| Phil, if you are talking about how to determine the proper 'polarity', there | are a couple of ways. One way, used by electricians all the time was | mentioned before, tieing one end of primary to one end of a secondary, | applying a low voltage to primary terminals and measuring the voltage | between the primary terminal and secondary terminal that aren't connected.
Yes, but not in the context of having a physical transformer present, rather, in the context of a theoratical construction. The test you describe would work to let someone hook up a transformer. But I'm looking to understand the orientation of transformer design with 100% confidence (it's not 100% yet ... and things that in the past were 95% turned out to be wrong about
5% of the time).
| But reading your followup posts, you seem to be looking for the 'other' | method. Such as how do manufacturers figure out just how to mark the | connections in the first place when they wind them.
I'm sure they can figure it out by the first method after a sample has been made. I wouldn't blame an engineer for testing it that way to be sure.
| Well, the answer goes back to the theory. Remember that when an isolated | load is connected to any secondary of an energized transformer, the | magneto-motive force created by the secondary current is opposing the MMF of | the primary current.
This I know. But there is some ambiguity to this. You said force and then current. The problem I see is that I don't know which it really is that determines the orientation. Given 2 identical windings, if the _voltage_ ends up being the same on the same ends, then the 2 _currents_ are flowing in opposite directions. This is explained by the fact that power is going in on the primary and power is going out on the secondary. Understanding the "opposing force" would help. But due to the loose usage of terms when many people speak about theory, especially "force" vs. "current", I just can't be sure what is going on there.
| So pick a hand, either hand (I like 'right hand'). Now, wrap your figures | of the 'chosen' hand, in the direction of current flow (for a half-cycle) | through the primary windings around the core. Note which way along the core | your thumb is pointing. Now, take your opposite hand (I hope you still have | two). Point the thumb of that hand (for me, it would be the 'left hand') in | the same direction along the core. Now your fingers of that hand wrap | around the core in the direction of current flow in the secondaries. | Regardless how many secondaries there are, and which way they are wrapped, | the current in*all* secondaries will go around in the direction of your | second hand's fingers. You're done.
I'm sure all the secondaries are like any other secondary. I'm sure all the primaries are like any other primary. It's the relationship between the primary and the secondary that I haven't pinned down.
Looking at the core from one end, I believe the direction of wire wrap, e.g. clockwise vs. counter-clockwise, is the issue. Whether the wire starts at the bottom and ends up at the top, or starts at the top and ends up at the bottom, is not. Or a wire could wind CCW going from top to bottom and then continue winding CCW in a new layer going back to the top, repeating until the needed number of windings are done. It would still be CCW from the referenced view.
For consistency, I would bundle all the windings together as a "cable" and wind this cable as described. While that may be a lousy way to construct a real-life transformer, I think it clearly shows the idea of everything in the same orientation.
When I energize the transformer, the primary current is opposed by the field. Whether that is an actually a current opposition or a voltage opposition would not matter (yet) since it affects the same winding. But when we look at the 2nd winding as a secondary, there would be a voltage potential there, but not being connected to a load, no current. Now if a load is connected, is the _current_ on the 2 windings going to be in the same direction? If I have described my scenario clearly enough, someone who thoroughly understands this should be able to say yes or no. If the currents are the same then the voltage on the secondary will be opposite because power is being taken out instead of being put in.
But is that so, that both primary and secondary currents go in parallel when power is drawn from the secondary? It seems that can't work because it would increase the field strength, and something I read suggest the secondary has to be tapping into the field strength, effectively lowering it, for power to be taken out.
So the other supposition is that the current in the primary and the current in the secondary will be going in opposite direction, cancelling each other out. This would then give identical voltage polarities.
While I'm still sitting on the fence, the "opposite current" scenario seems more plausible because it would have to be in order to correctly describe how 2 wires in a cable supplying power to some load will cancel each other's magnetic field. But I've yet to see clear, detailed, unambiguous wording that confirms this.
| If your primary and secondary windings make just one 'pass' down the length | of the core, and they wind around the core in the same direction, then the | 'dot marked lead' of the secondary is at the same end of the windings as the | 'dot marked lead' of the primary. You can pick either end for the 'primary | dot', just have to make sure you put the 'secondary dot' at the same end. | If the secondary winding twists around the core in the opposite direction, | then the 'secondary dot marked lead' is at opposite end from the 'primary | dot marked' lead. | | But be careful when talking about 'ends'. After all, if the primary has | many turns, it may wind clockwise down from the 'top', get to the bottom, | and continue to make windings in a second layer, working back up to the | 'top'. So both ends of the winding can come out at the same 'end', one at | the innermost layer, the other at the outermost layer. It is more important | to note which direction *around* the core the current flows when on a | particular half-cycle (i.e. which way do your fingers 'wrap').
OK, I think I can conclude, as I mentioned earlier, that the ends don't really matter. It's strictly the direction around the core.
BUT ...
If we are dealing with a core in the form of a loop, with a primary on one side of the loop, and a secondary on the other, then we have to reverse things because the field itself has been turned around.
Here is a more complex scenario which is one of those things that has given me the want to find out precisely about this. Suppose I have an "E" core transformer, which has 3 vertical bars crossing between a top bar and a bottom bar. This is the typical core used for a 3-phase transformer. Label the three vertical core segments A, B, and C. If I wind a primary around core A and energize it, the field will loop around through cores B and C. If I wind a secondary around core C, now what happens? Will the secondary around core C be able to get full power out, despite B being present? Or will the B core reduce the available power in some way?
Suppose I wind a tertiary winding around core B, and monitor the voltage by drawing a trivial few microamps. Will this voltage change as more and more power is drawn from the winding around core C?
What happens if I put a big load on B, or even solidy short it? How will that affect the power I can get from C? What if I put a resonant circuit on the B core, peaked at a high impedance at the power frequency?
If I had 3 windings around a single core, I can better visualize what might be happening, despite a few lingering doubts or ambiguities about how it all works. But with the E-core, things are "stranger" here.