voltage. So say at 400Vin we need a minimum of 50mA to have proper operation and at 700V we need 100mA. During this low laod range, I looked at EN pin, SS, COMP and g-s voltage of the top and bottom MOSFETS in HV side. EN voltage is ok and constant, when we operate in this low load area, we get some random noise at which SS voltage drops and comp pin voltage drops and gate source voltage from a very narrow pulse goes to a full duty cycle. Same happens on the g-s of both top and bottom mosfets randomly.  As this issue, noise is more severe (more repetitive) at High Vin then I am thinking it can be to do with the minimum duty cycle capability of the controller, however, as I have this issue in some boards and not have it in other boards, I take it it has to do with the tolerances in the parts, hence I should be able to solve this on the board.

as this minimum current level is different from board to board hence I dont believe it is a good idea to use a bleed resistor across the output to achive a minum load for the converter. The reason is that if consider the worse board with the highest minimum current requirement and highest input voltage to set the bleed resistor, then the power loss in the resisotr would be excessive around 9W. Can you think of anything else I can try please?

 

 

  • Hi Omid,

    I think the situation you saw is some kind of expected.

    This "issue"  would happen for most controllers though it could be half-bridge, full-bridge or Forwad topology.

    This is caused by the diode rectification, if the load is very low, it would means the duty cycle should be very narrow , and comp may be forced to low and high by the rise/fall of Vout. The burst mode would cause audible noise.

    To avoid this mode, Dummy load is a choice.

    I think another choice is using synchronization rectification instead of Diode Rectification. 

    Regards

  • Hi thanks,

    as I mentioned, the thing is that the minimum current needed to get proper operation is different from board to board, 50mA up to 200mA, hence the power dissipation in the bleeding resistor can be as high as 9W, which is unacceptable. Also in some boards there is no issue at all, so I believe I should be able to fix this on the baord without the bleed resistor. I think maybe we are on the borderline od stability (or minimum duty cycle capability of the chip or ... hence the tolerance of the components lead to this issue becoming quite random from baord to board. Any help or suggestions would be much appreciated.

    thank you

     

  • Hi,

    this issue is only at very load and the minimu load current to get past this low current unstable range is different from baord to board. in sme boards I dont have any issues at all and in some the converter needs a minimu load of 300mA

  • Hi,

    Will you change interchange the magnetic components between different boards to see whether the tolerance of Transformer/Inductor is the major cause?

    Regards