Speaker polarity

[odourboy]

So can someone explain why it matters other than "I tried it and it sounds better"?

'scottl' on TGP (he's dogears here) says wrote:
"Big feedback with reverse cable and just ok feedback with regular cable."

and

"Notes bloom into INFINITE sustain one way and fade out weakly the other way. There is also a slightly softer initial attack when the speaker moves forward first. Without enough gain to get that singing feedback, you may not notice."

Others confirm his observations. Personally, I don't believe it's got anything to do with the speaker moving 'forward first' - my interpretation is that the total loop gain (strings, pickups, amp, speaker, air, guitar body, back to strings) depends on which way you wire the speaker. Of the two ways, one will have less attenuation, and produce a livelier interaction with the guitar.

Funny - Since the phase relation between your mic/ears, strings and speaker vary as you move from inch to inch around the room, I was thinking the only possible difference was the initial impulse provided by the speaker moving 'forward first' - i.e blowing rather than sucking.

[tele_player]

Again, just my own thoughts on this... but I haven't heard any explanations which sound better to me -
1) I don't think location of mic/ears figures in at all
2) it's all about the complete 'system', from the guitar, through the amp, and back... I don't think it's about exact cancellation/support, in the usual fully-correlated way - rather, that, of the two phases selected by the speaker wiring, one will produce a higher average loop gain. The effect is said to be enhanced by OD, which, in my thinking, produces so much more overtone content that it increases the likelihood of accidental correlation yielding positive feedback.

As for the 'blowing' vs. 'sucking', it's only during the tiniest instant in the attack where that can even be detected. After that, the speaker is just moving back and forth. On average, the speaker blows as much as it sucks.

[dogears]

As discussed, ad nauseum on TGP, the current thinking is that the OD tone has a very assymmetrical waveform. The speaker moving forward on this intial transient is what is favoring the coupling. Just a theory.

All I know is that it is absolutely occuring. It is repeateable, and not 100% dependent on where you stand. It seems to be a real effect.

Seems to my uneducated brain, that you will lose something if the rich intial attack is reproduced by a poor bandwidth, no hf content, rearward firing speaker back.

Doesn't matter. I hear and feel what I know and I liek the reverse phase. As does every single other person who ever bothered to really try it. So, it does matter. Someday we'll have the real explanation!

[Luthierwnc]

Has anyone come up with a "D" type loop that has an extra phase reversal to bring the signal back to the way it went in without much change in tone? Maybe a very lo-mu triode?

Skip

[Tonegeek]

I did not read the TGP posts on this subject so pardon me if this has already been posited:

My feeling is the initial direction the speaker moves has nothing to do with getting more sing and feedback. I believe the guitar strings are being driven by an in phase (or mostly so) low or low mid frequency signal from the speakers . I am ruling out high frequencies for 2 reasons. Number one, high frequencies have less energy than low frequencies of the same percieved volume (bass amps need more power for this reason). This translates into less ability to drive your strings (unless your amp is really cranked). Second reason, is because of their shorter wavelength they would affect the feedback if you moved just a few inches. A 1k HZ signal will be 180 degrees out of phase if you move just 6.6 inches toward or away from it. In my limited experiments with the reversed wired cable, moving a bit closer or away from the speaker did not make that much difference. I believe for that reason that the lower frequencies are involved more than the high frequencies in this phenomena. A 200 hz signal requires you to move 33 inches for a 180 degree change. So you have more of a range of movement before some cancellation, but I think this is where it gets a little more complicated. Low frequencies being less directional as the freq. goes down will be more evenly dispersed around a room than higher freq. By the time you move a few feet away from your amp, you may also be moving into a sound wave coming off of a wall in the room. The signal could be weakened due to cancellations from returning echos or it could be stronger due to coupling with another in phase echo. As you move away from the speaker, room geometry, echo, dispersion, signal strength, etc. is going to affect the speaker/vibrating strings phase relationship. It seems to me that the distance you are from your amp and room geometry will determine which way you need to wire your speaker cable. It could change depending on the room. I would love to experiment some more with this phenomena to see how much room geometry, and room size makes a difference. If I played out more, I would consider putting a phase switch on my speaker box.

[ayan]

I did not read the TGP posts on this subject so pardon me if this has already been posited:

My feeling is the initial direction the speaker moves has nothing to do with getting more sing and feedback. I believe the guitar strings are being driven by an in phase (or mostly so) low or low mid frequency signal from the speakers . I am ruling out high frequencies for 2 reasons. Number one, high frequencies have less energy than low frequencies of the same percieved volume (bass amps need more power for this reason). This translates into less ability to drive your strings (unless your amp is really cranked). Second reason, is because of their shorter wavelength they would affect the feedback if you moved just a few inches. A 1k HZ signal will be 180 degrees out of phase if you move just 6.6 inches toward or away from it. In my limited experiments with the reversed wired cable, moving a bit closer or away from the speaker did not make that much difference. I believe for that reason that the lower frequencies are involved more than the high frequencies in this phenomena. A 200 hz signal requires you to move 33 inches for a 180 degree change. So you have more of a range of movement before some cancellation, but I think this is where it gets a little more complicated. Low frequencies being less directional as the freq. goes down will be more evenly dispersed around a room than higher freq. By the time you move a few feet away from your amp, you may also be moving into a sound wave coming off of a wall in the room. The signal could be weakened due to cancellations from returning echos or it could be stronger due to coupling with another in phase echo. As you move away from the speaker, room geometry, echo, dispersion, signal strength, etc. is going to affect the speaker/vibrating strings phase relationship. It seems to me that the distance you are from your amp and room geometry will determine which way you need to wire your speaker cable. It could change depending on the room. I would love to experiment some more with this phenomena to see how much room geometry, and room size makes a difference. If I played out more, I would consider putting a phase switch on my speaker box.

Totally agree with the math and the concept. I, unfortunately, also hear a difference with the OT secondary polarity reversed. However, one thing that is also a huge variable is where you play the amp. I have played the same amp/cab in a carpeted studio and have to struggle to get the thing to feedback for me; on the other hand, I have done gigs with glass or mirrors behind the stage and then my struggle is aimed to control my 335 from taking off all by itself.

I have thought of this in the past, but it it would be great to have a polarity switch that one can flip on-the-fly to accommodate the desired response of the amp depending on the situation (a trick that most acoustic guitars have, for example).

One thing that completely makes the theory fall apart, IMHO, is: how do we know that, as we strike a note on the guitar, the signal will have a positive peak insted of a negative one? Not all pickups are consistent from one to the next -- for example, it is a well-known fact that Fralin flips the magnet on all his humbuckers so they end up being out of phase with Duncans, Gibsons, DiMarzios, etc. So which type take off upward and not downward at signal rise? Let's not even discuss the deliberate wiring of two humbuckers out of phase with one another...

Gil

[dogears]

Gil,

You observation regarding pickups is dead on. My Dimarzio's behave 180 degrees different than all my others! So now, I use a regular cable instead of the reverse phase on my Dimarzio pup guitars.

After I intially installed them, I was shocked by how much more feedback I was getting if I shut my loop off. After tweaking and also swapping tubes, I swapped the cable and I suddenly got all the feedback with the loop in.

So, my ear and experience yet again confirmed the phenomena!

[ayan]

Gil,

You observation regarding pickups is dead on. My Dimarzio's behave 180 degrees different than all my others! So now, I use a regular cable instead of the reverse phase on my Dimarzio pup guitars.

After I intially installed them, I was shocked by how much more feedback I was getting if I shut my loop off. After tweaking and also swapping tubes, I swapped the cable and I suddenly got all the feedback with the loop in.

So, my ear and experience yet again confirmed the phenomena!

OK, that settles it, I am putting in a patent application for a "Dynamic Sustain Enhancer Switch" (a.k.a., a phase reversal switch) before Mr R. Smith beats me to it!

Gil

[JimiB]

what about a stompbox with isolated jacks and a dpdt (or a 3pdt if you NEED to know) to flip the phase going into the amp?

[ayan]

what about a stompbox with isolated jacks and a dpdt (or a 3pdt if you NEED to know) to flip the phase going into the amp?

The only problem with doing it that way is that your whole pedalboard would have to be isolated from the amp, including the line ground. All pedals are different, but many of them have the "small signal ground" be the same as the AC ground, and the case is at that same potential. If that is the case, things wouldn't work so well.

I believe the only places where we can flip the polarity cleanly are either before you hit the guitar jack, provided your pickups don't have have grounded shielding and/or covers (so, bascially this leaves only piezos and some single coils in the game -- and of course active electronics as well), or else after the speaker jack. Unfortunately, switching the latter requires a heavy duty switch.

Gil

[mlp-mx6]

So would the "solution" be to have the speaker jack(s) isolated from the chassis, and wired to the DPDT switch outer contacts in criss-cross fashion, with the OT secondaries wired to the switch center contacts? In this scenario I would have the OT "ground side" going from the center contact ground side to ground, and I'd only be switching what goes to the jack(s).

Make sense?

[ayan]

So would the "solution" be to have the speaker jack(s) isolated from the chassis, and wired to the DPDT switch outer contacts in criss-cross fashion, with the OT secondaries wired to the switch center contacts? In this scenario I would have the OT "ground side" going from the center contact ground side to ground, and I'd only be switching what goes to the jack(s).

Make sense?

No need to isolate the speaker jacks from the chassis. The speaker itself is not gounded (to a physical ground that will come in contact with anything else), so you just basically flip the wires at the speaker terminal and that's that. As you suggested, the DPDT will have both sets of diagonally opposed contacts jumped, and that's all.

Gil

[mlp-mx6]

OK.

I was concerned that connecting the "wrong" side of the OT secondary to ground would turn the negative feedback loop into a positive feedback loop.

I often isolate the speaker jacks anyway, so that I'm controlling the ground currents and not using the chassis to conduct those currents. And those are some of the largest currents in the amp!

Thanks again,
Michael

[ayan]

OK.

I was concerned that connecting the "wrong" side of the OT secondary to ground would turn the negative feedback loop into a positive feedback loop.

I often isolate the speaker jacks anyway, so that I'm controlling the ground currents and not using the chassis to conduct those currents. And those are some of the largest currents in the amp!

Thanks again,
Michael

Well, let's make sure I didn't mislead you: you will NOT change how the OT is grounded -- if you change that sure, you will get positive feedback in the power section etc. You will just change the speaker terminal to which the ground wire from the amp connects.

Gil

[JimiB]

Matchless did this in the DC-30 with isolated speaker jacks and a heavy switch. works like a charm. On the cab though you wouldnt need to isolate the jacks.