How do I RECORD a single acoustic guitar in stereo?

OK, so I've recently been experimenting with simulating a stereo sound starting with a single acoustic guitar track. The approach has been to copy the track, shift it forward or backward a few milliseconds, and the pan the two tracks left and right, respectively. See:

http://www.tdpri.com/forum/recording-progress/176520-how-get-stereo-sound-only-one-recorded-track.html

The result does in fact sound sort of like stereo, but what I'm finding is that whether I use a delay of just 2 or 3 milliseconds, or a delay of 5 or 10 or 20, this approach seems to spoil the lovely, pure, clarity of the original recording. It also seems to emphasize the harsh percussive sounds at the starts of notes (probably because we hear these sounds twice in rapid succession rather than once - I suspect this is the main reason why the sound does not seem as pure and clean as the original recording???).

This leads me to think that if I want really good stereo sound from an acoustic guitar, I need to record in stereo from the start.

But how?

My understanding so far is that the best acoustic guitar sound comes from a mic placed really close to the instrument near where the neck meets the body, and this seems to be working well for me (too close to the sound hole and the sound is too deep and boxy sounding). And then some folks put a second mic at a greater distance to pick up room ambiance and such, and the two tracks are mixed. I plan to try this two mic approach when my second mic arrives, but this sort of two mic approach won't really be stereo, right? I'm assuming the sound from from the second mic will be a lot softer in the mix, and it seems that panning these two sorts of tracks left and right would not really give you a stereo sound?

So, what sort of mic placements would I use to get a true stereo recording of a solo guitar?

Or is it silly to even be thinking in terms of stereo if you are recording a single instrument?

Why do I want a stereo recording from a single solo guitar? The reason is that I'm finding that if a recording has some multi-track parts panned left and right to give a stereo sound, and then you also have some solo segments that are single track, there is a noticeble change in overall sound when the single track part comes in. I want to have that stereo sound continue throughout a mix while preserving the pure clarity of the solo parts.

Thoughts????

But how?


Well, any number of ways...
Google "stereo recording techniques", and you'll get a whole bunch of info on M/S, Blumlein, etc. etc...
There are many theories out there as to what is the "best" way to make a good stereo recording...


I plan to try this two mic approach when my second mic arrives, but this sort of two mic approach won't really be stereo, right?

Not if you pan both mics to the center...
If you want a stereo recording, you need to pan your 2 (preferably matched) mics L/R...





Or is it silly to even be thinking in terms of stereo if you are recording a single instrument?



Thoughts????

Well, if the guitar is obviously the featured instrument, I would definitely try to record it in stereo to give it as much of the sonic stage as possible...

I'm not a big fan of using more than one mic on any instrument, but if the instrument is highly featured or solo, then yeah, stereo is the way to go...

I like a matched pair of KM84s, or Gefell UM70s, or for a quick and effective solution, Rode NT4 is a cool option...

But there are many ways to do this...

You need two mics (or a stereo mic - which is essentially two mics) to record a source in stereo.

A mic at the 12th fret and one on the body behind the bridge is a classic way to record an acoustic guitar (whether it is mixed in stereo or not).

You could pan them some without going full left/right. A touch of a good stereo reverb may help also.

What software are you using? Protools? Sonar? There are stereo-izer plug-in for most programs.

You can get a different sound by placing on mic near the 12th and placing another mic over your right shoulder pointed at the side of the guitar. Sort of mimics what you hear while playing.

I sometimes aim the mic at the end of the neck just ahead of the sound hole too.

You might also try copying the signal twice, time shift each of those slightly (but equally) and pan one wide left, one wide right and decrease the levels of each by about 50% to 75% and then leave the original signal at the center.

I might be missing something here but are you just trying to achieve an overall fuller sound from the one instrument?

If that's all you're trying to do then all I do is copy to original track to a separate track and then pan one slightly left and the other slightly right.

To my ears, if you spread them too far apart it sounds unnatural because the one instrument can only be in one place, therefore to have the sound spread to wide just souds wrong to me.

But I ain't no recording guru. I do know however that doubling the track by simply copying to another track does add some significant depth and fullness to the sound.

Every guitar sounds out differently, and needs it's own treatment. The only thing that's never worked for me is a mic directly over the sound hole. Way too boomy (as you said). GTR suggested the most common placement. 12th fret and behind the bridge. Too far up the neck captures excessive "zipper" noise. Experimentation is the key.

Not always necessary to pan both out opposite from center. Sometimes it's nice to pan like 7 and 9 o'clock, just getting a wider sound to one side of center. I guess it depends on how busy the track is already.

Sometimes, if I'm just placing one stereo acoustic in a busy track, I'll save the extreme left and right spots for the acoustic. That makes it very audible at a lower volume.

Never much cared for "faked" stereo, outside of applying a stereo reverb to a mono recorded guitar, but you need a pretty sparse track to hear the results of that. Works great with a guitar/voice, Neil Young kind of thing, if you're stuck with a mono acoustic.

One thing I have done is to re-record the guitar on a second track, mic'd from a high quality monitor. Gives the track a different timbre, and can make it sound like a different guitar. One way to separate the two further is to tap a delay signal (at 100%) from the re-recorded track, and introduce a slight bit of oscillation to the delay signal. A random pattern is best, but even a very slight amount of sine wave modulation helps to randomize things. If you start hearing tuning issues, back it off some.

A key factor here is to remember the difference between true stereo recording, and simply using two mics on a source.

True stereo recording involves specific mic placement technique that is intended to pick up sound in a manner that mimics a pair of human ears. Proper stereo mic'ing also eliminates phase problems between the two mics.

When pointing two mics at different parts of a source, the general idea is to capture the differing qualities of the different parts, and then blend them. If you then pan the mics wide, you end up hearing, (for instance) the sound hole of a guitar in one ear, and the strings on the neck in the other. This doesn't make for a very natural sound. Two mics used this way can also produce phase problems.

The technique that I would recommend depends on the room. If you're in say, a spare bedroom, I would recommend sticking with a single mic and experimenting with stereo reverb and/or delay. Any plugin that outputs to stereo should be fine. My reasoning is that unless you've got a room that makes the instrument sound particularly good, try capturing the instrument as directly as possible, and add ambiance later.

If you've got a cool room to record in, then go nuts with stereo mic'ing, which will inherently capture more of the room sound. The technique that I'm currently head-over-heels for is the Mid-Side technique, which requires at least one mic with a figure-8 pattern and another that can be almost anything, though a cardioid is usually used. My explanation of the M/S technique is spotty, so I recommend Googling it and reading up as much as possible - and then go nuts!

OK, so I've recently been experimenting with simulating a stereo sound starting with a single acoustic guitar track. The approach has been to copy the track, shift it forward or backward a few milliseconds, and the pan the two tracks left and right, respectively.

This will effectively widen the sound field but it's not stereo. Stereo would imply that there is information or imaging in one channel that is not in the other...in the setup you mentioned, both channels are identical.


This leads me to think that if I want really good stereo sound from an acoustic guitar, I need to record in stereo from the start.

But how?

I use the coincident pair or XY technique. It works quite well and all but eliminates the worries of phasing issues.

Coincident Pair How-To (http://www.recording-microphones.co.uk/XY-coincident-stereo.shtml)

Here is a clip of my Taylor using this mic technique. 110 degree spread, about 6" off the fretboard at the 12th fret. I don't have a mono version to compare it to, but you can compare it to your current recordings and see if it's what you're looking for.

Taylor 514CE Coincident Pair Stereo (http://www.crenshawweb.com/taylor_514ce_c2.mp3)

And then some folks put a second mic at a greater distance to pick up room ambiance and such, and the two tracks are mixed. I plan to try this two mic approach when my second mic arrives, but this sort of two mic approach won't really be stereo, right?

Correct. This is like using room mics on a drum kit. While the room mics themselves might be set up in a stereo config, producing a stereo image of the room, they're not producing a stereo image with the near field mics.

Peace,
Mark

One of my favorite micing techniques is to put a 2nd mic over the picking/strumming shoulder of the guitar player and aimed just behind the bridge. Make sure you keep a 3:1 ratio between mics to minimize phase (i.e. mics are 7" from guitar and 21" apart.) Then pan the tracks slightly right and left to taste and balance levels to taste.

BTW...that recording sample is essentially mono, both channels are the same.

BTW...that recording sample is essentially mono, both channels are the same.

Hmmm. Not sure how I did that...must have had them straight up center when I bounced 'em. I'll give it a look when I get home this evening. Sorry for the bad post. :oops:

Peace,
Mark

If you have two mics then you can do an M/S...for that if you can...and have like an AKG C414ULS put that in the figure 8 setting, that on top of it add a good cardioid mike in an omni pattern and you have it...

I might be missing something here but are you just trying to achieve an overall fuller sound from the one instrument?

If that's all you're trying to do then all I do is copy to original track to a separate track and then pan one slightly left and the other slightly right.

To my ears, if you spread them too far apart it sounds unnatural because the one instrument can only be in one place, therefore to have the sound spread to wide just souds wrong to me.

But I ain't no recording guru. I do know however that doubling the track by simply copying to another track does add some significant depth and fullness to the sound.Going a little further with Del's technique you can get a "poor man's" dual mic effect by adjusting the eq on each track a little differently - for instance, you could boost the high end a bit on one track to boost the percussive sound of the pick on the strings and boost the lower end on the other track to bring out the body resonance. This can add some complexity to the overall sound. Of course, as with many things, subtly is the key word.

A key factor here is to remember the difference between true stereo recording, and simply using two mics on a source.

True stereo recording involves specific mic placement technique that is intended to pick up sound in a manner that mimics a pair of human ears. Proper stereo mic'ing also eliminates phase problems between the two mics.

When pointing two mics at different parts of a source, the general idea is to capture the differing qualities of the different parts, and then blend them. If you then pan the mics wide, you end up hearing, (for instance) the sound hole of a guitar in one ear, and the strings on the neck in the other. This doesn't make for a very natural sound. Two mics used this way can also produce phase problems.

The technique that I would recommend depends on the room. If you're in say, a spare bedroom, I would recommend sticking with a single mic and experimenting with stereo reverb and/or delay. Any plugin that outputs to stereo should be fine. My reasoning is that unless you've got a room that makes the instrument sound particularly good, try capturing the instrument as directly as possible, and add ambiance later.

If you've got a cool room to record in, then go nuts with stereo mic'ing, which will inherently capture more of the room sound. The technique that I'm currently head-over-heels for is the Mid-Side technique, which requires at least one mic with a figure-8 pattern and another that can be almost anything, though a cardioid is usually used. My explanation of the M/S technique is spotty, so I recommend Googling it and reading up as much as possible - and then go nuts!

+1

Recording is an art form, it can be as complex or simple as you make it. There are so many variables to consider especially within the context of creative music making. Some good advice here.

Stereo micing for acoustic guitar is usually done for solo guitar work. If you record your acoustic as part of an ensemble, drums and bass etc., then it is usually preferable to single mic it. However let me repeat, recording is an art. There is no wrong or right if the finished product smokes.

There are some great recording forums out there. But be warned it is a deep rabbit hole prone to huge gas attacks.

play the same passage twice, hard pan those 2 tracks.

play the same passage twice, hard pan those 2 tracks.+1 - unless it's a solo guitar recording...