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.

How to improve the efficiency of switching power supply

Other Parts Discussed in Thread: UCD3138, UCC27211, UCC27324

Hello, I am currently working on a 200W full-bridge hard-switching power supply with a target efficiency of 92%, and now the efficiency is only 90%.
What aspects should I use to improve the efficiency of switching power supplies? Is there a corresponding document for reference?

  • Hi

    One aspect is to looking at driving the mosfets faster by reducing the value of the turn-on and turn-off gate drive resistors. It is also worth checking the with athermal camera to see which components are getting hottest and focusing your attention on how to reduce the power loss in those devices.

    Have to done a power loss calculation to determine what the efficiency should be in theory. What is the input voltage and output voltage of the design. Which TI controller are you using.

    Regards

    Peter
  • Thank you for your advice.
    The product input voltage is 24V, the output voltage is 5V/50A, the controller is TI's UCD3138, full-bridge hard-switching topology, the primary driver is UCC27211, and the secondary driver is UCC27324. The hottest devices in the prototype are the primary MOS tube and the secondary MOS tube.
    At room temperature full load test, the shell temperature of the primary MOS tube is 100 degrees, the shell temperature of the secondary MOS tube is about 80 degrees, the temperature of the magnetic element is about 50 degrees, and the temperature of other chips is about 40 degrees.

    The biggest loss is MOS tube. I want to lower the resistance of the drive resistor according to your instructions. I used 10 ohms before. I also find some better performance MOS tube, Rdson, Qg, Cds smaller MOS tube. I also saw from the Internet that a Schottky diode is connected in parallel with the MOS tube, which means that the loss can also be reduced.
    Thank you.
  • Hi

    I think you are following the correct path. The primary side and secondary side mosfets are much hotter than the transformer.

    If you monitor the gate drive waveforms and Vds waveforms while reducing the gate drive resistor values you should be able to tune the design for maximum efficiency. I would expect the turn on resistor values to be in the region of 1R to 4.7R and maybe even 0R for turn off.

    It is also worth trying out different mosfets with lower Rdson and Coss to get the balance between switching loss and conduction loss.

    Placing a schottky diode in parallel with the SR's may not help as the inductance in the diode package may prevent the current from commutating from SR body diode to the Schottky device.

    Regards

    Peter