• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

What's Active PFC good for?

paulsiu

Member
I looked at various article on what active PFC is. Many of them talk about reduction of harmonics, or increase power factor. In layman's term, what does this feature do? Does it handle AC with dirty motor noise? Does it fix problems like a outlet with a bad ground ( a common problem for renters)?

Paul
 
The primary purpose of APFC is to help out the power company. By pulling your system's power factor closer to unity, it reduces the reactive power drawn by your computer. Reactive power is the cholesterol of the power grid: everyone has it, no one wants it. So in concept, it reduces their operating costs, which might encourage them not to raise prices on you.

There is a side effect that APFC supplies sometimes handle "dirty" power a little better, but it's not a wide difference.

This article is not ridiculously technical, though I would hesitate to say that it's all in layman's terms.
 
In addition to what Aluvus already explained, from the customer point of view, APFC ensures better voltage regulation in relation to electrical load. With passive PFC, an increase in load causes a droop in output voltage. This droop is mainly caused by increase in reactive load.
 
On the user side, it's pretty much about regulatory compliance(at least in the EU). Large outfits, industrial and similar, are often billed extra by the power company for lousy power factor. Homes and light offices pretty much never are. Eats a little bit of efficiency in order to make the power company's life easier.
 
Better PSUs will have Active PFC. I only buy SMPSs with it. It will increase UPS run time
just a bit and why would I want to reflect garbage into my UPS??


...Galvanized
 
Active PFC is also an essential part of the design of a high quality, high efficiency PSU. The inclusion of active PFC allows the PSU designers to optimise the rest of the PSU in ways that are not possible without Active PFC (giving better efficiency and better power quality).

While Active PFC on it's own doesn't improve efficiency (early adopters of active PFC didn't get improved efficiency, and in many cases the performance of a PSU with active PFC option was worse than one without), a PSU designed with Active PFC in mind will usually have considerably better efficiency than one designd without active PFC.
 
Originally posted by: Mark R
Active PFC is also an essential part of the design of a high quality, high efficiency PSU. The inclusion of active PFC allows the PSU designers to optimise the rest of the PSU in ways that are not possible without Active PFC (giving better efficiency and better power quality).

While Active PFC on it's own doesn't improve efficiency (early adopters of active PFC didn't get improved efficiency, and in many cases the performance of a PSU with active PFC option was worse than one without), a PSU designed with Active PFC in mind will usually have considerably better efficiency than one designd without active PFC.

Which would mean that PSUs designed with APFC from the ground up are better designs then units that have it tacked on as an after thought?
 
Originally posted by: Operandi

Which would mean that PSUs designed with APFC from the ground up are better designs then units that have it tacked on as an after thought?

Yes.
 
One good side effect of active pfc is that it doesn't need a power selector switch and can handle the full range from 100-240V.

I think it handles brownouts better as well.
 
Originally posted by: zmaster
What's Active PFC good for?

One word: Europe

I dont think ive seen active pfc psu's anywhere else.

Active PFC PSUs are not particularly common in Europe. Unless you specifically look for an active PFC PSU, you can be pretty sure you won't get one. It's for the same reason that many high-end PC parts are less available in Europe than in the US.

It's certainly true that Europe mandates PFC on all PC PSUs - but traditionally, this has been the cheaper passive PFC. Active PFC is certainly not necessary in Europe. Possibly, as active PFC components become cheaper and manufacturers try to consolidate their designs - we may well see passive PFC become a thing of the past. However, we are not at that point yet.
 
Sorry - I can't resist. For decades, I have always believed that an active PFC was a private first class on active duty. 🙂
 
Originally posted by: Mark R

While Active PFC on it's own doesn't improve efficiency (early adopters of active PFC didn't get improved efficiency, and in many cases the performance of a PSU with active PFC option was worse than one without), a PSU designed with Active PFC in mind will usually have considerably better efficiency than one designd without active PFC.

Incorrect.

Any additional circuitry has power loss. Active PFC in NEVER inherent to the design of a switchmode power supply. It is BECAUSE a power supply uses switchmode technology that it requires PFC for better power factor. Because PFC requires additional circuitry there is always loss, therefore lower efficiency.

 
From an article I wrote that may never get published because the editor said I was "all over the place." (photos and diagrams missing for obvious reasons)

Power Factor Correction

Something that seems to need a lot of clearing up when it comes to power supplies is "power factor correction," or "PFC."

Some people say that power factor correction makes a power supply more efficient, but as you should now know, for a power supply to be more efficient it needs to pull less wattage from the wall. Power factor does not suddenly make a power supply pull less wattage from the wall. It does allow a power supply to pull less VA. And in some parts of the world that might save you on your electric bill.

When talking about DC, VA is the same as Wattage: V * A = VA. But when talking about AC, VA takes power factor in consideration: (V * A) / PF = VA. So VA can only equal Watts if power factor is 1.

Some electronics, like heaters, toasters, coffee makers, ranges, are what is called a "linear load" (or "simple load" or even "resistive load") and have a power factor of 1. While those with a "non-linear load", like a computer's power supply, have a power factor as low as .60.

Let's say you have a power supply that's non-PFC. We're going to assume that your PC typically puts a 250W load on this power supply. Let's also say that the efficiency of the power supply at 250W is 83%. That would mean it's pulling 300W from the wall. If you're billed by the KWh, like so many of us residential customer in the U.S. are, then that's how many Watts it's pulling from the wall is all you need to worry about.

But if you're in the EU, or even a commercial customer in some parts of the world, and are charged per KVA, then you'll be concerned with the fact that your power supply is drawing 500VA from the wall. If we were to switch this power supply out for one with active PFC and a power factor of .99, assuming it's the same efficiency at your typical load, you're only pulling 303VA from the wall.

So now that you understand what power factor correction is for, now I'm going to try to explain to you what power factor is...

Power factor is the ratio of real power consumed to apparent power. Or W over VA, which is another way to look at the equation I use above to figure out VA based on Watts and PF (simple algebra here, people.) Traditionally, PF is known as the phase difference between the sinusoidal voltage and current (amperage) waveforms.

Something that has a "linear load," like a simple resistor, has identical voltage and current waveforms, are sinusoidal and are in phase with each other.

Figure 1
Current and voltage in a linear load have both identical sinusoidal waveforms.

Some electronics, like motors, have a current waveform that lags behind the waveform for the voltage. The waveforms are still sinusoidal, but they are out of phase. This affects their power factor. These loads are called "inductive."

Figure 2
Current and voltage waveforms in inductive loads are sinusoidal, but out of phase.

Because a switch-mode power supply, like a computer's power supply, conducts current in small pulses, it has waveforms that are in phase, but the current waveform is not sinusoidal. So naturally, this isn't good for power factor either and this is why the load is called "non-linear."

Figure 3
A non-linear load has a non-sinusoidal current waveform.

Non-linear loads are probably the loads your utility company hates the most because a non-linear load draws harmonic currents from the AC mains. Luckily, the power factor correction used in power supplies reduces these harmonics. This reduction in harmonics can in turn clean up the power for everything else plugged into this circuit.

Active PFC uses a switching regulator with a good deal of complex circuitry to reduce harmonics. Components such as FET's, diodes and even IC's are used to improve power factor to 95% or better! Active PFC can also automatically adjust the input voltage of a power supply.

Passive PFC uses a passive element, such as a large ferrite core, to dampen the harmonics. Unfortunately, power factor using such means can only be corrected to about 80%.

Unfortunately, any means of power factor correction is going to use some of your effective power. So although your computer may pull less VA with power factor correction, it's going to pull more Watts. So I suppose in that sense your PFC power supply is actually less efficient than a non-PFC model, but this penalty tends to only be around 5%. And this use of power becomes less with the higher your input voltage is (because you draw less amperage at higher voltages and less amperage means less resistance.) This is why some power supplies in the EU, where 230V is used, they will sell 230V only APFC power supplies. To make a full range APFC power supply, considerations need to be taken to cool the PFC circuitry.
 
Originally posted by: jonnyGURU
Power Factor Correction/q]

Thank you! This is the first note that defined the acronym used - the cardinal rule of tech writing is do that immediately. Glad to know what PFC means today. 🙂
 
Originally posted by: jonnyGURU
Any additional circuitry has power loss. Active PFC in NEVER inherent to the design of a switchmode power supply. It is BECAUSE a power supply uses switchmode technology that it requires PFC for better power factor. Because PFC requires additional circuitry there is always loss, therefore lower efficiency.

I used to think like you, until I had a long discussion with a PSU designer about the role of active PFC.

It is certainly true that PFC consumes energy, and has its own intrinsic losses. That is not in doubt.

The issue is the design of the rest of the PSU.

A conventional line-powered SMPS has its design significantly compomised because it has to cope with variable line voltage and 100/120 Hz ripple. In order to ensure that regulation will be maintained at the ripple trough when the line voltage is at its lower limit of specification, then the main ferrite transformer must be significantly oversized. This oversizing results in increased core (and potentially copper) losses.

An active PFC circuit provides the rest of the PSU with a precisely regulated, low-ripple DC voltage. Redesigning the main PSU magnetics for this improved environment is a major reason for improved efficiency.


 
Originally posted by: Mark R
Originally posted by: jonnyGURU
Any additional circuitry has power loss. Active PFC in NEVER inherent to the design of a switchmode power supply. It is BECAUSE a power supply uses switchmode technology that it requires PFC for better power factor. Because PFC requires additional circuitry there is always loss, therefore lower efficiency.

I used to think like you, until I had a long discussion with a PSU designer about the role of active PFC.

It is certainly true that PFC consumes energy, and has its own intrinsic losses. That is not in doubt.

The issue is the design of the rest of the PSU.

A conventional line-powered SMPS has its design significantly compomised because it has to cope with variable line voltage and 100/120 Hz ripple. In order to ensure that regulation will be maintained at the ripple trough when the line voltage is at its lower limit of specification, then the main ferrite transformer must be significantly oversized. This oversizing results in increased core (and potentially copper) losses.

An active PFC circuit provides the rest of the PSU with a precisely regulated, low-ripple DC voltage. Redesigning the main PSU magnetics for this improved environment is a major reason for improved efficiency.


That might be true of the particular PSU manufacturer you spoke with, but typically varying line voltages and frequencies are managed by an IC. Yes, another component that uses the products efficiency, but most of the time a larger main transformer is not necessary anymore.

The lossage I'm talking about is from current SMPS design and is the lossage from the additional circuitry, not oversized transformers. The lossage is minimal, but it exists just the same.

I'm just trying to say that a power supply is NOT more efficient BECAUSE of APFC. That's a common misconception and it's simply not true. It IS less efficienct for the most basic, fundamental reasons.

 
While we're on this subject, does anyone know what the power factor of a monitor typically looks like? Does it make a difference whether it's a CRT or LCD?
 
That might be true of the particular PSU manufacturer you spoke with, but typically varying line voltages and frequencies are managed by an IC. Yes, another component that uses the products efficiency, but most of the time a larger main transformer is not necessary anymore.

My understanding is that this is standard PSU design practice - and isn't specific to one manufacturer.

Yes, the varying line voltages/load currents are managed by an IC, but the IC only controls the input to a transformer. The transformer, and the power electronics, has to be designed to cope with the extreme cases, which often means degraded performance under typical conditions.

In the case of active PFC PSU, the active PFC system provides the main PSU (IC, switches and transformer) with a tightly regulated environment. This can allow optimisation of the devices for 'typical' rather than 'extreme' scenarios.

There's another small bonus, because the active PFC system is a voltage booster (typically generating a regulated 400 V DC), the currents flowing in the switches, capacitors and transformer windings are lower than in an non-active PFC system, which may typically operate at about 300-310 V (with a subsequent reduction in I^2R losses of about 40-50%), and potentially as low as 285 V.

Yes, there are losses in an active PFC circuit - but they aren't necessarily that high. Active PFC systems have typical efficiencies of 94% or more.
 
It sounds more like the person that explained that to you were describing the benefits of passive PFC (which really is nothing more than adding a transformer to an existing SMPS unit) versus active PFC.

And it is actually typical for a power supply to run at 230V+, even when run on 115V. The 115/230V switch typically found on a non-PFC PSU really just bypasses a voltage doubler prior to the primary side of the PSU.

It is true that active PFC circuitry can clean up power so a power supply can, in theory, work more "efficiently," but it's still not "more efficient" since there are already, and always have been, a coil, varistors and a number of capacitors held to the task of filtering the AC input prior to the initial rectification of voltage, whether you have APFC or not.

In other words, no design that I am famiar with can compensate 100% for the efficiency loss of APFC because there is less transient filtering in place post power factor correction because it is not as necessary as a non-PFC unit.

I say "as I know of" because I'm aware anything is possible. But I do know that 90% of the brands out there do not implement such circuitry.
 
Back
Top