Transformer wiring

[Buffalo Tom]

I'm using OEP transformers with dual windings, which I'm wiring in series. Previously, I connected them as shown in version 2, but I'm considering switching to the version 1 configuration because the physical pins are closer together that way. Can you confirm that this change won't have any electrical impact?

[antonis]

It looks absolutely OK to me..

[amz-fx]

Appears to be the same with the primary windings in series. That's how I would wire it too.

Best regards, Jack

[mozz]

What about phase? What is it anyway? Power, audio, interstage?

[R.G.]

I'm using OEP transformers with dual windings, which I'm wiring in series. Previously, I connected them as shown in version 2, but I'm considering switching to the version 1 configuration because the physical pins are closer together that way. Can you confirm that this change won't have any electrical impact?

It will have no electrical impact. The dots on the transformer schematic indicate the ends of a winding that all swing the same direction at the same time. More on this below in mozz' question on phase. You can think of this as "all the dotted ends go positive at the same time". So to make two windings be in series, you connect one winding's dotted end to the NON-dotted end, exactly as you have done.

Another way to think of it is what if the transformer had only flexible wires, not pins? The wire leads can go in any order at all, so the dot notation is the only way to get phasing right.

What about phase? What is it anyway? Power, audio, interstage?

"Phase" in transformers is a little odd compared to its use in audio. In audio, it's continuous - signals can have a phase difference of zero to 360 degrees. With transformers, this gets simplified down to only two phases - in phase and out of phase. The dotted ends of a winding are all in phase with each other; the non-dotted ends of windings are all in phase relative to each other.

Imagine it like this. Pretend all the dots are "+" signs. Connect the NON-dotted ends to ground in your mind. Hook your mental oscilloscope to ground and all of the +/dotted ends. You can do this because your mental oscilloscope has as many channels as you want. Pick a winding and hook a (mental) 9V battery across it, negative side to "ground" and positive side to the dot/+ end. The instant you touch the battery to the winding, all the dotted/+ ends jump positive. They are all in phase.

In winding a transformer, the dotted end is often taken to be the start end of a winding. When you wind multiple windings on the same core/bobbin and the winding machine always turns the same direction, all the start ends are in phase, and all dotted. The dot is just a marker tag for either start-of-winding or end-of-winding. Either way is OK, you just have to be consistent.

Of course, the continuous phase matters, too. Mostly audio and effects work has only one primary winding, and does not involve three-phase multi-primary connections. I highly recommend that you ignore polyphase transformers unless you have a need to work with electrical power connections.

For single phase power and non-tricky audio, the thing to remember is that when the dotted end goes positive (relative to the non-dotted end), all the windings go dotted-positive relative to the non-dotted ends.

The reason I keep harping on polarity indications is that for single phase (single primary), it's all you get to play with. For a signal isolator transformer, you get a primary and a secondary. If you ground the non-dot of the primary and inject signal into the dot end, then the dot end of the secondary follows the primary and you just get isolation. If you ground the dot end of the secondary and take your signal off the non-dotted end, the signal is inverted. Every transformer winding can follow or invert the signal on the primary; this is in addition to transforming the voltage and current.

In using interstage transformers, getting the dot/non-dot wrong means you've added another signal inversion to the signal path. This can cause all kinds of havoc where the interstage windings drive power stages, for instance.

If you use a multiple-winding transformer, you can pick inverted/non-inverted as you choose.
A center tapped winding is really two windings, with one dotted and one non-dotted end connected together inside the transformer and only one wire brought out from the internal connection.

All of this applies to all transformers WITHIN THEIR SIGNAL PASS-BAND. Power transformers are designed to be really, really good at passing power line frequencies and no particular thought given to higher or lower frequencies. They may accidentally be good at passing higher frequencies, but almost never good at passing lower frequencies. Audio and interstage transformers are designed with thought about making the frequency pass band be a certain low-to-high frequency.

For audio, the low frequency range may be as low as 20Hz for super hi-fi transformers and as high as 20+ kHz for the high frequency pass band. For audio interstage transformers, the same applies. Radio uses interstage transformers too. In RF work, the frequency range is often deliberately narrowed to the point that the transformer is actually part of a very narrow filter to select only the desired frequency range. The analog 0-to-360-degree phase variations with frequency are still there, but the dot/non-dot signal inversion overlays it.

I get the feeling that I'm muddying things up more than clarifying now. Ask questions.

[Buffalo Tom]

Thanks R.G. 

I'm using OEP transformers with dual windings, which I'm wiring in series. Previously, I connected them as shown in version 2, but I'm considering switching to the version 1 configuration because the physical pins are closer together that way. Can you confirm that this change won't have any electrical impact?

It will have no electrical impact. The dots on the transformer schematic indicate the ends of a winding that all swing the same direction at the same time. More on this below in mozz' question on phase. You can think of this as "all the dotted ends go positive at the same time". So to make two windings be in series, you connect one winding's dotted end to the NON-dotted end, exactly as you have done.

Another way to think of it is what if the transformer had only flexible wires, not pins? The wire leads can go in any order at all, so the dot notation is the only way to get phasing right.

What about phase? What is it anyway? Power, audio, interstage?

"Phase" in transformers is a little odd compared to its use in audio. In audio, it's continuous - signals can have a phase difference of zero to 360 degrees. With transformers, this gets simplified down to only two phases - in phase and out of phase. The dotted ends of a winding are all in phase with each other; the non-dotted ends of windings are all in phase relative to each other.

Imagine it like this. Pretend all the dots are "+" signs. Connect the NON-dotted ends to ground in your mind. Hook your mental oscilloscope to ground and all of the +/dotted ends. You can do this because your mental oscilloscope has as many channels as you want. Pick a winding and hook a (mental) 9V battery across it, negative side to "ground" and positive side to the dot/+ end. The instant you touch the battery to the winding, all the dotted/+ ends jump positive. They are all in phase.

In winding a transformer, the dotted end is often taken to be the start end of a winding. When you wind multiple windings on the same core/bobbin and the winding machine always turns the same direction, all the start ends are in phase, and all dotted. The dot is just a marker tag for either start-of-winding or end-of-winding. Either way is OK, you just have to be consistent.

Of course, the continuous phase matters, too. Mostly audio and effects work has only one primary winding, and does not involve three-phase multi-primary connections. I highly recommend that you ignore polyphase transformers unless you have a need to work with electrical power connections.

For single phase power and non-tricky audio, the thing to remember is that when the dotted end goes positive (relative to the non-dotted end), all the windings go dotted-positive relative to the non-dotted ends.

The reason I keep harping on polarity indications is that for single phase (single primary), it's all you get to play with. For a signal isolator transformer, you get a primary and a secondary. If you ground the non-dot of the primary and inject signal into the dot end, then the dot end of the secondary follows the primary and you just get isolation. If you ground the dot end of the secondary and take your signal off the non-dotted end, the signal is inverted. Every transformer winding can follow or invert the signal on the primary; this is in addition to transforming the voltage and current.

In using interstage transformers, getting the dot/non-dot wrong means you've added another signal inversion to the signal path. This can cause all kinds of havoc where the interstage windings drive power stages, for instance.

If you use a multiple-winding transformer, you can pick inverted/non-inverted as you choose.
A center tapped winding is really two windings, with one dotted and one non-dotted end connected together inside the transformer and only one wire brought out from the internal connection.

All of this applies to all transformers WITHIN THEIR SIGNAL PASS-BAND. Power transformers are designed to be really, really good at passing power line frequencies and no particular thought given to higher or lower frequencies. They may accidentally be good at passing higher frequencies, but almost never good at passing lower frequencies. Audio and interstage transformers are designed with thought about making the frequency pass band be a certain low-to-high frequency.

For audio, the low frequency range may be as low as 20Hz for super hi-fi transformers and as high as 20+ kHz for the high frequency pass band. For audio interstage transformers, the same applies. Radio uses interstage transformers too. In RF work, the frequency range is often deliberately narrowed to the point that the transformer is actually part of a very narrow filter to select only the desired frequency range. The analog 0-to-360-degree phase variations with frequency are still there, but the dot/non-dot signal inversion overlays it.

I get the feeling that I'm muddying things up more than clarifying now. Ask questions.

[antonis]

A center tapped winding is really two windings, with one dotted and one non-dotted end connected together inside the transformer and only one wire brought out from the internal connection.

That's the major issue for connecting both windings in parallel (for doubling current capability)..

[R.G.]

A center tapped winding [...] only one wire brought out from the internal connection.

That's the major issue for connecting both windings in parallel (for doubling current capability)..

True. A center tapped winding can't be connected with both sections in parallel - the wires to do this are not available.

Connecting windings in parallel has some subtle problems. The two windings need to have >exactly< the same number of turns to be paralleled. If not, there is a voltage difference, and a current flows through the two windings. This current is limited only by the wire resistance of the windings and it can be BIG. Modern automated winding machines are better at this than pre-1980s manually controlled winders. It's only that automation that makes paralleling windings really practical in volume.

The professor who taught our transformers and motors class was clear about how you connect paralleled and "rings" of windings. You hook up one of the common points and leave the other end open. Then you use a meter to measure the voltage difference between the not-yet-connected ends. If this voltage difference is substantially zero, you can parallel the windings. If not, you can't. This is a sensitive test for whether the windings are truly the same number of turns.

This is even a bigger deal with three-phase power. Three primary- or secondary-windings on a three-phase transformer are commonly hooked up either in "Y" or "delta" format. In "Y", all three windings are connected together at a center point and the three ends have the three phases of AC voltage available. In "delta" the three phases are connected in a ring. If all the windings have truly the same number of turns, the delta ring will have zero volts across a gap between any two of the winding connections and you can close the "ring" without circulating currents. Circulating currents in 3-phase delta connections are BIG.

[PRR]

Three primary- or secondary-windings on a three-phase transformer are commonly hooked up either in "Y" or "delta" format.

[antonis]

For math lovers...



P.S. The above transformation formulae is used for Baxandall Max Treble boost calculation..

[R.G.]

Thanks Paul, Antonis! I was too lazy to go grab or create the illustrations. 

[Rob Strand]

You should always follow the manufacturers recommendation for connection of series windings.  For mains transformers there are safety issues.  For signal transformers it can affect capacitance.

This shows why the two connections are not the same.  They are physically different and the voltages across points in the windings can be different.  This can affect safety.