Just to clarify.

[coulter]

When solving these problems, I tend to stick with the _same_ set of sense designations rather than switching for different problems. ...

Okay. I'm good with you doing this. I work on enough different stuss that I have not seen a need for this constraint.

... I stick with the source neutral as the voltage reference, and use 'current flowing _out_ of each source terminal' as my current sense....

I'm translating this as generalized to 3ph grounded Wye. Again okay.

... But I find that it makes life easier when solving for (say) multiple loads connected to a wye source. ...

Ah - This is why I guessed the above quote was generalized to grounded Wye

... In the single phase center tapped system, the same current is flowing through the entire system. There is _no_ change in the current flow as it passes through the neutral point. The only thing that changes is the _description_ of the current flow. ...

This one baffled me for a while. But maybe I got it now. You are differentiating between a KVL solution where the current is in a loop - vs a KCL solution where the currents sum to zero at a node. For example, KCL at a node: Iin -Iout = 0 Hence the perception that the current swapped sense (direction? description?). Is this correct?

... You want to describe your voltages and your currents using the same 'sense', so that when you multiply current times voltage you get a positive number for components that supply power, and a negative number for components that consume power. ...

I will have to think about this a bit. It's pretty rare that I am concerned with generator power out and load power in in the same breath. But I think you are saying: Pgen - Pload = 0 Which as simplistic as that sounds, does come up when one can get into a suituation where the gens are exporting vars to a utility (and that one has come up once)

... If you elect to describe all of your voltages relative to the neutral point, then the sense that you use to describe current must also be oriented with the neutral point. If (as I do) you always describe current in terms of the flow _out_ from the neutral along the various branches of the circuit, then the sign or phase of the _description_ of the current flow will flip as you go through neutral. This change in description of the current matches the description of the voltage, and maintains the convention that (voltage * current) is power supplied by that component ...

Again, I'm translating this as generalized to multi-phase grounded Wye - which isn't what we are dealing with.

Excellent geography lesson. I'll keep it in ming next time I drive through North Pole (I do that a few times a month:smile:

carl

 

[rattus]

Yes it does:

Yes it does:

rattus -

Still no news. As I have agreed, with your solution setup, yes this happens. As I said, with my solution setup, it does not.

carl

Carl, the currents in your diagram still return to node L2 which is opposite the polarity mark. The voltage on L2 must be described as V2n which is 180 out of phase with I2.

Now consider an ideal, long, lumped transmission line. We can describe all loop currents as being CW, but at some point down the line the phase of the voltage reverses which means the current should be CCW.

This example illustrates the point, but I am still looking for a real world example.

 

[jim dungar]

I guess that I'd just prefer to think of a center tapped secondary in a fashion more similar to a wye secondary, even though the polarity dots will be arrayed differently in the two.

Jon,
Your concept simply ignores the real world physical connection of transformers. You have taken a base model (the grounded wye with the dots pointing out) and extrapolated the math (but not the underlying reason) to other systems with neutrals. Sure you can analyze a transformer using your concept, you just can not design a single core one with it.

 

[coulter]

Carl, the currents in your diagram still return to node L2 which is opposite the polarity mark. The voltage on L2 must be described as V2n which is 180 out of phase with I2.

Now consider an ideal, long, lumped transmission line. We can describe all loop currents as being CW, but at some point down the line the phase of the voltage reverses which means the current should be CCW.

This example illustrates the point, but I am still looking for a real world example.

Rattus -
This post is a train wreck. Possibly some good stuff, but reassembly required, and morbidly interesting.

Which post of mine is this disasterous diagram of mine in?

I am completely not understanding the example of a long transmission line with reversing voltage phases and currents changing from CW to CCW.

Could I suggest sticking with my disaster post? Perhaps I will better understand the advantages of your model.

carl

 

[rattus]

Rattus -
This post is a train wreck. Possibly some good stuff, but reassembly required, and morbidly interesting.

Which post of mine is this disasterous diagram of mine in?

I am completely not understanding the example of a long transmission line with reversing voltage phases and currents changing from CW to CCW.

Could I suggest sticking with my disaster post? Perhaps I will better understand the advantages of your model.

carl

Don't remember the number, but you had all CW loops, and that is perfectly alright. Now, when we start talking average power things get a little sticky. Some one pointed out that the return current, Ir, was out of phase with the voltage, V2n.

But, to compute power, we must take the product of the outgoing current and the source voltage. We simply reverse the direction of Ir by negating it. Then voltage and current are in phase and the product is positive.

Easier still, if Ir is in the direction of the voltage rise in the source, then the product is positive.

 

[coulter]

Don't remember the number, but you had all CW loops, and that is perfectly alright. Now, when we start talking average power things get a little sticky. Some one pointed out that the return current, Ir, was out of phase with the voltage, V2n.

But, to compute power, we must take the product of the outgoing current and the source voltage. We simply reverse the direction of Ir by negating it. Then voltage and current are in phase and the product is positive.

Easier still, if Ir is in the direction of the voltage rise in the source, then the product is positive.

rattus -
This does not read like part of the discussion, it reads like a diversion. I don't see anything in your last two posts that are part of the discussion.

I don't know where the power questions came from - certainly not part of this discussion. However if you are interested in complex power, it is the voltage times the complex conjugant of the current.

So how about back on topic?

Am I to assume that the unknown disaster post is now okay?

I still have no idea what you are talking about on the long transmission lines issues.

carl

 

[winnie]

Jon,
Your concept simply ignores the real world physical connection of transformers. [...] Sure you can analyze a transformer using your concept, you just can not design a single core one with it.

I don't see why one couldn't design a center tapped transformer using the neutral as the reference, but I see your point that using the coil orientation for the reference/sense probably makes good sense when designing the transformer itself.

When viewing the transformer itself as a black box source, I don't see any reason to use a one method for three phase wye and another method for single phase center tap.

The "wye" source could actually be a delta secondary combined with a zig-zag autotransformer to derive a neutral, or it could be a 'Scott connection' secondary with only two secondary coils and a neutral tap on the teaser coil. In these cases, would you use a different reference/sense for your voltage and current values, in order to match the perhaps confusing coil arrangement? Or would you simply stick with the same descriptions normally used for a conventional wye source?

-Jon

 

[rattus]

rattus -
This does not read like part of the discussion, it reads like a diversion. I don't see anything in your last two posts that are part of the discussion.

I don't know where the power questions came from - certainly not part of this discussion. However if you are interested in complex power, it is the voltage times the complex conjugant of the current.

So how about back on topic?

Am I to assume that the unknown disaster post is now okay?

I still have no idea what you are talking about on the long transmission lines issues.

carl

The power question did arise, but it is moot now, so we can forget it.

I see nothing wrong with your post. You just defined all your loops to be CW. I was just trying to demonstrate that CCW loops are equivalent if not intuitive.

Let me provide a diagram for the lumped transmission line. It provides delay, and delay provides phase shift, and a few milliseconds of phase shift create an inversion.

 

[LarryFine]

I see nothing wrong with your post. You just defined all your loops to be CW. I was just trying to demonstrate that CCW loops are equivalent if not intuitive.

But don't the loops need to be in the same clockwise direction to explain why the neutral current is a differential, and not an additive product? As I've said, intuitively, at least to me, this is a matter of polarity, and not a 'shift' (which at least implies a matter of timing), as it would be in the DC equivalent.

Two DC sources in series, with a grounded center tap, would be considered opposite in polarity when measured from ground. If we instead grounded the negative terminal, giving us two positive voltages, there would be no issue. Same with grounding one end of a center-tapped secondary.

To me, AC is the same in that regard. I absolutely agree that opposite polarity and a 180-degree phase shift are indistinguishable, but why it occurs is relevant to whether it's one or the other. Adding a negative and subtrating a positive may be equivalent mathematically, but electrically, I want to know why.

 

[rattus]

1. But don't the loops need to be in the same clockwise direction to explain why the neutral current is a differential, and not an additive product? As I've said, intuitively, at least to me, this is a matter of polarity, and not a 'shift' (which at least implies a matter of timing), as it would be in the DC equivalent.

2. Two DC sources in series, with a grounded center tap, would be considered opposite in polarity when measured from ground. If we instead grounded the negative terminal, giving us two positive voltages, there would be no issue. Same with grounding one end of a center-tapped secondary.

3. To me, AC is the same in that regard. I absolutely agree that opposite polarity and a 180-degree phase shift are indistinguishable, but why it occurs is relevant to whether it's one or the other. Adding a negative and subtrating a positive may be equivalent mathematically, but electrically, I want to know why.

Larry,

1. No they can be either direction. If I1 and I2 are in the same direction, one will carry a phase shift of 180 degrees which says subtract me!

2. If you drew a current arrow in the wrong direction and then wrote a loop equation, the resulting value of current would carry a negative sign indicating turn me around!

3. In some cases we subtract because we made a wrong assumption, say on the sense of the current. In other cases, such as in a string of batteries, we add voltage rises and subtract voltage drops.

Now, one may not use these techniques often, maybe never, but I would think that understanding such trivia adds to one's depth of knowledge.

 

[jim dungar]

I don't see why one couldn't design a center tapped transformer using the neutral as the reference, but I see your point that using the coil orientation for the reference/sense probably makes good sense when designing the transformer itself.

In these cases, would you use a different reference/sense for your voltage and current values, in order to match the perhaps confusing coil arrangement? Or would you simply stick with the same descriptions normally used for a conventional wye source?

Jon,

Yes a transformer could be designed with the neutral as a reference, but that is not the circuit that Rattus has consistently shown (he has been using the ANSI standard connection).

When setting up an initial analysis I do not pick a single reference point. I put my unknown source currents in the same direction as their voltages especially if polarity marks/dots have been provided.

 

[winnie]

But don't the loops need to be in the same clockwise direction to explain why the neutral current is a differential, and not an additive product?

Heh, I'd been thinking of using this example as the reason why it is better to use the neutral node as the reference and have 1 CW loop and 1 CCW loop *grin* To me it is more intuitive to have the arrows from the two current loops both pointing in to the neutral and to say that the polarity of one is the opposite of the other. Just goes to show me that once one is caught up in a particular representation, what is 'intuitive' will be whatever fits that representation.

As I've said, intuitively, at least to me, this is a matter of polarity, and not a 'shift' (which at least implies a matter of timing), as it would be in the DC equivalent.[...]

To me, AC is the same in that regard. I absolutely agree that opposite polarity and a 180-degree phase shift are indistinguishable, but why it occurs is relevant to whether it's one or the other. Adding a negative and subtrating a positive may be equivalent mathematically, but electrically, I want to know why.

I was thinking about the above points, after my response to Jim in a previous post.

Any phase difference clearly means a timing difference in the sine wave, but does not necessarily mean a timing difference in the source of that sine wave. A center tapped coil is a perfect example, one that we've bandied about more than enough times; there is _no_ timing difference at all in the generation of the output, but an apparent 180 degree phase difference in the outputs.

As I agreed with Jim, if designing the transformer itself, it makes the most sense to match the description of voltages and currents to the layout of the transformer.

But this leads me to the question: what do you do when you have systems with different _internal_ physical layouts, but with corresponding output. If you are just looking at the output, do you change your representation to match the internal layout, or do you simply use the same representation that you are used to?

I would contend that to answer most problems, you would simply use your standard representation.

To get to examples, I need to go to three phase. You can get an apparent 'wye' source using a real set of three secondary coils, energized 120 degrees apart.

But you can also get it with _two_ secondaries, one center tapped, the other with an 86.6% tap and a 36.6% tap. Connect the 'teaser' 0% terminal to the 'main' center terminal, and use the 36.6% tap on the teaser as your neutral. Energize the coils 90 degrees apart. The output is an apparent wye source, with 120 degrees between output terminals relative to the neutral. But internally it is a mess, with an inversion and 90 degree differences.

Going in the other direction, consider the connections used to get single phase output from a '12 lead' generator. At least one of the connections that I remember gives what looks like two small deltas connected corner to corner. Internally there is quite a bit more than a simple inversion going on, but the output looks like a simple inversion relative to the neutral.

-Jon

 

[rattus]

For Carl:

For Carl:

Carl, it is unlikely that you would find such a network in the power industry, but the attachment contains a textbook example which would likely produce incorrect assumptions about current sense. It certainly would if you assumed all CW senses.

 

[coulter]

You're right - I'd likely make loops 2 and 9 CCW.

carl

 

[rattus]

You're right - I'd likely make loops 2 and 9 CCW.

carl

Care to solve for all currents?

 

[coulter]

Care to solve for all currents?

Okay.
I1 = 11.6
I2 = 12.7
I7 = 10.4
I9 = 9.3

Now you pick out the values I used.:grin: I'll make it easy: Voltages sources are all phase angle zero, impedances are resistive.

carl

 

[coulter]

Just goofing with you. I didn't solve it. Maybe I can this afternoon.

carl

 

[mivey]

Since you are going to play, make the sources V1, V25, V78, V9. Get it right and you can have the 2 credits back that you owe me.

 

[mivey]

On second thought, make all the voltages 120<0 and the resistors all = "R"

 

[coulter]

Care to solve for all currents?

rattus - your last two posts appear to be another diversion. It's not really pertinent to the discussion.

Since you are going to play, ...

mivey - You're right, i'm screwing around. But its too late, I already picked the resistors at 1 ohm the voltage sources at 1V. i1, i4, i5, i7 CW and i2, i3, i6, i8, i9 CCW

Source i1 = .367
Source i2,i5 = .325
Source i7,i8 = .558
source i9 = .433

i1 = 0.367
i2 = 0.125
i3 = 0.067
i4 = 0.225
i5 = 0.200
i6 = 0.075
i7 = 0.333
i8 = 0.225
i9 = 0.433

Absolutely no guarantee these numbers have any resemblance to accurate calculations. :-? Didn't check the equations, didn't check the entry, didn't check the math.

This would fall under the, "If a thing isn't worth doing at all, it isn't worth making sure its right.":roll:

Even if this is close enough, I still owe the 2 credits. Enjoy

Back to work for me

edited to fix some of the more glaring grammar gaffs

carl