Phase flipping and transformers

[abbey road d enfer]

My hunch is that this a phenomenon inherent to transformers. Perhaps the way the primaries are wound? Or maybe it's component mismatches in the surrounding circuitry?

Transformers have parasitics, leakage inductance and stray capacitance. Each end of a winding has a different stray capacitance, due ti different distance with the other windings, with the core and with the chassis. This results in different loading of the circuit and potential different interaction between the windings and the rest of the circuit when polarity is flipped. This usually results in different HF extension; in a well designed circuit, that would happen outside the hearing range.

 

[abbey road d enfer]

Indeed listening to your own voice in cans can be conflicted by internal parallel conduction paths, and very sensitive to polarity and path delay. Never attempt critical listening that way.

I believe the problems related with this "method" are well-known and still people continue to use it! I think we should make a warning poster in red font size 28.

 

[rp]

Transformers have parasitics, leakage inductance and stray capacitance. Each end of a winding has a different stray capacitance, due ti different distance with the other windings, with the core and with the chassis. This results in different loading of the circuit and potential different interaction between the windings and the rest of the circuit when polarity is flipped. This usually results in different HF extension; in a well designed circuit, that would happen outside the hearing range.

It's nice to confirm that there's potential for such a thing. I wonder how compromised the design or tolerances would have to be for this to be a noticeable effect in practice.

 

[rp]

So you recorded  with a mic pointed at the speaker with one pass of each polarity and then nulled them,  and it didn't show anything outside of the expected???

It was very unscientific. The stuff that didn't cancel was the ambient sound - rumble from the street and hf differences. By ear, the difference between what was cancelling and what wasn't seemed substantial enough to make the call.

 

[abbey road d enfer]

It's nice to confirm that there's potential for such a thing. I wonder how compromised the design or tolerances would have to be for this to be a noticeable effect in practice.

It can be significant when the transformer designer has not taken these effects in consideration. Typically when a power transformer winder is asked to make audio transformers.

 

[CJ]

acoustic guitar amps usually have a phase (polarity) switch, the change in sound is easily noticed,
probably a lot to do with the speaker and the strings,

sometimes flipping the primary wires of a power transformer can reduce electrical noise,

 

[JohnRoberts]

acoustic guitar amps usually have a phase (polarity) switch, the change in sound is easily noticed,
probably a lot to do with the speaker and the strings,

Acoustic guitars didn't always use amplifiers. Indeed the resonant body can be sensitive to polarity feeding speakers in its vicinity, with it's output. 

sometimes flipping the primary wires of a power transformer can reduce electrical noise,

Many US outlets are wired with improper line and neutral assignments. Generally this is innocuous for safety as UL treats all primary wiring as hazardous, but one lead will have 0V nominal and the other 120VAC nominal, so lead dress around sensitive circuits can matter (shouldn't but could).

JR 

 

[CJ]

if neutral is also ground then swapping primary leads can put either the hot end or the cold end next to the secondary wind,

 

[JohnRoberts]

if neutral is also ground

neutral is bonded to ground at the panel so also 0V (more or less).

then swapping primary leads can put either the hot end or the cold end next to the secondary wind,

Are you talking about electrostatic coupling from primary to secondary inside the transformer?  I don't recall ever encountering this as a design issue in power transformers for product designs I was involved in.  I am always willing to learn.

JR

[edit] There are some inexpensive driven rail power amp designs where the audio output is literally the CT of the transformer secondary, but again not aware of issues.  [/edit]

 

[abbey road d enfer]

Are you talking about electrostatic coupling from primary to secondary inside the transformer?  I don't recall ever encountering this as a design issue in power transformers for product designs I was involved in.  I am always willing to learn.

This is typical in most guitar amps. If the amp is not earthed and the meat-puppet killer cap disconnected, noise is dependant on the oreientation of the mains plug. That is because the capacitance between either end of the winding and the secondary/core is different.

There are some inexpensive driven rail power amp designs where the audio output is literally the CT of the transformer secondary, but again not aware of issues.

The Crown MT1000 was guilty of that. When the earth conductor's inductance was too high (long power lines), there would be a tendancy to oscillation triggered by the program's HF content. The subsequent MT1200 did not have this issue.

 

[JohnRoberts]

This is typical in most guitar amps. If the amp is not earthed and the meat-puppet killer cap disconnected, noise is dependant on the oreientation of the mains plug. That is because the capacitance between either end of the winding and the secondary/core is different.

I will take your word for it that primary wiring polarity matters for tube guitar amps but AFAIK the notorious stinger cap switch just alternated the chassis cap between line and neutral, not swapped primary polarity too (which would require a more expensive 2 pole switch, and double insulated energized wiring).

The stinger cap in old school guitar amps bootlegged a chassis ground (capacitively) from the neutral (2 wire) line cord pin.  This was to shield the high impedance tube circuitry within. Obviously cap coupling chassis to the energized mains lead will have the opposite effect.

The Crown MT1000 was guilty of that. When the earth conductor's inductance was too high (long power lines), there would be a tendancy to oscillation triggered by the program's HF content. The subsequent MT1200 did not have this issue.

The driven rail power amp topology was widely used by many (most major?) amplifier companies. The main attraction is that the amplifier front end can use low voltage electronics, except for the final power devices. Back in the 90's for Peavey I even investigated, reducing this low voltage amplifier front end to a single IC, but the cost/benefit wasn't there... and the junior engineer the IC company tasked to work with us, knew less about internal IC design than I did (I had to explain to him how an OTA worked and why you could just use emitter degeneration resistors to deal with mismatched OTA LTP  : ).

I am unsure that the Crown model's driven rail instability is related to mains wiring polarity, but perhaps it could be. Seems to me more like a marginally stable amplifier design. The driven rail amplifiers were always medium to low fidelity because of having to swing the transformer winding  (and it's C to ground?) with the audio output, so relatively heavily compensated and slow.

Which is why I even mentioned them in this context.  I have used many driven rail amps inside power modules (always mono) and never had a concern about primary wiring lead polarity, but this was a mature technology by the time I got there ('85), so perhaps already sorted and off my radar screen.

JR

 

[CJ]

the electrostatic shield between the primary and secondary might come into play as far as primary lead polarity,

it is a fine point that is of interest to audiophiles and buliders  of sensitive test equipment,


if you really want to get picky you can wind the primary and secondary so that the eddy currents in the copper shield will be canceled,

 

[JohnRoberts]

the electrostatic shield between the primary and secondary might come into play as far as primary lead polarity,

it is a fine point that is of interest to audiophiles and buliders  of sensitive test equipment,

and microphone transformers

JR

if you really want to get picky you can wind the primary and secondary so that the eddy currents in the copper shield will be canceled,

 

[abbey road d enfer]

AFAIK the notorious stinger cap switch just alternated the chassis cap between line and neutral, not swapped primary polarity too (which would require a more expensive 2 pole switch, and double insulated energized wiring).

That is correct, but in the very frequent case of a two-conductor mains lead and a disconnected stinger cap, the unbalanced xfmr primary capacitance makes a difference. Many US guitar amps that came into Europe were just equipped with a step-down xfmr and the stinger cap removed. I would think any European aspiring guitar hero in the 60's knew that turning the mains plug would change the amount of hum.

I am unsure that the Crown model's driven rail instability is related to mains wiring polarity, but perhaps it could be.

That's not what I wrote. The problem is not related to primary polarity, it's just an issue of how the capacitive current flows.

Seems to me more like a marginally stable amplifier design. The driven rail amplifiers were always medium to low fidelity because of having to swing the transformer winding  (and it's C to ground?) with the audio output, so relatively heavily compensated and slow.

The issue is that the extra current has to be dumped into the earth conductor; if its inductance + resistance is high enough, it develops a voltage that can be in phase with the input signal and generate positive feedback.

 

[JohnRoberts]

That is correct, but in the very frequent case of a two-conductor mains lead and a disconnected stinger cap, the unbalanced xfmr primary capacitance makes a difference. Many US guitar amps that came into Europe were just equipped with a step-down xfmr and the stinger cap removed. I would think any European aspiring guitar hero in the 60's knew that turning the mains plug would change the amount of hum.
That's not what I wrote. The problem is not related to primary polarity, it's just an issue of how the capacitive current flows.
The issue is that the extra current has to be dumped into the earth conductor; if its inductance + resistance is high enough, it develops a voltage that can be in phase with the input signal and generate positive feedback.

OK gotcha... the swinging transformer winding has capacitance to chassis ground.  A floating ground or high impedance can support a voltage imposed on that ground.

Again voltage/current on chassis ground "should not" affect the audio path, but there are many less than ideal audio paths, and IIRC the Crown amps were popular for install use, so who knows what kind of signal wiring will be encountered in installs.

JR 

 

[CJ]

here is some blurbage on power transformer primary leads polarity, 

what you do is carefully put a ground lift adapter on the piece of equipment and measure AC leakage to the chassis between ground (center lug on adapter or cold water pipe or copper rod) and the chassis itself. flip the primary leads and measure again. use the orientation with the lowest voltage reading.

https://www.audioasylum.com/messages/general/449743/the-other-side-of-the-coin-long

http://forums.stevehoffman.tv/threads/vintage-non-polarized-two-prong-cord-identifying-hot-neutral.610589/

 

[JohnRoberts]

here is some blurbage on power transformer primary leads polarity, 

what you do is carefully put a ground lift adapter on the piece of equipment and measure AC leakage to the chassis between ground (center lug on adapter or cold water pipe or copper rod) and the chassis itself. flip the primary leads and measure again. use the orientation with the lowest voltage reading.

https://www.audioasylum.com/messages/general/449743/the-other-side-of-the-coin-long

http://forums.stevehoffman.tv/threads/vintage-non-polarized-two-prong-cord-identifying-hot-neutral.610589/

And since all outlets are not wired correctly (line and neutral sometimes swapped), confirm that the outlet you use for your test is correct, "and" that all the outlets you plug into for regular use are wired correctly.

Just to be clear are you talking about reversing the transformer primary leads, or line cord? (ok Hansen link mentions primary wiring). I have seen 2-wire consumer gear with a capacitor between chassis ground and line cord "neutral" or what should be neutral. Swapping around line cord in such a 2-wire product will alter line cord wiring polarity wrt such consumer stinger caps (but you'd have to grab the chassis ground for leakage measurement).  Old guitar amps often have stinger caps that can dump current into chassis.

Finally swapping primary polarity inside a unit with a stinger cap could be swamped out by the stinger cap so maybe coordinate both. 

I don't love the idea of messing around with the safety ground lifted inside 3 wire chassis that may be single insulated. Primary wiring is not the only source of shock hazards protected by a grounded chassis.

JR