Is PFC worth it??

[aboubkr90]

@Andy CMI not power factor. In this post I'm trying to separate the power factor from other properties of PFC to see if the power factor specifically is worth 5-10% of the world energy.

[aboubkr90]

@PGPG What I'm proposing is that the car doesn't need to move strictly horizontally.

[Phil1977]

Your statement does not get true by repetition. From the power plant to the consumer the system with PFC is more efficient than without. Because transmission losses are indirect proportional to the power factor.

[Andy Chee]

see if the power factor specifically is worth 5-10% of the world energy.

for the energy generators, power stations, and power companies, it is definitely worth it.

for the individual home consumer, not worth it.

[Siwastaja]

Simply put, the power factor has to be corrected somewhere. This is the classic distributed vs. centralized discussion. Distributed PFC (every device having decent power factor) avoids creating a problem to begin with. Poor power factor at devices might have a tiny cost saving at the device but then the PFC has to be placed elsewhere. Classic "somebody else's problem", which is why legislation gets involved.

Yes, grid-level power factor correctors are a real thing. In the past with simple linear phase-shift errors, they were simply switched capacitor banks (because inductive motor and transformer loads dominated). Nowadays more advanced semiconductor-controlled stuff because today's poor PF is a completely different thing than phase shift between current and voltage. And using home PV inverters to participate in a distributed way is a recurring discussion which has some reality in it.

It is usually a good idea to avoid creating the problem in the first place. I could accept exceptions but they need to be rationalized with real calculations for total costs (including grid level power factor correction and higher transmission losses).

[ejeffrey]

Sorry for the late reply. There are a million ways to get rid of harmonics that doesn't cost 10% efficiency.

Of course.  That includes PFC corrected power supplies. The idea that PFC costs 10% or even 5% is just wrong.   For instance, the Meanwell UHP-200-24: https://www.meanwell.com/webapp/product/search.aspx?prod=UHP-200(R) is 94% efficient.  You would be hard pressed to find a non PFC power supply of similar power and voltage with better than say 96% efficiency.  In fact, if you look through meanwell's product line, their non-PFC power supplies are generally targeted at low cost.  They actually have worse efficiency than their PFC products (more like 86-90%).  I might as well claim that PFC improves efficiency by 5%!  That would of course be wrong, the difference is not intrinsic to PFC vs non-PFC, it's just how the device is designed.

Basically the whole premise of your claim is flawed, PFC isn't significantly less efficient even measured strictly at the point of use.  When considering the generation and distribution costs PFC improves overall efficiency for medium and high power loads at least.

[madires]

https://en.wikipedia.org/wiki/Power_factor#Importance_in_distribution_systems

[schmitt trigger]

Ok everyone!
Your are explaining why we PFC. I know why we PFC.
I'm saying there are other ways that doesn't have much impact on the efficiency of our devices. Like ferrite beads. The rest of the slow CMI spectrum, if ignored, will become white noise that will average to zero (not exactly, but close).
For critical devices, where a human could die if it goes wrong, there we can PFC. and it will regulate the stable voltage the device needs to operate safely.

Again; your arguments demonstrate that you haven’t the foggiest idea how a large power generation and distribution network operates. Why don’t you research World Power Quality Standards instead of writing platitudes?

A perfect example of the Dunning-Kruger effect.

EDIT: I see that you registered only 2 days ago. So essentially you registered with the sole purpose of starting a controversy.

[PGPG]

@PGPG What I'm proposing is that the car doesn't need to move strictly horizontally.

I have only give a suggestive image showing how proposed by you ferrite beads are related to what PFC does.
I understand that you didn't suggest that car in my example has to move strictly horizontally, specially that you suggested to use tools (car shock absorber in my example) that can't do that.

[aboubkr90]

@phill I'm not repeating to make it true I'm repeating so you would understand my point. 'cause nobody answered my question.
@ejeffrey And some go as low as 2%. If they follow normal distribution they would average to 5%. But I can assure you they don't. The majority will be around the 10% area.
@schmitt I didn't join to star a controversy, I joined to ask a question.

[aboubkr90]

Conclusion. After siwastaja's previous comment I imagined some local devices forming a star connection, like a brushless motor, and yeah that's insane ('cause some of them started oscillating since it is not a real star and there is no core).

[Andy Chee]

'cause nobody answered my question.

Your question was answered multiple times.

Is PFC worth it?  For the power station generator, yes, it is worth it.

[Phil1977]

@phill I'm not repeating to make it true I'm repeating so you would understand my point. 'cause nobody answered my question.

In the first place you repeatedly stated that PFC causes losses. That´s just not true in general. PFC creates some losses at your site, but prevents mostly larger losses somewhere else in the grid. Yes, PFC is a burden for the consumer because it makes devices a little more expensive and marginally increases the accounted power consumption. But in the end the power plant needs less fuel or can supply more customers.

[MrAl]

Hello again,

One of the things that has not been directly mentioned yet is what power factor correction actually does, what it affects.

First, we can view a PFC circuit as being between the power line and the product.  That means it acts as an intermediary between the line and the product such as a power supply.  If we look at what it does for the product, it does not do much.  That's because it's not really made for improving the product, it's made for improving something about the line itself.  If we look at how it affects the line, we see that it has an important function and it affects the line by a significant amount.

This is why one method of correcting the power factor is done by placing a capacitor across the input line.  If we think about this, the capacitor does nothing for the sine voltage, so the product does not see anything different.  The line however has to constantly supply current to that capacitor and so the line gets affected by the change in current demand.

In the simplest case, the capacitor just has to change the phase angle of the current from the line.  That means the impedance the line sees changes, but the product does not see any change.

[madires]

PFC also helps the device/load by optimizing the power consumption. Any reactive power created by loads has an impact on the mains installation of your building/home, i.e. you could run into wiring/breaker limits sooner than with an optimal power factor.

[Siwastaja]

I don't like the term PFC used as widely as it is. It refers to power factor correction as if poor power factor is a necessity and then correcting it is clearly a separate component. Sometimes this is the case, but I like the more generic concepts like "good power factor design". Requiring designs to not pull operating current in nonoptimal spikes is a sane requirement and it helps everyone.

As such, I don't know any legislation which mandates use of PFC. Never heard of such requirement. They do however mandate some certain power factor (among MANY other things). You are not allowed to manufacture and sell any random crap design.

[Phil1977]

@Siwastaja: Of course - I think the term "PFC" comes from AC-DC SMPS like computer power supplies where PFC generally required either a bulky and quite lossy LC-filter or an additional voltage conversion step of the kind of a power/voltage controlled step up converter before the buffer cap. These building blocks now are called "PFC".

Many other appliances like induction hobs, VFDs or probably even AC inverters (roughly just a VFD with a more complicated user interface) can have a more integrated design. They either need no buffering, like the induction hob that just delivers no power during the mains zero crossings, or are designed to work with a non completely stabilized buffer voltage.

Just out of curiosity: What was the power factor of an old style 3phase mercury vapour rectifier???

[Andy Chee]

I think the term "PFC" comes from AC-DC SMPS

I think the term "PFC" might even pre-date SMPS.  Old fashioned inductor ballast fluorescent light tubes often had power factor correction capacitors fitted to them, particularly in factories, multi-level office buildings and shopping malls, where hundreds of fluoro tubes were used for general lighting.

Multiply those installations by the hundreds in an average city and the power factor losses add up.  Of course, lighting technology has since evolved to more efficient lumens per electrical watt.

The type of loads imposed on power grids has changed markedly compared to, say, 1970.  Fewer inductive ballasts, more capacitive SMPS.

[coppice]

I think the term "PFC" comes from AC-DC SMPS

I think the term "PFC" might even pre-date SMPS.

The term PFC predates semiconductors, going back to the earliest days of AC power engineering.

[Phil1977]

Yes, thanks... In my perception PFC was always this smoothing of rectifier inputs, but of course there have been these capacitor or inductor banks for shifting phases. In german language these devices were often called "Phasenschieber <🥞> phase shifter, I first heard the term "PFC" together with SMPS.

[Siwastaja]

Yes, thanks... In my perception PFC was always this smoothing of rectifier inputs, but of course there have been these capacitor or inductor banks for shifting phases. In german language these devices were often called "Phasenschieber <🥞> phase shifter, I first heard the term "PFC" together with SMPS.

Remember that power factor itself was defined directly from phase shift, PF = cos(phase_shift). That definition of course had to change a long time ago* so that the concept of PF could still remain valid with many forms of modern loads which can draw current peaks at any arbitrary point in time.

But the concept is simple: if you try to pull a lot of current while voltage is at low point, you are transferring less power yet transferring all that current. Optimally the instantaneous current should be directly proportional to instantaneous voltage. This gives PF=1, and this gives same RMS current as if were a DC distribution system with the same RMS voltage. Running at less than PF=1 is then equivalent of running a DC distribution at lower voltage, meaning higher current and more resistive losses.

*) or maybe for old-timers, it feels like it was recently

[MathWizard]

I've made DC-DC buck and boost converters on a breadboard with all discrete parts, (except maybe a 555timer) . One of these days I want to use a small transformer like 24V, and try to make a PFC stage and forward converter type SMPS, just on a breadboard. IDK what the transistor count would be, so I might want to use some real op-amp's, and timer/counter's.

[ejeffrey]

@ejeffrey And some go as low as 2%. If they follow normal distribution they would average to 5%. But I can assure you they don't. The majority will be around the 10% area.

I challenge you to find one in in production, general purpose power supply where you can demonstrate the use of power factor correction decreases the point of use efficiency by 10%.  I don't believe they exist, and I think your are just making it up, or repeating something you heard from somebody else making it up.  As an average, that is completely absurd.

Also, it is 100% categorically wrong that power factor correction increases total energy use.  Overall it definitely decreases energy use.  We burn less fuel and install less copper and iron because large loads are required to be power factor corrected.

[T3sl4co1l]

I've made DC-DC buck and boost converters on a breadboard with all discrete parts, (except maybe a 555timer) . One of these days I want to use a small transformer like 24V, and try to make a PFC stage and forward converter type SMPS, just on a breadboard. IDK what the transistor count would be, so I might want to use some real op-amp's, and timer/counter's.

You may find this of interest,
https://www.eevblog.com/forum/projects/power-factor-correction-discrete-control-circuit-(for-bcm)/msg4405327/#msg4405327

Tim