Half-bridge SMPS without current sense transformer.

[mtwieg]

It probably wouldn't be too difficult to use a sense resistor on the capacitor side of the primary coil. It won't be experiencing too much common mode interference (so long as the capacitors are big and low ESR). So a decent low cost diff amp might be doable. There are also specialized ICs meant for high side current shunt measurement which may be very useful (though they often have common mode voltage limitations). Often they are used for motor current sensing in half bridge drives, which is pretty close to this. If you do need fast current detection, then I would look at those sensor ICs.

 

[Orson Cart]

mtweig makes a good point, although I suspect the cost of the high side current measuring devices may be more than a small & cheap 200:1 CT. By the way, C3 is not needed in the sketch below,

Another alternative I have seen is to use C3 as shown in the circuit and use a bridge rectifier on this cap to drive an opto (via a suitable zener/resistor and filter cap on the opto LED) in this way high peak currents in the mosfets translate to high peak voltages on C3 which can be used to drive the opto to signal the control that the PWM should be reduced - still a bit slower than a CT but might work for this app.
Regards, Orson Cart.

 

[grizedale]

I see your point, i must have been dreaming...sorry about that.

I need to sense the current as its simply not safe to not sense primary current in most SMPS.

During transients, non-soft-starts non-soft-re-starts, serious primary overcurrents can occur, well above the fets rating.