May I ask a question about the single vs two phase stuff

[Besoeker3]

180 degrees in direction, as in polarity, yes; 180 degrees in timing, no. There is no timing differential.

Is 8.33 ms not a difference in timing?

 

[GoldDigger]

Is 8.33 ms not a difference in timing?

Yes, that is a apparent difference in timing. But whatever the waveform going into the transformer, the two output waveforms are always the polarity inverse of each other. It is only if the input waveform meets certain limiting conditions (like being a pure sine wave of a single constant frequency) that it appears to also be describable as an 8.33ms delay.

 

[winnie]

With a pure single frequency sine wave, an inversion and a half cycle time displacement are indistinguishable and equivalent. It is meaningless to say there isn't a time displacement, or only the 'appearance' of a time displacement, because there are completely equivalent.

Once you deal with multiple frequencies, then the underlying mechanism becomes relevant, and an inversion becomes different from a displacement in time.

-Jon

 

[jaggedben]

Let me pile on! Oh, wait, I already (crudely) illustrated the point back in post 946.

(And yes, the point applies to three phase, even.)

 

[jaggedben]

A case of an actual time shift might be, say, a damaged stator on a three phase generator?

 

[mivey]

But of course it does. The L2 voltage is a reflection about the X axis of the L1 voltage. It's the general case that covers any and all waveforms, and it illustrates how the two waveforms in question are actually created. L2 is most definitely not created by delaying (phase shifting) L1.

He is not saying the signal is created by delaying a different signal.

Boty signals start at the same time. One goes positive when the other goes negative. The positive of one is 180d after the positive of the other i.e., the positives occur at different times.

 

[mivey]

A case of an actual time shift might be, say, a damaged stator on a three phase generator?

What time shift?

A three phase set of voltages don't shift time either. All voltages start at the same time.

There is a difference as to when the positive of each signal peaks.

As with any signal, that will depend on what you use for a reference. Basic physics tells you that you can pick any reference.

In one case we use two different references and the other uses one reference. Both choices are valid.

 

[mivey]

Yes, that is a apparent difference in timing. But whatever the waveform going into the transformer, the two output waveforms are always the polarity inverse of each other. It is only if the input waveform meets certain limiting conditions (like being a pure sine wave of a single constant frequency) that it appears to also be describable as an 8.33ms delay.

Reality, not apparent. A reference is a choice, not something determined by someone else's opinion.

A transformer, three-phase or otherwise, can have multiple valid references and choosing one over another does not suddenly make all other signals "apparent" signals.

 

[jaggedben]

What time shift?

A three phase set of voltages don't shift time either. All voltages start at the same time.

There is a difference as to when the positive of each signal peaks.

As with any signal, that will depend on what you use for a reference. Basic physics tells you that you can pick any reference.

In one case we use two different references and the other uses one reference. Both choices are valid.

I think my hypothetical example of a bad stator was wrongheaded with respect to what I was talking about. I was trying to think of a physical defect in a real-world generator that could create a repeating distortion in the waveform of each phase that is in fact time shifted. Perhaps there is no such plausible defect.

(Going back to post 946 again might help make it clear what I'm trying to ask about. Is there a real-world phenomenon that could create something like the second graph?)

 

[jaggedben]

Reality, not apparent. A reference is a choice, not something determined by someone else's opinion.

A transformer, three-phase or otherwise, can have multiple valid references and choosing one over another does not suddenly make all other signals "apparent" signals.

Reality or perception, Golddigger's point is that with a center-tapped transformer the time shift interpretation can only be valid with certain ideal waves. And since real world waves are never completely ideal...

 

[mivey]

Reality or perception, Golddigger's point is that with a center-tapped transformer the time shift interpretation can only be valid with certain ideal waves. And since real world waves are never completely ideal...

Wind three equal coils along a generator shaft. Pick equivalent reference points and you get three equal voltages.

Rotate the second coil 120 degrees relative to the first and rotate the third coil 240 degrees relative to the first then you get a set of three voltages with a 120 degree separation. Our reference terminals are the same as when we started.

Similarly,

Wind two equal coils along a generator shaft. Pick equivalent reference points and you get two equal voltages.

Rotate the second coil 180 degrees relative to the first then you get a set of two voltages with a 180 degree separation. Our reference terminals are the same as when we started.

Nothing "apparent" is happening at all. What happens is a physical reality. A reference defines the signal and you are free to assign a reference. Plain and simple physics.

You can have a transformer bank with reference points at the end and in the center and it produces real working voltages. Been there, done that.

You can take a voltage in one direction for 1/2 the winding and the opposite direction for the other 1/2 of the winding and it produces real working voltages and can even create other phase angles. Been there, done that.

Nothing "apparent" about it. Take it from people who actually do these things in their work.

 

[Besoeker3]

Nothing "apparent" is happening at all.

The voice of reason.

 

[wwhitney]

Suppose due to harmonics the L1-N voltage is of the form sin(x) + a*sin(2x) [for some constant a and where I've set other constants to 1 for simplicity]. What's the L2-N voltage for an idealized center-tapped transformer secondary:

1) sin(x + pi) + a*sin(2(x + pi)).
2) - sin(x) - a*sin(2x)

In this case the two answers differ. And (2) is correct, (1) is incorrect.

This post is unrebutted and shows that negation is the more generally correct transformation. Phase shift does not work when any even harmonics are present.

Cheers, Wayne

 

[jaggedben]

Wind three equal coils along a generator shaft. Pick equivalent reference points and you get three equal voltages.

Rotate the second coil 120 degrees relative to the first and rotate the third coil 240 degrees relative to the first then you get a set of three voltages with a 120 degree separation. Our reference terminals are the same as when we started.

Similarly,

Wind two equal coils along a generator shaft. Pick equivalent reference points and you get two equal voltages.

Rotate the second coil 180 degrees relative to the first then you get a set of two voltages with a 180 degree separation. Our reference terminals are the same as when we started.

Nothing "apparent" is happening at all. What happens is a physical reality. A reference defines the signal and you are free to assign a reference. Plain and simple physics.

You can have a transformer bank with reference points at the end and in the center and it produces real working voltages. Been there, done that.

You can take a voltage in one direction for 1/2 the winding and the opposite direction for the other 1/2 of the winding and it produces real working voltages and can even create other phase angles. Been there, done that.

Nothing "apparent" about it. Take it from people who actually do these things in their work.

And your point is...?

The post you were responding to referred to a center tapped transformer, by the way, and not a generator with a rotating shaft.

 

[mivey]

And your point is...?

The post you were responding to referred to a center tapped transformer, by the way, and not a generator with a rotating shaft.

The generator winding example might be easier for people to visualize and picture physically.

In both cases the phase displacement is determined by the reference choice. The signals taken from the generator or the transformer can have a 0 degree displacement or 180 degree displacement. Both have a physical arrangement that determines the phase relationship. Both have a common driving source.

 

[Besoeker3]

In both cases the phase displacement is determined by the reference choice.

For 120-0-120 isn't the neutral the obvious and logical choice?
That makes L1-N and L2-N 180deg apart. This two different phases.

 

[winnie]

For 120-0-120 isn't the neutral the obvious and logical choice?
That makes L1-N and L2-N 180deg apart. This two different phases.

That is two different phase angles. It is entirely reasonable to say two different phase angles from a single phase supply.

We call them 3 phase motors even though there are 6 (or 12) phase bands in the stator.

We call it hexaphase when there are 6 phases 60 degrees apart.

The number of actual phase angles in the system is only weakly related to the phase name that we give it.

-Jon

 

[Besoeker3]

That is two different phase angles. It is entirely reasonable to say two different phase angles from a single phase supply.[/QUOTE. They are not in phase so they are different phases.

 

[jaggedben]

The generator winding example might be easier for people to visualize and picture physically.

In both cases the phase displacement is determined by the reference choice. The signals taken from the generator or the transformer can have a 0 degree displacement or 180 degree displacement. Both have a physical arrangement that determines the phase relationship. Both have a common driving source.

In the case of the transformer, unless that source generates a periodic waveform then there is no utility to the idea of a displacement measured in degrees or radians. That math would not perform any useful description of the wave if the transformer is the one that ggunn linked to in post 336 and its being used for its intended application. (We'll, not unless your playing one note on a synthesizer that went out of style in the 70s.)

 

[jaggedben]

For 120-0-120 isn't the neutral the obvious and logical choice?
That makes L1-N and L2-N 180deg apart. This two different phases.

The last sentence does not follow from the first two.