I backed into this via double tracked acoustic guitar, love the potential at some places along the fretboard, but am currently stuck in advancing, at the point of destructive (or, better put, so-so) phasing at certain frequencies.
 
Anybody else playing with this?
 
Any bodies of knowledge out there on this subject?
 
Any unproven theories that may help?

My gut says its possible to bandpass pleasing interaction on one side of the doubletrack, but I can see that causing imaging probs.
 
Wandering, left to right.
 
Is this approach going to get as deep, as fast, as I think it will?

got sound examples? 
 
Generally comb filtering is something to be avoided at all costs. It usually sounds bad.

Comb filtering is happening every where.
 

Perhaps what Jay Tee is referring to is the creative use of comb filtering to produce width in double-tracked instruments, where two identical tracks are differentiated by complementary comb filters. It's tricky, but with some experimentation it can make for nice-sounding stereo effects on rhythm guitars. With care it can create a sense of width while remaining mono-compatible.
 
The time-honored method for doing this is to clone a track and then apply complementary filters - usually using a graphic equalizer - to each track. A comb filter is doing the same thing but with a lot more bands. The challenge is that comb filters are not symmetrical, so it's pretty much impossible to have truly complementary filtering, resulting in some frequencies always being reduced or over-emphasized for a hollow sound. You can, however, mitigate this by placing the filters in parallel and dialing in just enough of the effect to be, uh, effective.
 

Actually, I haven't been cloning, but true doubletracking. It started with Seasons of Wither when I forgot to mute a take and had two going at once on playback.
 
It got a much wider spread and spaciousness then 'Smiths original, even pre reverb, a deep space swoosh better than flange or chorus plugs, but, like I said before, its better at some places on the neck than others, except...
 
I think I'm a midbrain, but not at the same time, and while I can sense the other side waiting with answers, I can't find the key to open the door when on the opposite side. Theory and math are partitioned off when I'm making music, subliminally there but not active.
 
Discussing things with others, like here,  makes a hardcopy I can reference from either vantage point, just like now. I'm seeing what I'm up against here.
 
First off, I have a feeling things are going south in mono with this technique, but haven't checked it yet. Theory says YES.
 
Second off, fretboard position probably is not the limiting variable here, differences in left/right track timing are probably orders of magnitude greater than fundamental variance. I think even a consummate studio musician will have trouble with timing sync on different takes at phase cancel/sum levels.
 
Still....I am enjoying this effect much more than any phase or flange plug or device I've ever used, so I'm going to continue to pursue it, but I have less confidence in mastering it than before.
 
Bitflipper might have a piece of the solution here, cloning identical takes and then introducing timing variation at controlled intervals to maximize positives and minimize negatives, but at that point, I'm back to an 'artificial" and mechanical system, essentially a phase or flange algorithm that lacks the sparkle (perhaps unpredictability?) of what I have now.
 
I think my next move is going to be to lay down a (third) center track to serve as the timing and audible "anchor", not necessarily panned zero but not pan wandering either. With that as foundation, it will mitigate mono deficiencies, and give me more control over when and how the two hard panned "doubletracks" comb filter the primary.
 
I can actually see a total of as few as five tracks...the center anchor, another take panned left, a third panned right, and the center anchor cloned left and right so I have the option of identical and disparate timing to work with via automated gain balancing.
 
The effect is yielding similar results with vocals, both background and lead,  so even though it currently seems a complicated mess, there is great potential value in advance.
 
Comments and suggestions welcome, thanks for those so far!

Mike Senior raves about the creative tone-shaping potential of comb filtering numerous times in his book "Mixing Secrets." 

One thing I've done is to take a send from a mono-recorded guitar track to a buss with a medium length, somewhat dark stereo reverb and apply Channel Tools widener preset after the verb--adjust the width to taste.  You can also tailor the dry/wet signal in the reverb on this buss (maybe you only want the reverb to be really wide).

Jay Tee 4303
I think I'm a midbrain, but not at the same time, and while I can sense the other side waiting with answers, I can't find the key to open the door when on the opposite side. Theory and math are partitioned off when I'm making music, subliminally there but not active.

I had an epiphany about that back in the 80's, when I was assigned by my employer to teach a class on creative thinking. I was working for a computer manufacturer, a company populated almost exclusively by left-brain types doing jobs in which creativity was neither a prerequisite nor often rewarded.
 
I took the assignment because it meant going out of town and kicking back in nice hotels on the company's dime. But the train-the-trainer course I took really opened my eyes.
 
I was already aware that I had two brain modes: computer geek during the day and playing in bands and piano bars at night. But the training course brought me to realize that I was never in both modes simultaneously, that one or the other was always dominant, and that it takes effort to switch between them. And that those individuals who do have the ability to straddle both at the same time have a rare gift. Some are insane.
 
After returning from the training, I went to my weekday gig, playing piano in a posh "athletic" club near my office. It was a great gig because it started at 5:00 PM and was 5 minutes from my office. But at 5:00 I'd always struggle to get warmed up, over-thinking everything I was doing. By 6:00 I'd forgotten about what my fingers were doing, letting improvisation flow out almost as a bystander. 
 
This mirrored the results of exercises we were taught in the creative-thinking seminar. The transition from one mode to the other is not gradual, it is a quantum state-change. It requires switching off one set of mental algorithms so that the alternate set can activate.
 
In the case of self-produced recording this presents a particularly difficult challenge, that of rapidly switching roles between performer and engineer. It's hard. It doesn't always happen reliably. But there's a solution: dedicate any given session to just one type of activity or the other, avoiding having to cross over. Decide to have dedicated tracking days and engineering days.
 
Before tracking day, set everything up and make a test recording, check for noise and distortion. Then quit. On tracking day, just track and don't even think about anything technical. Then quit. Next session, put on your geek hat and check how the tracks are sitting in the mix, how you're going to effect them, what kind of EQ they'll need, make your edits, put in your markers.
 
By separating creative and technical activities you give your brain time to make the switch and you'll be more productive in either mode. My #1 tip: stay away from these forums on tracking days. If you can't stay away, at least limit yourself to the Songs forum.
 
 

LMAO.
 
This is getting printed and posted, next to the patch map, room node spectral, and Pentatonic variants.
 
Thanks Bit!