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.

LM5025C: Clamp Capacitor Value

Part Number: LM5025C
Hi, we are developing Active Clamp 48V DC-DC but unstable, especially worry about broken a gate driver when load is haevy.
Attached is simplified primary schematic controller to main transformer, and we began to look at Clamp capacitor.
Now using 0.22uF, is this match to our application? Are there any calculations?

Another conditions are
Primary-inductance of transformer : 5uH
Switching freq. : 200kHz
RT=30kohm

Your prompt reply is highly appreciated.

Best Regards,
  • Hi

    I would recommend you check the dead time between OUT_B (clamp drive) and OUT_A. The gate drive transformer in the high-side should have 10-20 ns of additional prop delay which the low-side does not have - this will skew the leading edge to falling edge dead time between the two drives. The dead time will change as a function of load - is it enough at full load?

    Regarding the clamp capacitor, it is typical to allow ~20% ripple see at the low-side MOSFET VDS. If the ripple is lower than 20% or not measurable at all, the clamp capacitor is too large. Having too large of a clamp capacitor will inhibit transient response. If the clamp capacitor is too small, the ripple voltage on VDS>20%, then transient response will improve at the cost of higher VDS.

    Are you really only using one SR on the secondary at 200W? If the SR is self-driven, please verify the gate drive is stable as load is increased?

    If you take a look at SLUA535A, section 4.3.2 details the selection and calculation of the clamp capacitor for an active cl;amp forward converter.

    Regards,

    Steve

  • Hi,

    I have not heard back since the last reply. I assume the issue you were having is resolved and I am closing this thread. Thanks for using the e2e forum and if you need additional support, please open a new thread.

    Regards,

    Steve