PFC question

[stj]

does it do much on a 230v psu?
and does it add stress to the main filter caps?

i have a few chieftech psu's with active pfc built as an afterthought using an exta board mounted over the primary heatsink.
these things are unreliable and generate a lot of heat - i would rip it out if i knew it wouldnt cause problems.

[everell]

Re: PFC question

picture please

[stj]

Re: PFC question

damn, i was afraid of that - i'll have to go dig them out.

i want to rebuild them as backups, i recapped them already, and may change the output rectifiers - i'v not checked what's already there.
but the pfc really pisses me off because it blocks the airflow.
and these things generate a LOT of heat.

[SteveNielsen]

Re: PFC question

Experiment with one of them? Try it and see if it works out ok.

[stj]

Re: PFC question

o.k.

they are chieftech psu's re-badged.
thermaltake
pure power-480APD

the pfc board has "WAN NIEN" printed on it.

the pfc board is using ucc3818 controller, byc10 diode, 20n60 fet and some other big semi i cant see.

notice the bullshit wattage rating on the label, and the sticker i put on it with the wattage ratings from the chieftech badged unit!!!

pics.

[goodpsusearch]

Re: PFC question

I am sorry to deliver some bad news but this platform oscillates if it is loaded heavily at 12V. It was never designed for ATX12V in the first place, but a tweaked 5V heavy psu.

I have the PPFC version. You can easily remove the APFC, but the input caps Voltage would be lower. Q=C*V. Less Q means the psu will be able to output 350W.

If this thing has worked many years without failing or drying the input caps I would leave it alone.

[stj]

Re: PFC question

well i have 3.
one is in use 24/7 but caps are going because i can make the 5v line glitch by turning mains stuff on.

other 2 died.
one wih no symptoms visible, the other blew a resistor on the pfc because a zener shorted.
that's been cleaned up.

i'm not concerned to much about the 12v issue, i have lots of 5v based stuff i can use them with.

[momaka]

Re: PFC question

Gut the APFC circuit. You can thank me later.
(And No, it won't cause a problem as long as you don't mess up anything.)

APFC makes PSUs less efficient. And yes, it does put more stress on the bulk input caps.

However, APFC provides better hold up time - both due to being able to work from a wider range of input voltages and being able to store more energy in the bulk input caps (a 100 uF cap charged to 200V will hold more energy than a 200 uF cap charge to 100V... since the energy, E, in a cap is = 0.5*C*V^2). The hold-up time is important. High power PSUs need APFC to have good hold up times. Either that, or very very large bulk input caps - which can get pricey (and we know manufacturers like to save money).

As for PPFC, that's a different story. Leave those, as they also provide extra filtering and RFI/EMI suppression on the input.

WAN NEIN is a fairly large PCB manufacturer. You will see it on plenty of electronics.

[ben7]

Re: PFC question

Gut the APFC circuit. You can thank me later.
(And No, it won't cause a problem as long as you don't mess up anything.)

APFC makes PSUs less efficient. And yes, it does put more stress on the bulk input caps.

However, APFC provides better hold up time - both due to being able to work from a wider range of input voltages and being able to store more energy in the bulk input caps (a 100 uF cap charged to 200V will hold more energy than a 200 uF cap charge to 100V... since the energy, E, in a cap is = 0.5*C*V^2). The hold-up time is important. High power PSUs need APFC to have good hold up times. Either that, or very very large bulk input caps - which can get pricey (and we know manufacturers like to save money).

As for PPFC, that's a different story. Leave those, as they also provide extra filtering and RFI/EMI suppression on the input.

WAN NEIN is a fairly large PCB manufacturer. You will see it on plenty of electronics.

The PFC does increase efficiency in regards to power transmission losses, from the power station to your house/place.

-Ben

[stj]

Re: PFC question

maybe i will de-pfc one and then compare them to see how much heat is generated.
the one that had pfc failure may have trashed the controller - i dont know yet.
if it has - it's going!


btw, the 2 dead ones came from a friend.
he had been sold them to run athlon64 dual-core servers with 6 drives.!!!!!!!!!!!!!
maybe that's why they died in under a year.

he gets miss-sold a lot of shit it seems.

[SteveNielsen]

Re: PFC question

maybe i will de-pfc one and then compare them to see how much heat is generated.

That's what I meant by saying try it and see.

[Pentium4]

Re: PFC question

I'd gut the PFC.

The PFC does increase efficiency in regards to power transmission losses, from the power station to your house/place.

Isn't that just because the power factor is higher? I think at the PSU level, PFC decreases efficiency. Efficient designs just so happen to usually have PFC.

I think tom66 brought up a flawless argument in regards to PFC in ATX PSUs, if it increases efficiency, then why does all the PFC silicon have to be on a large heatsink?

[stj]

Re: PFC question

yes, if the pfc generates heat then it is using additional power.
why should i pay to help a private electricity company?

[momaka]

Re: PFC question

The PFC does increase efficiency in regards to power transmission losses, from the power station to your house/place.

Technically speaking, Yes, you want to have a high PF to increase the efficiency of the electric transmission.

But PF correction is not so simple. Even the power lines themselves appear slightly inductive, so PFC is needed at each substation. Thus, most substations do have PF correction apparatus. And they correct not only for the transmission/power lines, but also entire residential neighborhoods (with industrial and commercial, it could be a different story).

So, in the grand scheme of things, a PSU or two in your home will hardly make a difference to the power grid, if at all. This is because other loads in your house may offset that. A SMPS usually appears as a slightly capacitative load, whereas motors and relays appear as slightly inductive loads. There are plenty of devices in the house that appear inductive. The refrigerator is one of those devices, and it usually has a high ON-time duty cycle.

Here is an interesting read with some nice visuals:
https://cdn.badcaps-static.com/pdfs/...899f119c1b.pdf

PFC pretty much lowers reactive power at the cost of increasing the real power used. This is bad news for you and good news for the PoCo, since they only bill you for real power usage.

I believe PCBONEZ did a in-depth write up on this a long time ago and pretty much covered the whole topic.

I think tom66 brought up a flawless argument in regards to PFC in ATX PSUs, if it increases efficiency, then why does all the PFC silicon have to be on a large heatsink?

why should i pay to help a private electricity company?

^ Exactly! (to both quotes)

[ben7]

Re: PFC question

Technically speaking, Yes, you want to have a high PF to increase the efficiency of the electric transmission.

But PF correction is not so simple. Even the power lines themselves appear slightly inductive, so PFC is needed at each substation. Thus, most substations do have PF correction apparatus. And they correct not only for the transmission/power lines, but also entire residential neighborhoods (with industrial and commercial, it could be a different story).

Yeah, we have a PFC bank a block away from our house. It's got 3 oil-submersed switches, and 3 capacitors. A typical setup you see often around here.

So, in the grand scheme of things, a PSU or two in your home will hardly make a difference to the power grid, if at all. This is because other loads in your house may offset that.

... but they add up! :P

Here is an interesting read with some nice visuals:
https://cdn.badcaps-static.com/pdfs/...899f119c1b.pdf

PFC pretty much lowers reactive power at the cost of increasing the real power used. This is bad news for you and good news for the PoCo, since they only bill you for real power usage.

I believe PCBONEZ did a in-depth write up on this a long time ago and pretty much covered the whole topic.

LOL, I really don't know how to respond to this xD
Maybe you should look at the overall benefits. (Better power transmission reduces pollution)

Lower harmonics are also a good thing too.
The US's power grid is already stressed out, don't want to stress it any more.

I think tom66 brought up a flawless argument in regards to PFC in ATX PSUs, if it increases efficiency, then why does all the PFC silicon have to be on a large heatsink?

Mhm, even at a high efficiency, with high powers involved you need a big heatsink.
Take multi-gigawatt electric transmission transformers as an example. They get over 97% efficiency, but they still need HUGE cooling fins, with fans.

Basically, the ultimate question I am bringing up, is does the power saved in transmission lines offset the losses in the PFC circuit?
There must be a reason why they use PFC in power supplies (however sinister that reason may be - lol), otherwise, why would they do it?

-Ben

[stj]

Re: PFC question

i actually doubt the u.s. power grid is stressed out.
it was built to run the powerhouse of the 1960's and 70's of western manufacturing.
the u.s. looks post-apocalypse now, all those power hungry factory's are gone.
the grid system is probably virtually idling.

but if they told you that, they couldnt leverage smart-meters and price increases along with "carbon" taxes onto you.

the u.k. is pulling the same lies.
the government says unless we want rolling blackouts we have to reduce power use and build new nuclear power plants.

BULLSHIT.

they want the plants because they need to enrich the fuel for the "trident2" ballistic missile program.
the u.k. was full of steel smelts and car factorys etc that are now gone - we are running at under 20% capacity.

[kaboom]

Re: PFC question

Yeah, we have a PFC bank a block away from our house. It's got 3 oil-submersed switches, and 3 capacitors. A typical setup you see often around here.


... but they add up! :P

Apples and oranges. Those pole-top caps do not, and cannot correct for SMPSes. First, the caps on poles and in substations, like any other capacitor, only corrects for its opposite sign; they're for compensation against lagging (inductive) power factor, caused by the lines themselves under high load, as well as lightly loaded and/or non PSC motors.

This is why POCOs run the voltage on the high side during times of heavy loading. If the voltage is kept higher, there are more capacitive VARS cancelling the (inductive) transmission line VARS. At least semi-legitimate (meaning non "FE") POCOs.

Those single-phase hermetic compressors for residential/light commercial central air always have a run cap, phase-shifting an "auxilliary" winding to give a moving magnetic field. They don't run at 50% lagging.


Capacitor input rectifiers (ALL non-APFC SMPSes) suck out the peaks of the sine wave when their caps charge. Not only that, the peak currents, when 120 degrees out of phase (as in Euro 416Y/240V secondary distribution) add and create triplens.

In the case of a resistive load with equal phase currents, neutral current is zero. Worst case imbalance results in neutral current being the same as phase currents.

If A=1A, B=1A, C=0A, then N=1A.

But in the special case of cap-input rectifiers, even with equal phase currents, not only does the neutral current not cancel, it becomes 1.732 times the phase current!


Think about the effects of this in an old European village, with #6CU secondary service and the 500VA transformer, which serves everything/everyone, being hundreds of feet from the load(s).

It should be no surprise that APFC was "suggested" for areas with such distribution.

LOL, I really don't know how to respond to this xD
Maybe you should look at the overall benefits. (Better power transmission reduces pollution)

Lower harmonics are also a good thing too.
The US's power grid is already stressed out, don't want to stress it any more.

Not quite. There are "policies" in place to make certain generating companies look good to their stockholders. "FE" comes to mind. Yes, that "FE."

Rather than effect better transmission, with cap banks at major (230/69KV and up) substations, they use peakers near load centers. But not for watts, as they should be used. Their exitation is adjusted to lead the voltage coming in, making the generators "look" like caps to the transmission system. IOW, generator capacity is wasted (in K/MVARS) going towards compensation; that capacity could just as easily go towards K/MWatts instead, with caps doing VAR compensation... Watts may be coming from TMI, but in order to get those watts, you've got a turbine peaker in your backyard.

But "keeping the fuel/money flowing" looks better to "FE's" stockholders than springing for caps.

PECO, OTOH, has caps on everything 34.5KV and up, and does just fine with more centralized generation.

PSE&G does this too, but I'm not sure of the lowest voltage they use caps on, and in what areas. Even then, they both use peakers, as expected.

"FE" just spends money and "spins the gennies." It should be no surprise that they're prone to voltage collapse and cascading blackouts, without caps in critical locations.

Wasn't it heavy power flows and loss of transmission ckt's (too much current, then line inductance dominates) combined with lack of leading VARS that caused the "northeast blackout?"