The controller monitors the resonant ringing of the switchnode voltage and switches directly at the valley. This results in a lower voltage during the turn on of the MOSFET. Unfortunately this popular topology has some drawbacks. The input and output ripple currents of a QR-flyback are high because they are discontinuous. The primary and secondary windings are not coupled ideal.
This results in leakage inductance that stores energy during each cycle. This energy causes an overshoot (spike) of the switchnode voltage. To limit the peak voltage that the primary switch sees a snubber network is needed, that burns the energy stored in the leakage inductance, which lowers the efficiency. Another disadvantage of the standard flyback is that the MOSFET must withstand a very high voltage. During the demagnetizing time the drain-to-source voltage VDS of the MOSFET is equal to:
VDS = VBULK + VFLYBACK + VLEAKAGE
VBULK = voltage of the input capacitor
VFLYBACK = reflected output secondary voltage
VLEAKAGE = voltage overshoot due to leakage inductance
VLEAKAGE is typically in the range between 50V and 100V for offline designs.
The Two-Switch-Quasi-Resonant (2S-QR) flyback topology has the potential to overcome some of the drawbacks. First of all it recycles the energy of the leakage inductance. Secondly the maximum drain-to-source voltage of the MOSFET's is equal or less the input voltage. The tradeoff for using this topology is not only the higher design effort, but also the obvious cost of the extra components. Two MOSFET’s (Figure1: Q1, Q2), two diodes (Figure1: D2, D3) and a driver-IC or gate-drive transformer are needed. It depends on the specific application if the 2S-QR topology is a serious alternative to the standard flyback.