Synchronous Rectifier Controller Circuit for Flyback

Hi everyone,

For flyback converter, I am planning to change secondary side. I want to change output diode with mosfe that has low Rds(on). But there is any analog circuit to make it without special IC?

Hi,
I think you should consider an opto and a driver.

A self steered synchronous MOSFET rectifier can be made with an auxiliary winding controlling the gate.

Yes, with an auxiliary winding you can do it...its easy when the converter is in CCM, but when it goes into DCM, then you have to put in circuitry to ensure you dont get backflow of current through the secondary just after the secondary coil has discharged and before the primary has switched back on.

What you need for flyback to handle the DCM case is a signal to indicate when the primary FET has turned off…….then, following that, you want a delay of some 200ns, then you turn on your sync fet. Then you “look” at the aux coil and when it transitions and changes polarity, you then turn the sync fet off……..you then put the sync circuit at “rest” until the primary fet turn on comes again…….then you start waiting for the primary fet turn off signal again…..etc etc

There are chips for this - use these 1st - while you are developing and testing your own discrete circuit to do the same job ...

The main problem for synchronous rectification is the ringing. If you have a topology such as flyback, the stress and ringing on the output Mosfets and diodes are severe.

Because of that the discrete control should have a time delay to filter these ringing sequence.

Commercial ICs have a programmable ringing compensation delay so that the differential comparator acts just after this ringing time.

It means also low efficiency , because during this ringing time the diodes are active. If you have higher frequency this effect worsens. Because the mosfet will have less time frame to act.The diode is more active while waiting for the ringing to stabilize.

Subtract also the switching loss of the mosfet from the usable active time window where the mosfet is active.

Yes and generally the ringing on the sec diodes of an offline flyback is worse ther higher is ns/np....this is because the leakage inductance referred to the secondary is greater the higher is ns/np.
Then again, if your ns/np is relatively lower, then you can quite easily snub out the ringing, as the leakage inductance referred to the secondary will be low.