Just one question: why?
There's no reason I can think of for doing that. If you're going to do that, you might as well use a transformer, which also provides isolation and will be much smaller, than a 60Hz transformer, given the same power level. In short, just use a switched mode power supply.
This is what I was originally thinking...
"
If I'm boosting to correct PF, then inverting, then rectifying, then bucking, I figured I could eliminate some of the losses across components by reducing some. I figured if I started out inverting, and my series cap esr was low that my only major losses should come from switching and rectifying, this seemed like it might reduce components, reduce cost of components, provide decent PF, while essentially only having to control the switching Hz. That seemed less complicated than PFC control loops ect.
"
Maybe I need to rethink.
Dual boost bridgeless PFC + inverter + parallel capacitive droppers
Adjust boost inductor PWM based on bulk capacitor voltage. Adjust inverter Hz to manage the loads' currents.
^^^Will that work in theory? Should I be looking to replace (capacitive dropper + inverter) for HF step down transformer, then just switch the primary side of the step down xfmr being fed by bulk cap to manage current of the load? The bulk cap V should be pretty constant as I'm varying boost PWM to create a predefined voltage desired at bulk cap? Goal is efficiency. Any advice welcome. If I skip inverter then I don't have to worry about my top side switches, I didn't know which were best, PFETS or triacs, or transistors or what. If you want to give your opinion on what you'd prefer for an application like that, that'd be awesome. The more I can pick up the better.
I know I can buy a PFC driver IC, but I'm trying to understand if I understand. I want to use arduino to monitor line V and at ZC detect bulk cap V and then adjust boost PWM to refill bulk cap based on the ZC measurement. Maybe implement a PID control for the PWM so that it incorporates previous bulk cap V measurements as well as instaneous bulk cap V measurement at ZC. I'm not too familiar with control loops and terminology. Or PID control either, but think I can implement with a little effort. <--does this sound plausible and efficient? Any tips? Tricks? I'd work some safety into it too, some sort of current limit/shutoff, resetable, amoung other things.
I'm intending to run multiple loads off the bulk boost cap. So more than 1 xfmr would be drawing from the bulk boost cap. Or more than 1 capacitive dropper would be pulling off bulk boost cap.