Split phase service--one or two?

[mivey]

Actually, it is not a logical choice, because this is a 120/240 volt system, and your 240 volt loads do not share this common reference point.

It is probably the most common reference in the world.

 

[mivey]

But I don't think that there is any doubt in anyone's mind that what is physically happening in our all too familiar center tapped transformer is an inversion, not a time based phase shift (i.e., delay).

A phase shift does not have to include an actual time shift. A phase shift and a time shift can also be related through the reference frames.

For example, in a three-phase generator all of the voltages start at the same time so they all have the same t0. But from a different reference frame, the times of the positive peak (tpeak) will be different for the voltages. With the three-phase generator, we produce a phase shift by a physical shift.

You can get a phase shift by using transformer voltages from different terminals and in different winding directions but that does not produce a real time shift for t0 in one reference frame. But, if we look at the positive peak times (tpeak) for those phase-shifted voltages, the peaks occur at different times.

Another way to get a phase shift is to time delay the waveform. You could do this using delay boxes or by a propagation delay. This time delay is the phase shift you are thinking about, but is not the phase shift we normally talk about when discussing transformer phase shifts. Outside of the shift from the flux delay (which we routinely ignore), the transformers do not have that kind of time shift.


Time shifts are just one way to create a phase shift.

A time shift (time delay, etc.) changes the time relationship between one wave and a reference wave. This will also affect the physical relationship between this shifted wave and a reference wave. Source side noise will appear at different points in time, but at the same place on the waves. If we change the reference time for each wave to be the positive slope zero crossing of each wave, the noise will appear at the same relative points in time.

This is not the type of phase shift we are usually concerned with in transformers and generators.


Physical shifts are another way to create a phase shift.

A physical shift (taking voltages from different terminals and different directions, etc.) changes the physical relationship between one wave and a reference wave. This will also affect the timing relationship between this shifted wave and a reference wave. Source side noise will appear at the same point in time, but at different places on the waves. If we change the reference time for the positive slope zero crossing of each wave to be the positive slope zero crossing of the reference wave, the noise will appear at different relative points in time.

This is the type of phase shift we usually are concerned with in transformers and generators.

 

[mivey]

It's the same reason why they scurried into the shadows over the noise analogy. It reveals that the phase shift is apparent and mathematical, and not physical. Two cross sections cut in each of the windings will show that the electrons are flowing in the same direction, regardless of their chosen points of reference.

No one scurried away from a noise analogy. In fact, I provided two examples using noise analogies. The scurrying was done by those who chose to ignore the fact that a phase shift does not have to be produced by a time shift. Dozens of references were provided illustrating this fact and to ignore the fact is denialism.

 

[mivey]

As I pointed out, many in audio commonly use "180 degrees out of phase" to describe a signal inversion, as well. Although we all know what they mean by it, that doesn't make it correct.

Neither does the use make it incorrect. It is an industry-recognized use. I provided pages of references illustrating that fact.

You may continue treating it however you wish but it does not change the way a transformer works.

Exactly. And we commonly use transformers to produce phase shifts that are not the result of time shifts. These phase shifts are the results of physical shifts created by taking voltages from different terminals and in different directions in the windings.

 

[mivey]

So how many phases are there in a 120/208/240 high leg open delta?

Well, depending how one treats the split phase transformer, one can count 5 phasors with 5 phase angles. Technically that is 5 phases. Even if one of the transformers is omitted, its L-L voltage is still present. We still call it a three phase service though.

240 volt sources: 3 phases max at any given time
120 volt sources: 2 phases max at any given time
208 volt source: 1 phase max at any given time

So we have up to 6 phases at any given time, but they are not all part of the same system of phases (same frequency, same voltage magnitude, terminal pairs used only once in the count). At most, we have a 3-phase system because of the three 240 volt sources making up a three-phase system of voltages (which could also be used as three (3) single-phase sources).

The transformer is a source for multiple systems.

 

[jim dungar]

Mivey,
It looks like you are getting argumentative by misrepresenting the context of my posts rather than addressing any actual inaccuracies.

With a center-tap transformer we have more than two terminals thus more than two voltages.

Your reply was to my specific example of a single voltage across two terminals.

You are not using correct notation. By sticking an angle on the end of the name, you are mixing the name and value. It is like saying the "direction from x to y in the north direction is in the north direction".

The 'name' of a voltage, Vxy, does not contain any directionality except in the sense of "from node X - to node Y". When paired with an angle, it becomes simply an identifier which is then used to as a means to differentiate one voltage from another voltage with the same angle.

Also, while a pair of conjugate angles adds to 360?, mathematical conjugation is something completely different.

I was talking about the conjugate of an angle, so I am not sure what point you are trying to raise.

 

[Rick Christopherson]

No one scurried away from a noise analogy. In fact, I provided two examples using noise analogies. The scurrying was done by those who chose to ignore the fact that a phase shift does not have to be produced by a time shift. Dozens of references were provided illustrating this fact and to ignore the fact is denialism.

Actually, every one of you did scurry away, giving one reason or another that it wasn't applicable.

I'm not going to go digging back for your post, but I am pretty sure you did not provide any references that a phase shift can occur without a time shift, as you claim. If you did provide some sort of reference, then I can assure you that it was discussing a mathematical phase shift, not a real phase shift. And since that is the topic at-hand, any such reference would be circular and useless.

 

[mivey]

Mivey,
It looks like you are getting argumentative by misrepresenting the context of my posts rather than addressing any actual inaccuracies.

Really? Is it time to resort to petty accusations? How about we try to keep the discussion on point instead of going down the path of taking jabs at each other?

Your reply was to my specific example of a single voltage across two terminals.

And your reply was in reply to my discussion about the original topic: the voltages across three terminals.

The 'name' of a voltage, Vxy, does not contain any directionality except in the sense of "from node X - to node Y". When paired with an angle, it becomes simply an identifier which is then used to as a means to differentiate one voltage from another voltage with the same angle.

You are still mixing the name with the value as I noted.

I was talking about the conjugate of an angle, so I am not sure what point you are trying to raise.

It appeared you were referring to the "conjugate" of the phasor. Perhaps I misunderstood your point, but the conjugation of a complex value is not taking the negative of the angle of the complex value; rather it is taking the negative of the second term (or the imaginary term for complex numbers).

 

[mivey]

Actually, every one of you did scurry away, giving one reason or another that it wasn't applicable.

I'm not going to go digging back for your post, but I am pretty sure you did not provide any references that a phase shift can occur without a time shift, as you claim. If you did provide some sort of reference, then I can assure you that it was discussing a mathematical phase shift, not a real phase shift. And since that is the topic at-hand, any such reference would be circular and useless.

The reason given for it not being applicable is because a phase shift does not require a time shift. This was pointed out with many references and definitions of phase shift where time shift is not required.

Your definition of a real phase shift was one that was caused by a time shift. You said you did not really care what the industry considered to be a phase shift but chose to stick with the rule that a time shift was required. That is perfectly fine if you want to do that. I also agreed that a phase shift that requires a time shift is not the phase shifts we have in transformers. I also agreed that for a discussion between you & I concerning time shifts, that I could even agree to use that definition. There are cases where we discuss phase shifts that are only caused by time shifts.

However, with transformer phase shifts, the rest of the industry uses a different definition of phase shift and that is the phase shift we are discussing. The phase shifts we are discussing with transformers does not require a time shift.

 

[Rick Christopherson]

The phase shifts we are discussing with transformers does not require a time shift.

That's because they are only mathematical. You are using circular logic for an argument. Find a single reference that says you can have a phase shift that is not a time shift, that is not referring to the mathematical transformation of an inversion into a phase shift. You can't and you never have!

 

[jim dungar]

Really? Is it time to resort to petty accusations? How about we try to keep the discussion on point instead of going down the path of taking jabs at each other?

You have already crossed that line. You accuse me of not being able to apply subscripts correctly. You infer that I do not understand phasors and their conjugates.
I have no respect for your 'discussion' practices.

 

[mivey]

You accuse me of not being able to apply subscripts correctly.

I did not say you were not able. I said what you posted was incorrect notation.

You infer that I do not understand phasors and their conjugates.

You were the one who mentioned phasors and their conjugates and gave an example of with a negative applied to the angle. You were making some point about how conjugation and inversion were different. Well of course they are different but who was conjugating the complex number? Your point was not clear and I said so.

Whether you understand it or not has nothing to do with whether or not what you were saying was clear. Somehow you were getting two different numbers for your results and I don't see how that is possible because an inversion and 180? phase shift yield the same number. Infer what you want but your use of terminology was not clear.

 

[mivey]

You have already crossed that line.

I see it differently. I questioned your post that I believed to be in error. I stated why I thought your methodology was incorrect.

You questioned my integrity by accusing me of being deceptive and argumentative. Not quite the same thing.

I have no respect for your 'discussion' practices.

Get over yourself.

 

[mivey]

That's because they are only mathematical. You are using circular logic for an argument. Find a single reference that says you can have a phase shift that is not a time shift, that is not referring to the mathematical transformation of an inversion into a phase shift. You can't and you never have!

I can and have provided dozens of references. Here is just one:

Westinghouse Distribution Transformer Guide:

2. Angular Displacement (Phase Shift)
For standard three-phase connections the phase-to-neutral voltage on the primary side either leads that on the secondary by 30? or is in phase with the phase-to-neutral voltage on the secondary side.
...
The delta wye and wye delta connections produce the 30? phase shift.

This is known as a phase shift from transformer primary to secondary. If you look at the primary and secondary waveforms with noise on the primary, you will see that it is not a time shift.

As a bonus for you: connection diagrams in the reference are shown for both a 30? and a 210? phase shift. Guess what makes the difference in phase shift?

 

[Rick Christopherson]

I can and have provided dozens of references. Here is just one:

Westinghouse Distribution Transformer Guide:

What does that reference have to do with your assertion that a non-mathematical phase shift is not a time shift?

I'm in full agreement with Jim regarding your discussion techniques.

 

[mivey]

What does that reference have to do with your assertion that a non-mathematical phase shift is not a time shift?

What does your assertion that a phase shift must be the result of a time shift have to do with anything? Clearly, our industry also recognizes phase shifts that are not the result of time shifts. Whether or not you agree with that is your own personal preference.

 

[mivey]

I'm in full agreement with Jim regarding your discussion techniques.

Well there's a shocker.:roll: But considering your historic technique, I'll just consider the source.

 

[Rick Christopherson]

What does your assertion that a phase shift must be the result of a time shift have to do with anything? Clearly, our industry also recognizes phase shifts that are not the result of time shifts. Whether or not you agree with that is your own personal preference.

A minute ago you claimed you could, and have, produced dozens of citations. You haven't produced a single one. And now you want to scurry away by diverting the topic.

Produce a single reference that backs your assertion that a non-mathematical phase shift does not require a time shift.

 

[mivey]

A minute ago you claimed you could, and have, produced dozens of citations. You haven't produced a single one. And now you want to scurry away by diverting the topic.

Produce a single reference that backs your assertion that a non-mathematical phase shift does not require a time shift.

You specifically asked for:

Find a single reference that says you can have a phase shift that is not a time shift, that is not referring to the mathematical transformation of an inversion into a phase shift.

and I produced a reference that says you can have a phase shift that is not a time shift that was not referring to the mathematical transformation of an inversion.

Now that you got what you asked for, I suppose you want to pose a different question.


It does not make any difference anyway as the answer you seek is the same one that I have been giving you all along: A phase shift does not have to be the result of a time shift. I have shown this through dozens of references. You do not happen to agree with that industry definition. That is your own personal preference and does not change the phase shifts we are discussing.

If you want to discuss time shifts, those are not the shifts we are normally concerned with in transformers.

 

[Rick Christopherson]

...and I produced a reference that says you can have a phase shift that is not a time shift that was not referring to the mathematical transformation of an inversion.

No you didn't. Where in your reference does it support your assertion?