Center-tap Transformer Voltages

[mivey]

In a course on industrial electronics, a graduate level course, taught by Dr. Joseph A. Boyd, we were assigned the problem to hand sketch the transient change of flux (or flux density if you wish) as a function of time for the above 1/2 wave rectifier circuit with an assumed 0 residual flux density at the time of switch closure on a voltage zero crossing of a sine wave source voltage, and for several cycles after the switch closure. I got it approximately but not totally correct.

If I ever had to do that in a lab, I have long since forgotten.

I am using the free, 2D version of MagNet FEA software...you could easily simulate it here by...

Thanks. I may give it a shot. I've seen some other software to model transformers for supply design but I never took the time to investigate the limits of the models.

 

[rattus]

Doesn't matter where the winding starts. I choose to start at the neutral in my analysis. If one could stand at the neutral and look at the windings, one winding would appear to be clockwise; then turn around and look at the other winding; it would be counter clockwise. Think about it.

Consider a right hand screw. If you look at it from the head, the threads are clockwise; if you look at it from the other end, the threads are counter clockwise. Try it.

Or consider a piece of threaded rod. It is threaded right hand or left hand. Try it.

Forget all that. Head must have been loose. Instead, think of it this way. Consider the point in time when L1 is at its positive peak. That means L2 would be at its negative peak. That is they are 180 degrees out of phase. That is all there is to it.

 

[mivey]

For the last input I have on this subject please click here

Since there are two live conductors in the system, it is sometimes incorrectly referred to as "two-phase".

Not being debated. "Two-phase" is a name reserved for a quadrature system.

The two live or "hot" conductors waveforms are offset by a half-cycle, or 180 degrees offset, when measured against the neutral wire.

Some keep saying this results in zero voltage measured across the ends of the windings. I have yet to figure out why they keep insisting that is the case as we know that is not true.

The two outputs are properly called "legs", not "phases".

Some wiki-contributer's idea of trying to define a new standard. There are too many "standards" out there already to ever reach a consensus on one single use of the word "phase" and adding a different term won't help.

 

[mivey]

Let's say each winding turn adds 1 volt per turn. The secondary winding has 120 turns and 120 volts out. Starting from the winding end, which is either L1 or L2 the winding turns add 1 volt out each in series to the winding center, system grounded point, which yields 120 volt.

As I contiue adding winding turns past the neutral point, what happens. The winding turns continue adding identically in phase, in series. Output voltage progresses 121 volt, 122, 123 volt ..

Voltage fundamentals that say a rise in one direction is a fall in the other. Nothing says we have to go all in one direction.

The voltage is a potential difference between two points. It does no matter if you say "A" is higher than "B" or if you say "B" is lower than "A". The reference point is arbitrary because the voltage is an integral function and has a constant of integration that we can choose to be whatever we want.

If the subsequent winding turns past the neutral point were "180 out of phase" each added turn, the output voltage would go 119 volt, 118, 117 .. In fact, this paradigm fails this test.

Some part of your audience wants to know "what happens" and you have the O O P repeating gobbledegook 240 is "the voltage are 180 out of phase" when he could have been saying, the "the wiindings are in series (and matched)". It makes my ears hurt.

No. This is an AC voltage, not a DC voltage. I'm too tired tonight to feel like repeating it all over again but AC voltage changes direction every 1/2 cycle and thus also has a phase angle. DC does not have a phase angle.

I agree completely and when you get done the voltage between Line #1 and Line #2 would be zero

That is not what is being said. I marvel at why you and others keep going back to that. This is not a DC signal being discussed and the voltages are not being paralleled in phase.

 

[mivey]

The winding does not start at the neutral. It starts at an end, L1 or L2. This is the physical reality of the device as fabricated by the manufacturer. The neutral is at the middle of the winding and the physical turn direction continues homogeneous. The fact of the winding does not reverse.

What does that have to do with the voltage reference?

Reversal comes as an artifact of how the meter leads or loads are attached to the sources. The leads reverse after the output, between the source and the load. Winding turn direction does not reverse.

Again, that has nothing to do with the voltage reference.

 

[mivey]

Consider the point in time when L1 is at its positive peak. That means L2 would be at its negative peak. That is they are 180 degrees out of phase. That is all there is to it.

With the neutral probably being one of the most common references in the world, I find it incredible that so many struggle with this.

 

[mivey]

Most power Transformers we use in the electrical trade have two seperate
secondary windings.

They are marked x1, x2, x3 and x4

You can put them in parallel for twice the current at 120 VAC or in series
for 240 VAC.

The normal series configuration is x1 to line #1 x2 and x3 tie together for the neutral
Line #4 is line #2.

None of that is being debated. If you think it is, you are not understanding what is being said.

If you really reverse one of the secondary windings in the series configuration it will be
180 degrees out of polarity from the other one. Doing this would cause the secondary
to null and produce zero volts between line #1 and line #2.

Also known as the parallel configuration where #1 and #2 get tied together as well. This is not what is being said.

If you reverse it in the parallel configuration it will make a very loud pop and let the
smoke out of it.

No one is talking about changing the connections. The simple fact is that we can take the positive direction to be either way for either 1/2 of the winding. It is a fundamental principle of how AC works.

Like it or not, assigning a positive direction is an assumption that we are applying to the circuit. There is nothing that pre-defines that for us. The physical reality is that the winding voltages do not have a set positive direction for the complete cycle because AC voltages change direction every 1/2 cycle.

Polarity tells us the relative direction of rise or fall at a given instant. It does not tell us which direction is positive and which direction is negative.

 

[ronaldrc]

None of that is being debated. If you think it is, you are not understanding what is being said.

Also known as the parallel configuration where #1 and #2 get tied together as well. This is not what is being said.

I thought I would throw that in everyone on here might not know that the windings are not always a continuous winding.

It wouldn't be a parallel unless you did hook Line #1 and line #2 together.Please click here for more

 

[mivey]

I thought I would throw that in everyone on here might not know that the windings are not always a continuous winding.

And I used that very fact in my explanation here: Using electrically isolated winding halves

It wouldn't be a parallel unless you did hook Line #1 and line #2 together

Of course not.

But, we have not been talking about using them in parallel but in concert as explained in this post: Forces and directions

 

[rattus]

For ___dan

For ___dan

Remember the number line you studied in grade school. Positive in one direction, negative in the other. This is analogous to the transformer winding. Now freeze time at t = 0 when L1 is at 170V. Then as we move away from L1 toward N, the voltage falls to 0, then as we move toward L2 the voltage falls toward -170V. The equations for V1n and V2n are then:

V1n = 170V*cos(wt)
V2n = 170V*(cos(wt + 180))

Remember these are the potentials at L1 and L2, not at the neutral.

 

[pfalcon]

...That's why I don't get the fuss. Ground is probably the most common reference used in the world. I understand it is perfectly fine to use a different reference, but to say one is valid and the other is not is just crazy talk...

Because the ground should always be down, and you keep wanting to put it up

... Here is a quote from the Wikipedia page ... Hopefully this makes sense...

Wikipedia is not a legitimate "source" to quote. Bad mojo, bad practice.

As gar said, more or less, that those with an engineering background tend to think in terms of phasors, loop and node equations, and common reference nodes. Those without that background tend to think in terms that work for them and that is OK as long the job gets done...

Hence the engineers have typically been on one side while the electricians are on the other

The problem arises I believe because some are taught the relationship between the core flux and the induced voltage in the secondary. Then they are taught that adding a CT creates two identical voltages which, as they are defined, are in phase and therefore ADD.

They are not taught however that both voltages can be defined relative to a common reference in which case one is subtracted from the other to get the potential DIFFERENCE between the two points. Or course to do this one must study trig and phasors which are not for the faint of heart.

Actually we are taught all that. Mivey et al are working with "equivalent circuit design". Which is fine and all for performing the math, measurements, and in the real world - real work. As an engineering practice we're taught to use equivalent circuits that best reflect the original circuit. From a transformer all the secondary voltages will be unidirectional and in phase with each other for our 240/120 circuit - by definition. Anything else gets a big fat F on your test.

In practical terms, once you have the equivalent circuit you're never going back again so ... 120<0 or -120<180 - who cares? Maybe the electrician cause that's what he sees on the voltmeter.

Pardon my flippancy, but I can't help but make the observation. This discussion is kind of like going to the dance at an all-boys school. Has it not occurred to you guys that you are all standing alone on one side of the room? ... You've been carrying on an argument for 30 postings, and none of you have noticed that the other side isn't even participating? ... I'm sorry. Please carry on with what you were saying...

Just found the thread so ...

Did you have something useful to add?:? If so, I missed it.:roll:

Sure I do.

Unless you're designing transformers this thread is irrelevant. +<0, -<180, it all works out the same if you're working exclusively on one side or the other of the transformer, or even with a virtual transformer.

ONLY if you're going to include the physical transformer does the polarity of the secondary voltages matter. Then based on the actual physical turns in the transformer the secondaries will both be <0 or both be <180. Nobody reverses the turns midway through the windings.

In engineering we're taught not to abuse circuit equivalencies. So both <0 or both <180 will get you a pass on the test while one of each will get you an F.

 

[rattus]

Both ways:

Both ways:

Consider a real transformer.

There is no question that if

Vx1 - Vx2 = Vx3 - Vx4 = 120V @ 0

Then

Vx4 - Vx3 = - 120V @ 0 = 120V @ 180.

You can have it both ways.

 

[pfalcon]

Consider a real transformer.

There is no question that if

Vx1 - Vx2 = Vx3 - Vx4 = 120V @ 0

Then

Vx4 - Vx3 = - 120V @ 0 = 120V @ 180.

You can have it both ways.

Which once again only models the secondary side with an equivalent circuit. Which again is fine when you're not concerned with the physical construction of the transformer and not trying to pass an exam. So you're statement "Consider a real transformer." (emphasis added) is the bogus part. You're equations are neglecting the real part of a real transformer that induces a real single phase, single polarity voltage across the entire secondary coil set. It doesn't induce the top and bottom separately.

They are mathematically and functionally equivalent but physically different.

And that's more a fact than how people deal with gravity.
Original idea: Heavy objects fall faster.
Next idea: Gravity is a constant so objects fall at the same rate.
Next idea: But only if you neglect wind resistance et al.
Actual fact: Gravitation pull is dependant on TOTAL mass not just the EARTH mass. Therefore more massive objects actually do fall faster. (neglecting wind et al)
Practical fact: Faster out at a kazillion decimal points so who cares?
Exception: Rocket and quantum scientists.

So who cares about how you set up your equations? People designing transformers and engineering students trying to pass an exam. Oh yeah, some old engineers that like purity in expression.

 

[rattus]

pfalcon,

The equations presented simply make the point that a phase shift of 180 is seen between the ends of a secondary which point has been argued for and against ad nauseum. The equations are universally true. The actual design of the transformer is of no concern as is the application.

 

[ronaldrc]

I personnelly am embarrassed by this Thread for the green apprentice that might be reading it.
And please except my apology for the confusion.


If you are really interested in this thread and want to know the truth please go
to a good site that explains how a alternator works. And after you have the basic
idea of how a AC sine wave is produced with a positive and a negative side
then make your on decision on that knowledge.

There opinions here

(1)-Single Phase is really two phase with one side of the secondary 180 degrees out of
Phase with the other secondary side.

(2)-Single Phase is only single phase and both secondary windings are in sync and not
a 180 out of phase.

My self and most electricians will agree with # 2.

Ronald

 

[rattus]

(1)-Single Phase is really two phase with one side of the secondary 180 degrees out of
Phase with the other secondary side.

(2)-Single Phase is only single phase and both secondary windings are in sync and not
a 180 out of phase.

My self and most electricians will agree with # 2.

Ronald

First off ronald old buddy, no one is claiming that it is a two phase system. Thought we should clear up that point before someone nails us on it.

Second, if you specify phase angle, you must also specify the reference point(s). It is simply a matter of where you put the black lead.

Either approach is perfectly valid. Take your pick.

 

[ronaldrc]

Rattus if you use the neutral as a common for both sides of the secondary.
When you hook your oscilliscope positive to line #1 the scope will trace
backwards. Do you agree with that? If so the rest is semantics.


Ronald

 

[gar]

111119-1115 EST

ronaldrc:

How does an understanding of how an alternator work provide any insight on how to view or analyze the voltages from secondaries of a transformer?

I believe the problem in communication on this subject is that many electricians have not studied electrical circuit theory.

Divorce your thoughts from the specifics of a transformer. Consider two 1.5 V batteries connected in series. Maybe should be called cells, but I will use battery as a generic name for a DC voltage source.

First, the connection is + to minus at the midpoint. Define the free positive end as Point A. Define the free negative end as point B. Define the midpoint as N. The voltage drop from A to B is 3.0 V. The voltage drop from A to N is 1.5 V, from N to B is 1.5 V, B to N is -1.5 V, and B to A is - 3.0 V. Using N as a reference point the voltage at A is positive and at B is negative. Thus when N is used as the reference point you can consider B as opposite in phase of A.

Second, change the midpoint connection so both negative terminals are connected together. Now both points A and B are positive terminals of the respective batteries. The voltage drop from A to B is zero, and B to A is also 0. The voltage drop from A to N is 1.5 V, and so is B to N. While N to A and N to B are -1.5 V.

Now change battery B to be a 1.0 V source and do the same analysis.

When you have two AC voltage sources that are identical in frequency and phase, then if the voltages are identical the difference between these voltages is zero. If one has its phase shifted by 180 degrees relative to the other, then the voltage difference is twice one of the voltages. If the phase shift is 90 degrees, then what is the voltage difference? And what is the phase of this new voltage relative to one of the original voltages?

Also note if the frequencies were not absolutely identical you would have a continuously varying phase relationship.

.

 

[gar]

111119-1219 EST

I also want to point out that a phase shift in the general sense has nothing to do with time. A cosine curve is simply a sine curve with a phase shift of 90 degrees. In other words
cos x = sin (x+90).

Also consider a resolver. This is an electromechanical device where the phase of the output signal, relative to an input reference to the resolver, is equal to the mechanical position of the shaft. As you move the shaft from 0 degrees to 360 so does the output phase angle change from 0 to 360 degrees. There is no reason to say that when the shaft angle is 180 degrees that the phase shift is not 180 degrees, but that you can only call the output an inverted sine wave.

Yes a phase shift shift can result from a time delay, for example a transmission line, but a time delay is not required to have a phase shift.

.

 

[rattus]

Rattus if you use the neutral as a common for both sides of the secondary.
When you hook your oscilliscope positive to line #1 the scope will trace
backwards. Do you agree with that? If so the rest is semantics.


Ronald

No, the scope can't trace backward. But if you use a dual trace scope, clip both grounds to the neutral, clip probe A to one line, clip probe B to the other, and sync on probe A, you will see two sinusoids separated by 180 degrees.

CAUTION: Never connect a probe ground clip to a hot line!