# Merely flipping secondary winding shouldn't affect transformer polarity

The polarity test for a transformer determines whether it has additive or subtractive polarity. I understand that convention is to recognize X1 and X2 on the secondary winding as, respectively, the negative and positive terminals, but does the secondary winding actually have an "inherent" negative terminal? Does the voltmeter somehow detect that X1 is the negative terminal?

Given that the secondary winding is a symmetrical object, I'm confused how merely flipping the winding can affect the voltage measurement.

Source

• The negative terminal would be the one you connect to what you determine to be negative. There could be edge-cases for high voltage transformers where you are not free to choose, but for normal 120 V ones, they don't know nor care. Do you intend to use it as shown to buck down or boost up your 120 V? Also, please look up transformer dot convention. Jun 7, 2023 at 14:41

The winding itself is not quite symmetric, because it needs to be connected to the core and other winding to couple magnetic fields. However, schematic symbol drawn in that figure (without a dot marking the polarity of each winding following the dot convention) is misleadingly symmetric.

If you swap the X1 and X2 connections, as shown in your example image, you are flipping only the two wires leading into and out of the winding, not the way the winding is connected to the core:

disregard the difference in turn counts in the two transformers, that is simply a sloppy drawing error on my part. The winding positions and directions of winding are the important elements.

The core and windings don't quite have a positive or negative on their own, but they do have a relationship where the polarity of one winding, either matches or is opposite the polarity of the other winding, and which is the case is marked with a dot on the schematic (following the dot convention I linked above).

The exact polarity (and hence position of the dots) depends on the way the windings are wound (including "overhanded" vs "underhanded" direction of winding). Moving wires around doesn't change that, but physically unwinding and then rewinding a winding differently may.

• My book says that because "the high- and low-voltage leads are independent of the arrangement of the windings of the magnetic circuit, the polarity of a transformer can be changed by interchanging the position of the two leads of any one winding as brought out of the case." This appears to directly contradict your note but I'm not sure. Would you say you and the author are in agreement? Jun 8, 2023 at 15:00
• Source: p73 of Transformer Principles and Applications (NJATC, 2006) Jun 8, 2023 at 15:02
• @artist_and_not_EE_by_training I am in agreement with them. Interchanging the position of the leads, as brought out of the case, is what I drew, and it interchanges the polarity (as measured externally with voltmeters, oscilloscopes, etc). The drawing in your question describes a different scenario - the wires are flipped and the winding is flipped on the core. Jun 8, 2023 at 15:13
• I believe you meant to say that in the image in my question, the wires are flipped only. Your drawing makes that very clear -- thank you very much nanofarad Jun 8, 2023 at 15:52
• @artist_and_not_EE_by_training I think you're right - it's the wording "winding is symmetrical" that suggested flipping both to me. Jun 8, 2023 at 16:03

but does the secondary winding actually have an "inherent" negative terminal? Does the voltmeter somehow detect that X1 is the negative terminal?

Transformers are AC devices. None of the terminals are "negative" or "positive". But the secondary winding has one terminal that's on the "same end" (magnetically speaking) as H1, and one terminal that's on the "same end" as H2. The test is simply to find out which one is which.

The rest of the material on that page, I would consider not terribly useful. There is no such thing as an "additive polarity" or "subtractive polarity" transformer. There are different ways of packaging transformers, and there are different ways of connecting a transformer in a circuit to achieve different goals.

It is not generally true that something is "always referred to as H1 and H2" or X1 and X2. In fact it's much more common in electronics to use dots to indicate the polarity of a transformer in a schematic. If you need to translate, then as that article points out, you can consider the "1" terminals as corresponding to dots.

• To be probably overly generous to their terminology, you can have an additive or subtractive connection when using a transformer as an autotransformer to slightly buck or boost a just-out-of-spec AC supply. Jun 7, 2023 at 14:57
• @Hearth Agreed. You can have an additive or subtractive arrangement using a transformer... but most of the time, that terminology doesn't even apply! Jun 7, 2023 at 14:59
• @artist_and_not_EE_by_training no (and also that's not really what they said). You can place one dot anywhere you want; the second dot (or remaining dots if there are more windings) go where they need to based on the direction of the induced current. Reversing the handedness of the windings is one of the things that will reverse that, but so is just choosing to draw it with the connections the other way around. Jun 8, 2023 at 14:57

Does this make it "make sense"?

"Phase dots" were added to an arbitrary terminal.

Wiring those phases so the dots are together in the loop, causes the subtraction of the secondary voltage from the primary. This is because the secondary phase is opposite the primary, so dots-together causes them to subtract. Just like if you placed a 9 V and 1.5 V battery in series, but flipped the 1.5 V one around backwards.

Wiring those phases so the dots are series-in-line, causes the addition of the secondary voltage to the primary. Since the phases are the same, they add.