User talk:AMcEachern

Power factor
I think it's unecessary and even inaccurate to say that power factor ranges between -1 and 1. You have actually 4 quadrants - +/- real power, +/- reactive power. A single sign doesn't accurately convey all four possiblities (is it even meaningful to say that leading reactive power flowing to the load is the negative of lagging power flowing from the load?) Power factor meters I've seen tend to be marked "lead" and "lag" instead of +/-.

I don't even understand what a power factor of -1 means -what is happening in the circuit at -1 power factor that is different from a +1 power factor?

All the definitions of power factor in IEEE Std. 100 say something along the line of the ratio of real power to total power, with no qualifications as to direction.

With respect, citing credentials on Wikipedia is unproductive - we've had some quite amazing credentials claimed that turned out to be inaccurate. --Wtshymanski (talk) 17:23, 14 December 2008 (UTC)

Thanks for the response. (And I truly appreciate your comment about citing credentials; I offered mine to you only to assure you that I'm not a flake, and certainly not as an argument that I'm correct.  Heaven knows, even with my credentials I'm wrong often enough!)

You are absolutely correct: there are 4 quadrants for sinusoidal single-phase situations, and you are correct that they are +/- real power, and +/- reactive power. And we all seem to agree that true power factor is defined as the ratio of real power to apparent power; and we all agree that the definition of apparent power is RMS volts multiplied by RMS amps. Because of the "square" term in root-mean-square, RMS values can only be positive.

So I'm not sure which of the following statements you disagree with: (1) power factor is the ratio of watts to volt-amps; (2) watts can be either positive or negative; (3) volt-amps can only be positive; (4) watts is always less than or equal to volt-amps. In my opinion, those four statements imply that power factor can take any value between -1 and 1. That was my intention the last time I worked on the IEEE 100 definition of power factor, many many years ago, but perhaps we didn't get it clear enough. (Now, if you really want to get confused, join us on IEEE 1459 where we're writing the power definitions for balanced and unbalanced, sinusoidal and non-sinusoidal, three-phase systems - just try to get consensus on the meaning of "VAR" on an unbalanced, non-sinusoidal corner-grounded delta system!)

You ask an interesting and useful question: what is happening in the circuit at -1 power factor that is different from +1 power factor? So let's assume that we're asking this question standing in front of a revenue meter (American terminology - my apologies - but I mean the meter where the electric power company connects to your house). If you have a power factor of +1, then your house is acting like a nice, pure resistive load, and the losses in the grid conductors (as a percentage of the power delivered) are minimized. If you have a power factor of -1, then somewhere inside your house you've got a photovoltaic inverter, or a natural gas generator, or a hybrid vehicle functioning as a generator, and you're pushing power back onto the electric company's grid, and, again, the losses are minimized. A power factor of +0.5 means essentially that your house is drawing twice as much current as would be optimal for the amount of power your house is using. A power factor of -0.5 means that your house's generator, whatever it is, is supplying twice as much current to the grid as would be optimal for the amount of power your house is delivering to the grid. In other words, the distance from "1" in the power factor tells you how much extra current is flowing, and the sign of the power factor tells you the direction the power is flowing.

Of course, the direction is arbitrary; a "-1" just means the power is flowing the opposite direction from what's expected: from the nominal source to the nominal load. If we're looking at a meter on a building that houses a big rotating generator, we usually thing of the nominal source as the building, and the nominal load as the grid. If we're looking at a house, we usually think of the nominal source as the grid, and the nominal load as the building. In either case, it's possible for the power to flow in the opposite-from-expected direction: we can use the generator as a motor, for example, or we can install a micro-grid generator inside the house.

Going back to my historic discussion on your talk page, where I mentioned that in displacement power factor (not true power factor), we sometimes used the sign to indicate leading or lagging, even though that was mathematically incorrect. In that (incorrect) case, there was no difference between a power factor of -1 and a power factor of +1. But nobody does that any more, except maybe some of us old power engineers when we're in a reminiscing mood, and thinking longingly of the days when currents were sinusoidal. As you correctly observe, more modern meters tend to say "lead" and "lag", which -- assuming they are measuring displacement power factor and not true power factor -- is a more accurate description. Of course, true power factor doesn't have a lead or a lag; it's just the ratio of watts to volt-amps, and there's no angle associated with it at all.

The reason I'm putting so much effort into this is that one of my young engineers, a very bright young man named Chao Yang who has joined us from U.C. Berkeley, made an error in the firmware for a new power meter we're designing. He incorrectly took the absolute value of the true power factor before putting it on the meter's display. When I asked him why, he showed me the Wikipedia article that says "true power factor is in the range of 0 to 1." Young engineers take Wikipedia articles seriously, and sometimes even base part of their designs on what they read here. So I want to help correct the errors, if they are errors...

Again, it's quite possible that I'm wrong, and I would be glad (and rather excited) to learn that I'm wrong. Could you review these comments and see if you are persuaded? if you are not, could you educate me? Thanks! Alex McEachern AMcEachern (talk) 03:11, 15 December 2008 (UTC)
 * The references I use (IEEE Std. 100 7th ed., and "Standard Handbook for Electrical Engineers 11th Ed.) don't mention direction of power flow at all in the context of power factor. Neither of these references gives a range of power factor. (Which of the several definitions did you work on? They come from several IEEE standards.) I'd like a reference explaining how to use negative signs with power factor.
 * As for your engineer, I suggest he observes the process by which Wikipedia articles are made and thinks seriously about what that means for credibility; even textbooks can be wrong, and Wikipedia is particularly prone to errors (read Sword-skeleton theory, for example). I wouldn't trust an AWG table on the Wikipedia to be correct, myself.
 * Which power flow (real or reactive) would then determine the sign of the power factor, in your definition? Is a var flowing toward the load the negative of a var flowing back from the load? If I walk up to your power factor instrument and see it showing "-0.5", am I supposed to turn up the field amps or crank up the steam throttle? The purpose of attaching a sign to power factor is to convey information, but to me there seems to be no unique way of deciding what information is meant to be attached.  Relays that protect generator sets are called "reverse power" relays, not "negative power factor" relays. --Wtshymanski (talk) 14:37, 15 December 2008 (UTC)
 * Just a quick response - my apologies for not taking more time. Regarding which power flow (real or reactive) would determine the sign of the power factor:  the definition is pretty clear, I think; it's watts divided by volt-amps.  Watts is real power flow.
 * The reason there's no discussion about the direction of power flow, or the range of power factor, in IEEE 100 is that we thought that it was well understood - the sign of power can be either positive or negative, and the sign of volt-amps is always positive. Maybe we ought to clarify that, but, as far as I know, there really hasn't been any dispute or question until now about either statement.
 * You're absolutely right - the purpose of attaching a sign to power factor is to convey information. The sign conveys whether the real power is flowing in the expected, conventional direction, or in the opposite direction.  This has become a really important piece of information with the development of "green" microgrids, residential photovoltaic arrays, micro wind turbines, etc.  There's nothing wrong with conveying that information with a watts reading (positive or negative watts), or detecting it with a negative power relay - both of those methods are completely correct and justified.  But if one accepts that power factor is the ratio of watts to volt-amps, one must also accept that the information about the direction of real power flow is also conveyed in the sign of the power factor.
 * I appreciate your comments about the reliability of Wikipedia articles - from your Talk page, it's clear that, unlike me, you're an expert on how this thing works (I'm trying to learn from your methods; note that I'm using two colons to indent now, and I learned how to sign things by following your example). But there's hope.  With people like you, who clearly care deeply about keeping these articles as clean and as accurate as possible, it's bound to get better. With best wishes - Alex McEachern AMcEachern (talk) 16:34, 15 December 2008 (UTC)
 * I think the Wikipedia article would be more consistent with usage of the term "power factor" if it left out the -1 to 0 part of the range. Perhaps it should leave out the whole range...if the IEEE can't define it, how can Wikipedia? You shouldn't use a 1-dimensional measure to represent a 2-dimensional reality.
 * Again, if I see -0.5 on the plant power factor meter, do I go buy some capacitors or sweep the snow off the PV array? Or do I have a blown SCR somewhere in the plant?
 * Caring deeply about a Wikipedia article is one road to madness - inevitably some schoolboy will come along and revise an article according to what he understood his lecturer to say in some half-remembered course.
 * IEEE 1459 appears to say (Note 1, section 3.1.1.1) that power flows from a source to a load and cannot be negative.
 * Overloading the "minus" sign on a power factor meter to indicate that the source and load have switched identities is confusing, to me, at least. I'm unable to find anyone making a power factor meter that does this - I would be interested in a reference that points at a power factor instrument that works in this way. The power factor instruments I've looked at this morning say "Lead/Lag" or even "Capacitive/Inductive", none of them said "Import/Export".
 * Now I'm not even sure I understand how a power factor instrument can tell the difference between leading and lagging power factor and reversed power flow. The meter sees potential between two terminals, it sees a current between two others - how does it tell which way (real) power is going? Cos(theta) repeats - cos(45) = cos (-45), how to tell the difference between cos(45+180) and cos(-45-180)? . If I swap any two leads of potential or current circuits on the power factor meter, surely I just change the pointer from one side of "lead-lag" to the other? If source and load are allowed to interchange at will, surely a single instrument can't tell the difference? A generalized power factor instrument would have two indicators, one for lead/lag and one for direction of power flow...I've never seen that. (I've spent a little time trying to get the dots on the CT's connected to the right terminals on the meters...a frequent source of frustration in setting up a new set of meters.)
 * I'm not sure that heroic effort in generalizing the definition of power factor for arbitrary numbers of phases, unbalances and waveshapes is a productive use of human effort. For example, if someone tells me the "power factor" of some arbitrary 6-phase unbalanced inverter system is now 0.5, it's not like I can compare that to the situation where I improve the plant power factor by buying some capacitors for the main distribution - the physics are different and we're blurring together two different kinds of phenomena by trying to shoe horn them into the same measure. Once we get away from how hot the wires are going to get, we lose the usefulness of the concept of a circuit having a unique "power factor" that is comparable over all situations. --Wtshymanski (talk) 18:00, 15 December 2008 (UTC)
 * I have a 1949 edition of the GE "Manual of Electric Instruments: Construction and Operating Principles" which describes a 4-quadrant power factor meter of the polarized vane type - it works like a miniature synchronous motor, with a polyphase rotating field like a motor, energized with the phase voltages, and the rotor magnetized by one selected phase current; the angle of the rotor directly represents the angle between voltage and current. The book describes, but sadly doesn't give a picture of, a 360 degree 4-quadrant scale, said to be useful for tie lines where the source and load ends can interchange. The GE manual says to use a wattmeter and varmeter for any critical measurements, because a single-current-coil power factor meter produces meaningless results in an unbalanced system. --Wtshymanski (talk) 04:14, 16 December 2008 (UTC)

You've got mail
-- Red rose64 (talk) 21:56, 8 June 2013 (UTC)