Simple PFC algorithm in software

Hello,
We wish to make a 230VAC input dual, interleaved Boost PFC stage (3.5kW) whose output voltage can be tightly controlled from 330VDC to 440VDC.
Dual PFC controller chips like FAN9672 simply don’t allow this wide range of output voltage control. Therefore we need to make our own PFC controller with a microcontroller running code that we write.
In order to have good duty cycle resolution we would need the microcontroller to be running at a very high frequency. However, above 1kw in power level, the PFC regulations on mains harmonics get slacker and so a “Modified” PFC algorithm can suffice…as follows. We simply switch the Boost FET at just a couple of different duty cycles…if on maximum power, then we switch the FET right across the mains half cycle. As the power reduces, we simply have an interval of time near the zero crossing where the FET is not switched at all….and we increase this “dead-time” interval as the power throughput is required to reduce further.
Do you believe this is an acceptable PFC algorithm for this power level?
(its just that we must do this quickly and with minimal code complexity for our junior software engineer)

For active PFC correction, your DC output voltage *has* to be above the peak of the mains. You would normally have a post Buck boost controller if you wanted a large range of control.

For active PFC correction, your DC output voltage *has* to be above the peak of the mains.

Click to expand...

Thanks, very true, but we've seen supplies that 'cheat' and they have a power diode over the inductor and boost diode, and just let the mains peak come through like that.

As I understand it, poor power factor tries to draw high A during low V, and small A during high V. I don't think it can be handled effectively simply by disconnecting the supply during times when the voltage waveform is near zero.

This is not to say I have an effective PFC solution for this case.

the extra diode you refer to is only for start up, to stop the resonant LC ring up of the +HVDC at switch on, it is not to defeat some pfc requirement.
There is no easy software solution to control for good PF.
We have designed many circuits that allow control of PFC output voltage using a PWM signal from a uP, a clever analog designer can always get around the chip limitations...

thanks, I appreciate that is what that diode is supposed to be for, but we saw current coming through it when the pfc output voltage was set low, of course, I agree with you, there shouldn't be current coming through there after startup, but there was, during normal operation (the reason for sometimes setting the pfc output low is as discussed in top post).

My boss thinks we can do pfc control in software by just copying sample software from microchip pfc application note.
https://ww1.microchip.com/downloads/en/AppNotes/01278A.pdf

If we cannot alter the pfc output voltage from 340vdc to 435vdc then we have had it, because we need to do that to get the llc stage operating at resonance for all battery voltages during charge...(300v to 400v)

We can do it by just allowing to go away from resonance frequency, but then we are not making it as good (not as efficient) as currently available market solutions, and our investors will pull the plug

340 to 435 should be do-able as it is above 325, good luck with the soft ware for the pic dsp, interference can be an issue...

Thanks,
What is unfortunate, is that there are no PFC control chips out there which allow the user to actually set the overvoltage level. All the proprietory chips set it too low and we cant get up to the 435VDC level. (or down to the 340vdc level)
I can appreciate now that due to the legal limitation of single phase mains outlets to 16A of current, one absolutely needs to be fully power factor corrected otherwise the high reactive current will mean the fuse tripping even when one is not actually delivering the full 3kw to the load.

Hello
Does anyone know of a single phase PFC controller which allows the user to set the overvoltage level themselves (we have vin = 230VAC). We must be able to vary the Vout of the PFC from 340VDC to 435VDC.

UC3855 might lower your switching losses and RFI...

- - - Updated - - -

UCC3819A has set-able Vout on the error amp...

I have used the FAN6982 and you can set the output volts on that, again as long as the DC is above the peak of the AC incoming supply. It's also worth having quite a lot of DC capacitance too to aid controller stability.