Is it Single or Two Phase?

[winnie]

Winnie,

First of all, phase shift is only defined for pure sinusoids, so any other waveform has no bearing on this discussion.

Now, if you can't tell the difference, why is it not a phase shift or phase difference if you like that term better?

I do not believe that it is appropriate to limit the discussion to pure sinusoids; the discussion started out with a _real_ transformer, and IMHO that is the proper basis and limitation of the discussion. Real transformer outputs can include harmonic components, in which case you most certainly can distinguish the inverted signal from the phase shifted signal.

We can _approximate_ the output of the transformer by saying that it is purely sinusoidal. Once we make that approximation, then we can perform calculations on the output that presume a 180 degree phase shift, and the calculations will be as correct as the initial approximation.

I belive that my previous comments have shown a strong preference for using the sinusoidal approximation and treating the output as though it has has a 180 degree phase displacement.

I am saying that this 180 degree phase displacement is a convenient, useful, and accurate approximation, but not reality. The only reason that I am saying this is that the transformer is a single magnetic core, and the output terminals are in fact inverted relative to each other.

If you add the basis that phase shift can only apply to pure sinusoids, then we can no longer use the concept of phase shift in this discussion, because a real transformer will only have an approximately sinusoidal output!

As an aside, I see no difficulty with vectors having a negative magnitude. When you break a vector down into basis components, negative magnitudes are used all the time.

-Jon

 

[kingpb]

The magnitude of a vector is an absolute number. Keeping with mathmatical rules, an absoulte number is always positive. When you combine the magnitude with the angle, it gives you direction. Yes, you could say that -1@ 180deg is equivalent to +1 @ 0deg. Therefore, applying a negative value, but why not just say +1 @ 0deg.

The negative sign is commonly used as a mathmatical operator that is necessary for combining vectors, but it does not change the fact that the magnitude is an absolute value, and therefore always positive.

For those that have not had the math training, The concept of a phasor, or vector notation should not be confused with using "j" or "i" as found in a complex number, which is notation used to define the sqrt of -1. Complex numbers are easily converted to phasors, and when done so, will always have a positive magnitude, and some direction.

Rattus, is correct on this!

 

[mpross]

phase shift

phase shift

Is this correct?

A Phase shift can be applied to any signal, it depends on what goes on in the system. Either you have a delay, or an advance, and I think it depends on causality.

-Matt

 

[rattus]

Nonsense Winnie:

Nonsense Winnie:

Winnie,

The argument is whether the fundamental voltages on L1 and L2 exhibit a 180 degree phase shift. Usually these voltages are essentailly pure sinusoids, and even if they are not, we assume they are.

There is simply no justification in denying the obvious: There is a 180 degree phase shift between V1n and V2n in a 1-ph system--just like the 120 degree phase difference in a 3-phase system.

So far, I have eight votes saying I am right. I would have nine, but my old AC Circuits prof died a few months back, but I can get more.

Now let me say this without offending anyone: If you have not struggled through an AC Circuits course, you may not understand what I and others are saying. That does not make you a bad person though.

Now if anyone can prove me wrong, I will apologize to all and shut up.

 

[mpross]

180 degree phase shift

180 degree phase shift

rattus,

count me as #9

After following all of the posts, I am finally on board.

I agree this whole thing is explained with the mathematics. When you have a sine wave that is shifted 180 degrees with respect to another, their values are opposite to each other (instantaneously).

I kept thinking that there was something going on physically with the xfmr to explain the shift, and this is where I was wrong, among other areas One sine wave with respect to the neut, and another with respect to the neut, yep thats it!

Thank you for helping me out!

-Matt

 

[winnie]

Rattus,

You can essentially count me as #10. We are at least 95% in agreement, and the rest is almost in the noise of imperfect terminology and slightly different basis of discussion.

The argument is whether the fundamental voltages on L1 and L2 exhibit a 180 degree phase shift. Usually these voltages are essentailly pure sinusoids, and even if they are not, we assume they are.

As I have said above, if we assume that these are pure sinusoids, then I totally and completely agree that the N to L2 voltage is a 180 degree phase shifted sinusoid relative to the N to L1 voltage.

The only difference that I can see in what I am saying versus what you are saying is that I claim that if you look at the way that the output is _not_ a pure sinusoid, then an _inverted_ output can be differentiated from a phase shifted output.

Seems to me that this is at the level of a terminology difference: can we even talk about a phase shifted composite waveform? Do we even have to bother saying that we are approximating this single phase system as a sinusoidal system? You claim a 180 degree phase shift; I claim that the output of a single transformer core with a single coil is _inverted_, but that in the limit of a sinusoidal approximation that this is the equivalent of a 180 degree phase shift.

-Jon

 

[bcorbin]

Actually, Rattus....I never struggled through my circuits classes. I flew through them. And it really doesn't matter how many "votes" you get, your way of approaching this problem is what gets people hurt. When you use shortcuts in terminology, methodology, whatever...it leads to a lack of understanding for those who follow you, and they may not take the time to learn that for anything but a perfect sine wave, the inverse does not equal a 180 degree phase shift. I hope you're still patting yourself on the back when one of them gets fried.

 

[rattus]

Where is the reference?

Where is the reference?

Actually, Rattus....I never struggled through my circuits classes. I flew through them. And it really doesn't matter how many "votes" you get, your way of approaching this problem is what gets people hurt. When you use shortcuts in terminology, methodology, whatever...it leads to a lack of understanding for those who follow you, and they may not take the time to learn that for anything but a perfect sine wave, the inverse does not equal a 180 degree phase shift. I hope you're still patting yourself on the back when one of them gets fried.

BC, you still have not justified your position. You are merely putting up a smokescreen. Your hint that the standard way of describing a phasor may cause an injury is totally unfounded.

AC circuits are routinely analyzed assuming perfect sinusoids, and now you admit that the inverse amounts to a 180 degree phase shift.

You and Winnie need to get off that kick and understand that any phasor assumes an ideal waveform.

Now put out the fires and support your argument with a reference or some logic. Until you do, I must assume you know you are wrong and are just stonewalling.

 

[mpross]

Question

Question

Could someone explain why it has to be a perfect sine wave?

I think it would depend more on the period of the two signals, and their amplitudes... ...so you would get like 239_V, or maybe 241_V in cases where the wave is not pure. Also, these two signals both have the same frequency, so the mathematical 180 degree shift would never bother. Am I way off here ?

I aced my circuits courses by the way! Piece of cake (carrot)

-Matt

 

[jim dungar]

Rattus,

Your argument is valid only if you are discussing L1-N and L2-N references. there is no 180 difference if you are discussing loads consisting of (a series connections of L1-N and N-L2) in parallel with a L1-L2 load fed from a center tap single winding transformer.

Circuit analysis should be the consistent regardless of the point of reference. If I chose L2 as my reference point is there still a "phase difference" between L2-N and L1-N voltages?

Every circuit book I own (granted they are all over 30 years old) discusses the ADDITION of phasors. Every problem is solved using addition, and of course when applying Kirchoff's current law sometimes we are adding a "negative" phasor value (which seems to be your rational for calling it subtraction), but they never ever call it a phase shift or difference.

 

[rattus]

Back to Square One:

Back to Square One:

Rattus,

Your argument is valid only if you are discussing L1-N and L2-N references. there is no 180 difference if you are discussing loads consisting of (a series connections of L1-N and N-L2) in parallel with a L1-L2 load fed from a center tap single winding transformer.

Circuit analysis should be the consistent regardless of the point of reference. If I chose L2 as my reference point is there still a "phase difference" between L2-N and L1-N voltages?

Every circuit book I own (granted they are all over 30 years old) discusses the ADDITION of phasors. Every problem is solved using addition, and of course when applying Kirchoff's current law sometimes we are adding a "negative" phasor value (which seems to be your rational for calling it subtraction), but they never ever call it a phase shift or difference.

Jim,

We are using the neutral as a reference which is standard practice in any system with a neutral. That should be clear by now. It is just like using the neutral as reference in a 3-ph wye. This is the consistency you want!

Yes, one adds the currents into the neutral node to obtain the neutral current. No argument there.

Now, I challenge you to compute the line-to-line voltage in a 3-ph wye by adding phasors. Larry said he would, but he hasn't. No one can! Remember now, a phasor is described by a magnitude and angle. No cheating!

My textbooks are 50 years old, and they say you add the negative of one voltage to the other to obtain the line voltage. But that is tantamount to subtraction which we learned in high school.

 

[bcorbin]

If you want to try to pull technicalities on me, we can all play that game. Phasors are not strictly pure sine waves. They are typically RMS values, which in no way dictates they be pure sine waves. A Fourier series (which includes every waveform you?re ever going to see in the real world, including the single phase of the ORIGINAL POSTER?S 240V service) can also be 240V, with a phase angle. You can keep harping on me for ?not providing references? all day long, but that won?t make you right. My knowledge of hardware may be suspect, but this theory stuff is basic to me; you would hardly feel obliged to provide references that 2 + 2 = 4.

Rattus, I?m just asking you to help disseminate the ?general solution? approach, which contains all of the answers all of the time, not one answer under very special circumstances. Are you probably going to get the correct answer when you assume they?re the same thing? Hell yes?..but just because I know how to order a cheese omelette in Paris doesn?t mean I speak French. It means I?m imitating someone who speaks French in order to crudely arrive my desired result.

 

[rattus]

BC,

You cannot attach a phase angle to a complex wave. BTW, inverting a complex wave shifts the phase by 180 degrees for each harmonic. Of course the periods are different.

Phase angles can only be applied to a pure sinusoid. This means that a phasor comprises the magnitude (RMS) of a pure sinusoid and its phase angle.

Engineering is full of approximations. The trick is to know how far to approximate. We know the waveforms are not perfect, but they are good enough to call them phasors.

Now if I do things my way, do I get the correct answer?

BTW, I picked up two more votes today.

 

[johnny watt]

I was under the impression that phasor notation used peak values rather than RMS values.

I do not understand why phase shift could not be used to describe any periodic wave.
Just because in theory the Fourier series harmonics would not be shifted 180 degrees with respect to their own frequency they would still be shifted 180 degrees with respect to the period of the wave as a whole.

 

[rattus]

I was under the impression that phasor notation used peak values rather than RMS values.

I do not understand why phase shift could not be used to describe any periodic wave.
Just because in theory the Fourier series harmonics would not be shifted 180 degrees with respect to their own frequency they would still be shifted 180 degrees with respect to the period of the wave as a whole.

Johnny,

Rotating phasors are functions of time and use peak values. They clearly represent purely sinusoidal saves. e.g.,

v(t) = 170[cos(wt) + jsin(wt)]

This same signal, a pure sinusoid, may be represented as a fixed phasor, i.e.,

V = 120 @ 0

The fact that real world waveforms may not be quite ideal has no bearing on the discussion. If you attach a phase angle, then the waveform must be a pure sinusoid!

I think BC and Jim are hung up on the fact that we are discussing a 1-ph system and won't accept the fact that a phase shift can be observed.

 

[rattus]

Complex Waves:

Complex Waves:

Johnny,

You answered your own question. If we shift the fundamental wave by 10 degrees, the second harmonic would be shifted 20 degrees, etc. A single phase angle cannot describe the shifts of all the harmonics.

I know this notation is often used, but strictly speaking it is incorrect. It is also incorrect to assign a negative sign to the magnitude of a phasor.

 

[jim dungar]

Rattus,
The consistency I am looking for has to do with the methodology of solving networks, regardless the point of reference. The base formulae should not change if the reference point is chosen as L1 or L2 instead of N. So if two voltage sources are in series the resultant voltage is the summation of the two individual sources even if it means having to add a negative value. With this basic concept it is much easier to be consistent with the application of Kirchoff's Law for solving 120/240V multi-wire and 2-wire combination circuits.

 

[LarryFine]

Now, I challenge you to compute the line-to-line voltage in a 3-ph wye by adding phasors. Larry said he would, but he hasn't.

I'm still working on it!

My textbooks are 50 years old, and they say you add the negative of one voltage to the other to obtain the line voltage.

That sounds like addition to me.

But that is tantamount to subtraction which we learned in high school.

"Tantamount" is not the same thing as "is." (According to Bill Clinton, anyway)

BTW, inverting a complex wave shifts the phase by 180 degrees for each harmonic.

To me, a 180-degree phase inversion resembles a phase shift, but only a half-cycle delay actually is a shift.

. . . a phasor comprises the magnitude (RMS) of a pure sinusoid and its phase angle.

Aren't they each positive values in their own respective right? (I'm not sure what my point is . . . yet.)

Now if I do things my way, do I get the correct answer?

Does that apply to all of us?

BTW, I picked up two more votes today.

Big whoop!

 

[LarryFine]

Rattus,
The consistency I am looking for has to do with the methodology of solving networks, regardless the point of reference. The base formulae should not change if the reference point is chosen as L1 or L2 instead of N. So if two voltage sources are in series the resultant voltage is the summation of the two individual sources even if it means having to add a negative value. With this basic concept it is much easier to be consistent with the application of Kirchoff's Law for solving 120/240V multi-wire and 2-wire combination circuits.

Yeah, what he said!

If we can add three phasors to equal zero, there has to be a way of adding two of them.

 

[rattus]

Yeah, what he said!

If we can add three phasors to equal zero, there has to be a way of adding two of them.

Yes, there is:

120 @ -30 + 120 @ -150 = 120 @ -90

But,

120 @ -30 - 120 @ -150 = 208 @ 0; (this was wrong and no one caught it)

Oh yes, phase angles can be positive or negative. e.g., -30 = +330

Do it graphically if the math gives you a problem.

Likewise,

120 @ 0 + 120 @ 180 = 0

But,

120 @ 0 - 120 @ 180 = 240 @ 0

Same concept, either case.