Pc power supply gate drive trans

[dr pepper]

While reverse engineering a Pc supply looking for a gate drive transformer (too lazy to wind my own) I noticed that on its gate drive transformer the full power trans primary feed goes through the gate drive transformer, there are 3 'secondaries', 2 go to the base, & emitter of the switching trannies and the one mentioned carries the current to the switched side of the power trans primary from the switching trannys collector/emitter, its a half bridge topolgy.
I think I've seen this before, why do they do this is it to speed up the switching trannies?

 

[dr pepper]

---

I think the winding might be for a self oscillating startup circuit, like the kinda thing found in the old Cfl lamps.
I assumed the Pwm chip would be powered from the 5v standby supply, but maybe its self starts and powers itself.

 

[dr pepper]

Ok, the 5v standby part of the ps is independant, the main supply self starts.
On the Ps on signal going active the output stage self oscillates as I mentioned above, then the Pwm chip on its o/p powers up and takes control regulating the o/p's, interesting design.
The base drive transformer seems to work ok without the feedback winding connected.

 

[FvM]

I see this kind of current feedback winding also in an ATX supply demoboard schematic. The circuit has no features to start in self oscillating mode and the driver is supplied by the auxiliary 5V_SB supply. Respectively it's started by the SG6105 switch mode controller, mandatory for PC on-off-control. The schematic is a bit blurry but can be still read.

Purpose of current feedback is in my view to source most of the driver power by the output current and and achieve forced beta operation of the switch transistor.