RMS value of current spikes from bridge rectifier

[MacIntoshCZ]

Hello i have designed active pfc, but i want to enable it after 10% load is achieved. So i was thinking about measuring current and when it reaches some value, enable it than. AC line is 230V 50Hz.
Whats my options? Do i need to add mcu and use ADC? Or could it be done with some filter?
thanks

[MacIntoshCZ]

I think something like this could work.

[Circlotron]

I'd measure the DC current after the main HV filter capacitor and use that as an indicator of load. If you are measuring the value of the current spikes, the peak value will change when the PFC starts working. This makes accurate measurement of load much more difficult and there is the possibility of it dithering either side of the threshold as the measurement circuit changes what it itself is actually measuring.

Edit -> If you measure the DC current, when the PFC starts up the DC voltage will rise and if it is feeding a SMPS the DC current drain will decrease, possibly causing the PFC to turn off again... 

[Alti]

Hi. I do not think you need a shunt to figure out the load is high. Since frequency is constant 50Hz, you'll observe ripple on PFC input cap.

So just use high pass filter on the cap: ripple is high -> load is high.

I am not sure how much time you spent on the spice model but for sure this is not 230V AC.

[MacIntoshCZ]

Edit -> If you measure the DC current, when the PFC starts up the DC voltage will rise and if it is feeding a SMPS the DC current drain will decrease, possibly causing the PFC to turn off again... 

Yes, thats right. Nice catch. thanks

[MacIntoshCZ]

ok thanks for tip. Spice model is not accurated. I just open it after long time, and plays with it.

[MacIntoshCZ]

BTW this is hearth of my board.

[Circlotron]

Assuming the boost converter runs in continuous mode, for the main diode D4 use a SiC schottky instead. It will make life so much easier.

[MacIntoshCZ]

Can you explain why? Lower forward voltage ( lower losses) i guess... I got this one on my hand. Its mostly done from scratch.

[Circlotron]

A SiC Schottky actually has higher forward voltage drop than a normal diode, but the big deal is essentially zero reverse recovery. At the instant the mosfet turns on a normal fast recovery diode will allow current to flow backwards through itself for several tens of nanoseconds so this means at that instant the mosfet has both high voltage across it *and* high current through it so it is subject to switching stress and some amount of heating. By comparison a SiC Schottky just lets go easily and so the mosfet stays happy. The diode you have chosen will of course work, just that it and the mosfet are grinding each other’s gears behind the scenes.

If the circuit is running in discontinuous mode, that is to say, the inductor current runs out before the mosfet turns back on then no big deal. Your chosen diode will be okay. The inductor current will actually reverse a little and turn the diode off softly.