Misplaced Pages

Revision as of 13:44, 17 July 2010

Energy Unassessed

This article is within the scope of WikiProject Energy, a collaborative effort to improve the coverage of Energy on Misplaced Pages. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.EnergyWikipedia:WikiProject EnergyTemplate:WikiProject Energyenergy ??? This article has not yet received a rating on Misplaced Pages's content assessment scale. ??? This article has not yet received a rating on the project's importance scale.

bill for kVAh

The article states "The significance of power factor lies in the fact that utility companies supply customers with volt-amperes, but bill them for watts." At least in Australia, this is not correct. The electricity meters measure volt-ampere hours. The bill incorrectly says kWh, but really you are charged for kVAh. This means that it is in the customer's interest to have unity power factor, since this reduces the amount they pay for electricity.

Confusing

The article is confusing. What do the various values of the power factor mean? Are they the same (just magnitude) for both sources and sinks? A diagram with at least -1, 0, and +1 would be helpful. How does leading and trailing play into this? Do only sinks have leading or trailing characteristics or do sources too? Does the utilization depend on matching them, or is +1 always desirable for sinks no matter what the value of the source?

If this isn't the right place for comments about the entry I apologise in advance. Please direct me to the proper place.

Power factor is confusing and is difficult to explain in simple terms. I don't fully understand it myself so the following should be treated with caution. When the PF is 1 it does not have a sign. If the PF is (say) 0.9, it can be "0.9 trailing" (-0.9) for a reactive load, or "0.9 leading" (+0.9) for a capacitative load. The question about sources is a good one but I don't know the answer. Biscuittin (talk) 09:35, 2 November 2009 (UTC)

Don't think so. Power factor meters in my experience say "lead" or "lag" and never "+" or "-" - this has been discussed quite a lot in connection with this page (see below under "Sign"). Where does the article get confusing to follow? --Wtshymanski (talk) 19:32, 2 November 2009 (UTC)

My mistake. I saw the + and - signs elsewhere on this page and assumed they referred to leading or trailing power factor. Biscuittin (talk) 19:55, 2 November 2009 (UTC)

Illustration

I'm a rather bored power engineer, so i had a hack about, hopefully it makes a bit more sense now. let me know if it needs clearer explanation

Much better. It might be nice to illustrate the power triangle with something like the graphic at http://www.ibiblio.org/obp/electricCircuits/AC/AC_11.html.

Distortion power factor

This article only talks about "displacement" power factor. What about the effects of non-sinusoidal currents? --Wtshymanski 22:35, 16 Dec 2004 (UTC)

Non-sinusoidal (non linear)analysis is very difficult, and a bit too tricky to try and explain here. I have included some details about harmonics. --Euripides 15:53, 3rd Jan 2005

energy returns to the source

I think that this sentence could do with a bit more explanation: "Since this stored energy returns to the source and is not available to do work at the load" - why is this exactly? --TimSmall 13:16, 1 September 2006 (UTC)

The energy does not go into the load; it is reflected back (down the power lines) to the source, and so is wasted. — Omegatron 13:53, 1 September 2006 (UTC)

Current that is not in step with the voltage does not transfer energy from the source to the load but continually circulates energy back and forth between the source and the load. This energy circulation is not 100% efficient. During each "trip" from source to load or load to source, some energy is lost as heat in the wires and other parts of the power generation, transmission and distribution equipment. Does that help? --C J Cowie 14:24, 1 September 2006 (UTC)

The energy doesn't necessarily get bounced back and forth, either. It can disappear into the source. — Omegatron 14:42, 1 September 2006 (UTC)

Most of the reactive power flows back and forth between the source and load such that "On one half-cycle, the source supplies energy to the energy-storage element, and on the next half-cycle the energy-storage element returns energy to the source....currents required to supply the stored energy produce losses in the generating and transmission system..." Scott, Ronald E. (1960). Linear Circuits. Reading, MA: Addison-Wesley. {{cite book}}: Cite has empty unknown parameter: |coauthors= (help) “Low power factor means more current and greater ${\displaystyle I^{2}R}$ losses in the generating and transmitting equipment.” Fitzgerald, A. E. (1983). Electric Machinery (4th ed. ed.). Mc-Graw-Hill, Inc. ISBN 0-07-021145-0. {{cite book}}: |edition= has extra text (help); Unknown parameter |coauthors= ignored (|author= suggested) (help)

Explanation of Apparent and Reactive power

There does not appear to be any proper explanation of what apparent and reactive power are, or how to calculate them.

Mrwooster 02:42, 25 February 2007 (UTC)

Form Factor

In the case of non-sinusoidal currents and voltages, the term "form factor" is sometimes used to refer to the ratio between the mean value and the RMS (root mean square) value. The form factor can become large in the case of pulsed waveforms with a high mark to space ratio.

True, but I didn't think it clarified power factor at all. --Wtshymanski 03:04, 16 November 2005 (UTC)

Question re load sharing generators

dear sirs,

i've just ran into your discussion about <power factor>. would you participate in following:

two equal generators running in parallel load sharing mode, - both of them loaded equal due (as sample 2 x 150kW) but one has too less current than another one (difference is about 150 Amps).

if to rely on the fact that P = square root 3 x U x I x Cos Phi we face reducing power factor for one of generators ... it is clear.

while peaks of load, as a result, one with higher current is tripping due to overcurrent protection.

the practical question is where the the influence for <power factor> appeared from (?) if measurements taken around both generators' excitation systems shows no difference.

with other words: could it be that consumers' =low state of insulation= causes such a behaviour ?

thanks for yr participation brgds Alex

This looks like a question for http://eng-tips.com electric power engineering section --C J Cowie 15:24, 1 March 2006 (UTC)

Advertising links

The last four of the five external links seem to various extents to be advertising links. However each one seems to be a direct link to useful information related to the subject of this article. Should they all be removed? --C J Cowie 01:30, 23 August 2006 (UTC)

I was wrong about that. Learn about Power Factor provides forms to be filled out to request information by email. Very little information is directly available from the linked page. I am removing that one now. --C J Cowie 17:43, 23 August 2006 (UTC)

User:Wtshymanski acted while was still talking. --C J Cowie 17:46, 23 August 2006 (UTC)

Physical Explanation of Leading Power Factor

I was just wondering if anyone can explain how a leading power factor makes sense physically. Current is a result of voltage, isn't it? So how can an effect possibly precede its cause? Unless it is really lagging by >=3/4 of a period...? Anyone? GBMorris 22:00, 9 December 2006 (UTC)

I agree that the current through the power lines is caused by the voltage (in combination with other factors). The voltage is not "caused by" the current or any other factor in my house, since the voltage across the power lines to my house remains practically the same whether I turn everything on (lots of current) or turn everything off (zero current).

So it seems like a leading power factor would be impossible, since the effect (appears to) precede the cause.

Let me try to draw an analogy:

A mountain climber climbs a snow-covered slope for days, finally reaching the top at high noon on Wednesday. Journalists in the peaceful village nearby use binoculars to watch him plant a flag at the peak, rushing to print a story it ("Mountain Climber finally climbs Mount on Wednesday"). Alas, he dislodged a chunk of snow that triggers an avalanche on Tuesday. That avalanche roared for what seemed like hours, with the greatest volume around 11 am. How can the peak of an effect (the avalanche on Tuesday, peaking at 11 am) possibly precede the peak of the cause ("Mountain Climber finally climbs Mount on Wednesday" with the peak reached at noon)?

When we place a capacitor across the AC power lines, many things happen. I suspect the capacitor article has a better explanation. For simplicity, let's assume that at time T=0, we place a discharged ( V=0 ) capacitor across the power lines at the exact instant that the voltage across the power line is also zero (the "zero crossing"). Also, we watch "the voltage" v(t) (the voltage across the capacitor) and "the current" i(t). For the next few milliseconds, the voltage increases approximately linearly with a slope m = dv(t)/dt. However, during those milliseconds, the current stays approximately constant at i = C*m.

The *rising* voltage causes electrons to flow in one lead of the capacitor, and out the other lead.

Later, around the time the voltage reaches its peak, the voltage stays approximately constant for a few milliseconds.

The approximately *constant* voltage causes electrons in the capacitor to stay more-or-less in the same place. So during those few milliseconds, the current of electrons is approximately zero.

We can graph this voltage v(t) vs. current i(t) over several cycles.

On that graph, we see the power line voltage going up and down as always, completely oblivious to the capacitor. The current reaches its peak *before* the the power line voltage reaches its peak, so we say the current is leading the voltage. --68.0.124.33 (talk) 16:27, 15 August 2008 (UTC)

Amateur explanation of power lead/lag

Having never studied this stuff to Uni level but having been interested since childhood, here is my take on the logic of current lead/lag:

A capacitor is like a short-term rechargeable battery, only a rechargeable needs to be charged carefully. A capacitor will act effectively like a short circuit until it is storing the same voltage as is being applied. At the other end of the scale, a capacitor that is sitting across a 12v DC line and is charged to 12v presents an almost infinite (minus leakage) resistance across the 12v line. Should that 12v line waiver, the capacitor's charge will try to either boost the deficiency or grab the increase.

Now put this across rectified AC, empty capacitor, 0v line. Line voltage increases, capacitor absorbs. Load also starts to absorb. Although the load will not reach full current until the voltage peaks, the capacitor will grab every electron it can get its layers on and therefore the current through the capacitor will be hugely higher than that through the load. Therefore, the current the line supplies will be far more to do with what the capacitor wants than what the real load wants. By the time the AC peaks, the capacitor is fully charged and presents no load so the real load gets everything.

Now, the AC wave starts to diminish. The line starts to become less charged than the capacitor, so the capacitor starts to discharge into the line. The load is supplied by both the line and the capacitor. The current drawn by the line is therefore much less than would be expected, due to the capacitor's help.

This is why the current curve precedes the voltage curve by 0.25 of a full sinusoidal cycle with a capacitor across the line.

Coils are the opposite of capacitors in that they fight, or kick against, any change in voltage. Put a coil across the same rectified AC line and it will appear as a near infinite resistance whilst the voltage is rising. As soon as the voltage has peaked and started to decrease, however, the coil then becomes less and less resistant. The nearer the voltage gets to zero, the less resistant the coil is. Near zero it is almost a short circuit and so the line current is at its highest. Having said that, the strange part is that the coil is now fully charged. Once the voltage starts to increase again, the coil then starts to discharge. So although the load drawn by the load is xAmps, the large part of this will be supplied by the coil and only a small amount by the line. This is why the current curve lags the voltage curve.

Read up on loudspeaker crossovers and passive graphic equalizer units. Also look at power supplies which used to have a diode valve (American:Tube) followed by an inline choke (coil), which then became bridge rectifier diodes followed by capacitor across the line.

Then look at rechargeable batteries: put 4 in parallel and charge to 1.2v. Once charged, switch them to series and they will return 1.2v x 4 = 4.8v. Now apply this to capacitors and you have a switched mode power supply. —The preceding unsigned comment was added by APNorth (talk • contribs) 23:41, 21 December 2006 (UTC).

ELI the CIVIL ICE man

Having gone straight from ham radio in high school to physics in college I missed out on all the nice EE mnemonics like "Little star up in the sky, power equals R squared I" or whatever it is, so ELI the CIVIL ICE man is news to me. Mulling it over, I concluded there must be four types of EE.

Type 1. Depends on mnemonics to remember whether current leads or lags voltage in capacitors and inductors.

Type 2. Intuits that current lags voltage in an inductor by thinking of inductors as flywheels that respond sluggishly to an applied force. Doesn't intuit current lead for capacitors (how could a mere capacitor predict the future?) but does remember that capacitors are the opposite of inductors.

Type 3. Is equally at home with voltage causing current and current causing voltage. An AC current applied to a capacitor causes the voltage to rise from zero when the current is at its peak and causes the voltage to reach its peak just when the current has dropped to zero and is starting to go negative. Hence in a capacitor voltage lags current by 90 degrees.

Type 4. Understands flywheels by analogy with inductors. --Vaughan Pratt (talk) 08:21, 26 January 2008 (UTC)

I don't know if there are four types of EE; after a few years in the business you don't think about this any more than you think about how to tie a tie. The mnemonic is just there to backstop you on an exam. Analogies are tricky - best to learn the physics ( then you don't have to remember which property is which - is speed more like current or voltage? )--Wtshymanski (talk) 04:30, 27 January 2008 (UTC)

If you know that a capacitor tries to resist a change in charge (voltage), because it must receive or send electrons (current) in order to change it's internal voltage. Voltage (charge) is proportional to the amount of electrons stored within the capacitor. So, one can deduce that the current flow will be greatest when the voltage rate of change is greatest (dv/dt). This rate peaks out at the voltage zero crossing point in the AC sine wave, which is exactly 90° ahead of the the voltage peak (slowest rate of change).

The opposite is true for inductors. An inductor's current flow is lowest when the dv/dt is highest, because the expansion of it's magnetic field requires little current because the magnetic lines crossing the inductor during storage creates a back EMF that makes the inductor look high impedance. When dv/dt is at minimum at the top of the voltage wave, the magnetic field begins to collapse which induces current into the circuit that opposes the still positive voltage. Also, when there is little or no dv/dt the inductor looks like a piece of long wire across the voltage source (low impedance or short circuit) and thus highest current. —Preceding unsigned comment added by Moesdaddy (talk • contribs) 20:16, 24 December 2008 (UTC)

Distribution vs. Generation Losses

If a poor power factor is balanced by compensating complementary elements, the distribution losses only occur within that sub-system. Low power factors do not lead to generation losses unless they are not compensated somewhere, within the system as a whole. So, in general, small users with some equipment with poor power factor need not feel guilty. But if poorly designed computer power supplies are commonly used, then there could be significant system-wide issues.-69.87.193.242 12:49, 4 April 2007 (UTC)

Yes, I think this also applies to compact fluorescent lamps, which usually have a much lower power factor than incandescent lamps. Biscuittin (talk) 09:43, 2 November 2009 (UTC)

Power factor for AC

I know to calculate power in an AC circuit you have to multiply the current times the voltage times the power factor, but what if you are actually measuring current with a meter? the initial equation for power factor is based on ohm's law. If I stick a current meter in a hot AC circuit does it measure a current that can be multiplied by the actual voltage to get power? using cosine theta would seem to me to throw this off. Then again I'm not an expert by any means. Expert advise please, and also perhaps if I'm right and the power factor only accounts for inductance as a means of correcting ohm's law for an inductive load that it should stipulate that you can use ohm's law directly with measured variables on a physical circuit.

Can't give a definite answer, but I think it depends what sort of meter is used. A moving-iron meter would give a different answer from a hot-wire meter. As for electronic meters, I have no idea. Biscuittin (talk) 09:50, 2 November 2009 (UTC)

Not really, no. Average-responding meters are calibrated to display the RMS values of a sine wave (or close to it). The only time you'd see a difference in current measurement is if the waveform has a different form factor than assumed in the meter calibration. This has nothing to do with power factor - to measure power factor you must measure real and reactive power, or current, voltage and phase angle. --Wtshymanski (talk) 19:32, 2 November 2009 (UTC)

Graphs

The graphs are pretty nice, but I think they should show RMS (root mean square) power rather than average. RMS voltage, power and current is the quantity typically discussed with AC signals.

Showing that the average AC power is zero in the power factor 0 graph is true but misleading. The load does not see 0 watts, and this is what a casual reader may conclude. —Preceding unsigned comment added by 24.68.59.219 (talk) 21:22, 10 April 2008 (UTC)

I believe that average power is what the original illustrator intended to demonstrate; that is what I did when I reimplemented the graphs in Gnuplot/SVG. There is some justification in doing so; an ordinary integrating electric meter will also show zero energy used when presented with a zero-PF load, which is why industrial consumers are nearly always demand-metered. 121a0012 (talk) 06:24, 17 April 2008 (UTC)

merge

I suggest merging power-factor correction and power factor correction unit into power factor. That way we don't have to re-explain "power factor", its implications, and ways to handle it, in all 3 articles. --68.0.124.33 (talk) 17:53, 15 August 2008 (UTC)

Yep. Better one long (but well-structured!) article than a lot of repetition. --Wtshymanski (talk) 18:00, 15 August 2008 (UTC)

I see Wtshymanski has merged these 3 pages. Well done! --68.0.124.33 (talk) 16:53, 26 September 2008 (UTC)

in the news

Have you seen what the people at Tom's hardware guide say about this "power factor" article: ? --68.0.124.33 (talk) 16:03, 4 November 2008 (UTC)

Hmm. What gets regulated. Powerfactor, of course. What's a "mobo" ? Is there a crazy gadget - No, of course not. What are the facts? See the article. Does it shut things off in low use? Does what shut what off?

Sure, the article needs work, but "Idununnastanit" is not a useful guide to the editor. --Wtshymanski (talk) 16:27, 5 November 2008 (UTC)

Active PFC

Active PFC is becoming a more important technical characteristic when purchasing power supplies for personal home computers. Defining this term as it applies to PC's would be very beneficial to many people.

68.241.177.152 (talk) 05:43, 10 January 2009 (UTC)jonny

Umm...it's in the article, already - did you not get that far? --Wtshymanski (talk) 16:01, 10 January 2009 (UTC)

power factor correction for residential

some say it is a scam: http://nlcpr.com/Deceptions1.php

the places that sell these capacitors say they work for residential homes. Can someone help me out here? Thanks --OxAO (talk) 04:42, 11 July 2009 (UTC)

The world is a very big place, but in the few places I've seen residential electricity bills, the utility doesn't charge individual residences for reactive power. Perhaps an apartment complex with a large air conditioning plant would be charged for its reactive power, but otherwise if you're not paying for power factor, there's no benefit to wiring capacitors into your home. --Wtshymanski (talk) 14:26, 15 September 2009 (UTC)

At least in the U.S., it would be determined by tariff and state regulations. For example, my power company's general tariff (applicable to all customers) states:

Except as may otherwise be provided in a specific rate, a Customer taking service is expected to maintain a power factor of not less than 80% percent . The Company may require any Customer not satisfying this power factor requirement to furnish, install, and maintain, at no cost to the Company, such corrective equipment as the Company may deem necessary under the circumstances. Alternatively, the Company may elect to install such corrective equipment at the Customer's expense.

Boston Edison Company d/b/a NStar Electric (2006-01-20). "Terms and Conditions — Distribution Services (M.D.T.E. No. 100A)" (PDF). Retrieved 2009-09-15.

It would be fairly unusual for a residential customer to have a power factor below 0.8 given that most residential loads have historically been resistive.

The individual service tariffs also allow the company to compute demand (for demand-metered, commercial customers) on the basis of either 90% of apparent power or 100% of actual power, at the company's option. Presumably they choose the higher-revenue option. 121a0012 (talk) 02:36, 16 September 2009 (UTC)

Right, the utility may require customers to stay above a certain limit, but residential meters only meter kwh, not kvah - so it would be difficult for the utility to detect an individual residential customer with low PF. I haven't seen anything on it but it would be interesting to find a paper describing what a modern home's power factor really is...with all the fluorescent lighting, computers, microwave ovens, etc., the load isn't purely linear resistive incandescent lighting any more. Anyway, unless your utility bill shows both wh and kvah or equivalent, no residential power factor improvement is going to help reduce your bill. --Wtshymanski (talk) 16:31, 16 September 2009 (UTC)

Metering

I wouldn't know where to put this in the article, so I'll say it here.

A Thompson watt-hour meter, still in widespread use, only measures the energy that is delivered to the load. Even though a purely reactive load does not use any energy and does not cause the meter to turn, it still is a drain on the utility because the current flowing in the distribution system causes resistive heating of the distribution wires. That's real power that's being lost outside the customer premesis, but it does not show up on the meter.

Newer electronic meters are able to directly measure the voltage and current waveforms, and the utility potentially can charge extra for a low power-factor load---if their billing system is set up that way. —Preceding unsigned comment added by 71.199.121.113 (talk) 14:59, 10 August 2009 (UTC)

why do active PFC SMPS need two stages?

Given that a SMPS already regulates variable input and output voltages and currents, why is it not feasible to add the PFC stage to the existing feedback circuit? —Preceding unsigned comment added by 69.236.89.96 (talk) 02:04, 6 September 2009 (UTC)

Yes, a single SMPS stage can be designed to regulate either input current, or input power, or output current, or output power, or output voltage. Unfortunately, a single SMPS stage cannot regulate all these things simultaneously -- in fact, it cannot even regulate 2 of them simultaneously.

For example, say we have a typical active PFC SMPS converting (more or less) 230 VAC to (regulated) 12 VDC. Also, say we connect it to a typical load that pulls roughly constant power from the SMPS.

The feedback circuit of the final stage chooses whatever duty cycle is needed to force the output voltage towards exactly 12 VDC -- a duty cycle that varies to pull more current when the input voltage is low, and less current when the input voltage is high, in order to transfer a constant output power.

If the feedback circuit chose any other duty cycle, then the output voltage would no longer be 12 VDC.

Unfortunately, most homes do not have 3 phase AC which can deliver constant power at good power factor. Instead, most homes have single-phase AC. To get a good power factor from single-phase AC requires pulling lots of power at the peaks and pulling zero power at the zero crossings.

The feedback circuit of the initial stage of a PFC SMPS chooses whatever duty cycle is needed to force the input current to be proportional to the input voltage -- a duty cycle that varies to pull zero current at the AC voltage zero crossings and lots of current at the AC voltage peaks.

It is impossible to do both in a single stage, because that one stage would need a duty cycle to be *both* "a duty cycle that varies to pull more current when the input voltage is low, and less current when the input voltage is high" *and* "a duty cycle that varies to pull zero current at the AC voltage zero crossings and lots of current at the AC voltage peaks."

We need (at least) 2 stages in order to get 2 independently-controllable duty cycles in order to independently control 2 things: output voltage and input current.

Is there some way we can clarify this in the article?

--68.0.124.33 (talk) 05:13, 9 September 2009 (UTC)

Article was/is a conflation of DC and AC theory

The idea of stored energy is appropriate for DC circuits but isn't useful for explaining reactance. I replaced the explanation that referred to the latter with traditional AC theory because reactance, leading/lagging current and a phase relationship are applicable to AC. A step function, as implied by use of the term "step" where "phase" was meant, is was used incorrectly. So-called stepper motors are driven by a square wave, which appears as the cross-section of a step. Theory about stored energy explains how a DC voltage doubler works but instantaneous V and I in AC explain reactance accurately. Kernel.package (talk) 12:00, 9 December 2009 (UTC)

(months later) Stored energy is precisely useful to explain reactance - that's why there is reactance. Fundamentally there is no "ac" or "dc" theory, it's all Maxwell....--Wtshymanski (talk) 14:48, 15 March 2010 (UTC)

Merge from Distortion power factor

Distortion power factor should be merged here, it's small and it needs to be given a contenxt which this article already supplies. --Wtshymanski (talk) 15:45, 16 March 2010 (UTC)

Agree --Chetvorno 22:20, 17 March 2010 (UTC)

Negative Power Factor?

I'm having great difficulty with the following wording: The power factor of an AC electric power system is defined as the ratio of the real power to the apparent power, and is a number between -1 and 1 (frequently expressed as a percentage, e.g. 0.5 pf = 50% pf). Real power is the capacity of the circuit for performing work in a particular time; it can be either positive or negative, depending on whether the power is flowing from the nominal source to the nominal load, or vice versa. Apparent power is the product of the RMS current and RMS voltage of the circuit, which, by definition, is always positive. Which power flow then determines the sign of the power factor? Real or reactive? Apparent power is always positive? Why? Supposing I'm watching the Manitoba to Minnesota tie line and suddenly one of the Nelson River Bipole lines trips...suddenly instead of exporting mumblety-mumble megavoltamperes, the utility is importing mumble megavoltamperes. Would be very tempting to say the sign of the apparent power flow has reversed. Maybe I've been talking to t0o many transmission people, they seem to be more worried about vars than watts. "Lead" and "Lag" I can sort-of understand, I don't know what a negative power factor means in the case of distortion. --Wtshymanski (talk)

If my learned co-editor is in fact who he says he is, I am presuming greatly - but at least I have some references, whereas so far I've seen no references defining 'negative' power factor as 'power factor with watts flowing back into the source'. --Wtshymanski (talk) 04:22, 16 December 2008 (UTC)

I've never seen power factor expressed as a percentage. Could this be a UK/US difference? Biscuittin (talk) 19:59, 2 November 2009 (UTC)

From the Fluke 434 Power Quality Analyzer manual (which agrees with my experience as a working Electrical Engineer):

"Interpretation of Power Factor when measured at a device:"

"PF = 1: all supplied power is consumed by the device. Voltage and current are in phase."

"PF = 0 to 1: not all supplied power is consumed, a certain amount of reactive power is present. Current leads (capacitive load) or lags (inductive load)."

"PF = -1 to 0: device is generating power. Current leads or lags."

"PF = -1: device generates power. Current and voltage are in phase."

In most areas of electrical engineering a negative power factor is rare - so rare that some EEs have never heard of it. There are two areas where negative power factors come into play a lot; electric motors that are braking a massive load (this includes regenerative braking on electric vehicles) and devices that source or sink power so as to correct for overvoltages/undervoltages in a poorly regulated line.

BTW, this has nothing to do with the practice of displaying leading/lagging of a positive power factor with a minus or plus sign. I have seen meters that do this (there is no consistency as to whether leading or lagging gets the plus sign) and it can confuse a technician who also has access to a Fluke 434 that uses a plus sign for both. Guy Macon 17:22, 15 July 2010 (UTC)

We don't usually alter previous editor's headings on talk pages. I would appreciate an authoritative reference for "negative" power factor defined as your meter manual suggests. Those two areas you talk about are pretty darn common in electrical engineering and if there was a wide-spread convention that "negative" power factor means "power flowing back to the source", surely it would be well documented? "Negative" power factor is not described at all in IEE Std. 100, which is a pretty good overview of IEEE standards practice. And there is IEEE Std. 1459, which says (Note 1, section 3.1.1.1) real power only flows to the load and can never be negative. Leading and lagging PF I've seen marked on scales, but "negative" PF is not something I've seen displayed on a power factor meter. There are 4-quadrant electrodymometer type meters, but they don't seem to mark the scaled as "negative". --Wtshymanski (talk) 18:11, 15 July 2010 (UTC)

Re: IEEE Std. 1459, I don't have it in front of me (I am at home) but I suspect that it defines "load" in a way that precludes generators of electrical power.

Re: "Those two areas you talk about are pretty darn common in electrical engineering and if there was a wide-spread convention that 'negative' power factor means 'power flowing back to the source', surely it would be well documented?", you appear to have missed what I wrote: "In most areas of electrical engineering a negative power factor is rare - so rare that some EEs have never heard of it." Again, AC loads that are energy sources are rare, not common.

Re: "but 'negative' PF is not something I've seen displayed on a power factor meter", the Fluke 434 Power Quality Analyzer does exist. I assume that you have only used the usual low-cost PF meters and have never needed something more sophisticated. IIRC, Hotektech, Dranetz and Extech power analyzers also measure negative power factor.

Re: "I would appreciate an authoritative reference for 'negative' power factor defined as your meter manual suggests", here are three, two from the dawn of electrical engineering and one modern. Plus, of course the Fluke manual, which I consider to be authoritative.

From The Electric Journal, Volume 5 (1908):

"Negative Power-factor: When a generator is connected to a circuit having in series only resistance, inductance and capacity, the current cannot be more than 90 degrees out of phase with the e.m.f., but if a synchronous motor (or a rotary converter or another generator) is in the circuit, the difference of the e.m.f.'s of the two machines sends current from the generator through the other machine. the generator is not delivering positive power to the other machine which, therefore, cannot run continuously without changing phase relation, or receiving power from some other source."

http://books.google.com/books?pg=PA480&dq=%22negative%20power%20factor%22&ei=imQ_TLywFpK4sQOZ-cH2CA&ct=result&id=lrESAAAAYAAJ&output=text

From the Philosophical Transactions of the Royal Society of London, Volume 203 (1904):

"The fact that the solid arc has a negative power-factor at frequencies below the critical frequency of 1950 indicates that the arc is under these conditions supplying power to the alternating current circuit, and that this is the fact can easily be shown experimentally by connecting a wattmeter so as to measure the power supplied to the solid arc by the alternating current, when it will be found that at low frequencies the solid arc is actually supplying power to the alternate-current circuit, while at frequencies above the critical value the alternate-current circuit supplies power to the arc. This observation is of course not in any way at variance with the principle of conservation of energy, since the alternating energy given out by the arc is derived from the direct-current energy supplied to it, the arc acting as a converter."

http://books.google.com/books?dq=%22negative%20power%20factor%22&pg=PA322&id=eG0OAAAAIAAJ&output=text

From "Analysis and Performance of 3-Phse Grid-Connected Induction Generator via Transistorized Ac Voltage Controller" EE Dept.- College of Engineering, Cairo University (2008):

"...Accordingly, the generator will absorb active power from the grid, which leads to negative power factor."

http://faculty.ksu.edu.sa/Alolah/Documents/Files%20of%20papers/C033.pdf

From the Fluke 434 Power Quality Analyzer manual:

http://assets.fluke.com/manuals/434_435_umeng0300.pdf

Search for "interpretation of power factor"

Also see:

http://us.fluke.com/fluke/usen/Power-Quality-Tools/Three-Phase/Fluke-430-Series.htm?PID=56077

I would also ask what you would say the power factor is for a line that has the current 180 degrees out of phase from the voltage. That's a test that I have run many times on AC power supplies using another, larger AC power supply as the "load." Some AC power supplies cannot handle such a load. Guy Macon 21:17, 15 July 2010 (UTC)

Not comforting... a student paper that misspells "phase" in the title does not fill me with confidence. A 1908 paper? Physicists? And overhauling loads are described in every undergrad machines lab. --Wtshymanski (talk) 02:39, 16 July 2010 (UTC)

Again I ask; what do you say the power factor is for a line that has the current 180 degrees out of phase from the voltage? Guy Macon 09:45, 16 July 2010 (UTC)

There's no power being transferred to the "load" if the current is going the other way, so it can't be defined. Or, we've got the source and load interchanged, so the PF is 1. Which one do the authorities prefer? This situation arises many times in transmission...some times the electricity runs from Winnipeg to Minneapolis, but some times the electricity runs from Minneapolis to Winnipeg, too. --Wtshymanski (talk) 13:07, 16 July 2010 (UTC)

There's no power being transferred to the load if the load is a pure capacitance, yet we have no trouble determining the PF in that case. You can't just say "it can't be defined." Such circuits exist and they have measurable voltage, current, and phase. If your method of calculating PF cannot handle a particular real-world combination of voltage, current, and phase, yet standard test equipment can, then something is wrong with your method of calculatiing PF. The PF still exists in the real world.

The concept of negative power factor has been in use for over a hundred years. I grant that it is rarely used and many engineers have never heard of it, but the concept of a using a flickermeter to obtain a numeric value for Pst is also something that many engineers have never heard of, yet Pst and Negative PF are measurable by most or all power quality analyzers. (See IEC 61000-4-15 and IEC 61000-3-3).

Invoking IEEE Std. 1459 proves nothing. It specifically defines real power as being only that which flows to the load, and thus by definition cannot be used to describe a circuit where real power flows from the load. Guy Macon 10:23, 17 July 2010 (UTC)

Well, yeah, that's kind of the point; "reliable references" are what we use here on Jimbo's dream. IEEE 1459 is a set of *definitions* of terms used in describing AC power and defines (for those who chose to use it) what "power factor" means. If the committee that wrote up 1459 thought that, for defining power factor, power only flows from a source to a load, that's good enough for me. Are you saying the Fluke company's designer who labelled something 'minus' instead of 'reverse power' is a higher authority than the IEEE? As you point out, power often flows either way down a wire so this is a common situation, and yet i don't ever see anyone calculating a negative power factor in my meager collection of texts. Could you find me a worked example somewhere where a prof is telling his students " -1 MW over +2 MVA means the power factor in this circuit is -0.5, lagging" or something to that effect; preferably an authority who can spell "phase" correctly in the title of a document.

I had a similar discussion with another editor some years ago and asked for some documentation, and he hasn't got back to me yet. If it's defined, why is it so hard to find anyone talking about it? Negative voltage, all the time. Negative resistance, sure. negative power, in some contexts, sure. Negative power factor - only in papers from 1908? Sounds fishy to me. (You can proably find more places talking about power factor greater than 1 than power factors less than 0.) ( I worked in an arc furnace shop so had to get familiar with flicker and flicker meters and the difference between 120 v bulbs and 240 v bulbs...) --Wtshymanski (talk) 13:44, 17 July 2010 (UTC)

• Unassessed energy articles
• Unknown-importance energy articles