phase polarity, feedback assisted sustain... does it matter?

[pinkjimiphoton]

forgive my public musing yet again...
but i was thinking in my curious way about feedback. and phasing.
it seems to me that when exploiting the hell out of feedback, perhaps it would be useful to change the phase of the guitar..

ie, make the speakers suck instead of blow. i suspect that one way feedback may be easier to manage than the other.
so, the question is... and no, i haven't tried this yet or even messed with it... but i'm wondering if perhaps adding a stomp to switch between +/- inputs on an opamp or something may be worth a play.

the way i figure it, and i am likely WRONG as usual, but if you're sustaining a note via feedback, and you wiggle it or move, you can catch different harmonics, right?

what if you could use the phasing of the signal itself to determine how sensitive to harmonic feedback the guitar may be?
another crackpot idea, or do ya think it may make a difference?

i may experiment a little in the dungeon with this.. but am curious what other's thoughts are.

all i can think of is the harmonic/normal modes on an ebow.  i'm betting there will be a difference between "suck" and "blow"...

like, a marshall seems better suited to feedback... it is an amp that historically "blows" or moves the air in phase with the input.

fender reverbs are historically out of phase, so they "suck"... the air moves out of phase with the input... fenders do NOT seem to feedback the same way as a marshall does... obviously for many and varied reasons. but i'm wondering if the phasing has more to do with it than you may first think....

i'm postulating that in/out of phase may make a difference in the way distortion reacts with a guitar signal. i bet one way will have increased harmonic access content, and the other more control.

anybody ever mess with this? too much lds at berklee again? wondering if it may lead somewhere cool...

[stallik]

I'm not sure that the phase relationship with the input will make any difference. What I found does make a difference is other sound sources of the same frequency being out of phase with yours.
Like when your running 2 amps out of phase and get cancellation. I've wondered if this might occur with 2 guitarists being out of phase with each other.
I've made a simple switch to go between amp and speaker which reversed the wires to get round such a problem with my dual amp setup. It's very informative

[Phoenix]

I don't think it would help, the act of moving to stomp the footswitch would take you in-and-out of the "sweet spot" a couple of times just in the distance you moved to plant your foot anyway.

Your thoughts about Fender and Marshall, well, Marshall's are associated with higher gain styles generally, and that's where you get more feedback. I'm not sure if you're refering to Fender Reverb's as in the amp model, or just generally all models with reverb, but they're not "out-of-phase", because reverb isn't time-correlated with the original signal, its phase relationship has no bearing on feedback. So neither Fender Reverbs or Fender reverb is out of phase.

Controlled feedback is about a few things, volume, gain, compression, proximity to the speaker, any degree of microphony in elements like pickups, tubes, ceramic caps or even cables (looking at you PlanetWaves!).

But, it's an easy enough thing to try if you want.

[pinkjimiphoton]

I don't think it would help, the act of moving to stomp the footswitch would take you in-and-out of the "sweet spot" a couple of times just in the distance you moved to plant your foot anyway.

Your thoughts about Fender and Marshall, well, Marshall's are associated with higher gain styles generally, and that's where you get more feedback. I'm not sure if you're refering to Fender Reverb's as in the amp model, or just generally all models with reverb, but they're not "out-of-phase", because reverb isn't time-correlated with the original signal, its phase relationship has no bearing on feedback. So neither Fender Reverbs or Fender reverb is out of phase.

Controlled feedback is about a few things, volume, gain, compression, proximity to the speaker, any degree of microphony in elements like pickups, tubes, ceramic caps or even cables (looking at you PlanetWaves!).

But, it's an easy enough thing to try if you want.

cool beans, thanks phoenix.

fender and marshall amps are out of phase with each other generally. particularly in the case of fenders with reverb... and its a legit thing.

the reason is every tube stage you go thru, reverses the phase of the input to it at its output.

so with fender amps like a bassman, the speakers will "blow", but due to the extra stages for the reverb, fender amps will "suck".

i've noticed differences with even stompboxes due to the polarity of the power supplies even. i use bricks to power my board, and in some cases find the whole thing will work  better.. .les noise, interference etc... if i reverse one of the power supplies. shouldn't matter, but definitely does... despite the fact that we're dealing with dc thats filtered and regulated.

feedback can be done myriad ways. i tend to be able to get it at almost any volume or position, but am curious to see if it will matter changing the actual phase of the guitar itself.

i don't run marshalls or fenders dirty... i tend to run 'em pretty clean and generate everything thru stompboxes.
its not the gain of the amps that seems to make the dif as much as whether the speaker is sucking or blowing.

not in all cases, but generally in the classic amps, the phasing does come into play, right down to your average fender twin having the speaker wires "backwards" electically from what's expected. marshalls, not having the extra stages for the reverb and mixer stages tend to play "forwards".

fender's cybertwin lets you choose whether it sucks or blows, and it definitely sounds and responds differently depending on which way the speakers phase. i haven't touched it in years <after dick wagner using it, its more an object of veneration than "gear"> but perhaps its worth investigating while the ol' lady's out

again, thanks for the schematic bro, i will take a better look now that i'm done blatherskiting

[ashcat_lt]

Feedback is all about phase and relative polarity of course.  The frequencies coming from the amp that are in phase with the string movement will tend to feedback most easily.  That's why moving the guitar can change the note that rings.  But this is all pretty unpredictable, and is almost arbitrary.

BUT Brian May has a phase switch for each of the pickups on his guitar.  In general, when he's only playing one through one pickup, there's no real practical difference.  EXCEPT when he's holding a note, and it's sustaining thanks to feedback from the amp, and he FLIPS the switch, it can sometimes cause the string to break to a harmonic of the note.  We all know the sound.  We can get it sometimes by just moving the guitar, but that's a bit of a hint.

ALSO, the magnetic sustainer systems - Fernandez, EBow... - have a "harmonic" mode, that usually causes it to feedback on a harmonic of the held note.  This is usually accomplished by inverting the polarity of the driver coil.  So there's that...

[pinkjimiphoton]

thanks ash, exactly kinda stuff i'm wondering about.

could a simple footswitch switching polarity of the input of a distortion with a fair amount of sustain make it bloom?

in a semi-predictable way?

i think it may be an unexplored avenue for the utility fuzzbox research kitchen to take on.

i've done - and found - the phase switch on one of my old firebirds to really make a difference with feedback assisted sustain sometimes.

and in phase? that means whats going out the speaker is the same phase as whats going into it. out of phase reacts differently. it really does suck and blow. and i bet that can be exploited in a completely unexpected way.

thanks for the thoughts. i think it may be kinda cool, if it can be added to a fuzzular as a bloom switch. hit the switch, harmonic changes to the opposite phase of the same frequency, right?

thats gonna either reinforce or cancel harmonics. i hope all the guys way smarter than i'll ever be can chime in on this one.

maybe its mojo snake oil boolshite too. but i dunno... like ash says, thats how a sustainer works like an ebow..

[thermionix]

fender and marshall amps are out of phase with each other generally. particularly in the case of fenders with reverb... and its a legit thing.

In the classic 2-channel Fender (Twin, Super, Deluxe) Reverb amps, the two channels are out of phase with each other because the reverb side goes through an additional gain stage.  For Fender vs. Marshall, it would depend on the specific models.  A JTM45 and 5F6A would be in phase with each other (same circuit)...until the signal reaches the speakers, anyway.  Jensen reversed the polarity of their speakers sometime around 1960.  Easy to flip at the speakers either way, make any two amps work together.

[pinkjimiphoton]

yeah, thats it. i remembered it from somewhere or other. thanks therm!

[Phoenix]

Ok, I'm sorry, but I don't think you actually understand what phase actually means.
Absolute phase doesn't actually matter at all - it's only RELATIVE phase that matters.
For example, you seem to be concerned with the number of inverting stages in an amp/stompboxes/polarity of speakers/whatever. But in reality, once we go from the electrical domain into the acoustic domain, we're not only dealing with absolute phase, but also spatial positioning. Due to the speed of sound and its wavelengths at different frequencies, whether any frequency is acoustically in phase once it reaches your pickups is entirely determined by your distance from the speakers, and this is why moving the position of your guitar relative to the speakers can alter which harmonics are fed-back. This also means that when playing through a dual amp setup, even if the two signals are electrically out-of-phase with each other when they reach the speakers, if you move the amps varying distances from each other you can bring them ACOUSTICALLY in-and-out of phase again. On a small stage or in a bedroom this is likely not practical, and an inverting stage is almost universally more convenient, but ultimately, the absolute electrical phase doesn't mean a thing.

So to recap, phase only comes into it when you have two or more signals, be that two different pickups, two different effects chains, or two different amps. Try just flipping the speaker wires in an amp sometime and see if it behaves any differently.

And, as ashcat said:

But this is all pretty unpredictable, and is almost arbitrary.

[pinkjimiphoton]

yer right greg, i don't know poo in fact, that's why i'm asking about this stuff. seriously.

so what is it when you change the phase of a pickup, then, accoustic, or is it electronic?

its kinda both, seems to me. the fender/marshall thing is kinda irrelevant i guess, unless ya play them things regularly and loud. having a fender and marshall side by side can be different i think than ya realize. or i am probably wrong, which is one thing i can usually be sure of

but still... if it doesn't matter, why does it matter with an ebow? its gotta be accoustic phase cancellations, right? either its in or out of phase with them, or off. one way primarily sustains the fundamental, but the other you usually get a pronounced upper octave effect.

i guess basically what i was wondering was if there were any value to exploring the idea in a stompbox setting rather than a phase switch on a guitar, or an ebow. it seems like its the sudden "break" when flipping the switch that makes the harmonic popout.

but looking at it after your description i guess you're right, as the output of an opamp is still gonna be the same regardless of which input you were to use on it, and for SURE there's a diff between pickups in or out of phase. just a dumb idea dude, its all good.

ya know i gotta mess with this and try and see if it will do something sonically audible, remember, i am the farthest thing from an EE, i got little issue with letting the magick smoke out of expensive components, and, as a fuzz-monger am looking to create distortions and exploit shit that shouldn't even work

rock on man

[Phoenix]

so what is it when you change the phase of a pickup, then, accoustic, or is it electronic?

That's electrical. The strings process (rotate) above the pickups, the fundamental frequency and its phase is the same all the length of the string (at least for the fundamental). Flipping the electrical phase (more properly polarity) of one of the pickups flips the output, and if somehow we were able to have two matched pickups in exactly the same position below the string, we'd get perfect cancellation, which is of course exactly what we get if we feed the signal from one pickup into a non-inverting buffer and also an inverting buffer and then mix the outputs.
But when we position pickups at different points along the string, we get different relative proportions of harmonics, more harmonics towards the bridge, more fundamental towards the neck. When two pickups are out of phase, they share fundamental and lower order harmonics, so these get cancelled the most, while higher frequencies get cancelled less and less as they increase in frequency. This is also assisted by inharmonicity, the phenomenon whereby the harmonics get progressively sharper closer to the bridge due to strings being stiff and not perfectly elastic (this is also mostly why we need compensation, though action does contribute a tiny amount too), so the upper harmonics are less frequency matched between pickups too.
So we get lower frequency cancellation, and harmonic emphasis, that's what gives you the nasal "honk" you get from out of phase pickups.

[thermionix]

This also means that when playing through a dual amp setup, even if the two signals are electrically out-of-phase with each other when they reach the speakers, if you move the amps varying distances from each other you can bring them ACOUSTICALLY in-and-out of phase again.

But frequency (wavelength) dependent, right?  So some notes could be brought back in phase by speaker offset, but not all.  Might be a desired effect for some purpose, but I think it's vastly simpler to get the two signals electronically in phase.  There's phase (as in time) relationship vs just signal polarity (without delay).  I'm saying that wrong, but you probably get what I mean.

[Phoenix]

But frequency (wavelength) dependent, right?

It's all ALWAYS frequency dependent. Even with two identical amplifiers in phase, move about the room at all and you'l get null points at different frequencies all over the place - same deal with two out of phase amps, you'll find plenty of peaks and nulls around the room. I did say that it's generally much more practical to flip phase, the wavelength of a low E (82Hz) is about 4.16meters, so you'd need to have 2.08m (half wavelength) of wiggle room, which most of the time you don't have. But if you did, yeah, you could get two "out of phase" amps to sound as much in phase as two "in phase" amps no problem.

[thermionix]

the wavelength of a low E (82Hz) is about 4.16meters, so you'd need to have 2.08m (half wavelength) of wiggle room, which most of the time you don't have. But if you did, yeah, you could get two "out of phase" amps to sound as much in phase as two "in phase" amps no problem.

Yeah but only on the low E, and other notes with certain harmonic relationships to low E, right?  I think an F (for example) wouldn't sound in phase, unless you moved the amps again, or your listening position.

(I'm not trying to argue, if it seems that way, I think we're in agreement here.  Just discussing, always trying to learn.)

[Phoenix]

the wavelength of a low E (82Hz) is about 4.16meters, so you'd need to have 2.08m (half wavelength) of wiggle room, which most of the time you don't have. But if you did, yeah, you could get two "out of phase" amps to sound as much in phase as two "in phase" amps no problem.

Yeah but only on the low E, and other notes with certain harmonic relationships to low E, right?  I think an F (for example) wouldn't sound in phase, unless you moved the amps again, or your listening position.

(I'm not trying to argue, if it seems that way, I think we're in agreement here.  Just discussing, always trying to learn.)

Yeah, it's good discussion. A couple of times I've stopped to re-think my position, which is always good.
The point I'm trying to make, and not sure if I've gotten across yet, is that the same thing happens with two IN PHASE amps, that unless you're positioned exactly the same distance between them you'll still have *some* phase issues. If you're offset between the amps, sure, you might have the low E in phase, but what about C#? This is of course all easier to deal with if both amps are in phase, but try this sometime, if you've got two amps out of phase, flip one back-to-front (works best if one is directly behind the other, oh, and I should specify I mean the combo/cab, flipping a head will do nothing!). Of course, it's not particularly practical to point one amp at the back wall of the stage, but your phase issues are gone!

[thermionix]

the same thing happens with two IN PHASE amps, that unless you're positioned exactly the same distance between them you'll still have *some* phase issues

Right, I get you, the sound doesn't all emanate from a single point in space.  Two speakers in the same amp still require you to be centered (and maybe your ears to be calibrated).

and I should specify I mean the combo/cab, flipping a head will do nothing!

And an open back cab at that.  Some combos are amazing from the back, even enough to piss you off that the front doesn't sound as good.

[antonis]

the absolute electrical phase doesn't mean a thing.

Perhaps because there isn't any "absolute" phase.. 

[PRR]

If the path goes through air, the "phase" is totally scrambled. If there is 10 feet of air, deep bass may have little phase-shift but treble will be shifted many wavelengths. At any specific frequency you "can" compute the phase of the direct path; but not while you are playing many-frequency notes and chords! The bounce-paths are even more randomly phased.

The *easiest best* way to try reversed polarity is a DPDT switch in the speaker lead. This is typically the only place in a g-amp where you can swap "hot and ground" without trouble or added parts.

In one polarity, feedback is strongest at certain frequencies. In reverse polarity feedback is strongest at different (generally in-between) frequencies. There's NO essential difference, and no way to predict which you will like better in a given room, position, or key.

In general PA, flipping polarity means going-through the 29-band EQ all over again, chasing different rings, and generally getting to a condition no different from before.

[pinkjimiphoton]

If the path goes through air, the "phase" is totally scrambled. If there is 10 feet of air, deep bass may have little phase-shift but treble will be shifted many wavelengths. At any specific frequency you "can" compute the phase of the direct path; but not while you are playing many-frequency notes and chords! The bounce-paths are even more randomly phased.

The *easiest best* way to try reversed polarity is a DPDT switch in the speaker lead. This is typically the only place in a g-amp where you can swap "hot and ground" without trouble or added parts.

In one polarity, feedback is strongest at certain frequencies. In reverse polarity feedback is strongest at different (generally in-between) frequencies. There's NO essential difference, and no way to predict which you will like better in a given room, position, or key.

In general PA, flipping polarity means going-through the 29-band EQ all over again, chasing different rings, and generally getting to a condition no different from before.

this is essentially what i was talking about before we got off on the phasing amp tangent. it would seem reversing the phase of the guitar or the speaker would kinda do the same thing.. in some cases, it will reinforce, in other cases, it will neutralize frequencies. this is what i hoped could be exploited as a stompbox "feedback helper" kinda deal

tho a reversed guitar or speaker may "sound" the same, electrically, they are polar opposites, no? thats where i was asking about speakers sucking or blowing. to me, a stock fender amp played clean with a shitload of dirt will feedback and react differently from my marshall set up the same way. the only real dif is the way the speaker plays. again, each stage of tube inverts the signal. if you invert a signal, you put it out of phase with where it was, right?

if ya haven't tried an ebow, try one. try the normal and harmonic mode. out of phase with "direct string synthesis" most def causes different.... and predictable... harmonics to bloom.

fascinating discussion, thanks gents...