Talk:Power factor/Archive 3

Negative PF issue
The material in the previous section is off topic for the section heading and is troubling in other respects.

PF is a defined term, and its value depends on that definition. In one view, the actual source matters and PF is never negative; in the other view, there's a nominal source and the PF may be negative. Either definition is possible/plausible. Viewing PF as an efficiency (where direction of flow irrelevant) is not absurd on its face.

I think most of the editors here would follow the nominal source/negative power factor view. Personally, I prefer that view, too, but I'm not willing to discard the other viewpoint as wrong (definitions cannot be wrong). I'm also not willing to concede that an unambiguous IEEE definition should be the last word on any issue. If the IEEE definition were PF is never negative, then I'd view a Fluke meter's PF < 0 reading as a different definition from a significant reliable source.

McEachern's paper is in some sense self-defeating on the definition because it acknowledges that at least some engineers believe that PF is never negative. Furthermore, the paper points out places where the IEEE spec is confusing or ambiguous on the issue. If an insignificant minority held the never negative view, then there would be little need to write the paper. To me, there is a minority viewpoint for PF is never negative.

My take is the WP article should adopt the nominal source/negative PF definition, but it should briefly acknowledge the never negative definition. I have no trouble with using McEachern's draft paper as the source for both positions. Even if the paper is currently unpublished, I can accept McEachern's membership on the 2000 and 2010 committees as bestowing prominence. WP:DUE. However, neither McEachern nor the paper speak for the IEEE. The paper describes a position that McEachern advocates, but the IEEE is currently silent on the ambiguity of its definition. Glrx (talk) 20:09, 10 June 2013 (UTC)


 * The minority viewpoint for PF is never negative is clearly a WP:FRINGE view, and the "some readers" the paper refers to appear to consist of Wtshymanski and a few students who got the wrong information from from Wikipedia before we corrected it. No source anywhere has been found that repeats the error in IEEE 1459 -- probably because the rest of that document -- especially figure 1 -- makes no sense if you believe that PF is never negative. No textbook teaches that there is no such thing as a negative power factor. No scientific or engineering paper makes the claim. Holding that theory makes it impossible to do a power factor calculation when the current switches from being in phase to 180 degrees out of phase with the voltage.


 * I say we treat it like any other fringe view, and give it the WP:WEIGHT we would give any other fringe theory where we have zero evidence that it is held by anyone but one Wikipedia editor and some students who got the information from Wikipedia.


 * The fact that at one time there were instruments that used + and - for lead and lag, on the other hand, is worth adding if we can find sources, which should not be difficult. --Guy Macon (talk) 02:00, 11 June 2013 (UTC)


 * I also agree that that negative power factor does not exist. Negative power factor is a reading on an instrument that we all understand the meaning of but in reality is impossible. When load power factor is measured and it becomes negative it means that the instrument manufacturer was tired of the "---" reading complaints and allowed it for the convenience of the user. A load never feeds back negative power into it's source. This is impossible in simple logic. Power direction determines source and load. What has happened is the load has become the source and the instrument has not been reversed to measure load power. People seem to be confusing negative power factor with the artificially created positive and negative RVA statements, arbitrarily assigned, for convenience of communication . This has never applied to power factor. A further problem with this method of validating technical issues, and other issues, for that matter, in WP is outside references are based on repeated misconceptions becoming de facto standard and then accepted over time, ain't it so? Where does that leave us? Only using references that are over forty years old? :) 174.118.142.187 (talk) 16:53, 13 June 2013 (UTC)


 * Do you have any sources that say that? Any textbooks that say that there is no such thing as negative power factor? Class handouts? Other than that one where the author told us it was a mistake, any IEEE documents? Research papers? Magazine articles? Any manual for any power analysis instrument? Anything at all? "Because someone using IP address 174.118.142.187 says so" doesn't quite meet our WP:V requirement.


 * Imagine that you are measuring the power factor of a transmission line used by a small factory. You measure 120 volts and 120 amperes, 0 degrees of phase shift and 0% voltage and current distortion. You hook up your Fluke 434 Power Quality Analyzer and is says that the power factor is +1. Just to be sure, you hook up your Voltech PM1000+ Power Analyzer It also says that the power factor is +1.


 * Having worked most of the night analyzing power, you take a break, eat breakfast, and enjoy the eastern Canadian sunrise. When you return, a huge photovoltaic array on the roof of the factory has kicked in. There is no change to the voltage, current, or distortion, but the voltage and current are now 180 degrees out of phase, the utility's watt-hour meter is running backwards, and both analyzers show a power factor of -1. How can this be? You round up a dozen other instruments, and every single one of them reads -1 So you call a university professor, who tells you "I am heartily fed up with students telling me that negative PF does not exist". You claim that all of the instrument manufacturers are purposely telling lies, both in what they display and in the explanations in their manuals. Do you have any sources for that claim?


 * This reminds me of the drunk driver who was driving the wrong way on the freeway. Upon hearing on the radio (over the honking horns) that there was a drunk driver who was driving the wrong way on the freeway, he peered through his windshield, noticed all of the headlights heading toward him, and exclaimed "My God! There are DOZENS of them!!" --Guy Macon (talk) 23:15, 13 June 2013 (UTC)


 * This reminds me of that drunken driver also. Was there any signs that say it is illegal to drive on that side of the road when you are drunk? Do you have any cites or references that state that? Class handouts? Magazine articles? Perhaps a driver training manual? Let's see one. That's what the discussion is all about and because somebody with a twisted example doesn't justify it to be accurate enough to be in WP. The only thing that can happen here is put in what we have references for. In light of conflicting ones we use the ones with the most weight. I weigh about 95kg. :).  When your meter goes backward you state the absolute value of the power factor and mark it as being a generation point. If the kWh meter is going backward you install one with a detent. Utilities don't buy back energy at the same rate they sell it and the customer has to disconnect his source until the proper approval is obtained and  metering installed. What happens when your meters indicate a lower VA than your watts? Do you have power factor higher than unity (1.0) or do you conclude you have a metering error? Is your VA reading arithmetic VA or phasor VA? Do we now have arithmetic power factor? Legal Metrologists will tell you they do and know when to use it and when not to. A silly example doesn't prove any of the argument either way. I only stated a logical observation conclusion in light of disagreeing references. There are instrument readings and there is knowledge how to apply them. Your example used improper terminology in the analysis to justify the negative power factor fallacy. Power doesn't flow from a load to a source and we never measure the transmission line, we measure the load with load wattmeters. Your meters are backwards. WP demands the best reference source, wright or not. Who knows? 174.118.142.187 (talk) 00:38, 14 June 2013 (UTC)
 * Checking one of your references where the professor says "Sandy McEachern who is, after all, the acknowledged expert". If this is true where is her published papers? 174.118.142.187 (talk) 00:53, 14 June 2013 (UTC)
 * HIS website, which includes a list of publications and biographical information, can be found here (as has been indicated several times already): http://alex.mceachern.com/ ElectricTattiebogle (talk) 13:16, 14 June 2013 (UTC)
 * So Sandy is Alex? That name has been used multiple times without confirmation I can find. I am not well versed in reference quality for WP but would a source with products advertised supported by the very same source documents be a COI or meet the WP standards? I only see promotion in the linked site provided. 174.118.142.187 (talk) 14:02, 14 June 2013 (UTC)
 * Both Alex and Sandy are familiar forms of Alexander. -- Red rose64 (talk) 14:11, 14 June 2013 (UTC)
 * Until we get someone else with the last name of McEachern posting to this talk page we safely assume that any references to McEachern are to the only McEachern in this conversation. --Guy Macon (talk) 18:32, 14 June 2013 (UTC)
 * I guess we will have to wait for "Mac" to input. :) Talk about confusion. Geesh! 174.118.142.187 (talk) 04:04, 15 June 2013 (UTC)
 * Your point about the rate at which utilities buy back energy is probably an America-centric view. Here in Europe, any electricity generated by a photovoltaic scheme (or wind mill, water wheel etc.) is sold back to the utility at a higher rate than you pay for the electricity supplied.  Here in the UK, I pay 13.68 pence for every kWh of energy supplied to me by my electricity supplier.  When the sun shines (as it is at the moment) and my photovoltaic panels are producing around 4 kW of power, I am paid 42 pence for every kWh that I feed back through my meter to the electric company (the amount I generate is separately monitored by the PV equipment).  My electric meter is currently running backwards so that the accumulated kWh is reducing indicating that it is integrating negative power.  If my electricity supply company is the source of my power, and my home the load, then during the day the power flows from the load to the source and the electric meter integrates that power by subtracting this negative power from the (positive) accumulated reading.  Of course, the power isn't really negative, or my washing machine heater would freeze the water rather than heat it (an absurd concept).  It is just an (almost) universally used convention to indicate reverse power flow.
 * I doubt Europe has any basic principles that differ from North America's. Your meter will be indicating backwards but both register accumulators will have to be calibrated and certified separately and the independent results have math applied to get your bill. We have incentive programmes (note spelling :)) that force the utilities to pay out more than the energy is worth. Our utilities have a few ways of doing it. I only get credit for any net energy fed back into the grid when I am a "source" to the system. Others get big rewards for everything produced at exorbitant rates paid by the government, even if they consume it all at the lower rate. The utilities don't pay for it. 174.118.142.187 (talk) 14:13, 14 June 2013 (UTC)


 * I don't understand your reference to, "both register accumulators". Electric meters here have just one 'accumulator' for the watt-hours consumed.  If the power flow is from the utility to the consumer, the reading on the register dials increases as energy is supplied.  If the net flow is from the consumer to the utility, the induction disk rotates in the opposite direction and the reading on the register dials decreases as energy is returned (i.e. the meter integrates negative power).   The feed in rate is paid to me by the utility (and separately metered by the inverter system).  It is true that they get it back from the government, but that is invisible to me.  DieSwartzPunkt (talk) 11:18, 16 June 2013 (UTC)


 * Since an electric meter does not indicate VA or watts, it would be impossible for it to indicate a lower VA than watts. DieSwartzPunkt (talk) 12:14, 14 June 2013 (UTC)
 * Your meter doesn't indicate watts or VA that you can see on the front. Perhaps power factor doesn't exist at your European location. :) — Preceding unsigned comment added by 174.118.142.187 (talk • contribs) 14:31, 14 June 2013 (UTC)


 * My meter doesn't indicate or measure VA, watts or even power factor at all because it has no provision for doing so. It's sole ability is to integrate power consumed, both positive and negative, with respect to time (i.e. it measures energy and nothing else).  Since the electric meter article contains a very nice picture of an induction type watt-hour meter, perhaps you would care to point out, where in its structure you believe that there is a hidden mechanism for measuring VA or power.  DieSwartzPunkt (talk) 11:18, 16 June 2013 (UTC)


 * I think 'straws' and 'clutching' come to mind here. The article contains no less than a dozen references supporting the concept of negative power factor (and it has to be understood that it is only a convention for reverse (positive) power flow).  Wtshymanski tried to defend his (up until now) unilateral stance that negative power factor does not exist by claiming that all 12 references were bogus ( The edit summary is the telling feature).  But as Guy points out, so far, there have been no references supplied that specifically state that there is no such thing as negative power factor (beyond the IEEE paper which actually contradicted itself).  If suitable and reliable sources could be found which support that viewpoint beyond a fringe theory, then the article could legitimately document the alternative concept.  But so far, nothing has been forthcoming.  DieSwartzPunkt (talk) 11:53, 14 June 2013 (UTC)
 * Well as much as I agree with you and "a negative can never be proven" (Quote from you, I believe) it has to come down to the best reference for the article. Consensus has not been used on this issue as currently the article is locked down with the onus on the new edits editors to provide convincing discussion and haven't. We hardly can expect a few, less than half a dozen edits, editors with only this article and reverts of Wtshymanski edits experience, to participate intelligently in the WP process and prove the fringe theory that negative PF does exist. OTOH I have to agree with Wtshymanski that the 12 references have no substance, possibly imaginary or not suitable referenced to be of value. 174.118.142.187 (talk) 14:31, 14 June 2013 (UTC)


 * Since you mentioned references, the solution here is very simple. Open a section at the bottom of this page to propose a change to the article.  State what you want changed and what you want it changed to.  Most importantly, include those all important references that support the change you want made.  But that's the problem isn't it.  In spite of the continued requests made here, no reference has been forcoming apart from one IEEE document that contadicts itself on the issue.  As for the existing references, how good or bad everyone believes they are somewhat immaterial.  Even if they are poor references, they are better than no references (and that is a slight reword of a quote from Wtshymanski himself - but not in relation to this discussion).  DieSwartzPunkt (talk) 11:16, 17 June 2013 (UTC)


 * Is it just me, or can anyone else smell a sock here? Not only is 174.118.142.187 supporting the lone view expressed by Wtshymamski, but he is also supporting Wtshymanski's view that the 12 perfectly good references are inadequate.  Or is it just a coincidence?


 * Well here is another coincidence. Wtshymanski's location is a matter of public record (I won't repeat it here for fear of falling foul of WP: OUTING).  174.118.142.187 geolocates to the same place.  212.183.140.50 (talk) 14:44, 14 June 2013 (UTC)


 * Strangely, the thought had crossed my mind. As a further coincidence, 174.118.142.187 has been compiling some interesting information of his own [here] (My thanks the Mr Wright for bringing this to my attention).  The additional coincidence is that he is clearly supporting Wtshymanski's tendentious editing by compiling a list of some of those opposing him.  174.118.142.187 also has other similar characteristics.  He exhibits tendentious editing traits of his own; he blanks embarrasing remarks from his talk page in short order   (relatively few Wikipedia editors actually do so).  And when the deletion has taken place, that sarcastic edit summary that is so typical of Wtshymanski.


 * While formulating this response, 174.118.142.187 blanked the previous comment, so it obviously touched a nerve. Since he is making allegations, but blanking anyone else's attempt to do so, we have that Wtshymanski charecteristic that he is right and entitled to do what he likes, and everyone else is wrong and cannot.DieSwartzPunkt (talk) 15:21, 14 June 2013 (UTC)


 * I don't buy it. The "geolocates to the same place" is a fabrication. 174.118.142.187 geolocates to Collingwood, Ontario Canada (closest major city: Toronto). Per Wikipedia policies, I cannot reveal any personal information about Wtshymanski that he has not himself revealed on Wikipedia, but at File:RedRiverFloodwayInletStructure.jpg, File:HP 95LX Pocket Computer.JPG and File:Mercury emissions by light source EPA 2008.svg he reveals that User:Wtshymanski is W. T. Shymanski. The location of W. T. Shymanski is easily found by a web search and it is over 2000 kilometers (1200 miles) away from 174.118.142.187's location. Plus, Wtshymanski has reverted 174.118.142.187 at least once. --Guy Macon (talk) 19:43, 14 June 2013 (UTC)


 * In other words, a registered editor and an unregistered editor who argue the same patent nonsense are not necessarily sockpuppets. They may just be two incorrect disruptive editors.   Robert McClenon (talk) 22:58, 14 June 2013 (UTC)


 * Exactly. I admit, when I saw an IP editor agreeing with Wtshymanski despite the utter lack of sources for the fringe view that say that there is no such thing as negative power factor, sockpuppetry crossed my mind, but checking the geolocation showed no connection. In the process of writing up the above I looked very carefully at the articles they have both edited, and it became clear they both hold to a combination of mainstream and fringe beliefs about engineering -- but not the same fringe beliefs. --Guy Macon (talk) 02:29, 15 June 2013 (UTC)


 * As one engineer to another, sometimes engineers have to make fine distinctions between different sorts of patent nonsense. Robert McClenon (talk) 03:21, 15 June 2013 (UTC)

Sorry but you have the observation wrong. To my knowledge I was not agreeing with Wtshymanski, at the time. When I first commented I didn't know who was on what side. Now can we please get on with the discussion or is it important to distract from it? Is there anything else to state? Just for the record I can see both sides of the argument. The negative PF doesn't feel right, to me, as it violates basic source/load logic and the way I was taught, but simple math tells us that Neg. PF exists mathematically. Instrument readings and patents are worthless, and bogus references do nothing to further a conclusion. One reference refers to a patent for a voltmeter and makes no reference to power factor, (it refers to the phase angle between two different currents in the coils) period and only makes readers suspect of other falsehoods. It needs to be removed, ASAP. More reading may prove WT correct about these references. It would be interesting to hear about his research on these references that stand so high on the one side of the scale. 174.118.142.187 (talk) 03:57, 15 June 2013 (UTC)


 * The issue of "negative" power is actually very simple. First, we need to remember that power factor is a combination of displacement factor and distortion factor.  Displacement factor is a simple mathematical equation that defines the phase angle difference between the fundamental current and voltage waveforms (i.e. at nominal frequency). The equation is just the cosine of the phase angle between the fundamental, which has a range between +1 and -1. Distortion factor is the factor that allows for the effect of harmonics (i.e. non fundamental frequency) in the overall waveform.  Nonetheless, the fact that the displacement factor can vary between +1 and -1, clearly means that the overall power factor is also between +1 and -1.  Most students are taught that anything between +1 and 0 is leading (current waveform reaches its peaks before the voltage), and anything between 0 and -1 is lagging (voltage waveform reaches its peaks before the current). What we are actually talking about is the direction of the flow of energy, which can clearly be in two directions; thus the postive and negative. I think the real problem with "negative PF" is how people are taught when they are students.Bhtpbank (talk) 09:43, 20 June 2013 (UTC)


 * You might want to read http://powerstandards.com/Shymanski/draft.pdf . Phase angles (leading and lagging) in the range of -90&#x00b0; to 0&#x00b0; and +90&#x00b0; to 0&#x00b0; have a positive power factor. Phase angles in the range of -90&#x00b0; to 180&#x00b0; and +90&#x00b0; to 180&#x00b0; have a negative power factor. --Guy Macon (talk) 10:23, 20 June 2013 (UTC)

Negative power (and hence negative power factor) from first principles
I find the notion expressed above that there can be no such concept as negative power to be completely absurd. That its non existence could be classed as a "fringe theory" just does not work. I am not aware of any other professional electrical engineer who would see it any other way.

Consider a length of transmission line energised at V volts DC (keep it simple to start with). There is flowing a current of I amps. Thus energy is flowing along said line. Energy is an absolute concept and can only be positive. But the flow of energy is a directional concept and must therefore be either positive (flowing one way) or negative (flowing the other). Mathematically, the power in our considered line is VI watts. If the direction of current flow is reversed (to become -I amps), the direction of power flow is reversed and becomes -VI watts (negative power).

Of course, which direction is positive and which is negative is purely an arbitrary choice. Without any knowledge of what is connected to the ends of our line it may be determined by nothing more that which way around we connected the ammeter. In a practical circuit, we assign labels to the two ends, "SUPPLY" and "LOAD". By convention we regard positive power as flowing from the SUPPLY end to the LOAD end, if only because it is logical to do so.

In an AC transmission line nothing changes. All the relationships hold true except for the complication of power factor. The power itself still flows either one way (positive) or the other (negative). Our positive current is a current more or less in phase with the voltage. Negative current is replaced by a current more or less 180 degrees out of phase with the voltage. So here again mathematically, where the current is 180° out of phase, the power is -VI watts. I could explain this better if I resorted to complex notation, but I am loath to do so as not everone may be familiar with it.

Power factor is defined as the VA in the circuit divided by the power. Since VA is the RMS volts multiplied by RMS amps, VA can never be negative (because the RMS value can never be negative). Thus in our example, power factor is -VI/VA = -1.

A practical application of this negative power malarky is in my own speciality of railway traction. Once again which direction we regard as the positive flow is arbitrary, but (positive) power is regarded as flowing into railway locomotives because that is all it has done up until the second half of the last century. Since that time, regenerative braking means that locomotives now also convert some of their dynamic energy back into electrical energy and return it to the overhead wire (or third rail). In this instance, according to the convention, the power flow into the locomotive is negative (i.e flowing out). Modern AC locomotive specs specify the power factor for when motoring (absorbing power) and the power factor for regenerating (supplying power). The power factor specified for when regenerating is specified as a negative quantity (because the power is negative). It is difficult to cite these because most readers here are unlikely to have access to them as they are considered commercially sensitive.

As for those 12 references one after the other in the article (is it not possible to pick the two best ones and bin the rest?). Although they are not references that negative power factor is accepted as a concept per se, they are references that power factor is used in the electrical engineering world (i.e. that the concept is accepted and used).

As a practical example, consider a locomotive climbing an grade and consuming 2P kW from the overhead line. Nearby are two locomotives descending and using their regenerative braking and are consuming -P kW each from the same overhead line system (i.e. they are actually supplying power as it flowing in the reverse direction). To find the power being supplied by the sub-station all you have to do is add up the powers being drawn by the individual locomotives (2P+(-P)+(-P) = 0) which is the correct answer. This would not work if power could only be considered as a positive concept.

In our workshop we still have a pair of fairly old Crompton electro-dynamic power factor meters (one single phase and one three phase). Both have circular 360° dials and the pointer on either can move around all 360° of it. In both cases the quadrants are marked; "+ lead"; "+ lag"; "- lead" and "- lag". Our modern electronic power analysers also indicate both positive and negative power and power factor. – Live Rail    &lt; Talk &gt;  11:45, 19 June 2013 (UTC)


 * "Fringe theory" has a very specific meaning on Wikipedia. See Fringe theories and Why Wikipedia cannot claim the earth is not flat. How we treat this particular fringe theory should be based on two factors; the complete lack of reliable sources that reject the mainstream four-quadrant description, and the notable lack of adherents of this particular fringe theory. Both of these factors lead me to the conclusion that this article should define power factor as being in the range between +1 and -1, and that there is no justification for even mentioning this particular fringe theory. --Guy Macon (talk) 17:26, 19 June 2013 (UTC)


 * I've been watching this (sad, wasteful) dispute for some time. I concur with Guy Macon unequivocally. No mention should be made in the article of this "positive only" claim. It has not been discussed as such in reliable sources over time, and is unlikely to ever be. It is an error of comprehension, compounded by an erratum, expanded to a colossally zealous refusal to accept standard industry-wide mathematically correct practice. Wikipedia is WP:NOTFORUM, WP:NOTBATTLEGROUND, and is not to be used for WP:OR. --Lexein (talk) 08:55, 21 June 2013 (UTC)


 * Please verify one detail. There has been the claim made that power factor can only be positive.  That argument has been verified, on this page, to be fallacious.  Has that fallacious claim been published?  If so, it is a fringe theory, and may need mention.  Or is it only the argument of one or two Wikipedians who don't understand?  If so, it is original research, and incorrect original research at that.  Robert McClenon (talk) 12:28, 21 June 2013 (UTC)


 * As far as I can determine, and I think Wtshymanski agreed above, that the claim that "power factor can only be positive", has been published (or rather, not demonstrated false) only unintentionally, and only in one place (that we've found so far), in the form of an error in a single IEEE standards document. This was discussed at length above, and sent on its standards-committee voyage to being corrected, by AMcEachern, above. --Lexein (talk) 15:19, 21 June 2013 (UTC)


 * I spent a lot of time looking for a source for this, because I wanted to be able to mention the "positive only" fringe theory. I not only looked for a reliable source that we can cite, but even for unreliable sources such as blogs, Youtube comments etc. so that at least I could say on the article talk page that someone besides Wtshymanski and 174.118.142.187 holds the theory. I found nothing except the usual mirror sites that repackage Wikipedia articles (we had it wrong until recently), and I really did spend a fair amount of time searching. I would assume that Wtshymanski and 174.118.142.187 also did that search and would have told us if they had found anything at all that supports their theory. -Guy Macon (talk)


 * This is a fascinating discussion. It seems to me that All About Circuits (an online book) is an excellent discussion (ranging over several pages) of power factor that might be a useful additional source for clarifying the issues. I do not have sufficient experience to venture an opinion myself, but I do find the writing here to be very clear. David Spector (talk) 13:08, 22 June 2013 (UTC)


 * Thank you for your interest and the link. Unfortunately, the material that you referred to does not move us any further forward because, although it gives a very good discussion of power factor, it is restricted to circuits where a 'supply' always supplies the power to the 'load'.  As already discussed, this whole business of negative power and power factor arises from a situation where the 'load' generates power and feeds it back to the 'supply' (Liverail's railway locomotives above are a superb example of this in action).  Sadly, the allaboutcircuits material does not discuss a circuit where the power flow reverses and is thus silent on the concept of reverse power flow and how is dealt with numerically.  This still leaves us with the situation that the point of view advanced above, that power and power factor cannot be negative, is unsupported by any references.  DieSwartzPunkt (talk) 11:48, 23 June 2013 (UTC)


 * I take your point. Since this seems to be an argument between experts, I'm afraid my opinion as someone who understands the basic physics of AC circuits is irrelevant. Probably we need an additional expert, such as someone who teaches this topic professionally. Or perhaps some textbooks that are explicit on this topic. David Spector (talk) 22:57, 17 July 2013 (UTC)
 * It is an interesting question as to how one can properly evaluate a dispute between experts. We see this in a lot of areas; climate change, creationism -- I have even seen spirited disputes about WWI battle tanks. One problem is that those who push fringe views often have a huge amount of "evidence" they can bury you with. Just try arguing with a holocaust denier or a World Trade Tower/9-11 conspiracy theorist some time.
 * So, how does a non expert tell who is correct? There are two methods, consensus and citations.
 * Consensus: Every engineer who has looked at this disagrees with Wtshymanski on this. Many are the usual anonymous screen names, but I use my real name and my resume is on my web page, and Wikipedia user AMcEachern is Alex McEachern, fellow at IEEE, author of (draft) Some corrections to IEEE Standard 1459-2010, and owner of the alex.mceachern.com and powerstandards.com web sites.
 * Citations: Wtshymanski has provided exactly zero citations to reliable sources that back up his position. That's always a bad sign. He does claim that IEEE 1459 supports his position, but it has been explained to him multiple times that he is misinterpreting IEEE 1459. One of the many engineers who have explained this to Wtshymanski is Alex McEachern, one of the principle authors of IEEE 1459. --Guy Macon (talk) 01:49, 18 July 2013 (UTC)
 * I don't understand why we are still discussing this. I thought we reached a consensus days ago, one that does not favor Wtshymanski's position. Binksternet (talk) 02:50, 18 July 2013 (UTC)

is the top diagram correct?
In the top diagram titled "Instantaneous and average power calculated from AC voltage and current with a zero power factor", shouldn’t the voltage and current waveforms be in sync? (for a zero pf) It appears the current lags the voltage by 90 degrees. Shane — Preceding unsigned comment added by 112.213.179.150 (talk) 22:27, 8 August 2013 (UTC)


 * The diagram is correct. Power factor is the cosine of the phase difference between voltage and current (valid for sinusoidal waves only).  When the voltage and current are 90 degrees out of phase, the cosine of that 90 degree phase shift is 0.  If the voltage and current were in phase, the the phase shift would be zero and the cosine of zero degrees is 1.  DieSwartzPunkt (talk) 15:31, 9 August 2013 (UTC)


 * Yes you are correct, my misinterpretation of the diagram, For some reason after viewing the image and reading the caption, I somehow automatically assumed a pf of 1 (unity), I don't know why despite it being clearly worded.  In my days of electrical theory 40 yrs ago it was always displayed a unity pf diagram first (using purely resistive load) along with some theory, then introducing leading and lagging diagrams after one had come to understand the theory a little. BTW... my question and comments may be deleted by anyone if need be, however it may be prudent to include another diagram showing the in-phase relationship of a unity pf.
 * Shane — Preceding unsigned comment added by 112.213.170.206 (talk) 20:55, 9 August 2013 (UTC)


 * A perfectly valid and reasonable point. If anyone has access to 'gnuplot', the original contributor of the diagrams kindly left the source code used to generate the plot on the file description page.  Can anyone tweak as required and oblige?


 * Incidentally if you add four tildes to the end of your post ( "~" ), The editor will automagically add a signature to the end of your post. DieSwartzPunkt (talk) 15:46, 10 August 2013 (UTC)


 * Correct correction. To be pedantic, three tildes works as well - you just don't get the date and time (but it is possible that some bot will remedy the omission).  DieSwartzPunkt (talk)


 * Yes Three tildes works, but is not recommended:
 * "Since typing four tildes adds the time and date to your resulting signature, this is the preferred option for signing your posts in discussions ... since [three tildes] does not date-stamp your signature, you may wish to sign this way only when leaving general notices on your user page or user talk page." --WP:SIG.
 * Useless trivia: you can also use three tildes, a space, and five tildes, as I have done with this post. --Guy Macon (talk) 20:59, 19 November 2013 (UTC)

Added section expanding upon distortion power factor
I recently added a section titled "Distortion in three-phase networks" below "Distortion Power Factor" to describe the local effects of non-linear loads, which I felt was a needed topic for clarity when it was later assumed that harmonic distortion is harmful to the network and connected loads. I am worried that this discussion is outside the scope of the distortion power factor itself, but I felt it would help with overall clarity.

I also added more detail in the Distortion power factor section to describe the distortion of voltage and why it can be neglected. The previous statement that it can be assumed that voltage is sinusoidal could be misunderstood by the reader to mean that there is no voltage distortion. On another note, I want to point out that the definition of distortion power factor is probably incorrect. To my understanding, current harmonics do not change the average power transferred to the load, but it affects the instantaneous power being delivered. I don't know exactly what the definition should be, but someone with more background should consider editing it, at least to make it more clear. Lastly, I am not sure if my citing of sources was acceptable, since I used a single source in multiple areas.Kmuster (talk) 07:29, 27 November 2013 (UTC)


 * Nice addition. Citing sources is always acceptable.  If your single source covers all the points made then fair enough.  My only question to you is: do not the 6th and 12th (etc.) harmonics also contribute to to the triplen currents as they are also in (angular) phase with each other in each (electrical) phase?  I B Wright (talk) 12:33, 28 November 2013 (UTC)

For nonlinear loads, the general PF is often defined to be average input power divided by RMS input power. The reason for this is that if the input voltage is sinusoidal and the period of the input current is an integer multiple of the fundamental frequency, the math works out such that the power factor defined in this way is equivalent to dividing the amplitude of the fundamental Fourier coefficient of the input current by the RMS of the input current, and then dividing again by square root of two. The distortion power factor formula you have listed is correct, but I think some additional derivations as per are in order. 128.174.115.178 (talk) 17:37, 3 December 2013 (UTC)

Not consistent
Not sure if this is by design, but note that on this page you say apparent power is simply S but in the wikipedia article for apparent power, you define apparent power as the magnitude of S, i.e. |S|, while S is defined as the complex power. It would be nice if these two entries were more consistent -- i.e. |S| is apparent power (the magnitude of the complex power, |S| = sqrt(P^2 + Q^2)) and S is complex power for both this article as well as the article for apparent power. — Preceding unsigned comment added by 50.58.26.98 (talk) 22:02, 15 January 2014 (UTC)


 * Apparent power can be quantified either way. It can be quantified as simply its magnitude (and often is), or it can be quantified in complex form. Which is chosen by engineers is based on what he proposes to do with it. In this article, the magnitude is sufficient for the discussion at hand and going into complex notation only serves to obfuscate the points beng made. The apparent power article, by necessity has to go into more detail and needs to reflect that apparent power is, in reality, a complex quantity. 86.171.45.200 (talk) 17:27, 16 January 2014 (UTC)


 * In reality, the magnitude of the complex power is |Q|. |S| is the magnitude of the total power. Bhtpbank (talk) 12:11, 17 January 2014 (UTC)


 * Not correct. S (or |S| if you prefer) is the apparent power (expressed as either magnitude 'S' or in complex form as 'P+jQ').  Q is the reactive power or the imaginary part of the apparent power.  P is the actual power or the real part of the apparent power.  This is covered in more detail at apparent power.  86.171.45.200 (talk) 13:48, 17 January 2014 (UTC)

Another non-compliant example
Because the image below has been nominated for deletion, I am posting a description of the content so that anyone reading this in the future will be able to see what the deleted image contained. - |               ELECTRIC              | |-| | Uab  14.947  kV  |  Ia     4352 A   | | Ubc 14.947  kV  |  Ib     4328 A   | | Uca 14.947  kV  |  Ic     4396 A   | | P    112.3  MW  | cos*  -1.001     | | Q     -8.0 MVar |  f     50.36 Hz  | - Descriptive text on file page: An illustration of a power factor display, incorrectly showing a value of -1.001 Original caption: A power factor display that is not calculated in compliance with IEEE 1459. Date: July 16, 2013 Author: Wtshymanski GNU Free Documentation License Creative Commons Attribution-ShareAlike 3.0 License.

--Guy Macon (talk) 16:11, 29 January 2014 (UTC)



Maybe it's a hyperbolic cosine? --Wtshymanski (talk) 19:40, 17 July 2013 (UTC)


 * There is a big difference between "not calculated in compliance with IEEE 1459" and "not calculated in compliance with Wtshymanski's misinterpretation of IEEE 1459". This has been explained to you in detail by one of the principle authors of IEEE 1459, who has started the process of correcting the ambiguity in IEEE 1459 that led to your misinterpretation. You were wrong. You need to drop the stick and back slowly away from the horse carcass now. --Guy Macon (talk) 00:52, 18 July 2013 (UTC)


 * But you would have at least expected that the idiot who wrote the software behind the display might have realised that the power factor is bound by the limits of +1.000 and -1.000 and applied those limits to the posible displayed values rather than permit the erroneous display of -1.001. One cannot help but wonder if a bad choice of current and/or voltage waveforms would produce a number even more outside of the bounds.  My own power factor meter (part of a modest priced plug in energy monitor) claims an accuracy of ±0.03, but I have never seen it display 1.01, 1.02 or 1.03.  DieSwartzPunkt (talk) 11:16, 18 July 2013 (UTC)


 * Related: File talk:Cosphi not IEEE 1459.jpg --Guy Macon (talk) 13:29, 18 July 2013 (UTC)


 * The picture shows a power factor not compliant with Wikipedia's definition either. Negative power factor is equivalent to saying real power reverses direction but apparent power doesn't. Seems arbitrary to me. --Wtshymanski (talk) 12:38, 28 July 2013 (UTC)


 * What part of "nobody agrees with your fringe theory" are you having trouble understanding? What part of "you have provided zero citations backing up your fringe theory" are you having trouble understanding? DROP THE STICK. DO IT NOW. --Guy Macon (talk) 22:46, 28 July 2013 (UTC)

I have come back here after over a month to discover that this is still rumbling on. I might have assumed that it had been long resolved, so what am I missing? I was a bit worried when I saw Wtshimanski's comment that the picture that he supplied shows a "power factor not compliant with Wikipedia's definition either". This somewhat alarmed me and so I checked what the article said. The section entitled "Definition and explanation" does not contain what I would recognise as a definition but does have an explanation. The nearest to a definition is the opening sentence of the article which is as good as any and makes it quite clear that power factor can be between -1 and +1. The "Definition and explanation" section discusses how power factor is arrived at, though does not explicitly state that apparent power (in VAs) is only ever positive or even how it is calculated.

However the link to Volt-amperes is quite clear in that it (correctly) states that the apparent power is the RMS volts multiplied by the RMS amperes. Since the RMS value (of anything) can only ever be positive, the apparent power in VA can therefore only be positive and thus is not considered to have direction. Wtshimanski is correct when he says, "real power reverses direction but apparent power doesn't". As for whether it is arbitrary? It probably is, given the definitions that electrical science has settled on for real and apparent power. The convention that we are left with is that apparent power does not have direction but real power does. I, as an elecrical engineer, have managed quite satisfactorily with that convention as have all of my learned colleagues. – Live Rail    &lt; Talk &gt;  12:46, 30 July 2013 (UTC)


 * If Wtshymanski was pushing his fringe theory by constantly arguing for it, that would be a behavioral problem to be addressed with warnings and blocks, but as long as he just pops up every so often, just post another "you have no sources/drop the stick" response. At this point it is really no different than the way we deal with creationists posing to the Evolution talk page.


 * As a (pointless but interesting) aside, try calculating power factor from the basic definition of RMS voltage and current without following the convention of using the (positive) principal square root. Totally useless for electrical engineering work, but calculating power factor using the negative square root (or even the square root of negative numbers, using complex numbers) is an interesting exercise.


 * In the real world using the normal conventions, adding a DC voltage offset -- positive or negative -- to an AC voltage waveform always increases the RMS voltage, which reflect the physical reality that a resistive load will get hotter whether you add positive or negative DC. Of course if it really is a resistive load, power factor cannot go negative either. --Guy Macon (talk) 16:26, 30 July 2013 (UTC)

For the generating unit in question, I saw it display a negative number in the field labelled "cos phi" when P was negative and Q was positive, and also with P positive and Q negative. Though I checked it each time I visited the powerhouse, I was never able to catch what it would say when both P and Q are negative. ( Asking: What is the generator doing when both P and Q are negative? And why would you run a hydro generator that way?)  What does Wiki consensus say it should display then? What does IEEE 1459 say it should display? But even Wiki consensus agrees that cos phi must be strictly contained in the closed interval -1, +1. --Wtshymanski (talk) 16:51, 24 August 2013 (UTC)


 * There seems to be something wrong with either your observation or the instrument in question (I suspect the former). Since P=Irms*Vrms*cos Φ, if the cos Φ reads negative when P is positive then there is something wrong as neither Irms nor Vrms can be negative by definition.  But neither your faulty observation nor faulty instrument constitutes a reliable source supporting your unilateral theory so  DROP IT .   DieSwartzPunkt (talk) 12:31, 25 August 2013 (UTC)
 * I was worried about such nasty suspicions, which is why my collection of pictures of HMI screens showing silly things for power factor continues to grow. For what it's worth, I've never seen sillyness from old-fangled pointer on scale instruments; it's only this new digital era that has lead to instruments in powerhouses saying ridiculous things. We were doomed when we were conditioned to click on "start" to shut down a computer; we've given up expecting computer-generated screens to tell the truth.  Wikipedia is consistent with this lowered expectation. --Wtshymanski (talk) 22:35, 25 August 2013 (UTC)
 * Clearly I need to be even more blunt; Wtshymanski, you have two choices. You can drop the stick, which means that on all pages where you have [A] every other editor opposing you and [B] zero sources (the one implies the other; if you had reliable sources to back up your fringe theories then the other editors wouldn't oppose you) you need to stop posting again and again about how wrong everyone else is. Or you can refuse to drop the stick, in which case I will personally write up an RFC/U. I think we both know that the result of that would be very likely to be a topic ban or a set of longer and longer blocks. Actions have consequences. --Guy Macon (talk) 22:47, 25 August 2013 (UTC)

Repetition has nothing to do with truthfulness. I have a reliable reference, the IEEE 1459 standard. --Wtshymanski (talk) 17:31, 14 November 2013 (UTC)


 * A reference that is so reliable, that in one place it states that power factor cannot be negative and in another place it states that it can. The fact that this reference is in error has been freely admited by its author above and corrective action put in train.  I B Wright (talk) 19:20, 14 November 2013 (UTC)


 * That's it. I have had more than my fill of Wtshymanski's fringe theories, battleground mentality and tendentious editing. Just as there is science and pseudoscience, there is engineering and pseudoengineering. In the next week or so I will write up a second RFC/U asking for a topic ban. Enough is enough. --Guy Macon (talk) 10:55, 19 November 2013 (UTC)
 * Ping me when you're done. I will want to participate. Binksternet (talk) 13:32, 19 November 2013 (UTC)
 * Me too. --Lexein (talk) 14:46, 19 November 2013 (UTC)
 * (At the risk of imitating a parrot.) Me too.  I B Wright (talk) 12:40, 28 November 2013 (UTC)

You are all on my notification list. I will let you know when I think the draft version is ready so you can suggest improvements, and again when it gets posted. I am also going to make a point of notifying all of W's supporters (Sound of Crickets...). --Guy Macon (talk) 13:45, 28 November 2013 (UTC)

Confusion about Negative Power Factor
I came to this page to try to understand why a power datalogger was giving me a negative power factor. In the page itself, it states that PF can be negative but does not provide any interpretation of this. In the calculation method given, the only way to get a negative PF is if the power flow is in the direction opposite the originally-selected convention (in other words, if you define point A as SOURCE and point B as LOAD and you have power flowing from B to A, then PF is negative). This jives with my understanding of PF but it does not make sense on my datalogger. It turns out that the datalogger uses the convention that negative PF means current lagging the voltage. This makes no sense at all to me, mathematically, and indeed I found a paper online by one of the authors of the IEEE standard on measuring power factor which says that this convention (sign indicating phase shift direction) should not be used as it is confusing.

I think that this subject should be addressed in the article. In other words, it should state that there are various interpretations of negative power factor but that only one interpretation is consistent with the mathematical formula for calculating PF (that is, that negative PF indicates reverse power flow).

I was about to edit the page to add this but then I saw this talk page. All I can tell from the talk page is that there was a huge argument and one person is continually accused of espousing a fringe theory - but I cannot tell what people think is fringe versus mainstream. What is the consensus here about how to interpret negative PF and do you think that should be made explicit in the article?

Thanks Sbreheny (talk) 00:13, 8 April 2014 (UTC)


 * Fringe theory: "There is no such thing as a negative power factor"


 * Obsolete convention: Phase angles (lagging) in the range of -90° to -0° have a negative power factor.  Phase angles (leading) in the range of  +0° to +90° to  have a positive power factor. Phase angles in the range of -90° to -180° and +90° to +180° are undefined. Some meters using the obsolete convention incorrectly show -90° to -180° as having a power factor in the range of -1.0 to -2.0. Some incorrectly "peg the needle" at -1.0. Some display "error". Some incorrectly give the same reading for -45° and -135° (this appears to be the source of the fringe theory). Also, some meters using the obsolete convention display "-" for leading and "+" for lagging.


 * Alas, the above is from personal experience, and is thus WP:OR. If we could find references, we could add a section on this to the article.


 * Current, standards-based definition: Phase angles (leading and lagging) in the range of -90° to -0° and +0° to +90° have a positive power factor. Phase angles in the range of -90° to -180° and +180° to +90° have a negative power factor. Power factor readings do not indicate leading or lagging. Modern meters often have a separate display giving the phase angle.


 * Of course -0° and +0° are the same angle and -180° and +180° are the same angle. One could use 0° to 359.999...° (has anyone ever done this when talking about AC power?) to remove this minor quirk.


 * There is more on this at http://powerstandards.com/Shymanski/draft.pdf


 * --Guy Macon (talk) 02:57, 8 April 2014 (UTC)


 * Oh My God, some idiot is starting up this debate AGAIN. Better check the archives,  Mario Castelán Castro. Millions have died and lands have been laid waste over this question. -- Chetvorno TALK 00:27, 11 October 2014 (UTC)


 * If you see a minus sign when your datalogger is displaying power factor, it means current is flowing the opposite way than expected. Reverse the dot and non-dot connections to each phase CT and the negative sign should go away. Not all instruments behave this way, and I have several amusing pictures of randomly flickering negative signs (and even power factor displays showing values over 1).  --Wtshymanski (talk) 21:23, 31 October 2014 (UTC)


 * [Big Groan] DieSwartzPunkt (talk) 16:43, 9 November 2014 (UTC)


 * Wtshymanski, as it clearly states at the top of this page, this is the talk page for discussing improvements to the Power factor article. This is not a forum for general discussion of the article's subject. --Guy Macon (talk) 19:40, 10 November 2014 (UTC)


 * WARNING***WARNING***WE ARE AT DEFCON-3 Please, for the sake of the world, pick up the red phone, stand down the bombers, pull back from the brink before it is too late. -- Chetvorno TALK 02:40, 11 November 2014 (UTC)


 * I suppose I'm out of date but I've never heard of a negative power factor. The normal power factor (when the voltage and current curves are in line) is 1. When they are out of line the power factor is either (say) 0.9 leading or 0.9 trailing, but not negative. If you are going to use the range -1 to +1 then you would have to re-define the normal power factor as zero. Biscuittin (talk) 19:07, 11 March 2015 (UTC)


 * If you are the typical engineer working on things with typical power supplies you may have never ran into a negative power factor. After all, the point is usually to get power from the wall, not feed power into it.


 * If you ever work in any environment that requires you to use a power analyzer, you know that the current, standards-based definition is that phase angles in the range of -90° to -0° and +0° to +90° have a positive power factor and that phase angles in the range of -90° to -180° and +180° to +90° have a negative power factor.


 * http://powerstandards.com/Shymanski/draft.pdf -- a paper written by by Alex McEachern, (Author of Electric Power Measurements and Handbook of Power Signatures, Chair of IEEE 1159.1 (Power Quality Measurements)]), explain all of this better than I could. --Guy Macon (talk) 22:13, 11 March 2015 (UTC)
 * I've read the paper but it still isn't clear. If you are not going to use + and - signs, how do you differentiate between a power factor of 0.9 leading and 0.9 trailing? You can't have a power factor of 1.1 Biscuittin (talk) 23:17, 11 March 2015 (UTC)

You may find 's explanation and practical example might make things a little clearer (here). He does not go into leading and lagging power factor though it is touched on in the last paragraph. DieSwartzPunkt (talk) 12:44, 12 March 2015 (UTC)


 * Could not find anything relevant there. Am I right in assuming that it is no longer acceptable to express power factor as a decimal?  It must be expressed as a phase angle? Biscuittin (talk) 19:28, 12 March 2015 (UTC)


 * That is an incorrect assumption. Power factors are expressed as decimals in the range of plus one to minus one. Phase angle is a separate measurement, expressed in degrees, and is only really applicable when both current and voltage bear at least some resemblance to a sine wave. Consider a case where the voltage is pretty close to a sine wave and the current is pretty close to being a square wave, but the top has a bit of a dip in the middle. Can you calculate the power factor? Easily. Can you calculate the phase angle? Only by making an assumption about where the "peak" of the square wave is. --Guy Macon (talk) 23:21, 12 March 2015 (UTC)


 * Could you please give me some examples? How is a power factor of 0.9 trailing now shown? How is a power factor of 0.9 leading now shown? Biscuittin (talk) 19:36, 13 March 2015 (UTC)


 * +0.9, +0.9.


 * Your question is the equivalent of asking "How is length of one meter heavy shown? "How is length of one meter light shown?". Like length and weight, power factor and phase angle measure different things and cannot be combined into a single number. --Guy Macon (talk) 21:17, 13 March 2015 (UTC)


 * You seem to be avoiding the question. If leading and trailing power factors are both shown as positive, how do I know whether the power factor is leading or trailing? Biscuittin (talk) 19:22, 14 March 2015 (UTC)


 * Your question makes zero sense.


 * QUESTION: If the weight of bleuberries and strawberries are both positive, how do I know whether the berry is blue or red?


 * ANSWER: Quit trying to measure color with a scale.


 * QUESTION: If leading and trailing power factors are both shown as positive, how do I know whether the power factor is leading or trailing?


 * ANSWER: Quit trying to measure phase angle with a power factor meter.


 * I am going to stop responding now. Either you lack the competence to understand power factor or you are trolling. You can reply to this if you wish, but I will not read it. --Guy Macon (talk) 13:05, 15 March 2015 (UTC)

There is plenty of potential for confusion here if you are not fully conversant with the subject. Power factor can be leading or lagging. Power factor can also be positive or negative. However, the two concepts are not interchangeable because they are precisely that - different concepts. If we consider the case where both voltage and current are sinusoidal, then power factor is considered leading if the phase of the current leads the voltage by anything between 0 and 90 degrees. Similarly, the power factor is considered to be lagging if the current lags behind the voltage by anything between 0 and 90 degrees.

On the other hand: power factor is considered to be positive if the phase between the current and voltage is anything between 270 and 90 degrees (through zero degrees). Similarly, power factor is considered negative if the phase between the current and voltage is anything between 90 and 270 degrees (through 180 degrees). Since the power factor is the cosine of the phase angle, power factor for any circuit is expressed as a decimal number between -1.0 and +1.0 inclusive. The relationship that gives rise to leading and lagging power factor applies equally when power factor is negative.

Thus it quite possible to have a leading positive power factor; a lagging positive power factor; a leading negative power factor or a lagging negative power factor. A non-sinusoidal current complicates the calculations somewhat, but all four outcomes are still possible. I mentioned elsewhere that we still have in our workshops, a pair of very old Crompton electro-dynamic power factor meters. The needles on both can fully move through 360 degrees. The circular scale is divided into 4 quadrants, with two opposite quadrants being marked 'leading' and the two remaining opposite quadrants marked 'lagging'. Two of the adjacent leading and lagging quadrants are identified with as negative and the two remaining as positive. Thus the instruments are capable of displaying all four power factor conditions outlined above. – Live Rail    &lt; Talk &gt;  13:18, 15 March 2015 (UTC)


 * I'm afraid that Guy Macon has not understood my question and I am certainly not trolling. I am trying to make the article comprehensible to the layman.  I'm afraid we are in a situation where the experts are so expert that they cannot understand the problems of the layman.  I will re-read the article and write again afterwards. Biscuittin (talk) 18:26, 16 March 2015 (UTC)


 * I think I now understand. We can still use the terms "leading" and "trailing" but we must not replace them with + and - because that would lead to confusion with negative power factor. Is this right? Biscuittin (talk) 19:25, 16 March 2015 (UTC)
 * Yup: you've pretty well got it. As I said: leading -v- lagging and positive -v- negative are different concepts.  It is not uncommon to have to read a technical article through more than once to grasp the points that it is making.   –  Live Rail    &lt; Talk &gt;  16:49, 19 March 2015 (UTC)

Closed interval or open interval
Are we certain the lead should say closed interval? All the best: Rich Farmbrough, 15:31, 2 April 2015 (UTC).


 * Yes. The power factor of a ideal load or source accounted as such is 1, and accounted as the other type (load as source or source as a load) is −1. I suppose that it is not physically possible to get to exactly 1 or exactly −1, because there is always inductance and noise (at least shot noise) which prevents current and voltage from being exactly within a mulitplicatie constant of eachother (or any other reference signal). However, in practice, the difference is usually negligible even for a real resistive load which has inductance. When analyzing systems, it is commonplace to consider power factos of 1 and −1, similar as to how we may say “the mass of [...] is 70 kg”, despite that it is just an approximation. 1 and −1 are valid power factors, just like 1 is a valid efficiency, even though it may be unattainable.
 * In principle, we could replace “closed interval” with its description, but pepole who don't know what it means may already consult that article; likewise, we could do the same for “dimensionless number”, “AC” (alternating current) and so on, I don't think that doing so would improve the article, it would just be digression which would duplicate effort and text.
 * Thanks for contributing.
 * Mario Castelán Castro (talk) 16:06, 2 April 2015 (UTC).

Power
Back in the 1960s when I was a stripling EE undergraduate at IC, London, it was impressed upon us with some force that power is simply power, and needs no qualification (real, imaginary or apparent). To qualify it is to add confusion. The quantity described here as apparent power is simply VA - voltamperes, while so-say "imaginary power" or "reactive power" is simply VAR - volt-amperes-reactive. In the case of VAR there is no connection with power at all, so to introduce the idea of power in this case will inevitably confuse. Is this power, VA, VAR terminology alien to Wiki authors - it seems to me to be preferable.G4oep (talk) 18:51, 9 January 2015 (UTC)


 * Sadly, for every page in your prof's textbook, there's a shelf full of textbooks cheerfully nattering on about real, reactive, active, inactive, virtual, apparent, wattless, etc. "power". It's bad terminology and we're stuck with it. --Wtshymanski (talk) 19:09, 9 January 2015 (UTC)


 * "there's a shelf full of textbooks" wins, and "it seems to me to be preferable" loses. As WP:FLAT says,
 * "If Wikipedia had been available around the fourth century B.C., it would have reported the view that the Earth is flat as a fact and without qualification. And it would have reported the views of Eratosthenes (who correctly determined the earth's circumference in 240BC) either as controversial, or a fringe view. Similarly if available in Galileo's time, it would have reported the view that the sun goes round the earth as a fact, and Galileo's view would have been rejected as 'original research'. Of course, if there is a popularly held or notable view that the earth is flat, Wikipedia reports this view. But it does not report it as true. It reports only on what its adherents believe, the history of the view, and its notable or prominent adherents. Wikipedia is inherently a non-innovative reference work: it stifles creativity and free-thought. Which is a Good Thing."
 * G4oep, please don't assume that a particular terminology is "alien to Wiki authors". Several of us are very familiar with the usage you mention. It isn't in the article because it fails basic Wikipedia standards such as WP:V, WP:RS and WP:OR. --Guy Macon (talk) 04:12, 10 January 2015 (UTC)
 * The assumption "that power is simply power, and needs no qualification (real, imaginary or apparent)" has led to actual cases where wires have caught on fire inside walls when feeding an office full of older PCs that only draw current -- a lot of current -- at the peak of the sine wave (imagine a bridge rectifier charging a big cap times a couple of thousand PCs). --Guy Macon (talk) 23:29, 12 March 2015 (UTC)


 * G4oep: I agree that “imaginary power” and “reactive power” are not ideal names, but they are not the same as volt ampere (VA) and volt ampere reactive and therefore should't be used interchangeably. The BIMP classifies the former as “physical quantities”; the later are the units usually used with the former, but are not SI units. Note that “voltage” and “volt” have similar names, but they are different and bear roughly the same relation to eachother as apparent power to volt ampere and reactive power to volt ampere reactive.
 * Science and engineering relates physical quantities without regard to their units; speaking of equations for volt amperes and volt amperes reactive instead of equations for apparent and reactive power would be a mistake. Units have fixed values when expressed using other units; physical quantities what may vary and the ways in which they are related may be described by equations such as $$S=\left|I\right|\left|E\right|$$ (this is an equation that relates physical quantities, not the units).
 * Regards. Mario Castelán Castro (talk) 17:37, 2 April 2015 (UTC).

Leading and lagging current
I think there might be a case for merging Power factor and Leading and lagging current. I made some criticisms at Talk:Leading and lagging current in May 2012 but nobody has answered them. Biscuittin (talk) 19:34, 14 March 2015 (UTC)


 * First you have to convince the entire world to destroy all electrical loads that are not a combination of capacitors, inductors and resistors. Then and only then would those two articles cover the same topic. Meanwhile, back here in the real world, loads are often nonlinear (example, a bridge rectifier and a large capacitor), and so we need a separate article on power factor. --Guy Macon (talk) 20:49, 14 March 2015 (UTC)


 * OK, I'm only trying to understand the subject but nobody seems to be able to explain it in a way that is comprehensible to the layman. Isn't this what Wikipedia is for? Biscuittin (talk) 22:03, 14 March 2015 (UTC)


 * As an engineer who lives and breathes this stuff, I am often poor at explaining things in a way that is comprehensible to the layman. I am not trying to mess with you; it is just a simple fact that I am better at communicating with other engineers. So, where can you get help that doesn't suck quite so badly as my help often does? Searching on Google for [ simple power factor explanation ] and [ power factor explained ] brings up a number of webpages and videos. Perhaps one of them may be able to help. Also, Reference desk/Science is full of fine folks who excel at explaining things. BTW, I agree that Leading and lagging current is a poor-quality article. Let me see what I can do about that. --Guy Macon (talk) 00:54, 15 March 2015 (UTC)
 * We've Wikied it up beyond hope. Where does the explanation go off into the weeds?  And can we fix it short of making the article a tutorial on the entire  history of electrical engineering since 1831? --Wtshymanski (talk) 05:17, 15 March 2015 (UTC)
 * I support merging Leading and lagging current, either here or in AC power. Note that it is not required that the articles cover the same topic in order to merge. That is a possible reason, but not the only one; see Merging. Merging meets the other 3 reasons mentioned (1 is enough already). The explaination of the current to voltage difference of phase is part of the context of both AC power and power factor. Either article would be more complete talking about this phase difference because it is a fundamental characteristic of AC power systems and a very common reason for power factor ≠ 1, and therefore a reason to consider the concept of power factor in the first place.
 * Biscuittin: There is a good explaination of power factor, leading and lagging current in All About Circuits which is free as in freedom like Wikipedia and GNU/Linux. Bear in mind that Wikipedia task is not “to explain it in a way that is comprehensible to the layman”; there is the guideline (not policy) Make technical articles understandable which is related but not the same.
 * Regards. Mario Castelán Castro (talk) 19:23, 2 April 2015 (UTC).

Unclear sentences
I read the article and I am somewhat confused about two sentences.

1. "This is "displacement power factor". The concept can be generalized to a total, distortion, or true power factor where the apparent power includes all harmonic components."

This sentence doesn't make sense to me. I don't know if it's the content or the punctuation (excess of commas). This sentence should be reworked to improve clarity.

2. "A typical multimeter will give incorrect results when attempting to measure the AC current drawn by a non-sinusoidal load"

Non-sinusoidal? Maybe it should say "non-resistive"? Please modify as needed to clarify.

ICE77 (talk) 04:05, 21 July 2015 (UTC)


 * 1. It's too bad there is no source cited here. I could edit this to what I think the editor is trying to say. But it would be better to find a source.


 * 2. I'm pretty sure I know what the editor is trying to say here. It's not the load that is non-sinusoidal, it's the current drawn by the load. So it could be a resistive (non-reactive) load, but one whose resistance changes through the cycle. But again, a source would help. Kendall-K1 (talk) 09:53, 21 July 2015 (UTC)

Kendall-K1, I entirely agree with you.

ICE77 (talk) 04:33, 22 July 2015 (UTC)


 * I find 2 Google book sources. one for each sentences. I hope it will help.--Wolfch (talk) 12:51, 23 July 2015 (UTC)

Non Sinusoidal forms
The section as its stands is naive. Nonlinear loads certainly create unwanted harmonics, and often only just odd overtones. Exact full rectification produces even harmonics. However, the distorted waveform itself is a summation of pure sinusoids, if one believes the validity of the Fourier series. A square wave has the ultimate richness in harmonics (the amplitude dropoff is 1/f) although it's generation may be deliberate with on/off switch. The type of equipment being powered determines if overtones cause reactive or reflection imbalances. Purely restive loads do not create such problems.14.202.249.249 (talk) 06:29, 11 November 2015 (UTC)
 * I think the point is that non-reactive non-linear loads supplied by a sinusoidal voltage will have a power factor other than 1, because the current and voltage, while they are in phase, don't have the same waveform. The picture helps. I'm not sure where the confusion lies, can you suggest alternate text? Kendall-K1 (talk) 18:00, 19 February 2016 (UTC)

Do you mean?
Says: "That means that the input current has energy at harmonics of the frequency of the voltage."

Do you mean: "That means that the input current has energy at harmonics of the frequency of the input AC voltage." ?

ie Harmonics of say 50 (Hz)? Cinderella157 (talk) 10:43, 15 March 2017 (UTC)


 * Yes, but the entire article is written in the context of AC, so I don't think it's necessary to say that. Kendall-K1 (talk) 13:01, 15 March 2017 (UTC)

lede improvement suggestions
The lede could be much improved. (I am just encountering this article for the first time. I am familiar with the subject matter.)


 * "In electrical engineering, the power factor of an AC electrical power system is defined as the ratio of the real power flowing to the load to the apparent power in the circuit,[1][2] and is a dimensionless number in the closed interval of −1 to 1."

The current first sentence is inappropriately worded. "dimensionless number" and "closed interval" are quite distracting, overly math-tech-weird. They would confuse and put off a typical browser-reader, needlessly. Let's use more ordinary language. Can't we just say here in the lede opening that it is a number ranging from 1 down to -1?

I fear that it is also unfortunate to have so much coverage of PF less than zero. I have no reason to doubt the truth of this, but over many decades of applying the concept of PF in real life, I have never encountered this aspect. I assume it is significant enough to deserve mention in the lede, but it seems to me that it muddies the concept not to clearly partition this more theoretical/odd variation of the concept.

I think the lede should more clearly present the basic idea by starting with the standard reference point: PF is one for ordinary plainest pure passive resistive loads. PF goes down to zero for somewhat inductive or capacitive loads, and would be zero for ideal pure such loads (which may not exist in real world).


 * "Linear loads with low power factor (such as induction motors) can be corrected with a passive network of capacitors or inductors."

This introduction to PF correction could be more clear/helpful by rewording in a way that better presents basic complementary balancing correction: inductive loads can be balanced by adding capacitance, and vice versa. This may be "obvious" to people who understand the concept, but would not be to newcomers, and is one of the keys to building an understanding of the basics.

(As a newcomer to this article, looking only at the lede, I am thinking that there are at least two aspects here that I don't know about: what are the tech math tools for telling the difference between leading/lagging cap/ind PFs that might both be say 0.5? If loads that supply power can drive the PF negative, why would the number not be able to go more negative than -1? Maybe the answers are awaiting in the rest of the article...)-71.174.175.150 (talk) 14:49, 3 February 2017 (UTC)


 * I completely agree. Please be bold and fix this. Kendall-K1 (talk) 16:13, 3 February 2017 (UTC)


 * Be Bold and fix it. Negative power factor is the greatest Wikipedia invention since the ceramic gas. --Wtshymanski (talk) 19:44, 11 April 2017 (UTC)


 * Before you follow the stunningly bad advice ("Negative power factor is the greatest Wikipedia invention since the ceramic gas.") above, please read Talk:Power factor/Archive 3 Key quote: "You might want to read http://powerstandards.com/Shymanski/draft.pdf . Phase angles (leading and lagging) in the range of -90° to 0° and +90° to 0° have a positive power factor. Phase angles in the range of -90° to 180° and +90° to 180° have a negative power factor." Note the filename; that document was written by Alex McEachern, in direct response to Wtshymanski's repeated attempts to introduce pseudoscience to this page. Alex McEachern is not just an IEEE member, but rather is an IEEE Fellow (awarded for "contributions to power quality measurement"), and is the co-author of IEEE 519, IEEE 1159, SEMI F47, IEC 61000-4-30, IEC 61000-4-34, IEC 61000-4-11 and, most notably for the purposes of this discussion, co-author of IEEE 1459. He is also the author of the Handbook of Power Signatures (generally considered to be the definitive reference on the subject) author of the Electric Power Measurements chapter of the Encyclopedia of Electrical and Electronics Engineering, and chairman of the International Electrotechnical Commission (IEC) TC77A Working Group 9, which sets the standards for power quality instruments. He is also the former Chairman of IEEE 1159.1 and a voting member of the IEEE Standards Coordination Committee on Power Quality. Also, he holds 30 U.S. patents in the area of AC Power, and is the owner of the alex.mceachern.com and powerstandards.com web sites.


 * That being said, I do agree that the lead can be improved a lot. Much of it is simply there to counter Wtshymanski's disruptive editing. --Guy Macon (talk) 04:44, 12 April 2017 (UTC)


 * Apart from attempting to resurrect the long dead and defeated "there is no such thing as negative power factor" argument, our argumentative friend has just made an edit giving a perfect example of negative power factor. 86.153.129.112 (talk) 12:27, 12 April 2017 (UTC)


 * That same edit makes the incorrect assertion that transformers are not loads. This is only true for ideal transformers, which I can't even find in the Digikey catalog. I'm tempted to revert but maybe some other inductive load that is intended to draw power, like a fluorescent light ballast, should be used instead. Also, just to be clear, I do not dispute the existence of negative power factor as a useful concept, I just think it's too much detail for the lead. Kendall-K1 (talk) 00:32, 13 April 2017 (UTC)


 * Even if you assume that the transformer is perfect, if there is a load on the other side the AC source sees a load. --Guy Macon (talk) 01:33, 13 April 2017 (UTC)
 * Yes but if you read the text it pretty clearly is talking about the transformer, not what is connected to it. "Inductive loads such as transformers and motors (any type of wound coil) consume reactive power with current waveform lagging the voltage." Kendall-K1 (talk) 01:46, 13 April 2017 (UTC)


 * So, if I start with an ideal transformer (no no winding losses, no hysteresis, no eddy current losses, etc.) connect it to an ideal resistance on the secondary with perfect impedance matching, and drive the primary with an ideal sinusoidal AC power supply (again, with perfect impedance matching), will the phase angle between voltage and current in the primary be leading, lagging, or in phase? Would the combination of the perfect transformer and the perfect resistor on the secondary be a resistive or inductive load? (BTW, I just Googled this, and found that "transformers are things that transform, usually from cars to giant robots".) :) --Guy Macon (talk) 06:12, 13 April 2017 (UTC)
 * BTW, here is a photo of William Stanley's first transformer, built in 1885: http://www.edisontechcenter.org/Transformers.html --Guy Macon (talk) 06:17, 13 April 2017 (UTC)


 * If the transformer is ideal, then it would appear to the source as a resistor when loaded with a resistor. The resistance presented is the secondary load resistor multiplied by the turns ratio squared. However, real world transformers are not ideal and a combination of factors means that the load presented to the source will have a power factor a little less than unity (usually lagging due to the leakage inductance of the windings). 86.153.129.112 (talk) 16:58, 13 April 2017 (UTC)


 * The way the text was originally written, it implied that a transformer was a load in its own right. I have to support Wtshymanski's removal of it as an example. It should be noted that: if you load the secondary of the transformer with a capacitive load, then the combination of transformer and load will draw power at a leading power factor. I do agree that it might be useful to substitute an alternative example of an inductive load. I'm not sure that a fluorescent ballast is a good choice because most fluorescent systems are power factor corrected (and in Europe it is mandatory). 86.153.129.112 (talk) 15:54, 13 April 2017 (UTC)
 * A real (non-ideal) transformer is a load in its own right, that's my point. But not a good example, I agree.
 * Fluorescent ballasts (real ones) are obsolete and easily confused with their electronic replacements so I agree it's not a good choice. I only mention it because we do have a list of examples elsewhere in the article: "electric motors, solenoid valves, transformers, fluorescent lamp ballasts, and others". Kendall-K1 (talk) 18:47, 13 April 2017 (UTC)
 * I don't know about you, but I am not in the habit of plugging in transformers that are otherwise unloaded for the sake of it.;-) 86.153.129.112 (talk) 13:13, 14 April 2017 (UTC)

This seems wrong!
Says: "A power factor of less than one means that the voltage and current waveforms are not in phase..."

Proposed:

"A power factor ≠ ±1.0 means that the voltage and current waveforms are not in phase..."

Burt Harris (talk) 17:47, 8 June 2017 (UTC)

Wouldn't it also make sense to define power factor as the cosine of the phase difference between voltage and current in a circuit? This explains why negative power factors are possible, and why for a power factor, 1 == -1 The relationships to power (real and reactive) is secondary. — Preceding unsigned comment added by Burt Harris (talk • contribs) 18:17, 8 June 2017 (UTC)


 * At Electric power it says "The ratio of real power to apparent power is called power factor and is a number always between 0 and 1." This appears to contradict what we say here. Kendall-K1 (talk) 11:55, 3 August 2017 (UTC)
 * Fix it!.--Wtshymanski (talk) 16:15, 3 August 2017 (UTC)

What's lurking behind your elecric meter?
The article originally contained the meaningless sentence, "However if there are high-resistance lines connected between the utility meter and a relatively high load then resistive transmission power loss behind the meter can be reduced by a small amount with power factor correction" (my emphasis). I initially changed the word 'behind' to 'after' (as in: on the consumer side of the meter). Another editor (Wtshymanski) reverted it back to the meaningless form with a comment, "talking about losses on the utility side of the meter here". Without thinking, I changed 'behind' to 'prior to' trying to clarify what was actually meant by 'behind'. As I walked away, I was troubled that it couldn't possibly be right and that my original change was correct. Any resistance in the wires feeding power to my house cannot make any difference. If I were to install power factor correction, the utility company may make some small reduction in I2R losses, but my electric meter, as you say, would not register it.

The sentence started as vague, and I had contemplated deleting it forcing any restorer to provide a reference to support it. 86.149.143.134 (talk) 17:30, 13 October 2017 (UTC)


 * What do you think of the wording I just put in? --Guy Macon (talk) 19:19, 13 October 2017 (UTC)


 * Certainly clearer than the original wording. 86.149.143.134 (talk) 11:14, 14 October 2017 (UTC)

θ and φ
Cosinus phi is meant to be spelled with a Greek letter "phi"; φ, not "th"; θ. — Preceding unsigned comment added by 217.122.166.57 (talk) 16:41, 20 December 2017 (UTC)