Stereo Mic Techniques Acoustic Guitar Advice

[woodnut]

Hello folks,

I have been experimenting with stereo mic techniques for the acoustic guitar as I am shortly going to be recording a very sparse E.P. with just fingerpicked acoustic and vocal, recorded simultaneously.

Since the production is so sparse I am looking to fill the space a bit and stereo miking the acoustic seems the sensible choice. I thought I had settled on X/Y but I don't seem to quite be getting the width that I would hope for. I am looking for a nice bit of natural sounding width but nothing unrealistically large.

With the capsules on top of each other in a classic X/Y arrangement (approx 1-1.5 feet from the 12th fret) the image naturally pulls off largely to one side as the mic that is towards the body picks up the larger signal. What is the usual protocol here when mixing this? Hard panned left and right pulls the guitar too far to one side. Adjusting the level of the weaker signal then gives an unnatural imag of the guitar, narrowing the image seems to introduce some phase issues and the overall sound does not seem as natural.

Should I be attempting to pull the X/Y mics back much further than 1 to 1.5 feet? I seem to loose a lot of bass and warmth due to the proximity effect when pulling the mics back.

I also tried not having the capsules on top of each other so that they just meet at right angles on the same level and this seemed a little better although the width is very narrow and it seems like both mics are pretty much getting the same signal.

I plan to try ORTF but this does seem like the 30cm spacing between mics may be too wide for close miking, I have tried mid side but the side mics got too much bleed from the vocals, I didn't really like spaced pairs when I tried this method.

If there is any advice or methods anyone would like to chime in with I would be very grateful! Many thanks!

[Hugh Robjohns]

With the capsules on top of each other in a classic X/Y arrangement (approx 1-1.5 feet from the 12th fret) the image naturally pulls off largely to one side as the mic that is towards the body picks up the larger signal.

And that's the whole point of the XY technique -- it is a naturalist method of capturing a sound image in an acoustic space. The sound of an acoustic guitar comes mainly from the resonant body, and very little from the neck -- hence if you position the mic array as you have most of the sound will be reproduced left of center. What else would you expect?

If you want a more balanced image the mic will need to be positioned in front, or just left, of the sound hole, being careful not to provide undue emphasis of the sound hole's boominess.

Also, the width of image produced by the XY technique is dependent on the polar patterns of the mics in use. You don't say what pattern mics you're using, but I presume they are cardioids which are the worst possible option for an XY array in front of an acoustic guitar because the stereo acceptance angle is about 180 degrees, and the guitar will occupy only about 25 degrees or less.

Consequently the stereo image will always be very narrow unless you place the array very close, and if you do it will sound exaggerated, and unnaturally boomy!

Crossed hyper-cardioids (or even fig-8s) will deliver a much more appropriate image width with the mic array at a sensible distance.

Alternatively, abandon the idea of capturing an accurate stereo image (which is pretty meaningless and pointless in this context anyway). Instead, space and balance the mics as separate entities to capture a sonically pleasing spread of sound. Typically, that would involve one mic near the junction of neck and body, and the other under the body looking up at the tail -- although both positions should be arrived at through careful experimentation as every guitar radiates sound differently.

Adjust the levels and balance the two mic signals independently to achieve an acceptable blend of character, and pan apart for the appropriate image width. This works best if the recording environment is pretty dead, and you can then add in a fake room ambience with a reverb heavy on early reflections.

I seem to loose a lot of bass and warmth due to the proximity effect when pulling the mics back.

Yep, that's what happens with pressure gradient mics, and that's what EQ is for!

I plan to try ORTF but this does seem like the 30cm spacing between mics may be too wide for close miking

The correct spacing for ORTF is 17cm. 30cm is the spacing for the alternative NOS technique. I wouldn't recommend either in this application.

I have tried mid side but the side mics got too much bleed from the vocals

Assuming the vocalist was also the guitarist, this is not possible. The null of the fig-8 side mic -- there should only be the one mic by the way! -- will always face the performer/instrument in any normal MS arrangement that I can conceive of, and therefore should not pick up any direct vocals at all. Perhaps you have strong reflective surfaces nearby which are degrading the separation.

H

[woodnut]

Thank you Hugh for taking the time to answer my post! There is alot there for me to think about.

It is not so much that I wan't expecting the image to pull off to one side with X/Y, it was that I didn't know how to center this more and how this is done by those with much more experience than me. Whether I should be changing the placment or simply panning to taste.

Yes the two mics I am using are Oktava MK012's with cardioid capsules, I hadn't realised that these were not ideal for X/Y on an acoustic guitar but I think I do understand your explanation on this.

As for the mid side bleed, yes it is possible that the reflective surfaces are to blame. I have several DIY bass traps that I use to create a dry sounding area for recording, but I then found that putting wood panels on the floor underneath the guitar helped the guitar breathe a little as it was a bit congested sounding. Perhaps these panels were contributing o the bleed of vocals into the side mike. I think I should revisit the mid side technique again.

I have possibly gotten a bit too focused on getting a sound right at source without EQ, hence the comment about the warmth and bass dropping off when moving the mics further back. I wondered if I was doing anything wrong here.

Thanks.

[Sam Inglis]

What mic were you using as the Sides mic in your MS array?

[Hugh Robjohns]

Yes the two mics I am using are Oktava MK012's with cardioid capsules, I hadn't realised that these were not ideal for X/Y on an acoustic guitar but I think I do understand your explanation on this.

Personally, I don't think XY cardioids are much use for anything other than ambient/atmospheric sound effects! If you have to use cardioids in a stereo array, then ORTF usually gives better imaging.

I have possibly gotten a bit too focused on getting a sound right at source without EQ, hence the comment about the warmth and bass dropping off when moving the mics further back. I wondered if I was doing anything wrong here.

All pressure gradient mics are designed to give a flat response at a given distance, becuase of the proximity effect. Used closer than that and they exhibit bass lift, used further away and they tend to lack bass. So the real question is, are you using the proximity effect as free bass-lift EQ?

Switching to omnidirectional mics (which have to be used as a spaced pair if you want a stereo effect) maintains a natural bass response at any distance, and also doesn't suffer the odd LF phase effects that afflict most cardioids.

H

[Tomás Mulcahy]

Partly agree with Hugh here, though I bow to his greater experience! Yes XY is excellent for ambience, though I have had some success with it fairly close to an acoustic guitar, and over a drum kit. I've some examples on line but they suffer from 128k mp3 streaming artefacts...

MS is not suitable for close miking.

ORTF is OK I prefer DIN myself. The stereo setup depends on the characteristics of the mics used, and it's an illusion anyway so a good mix of science and critical listening is required. We're assuming the room sounds nice as well...

Here's a good stereo miking primer:
http://www.dpamicrophones.com/en/Mic-University.aspx

[Sam Inglis]

MS is not suitable for close miking.

I'm curious as to why you say this... MS has an acceptance angle like any other stereo technique, but as long as you stay within that angle, why would it be worse than any other coincident technique? I've used MS on acoustic guitar before (with single point stereo mics) and it sounded OK to me. I like it a lot on overheads too.

[Tomás Mulcahy]

Hi Sam. If you're too close to the source it will be in the deaf part of the sides mic's lobes won't it? So you won't have much direct sound on the sides. To be honest I don't really understand the acceptance angle, but also there's a bit more to it than just that one parameter, I'd have thought. From experience, MS always comes out lop sided when positioned for the sweet spot of the guitar. I've only ever done it with two mics so maybe with capsules closer together in a single MS mic it might work better?

[Jack Ruston]

JMHO, but I would abandon the idea of a 'true' stereo technique for acoustic guitar. Use two mics, one near the neck and one near the body. Don't worry about spacing or any of that 'standard' stuff. See if you can get a nice phase relationship that sounds good panned. See what you think of it. If you're not convinced don't be afraid to go back to a single mono mic.

The guitar itself, the strings, their age and the technique in relation to those are going to have a far, far greater effect on the emotional impact of the recording than whether the guitar is stereo. It's causes complications when you sing at the same time. Try not to become too attached to the idea of this massive wide guitar sound.

Personally, I'd put a stereo ambient pair up, and then mic the voice and guitar each with one mic, looking for best rejection of the other source.

J

[Hugh Robjohns]

Hi Sam. If you're too close to the source it will be in the deaf part of the sides mic's lobes won't it? So you won't have much direct sound on the sides.

It doesn't work me that, and there are no restrictions or quality issues to using an MS array close to the source (other than proximity effects).

If you increase the gain of the side mic you narrow the acceptance angle and the image will be whatever you want, even when positioned close to the source.

To be honest I don't really understand the acceptance angle

The acceptance angle is the angle in front of the stereo mic array over which sounds can be captured and reproduced over the arc between the speakers. With XY cardioids, that angle is about 120 degrees. With fig-8s it is 90 degrees (with a second 90 degree span behind the mics).

In the case of an MS array, imagine the superimposed polar plots of mid mic (let's say cardioid) and side fig-8. If you draw straight lines from the coincident centres of the two mics to the points on either side where the curves of the two plots cross, that defines the acceptance angle. If you increase the gain of the side mic, the fig-8 plot enlarges and the crossover points move toward the front, narrowing the acceptance angle. Consequently a source in front of the ms array will take on a greater proportion of the stereo image.

H

[Sam Inglis]

From experience, MS always comes out lop sided when positioned for the sweet spot of the guitar. I've only ever done it with two mics so maybe with capsules closer together in a single MS mic it might work better?

True enough -- it does come out lop sided if you point it at the neck/body join -- but then as Hugh says, so will any coincident pair in that position, because one side is picking up the guitar body and the other isn't. I guess with XY you can get away with putting the mics nearer the sound hole because they're not pointing directly at it?

[The Elf]

Unless I am certain that an acoustic guitar is going to sit in a busy mix I usually try to get a stereo recording. I really don't like a mono-recorded guitar in isolation - to me it just sounds thin and 'wrong'. A 'stereo' recording of an acoustic, for me, is not about achieving a 'wide' guitar sound - I'm simply looking for something that 'blooms' between the speakers.

I use a variety of stereo techniques, depending on context, and I really don't mind if the image is a tad off to one side - usually that helps to leave space for other instruments, or the lead vocal. As long as the recording sounds 'honest' it doesn't make much difference to me.

My favourite technique is a pair of omnis at neck and body, as has already been described, but I'm not averse to M/S and ORTF - and I have even achieved decent results in XY with a pair of cardioids from time to time!

[Mike Senior]

I don't think XY cardioids are much use for anything other than ambient/atmospheric sound effects!

I think I'll frame that! Funnily enough, the main thing I like about XY is when you largely ignore the stereo issue and instead use the array as if it were a single mic with wider horizontal pickup. This is why I like it on acoustic guitar in particular.

Switching to omnidirectional mics (which have to be used as a spaced pair if you want a stereo effect) maintains a natural bass response at any distance, and also doesn't suffer the odd LF phase effects that afflict most cardioids.

A definite +1 on this! I've just done a Mix Rescue where I had to have a bunch of acoustic guitar parts rerecorded, and the results with an omni mic knocked spots off the cardioid-miked originals. Interestingly from your perspective, it was the MK012 omni capsules that were used, in conjunction with a Studiospares mic baffle to allow a more distant positioning without too much room sound.

Omni mics are lovely things, and rather underused in small studios.

[Mike Senior]

(other Han proximity effects)

You need to switch off your spell-checker's 'Star Wars' mode, Hugh...

[Mike Senior]

he acceptance angle is the angle in front of the stereo mic array over which sounds can be captured and reproduced over the arc between the speakers. With XY cardioids, that angle is about 120 degrees. With fig-8s it is 90 degrees (with a second 90 degree span behind the mics).

I thought the acceptance angle with fig-8s was a little narrower than that, given that you only need about 15dB L-R level difference for hard panning. Does that make sense? I seem to remember you get that kind of level difference at around 35 degrees off-centre with crossed fig-8s, theoretically speaking -- although of course practical realities are another thing entirely, given that real mics (especially large diaphragm condensers) will deviate quite considerably from the 'ideal' polar response.

[Mike Senior]

I guess with XY you can get away with putting the mics nearer the sound hole because they're not pointing directly at it?

That's an interesting perspective -- not one I'd considered. My instinct is that this would be more dependent on the XY array's mutual angle or the MS array's mid-side level than on the array itself. Even assuming matched acceptance angles, though, I'm not sure it would make that big a difference: it's the HF that always seems the first thing to suffer off-axis, rather than the LF. The greater proximity effect bass boost of the MS rig's fig-8 mic might be an issue, though.

Hmmm... <thinking cap on>

[Hugh Robjohns]

Omni mics are lovely things, and rather underused in small studios.

Absolutely. Let's mount a campaign for the return of first order mics: omnis and fig-8s. If they were good enough for Mr Blumlein, they should be good enought for all of us! Cardioids are the work of the devil!

H

[Hugh Robjohns]

(other Han proximity effects)

You need to switch off your spell-checker's 'Star Wars' mode, Hugh...

Arrgghh! I spend half my life fighting the bloomin iOS auto-correct nonsense. Can you turn it off on an iPad?

H

[Hugh Robjohns]

I thought the acceptance angle with fig-8s was a little narrower than that, given that you only need about 15dB L-R level difference for hard panning.

It is a little -- I was talking in general round figures to make the comparitive points -- and is obviously dependent to some extent on the specific mics in use and their mutual angle. A lot of people prefer to use a mutual angle of 80 degrees rather than 90 to provide a wider acceptance angle, for example.

H

[The Korff]

You can get rather decent Bluetooth keyboards for them for not much ££, which might help? I know that iOS is clever enough to hide the onscreen keyboard when you've got one connected, but not sure how it then handles the infuriating auto-correct feature.

first order mics: omnis and fig-8s

I've not heard this term applied to mics before, could you expand?

Cheers!

Chris

[Hugh Robjohns]

first order mics: omnis and fig-8s

I've not heard this term applied to mics before, could you expand?

The two fundamental ways of making a mic are pressure operation or pressure gradient operation -- omni and fig-8 polar patterns, respectively. These are referred to as 'first order microphones' because they are the basic building blocks.

Cardioids, hyper-cardioids, super-cardioids, and sub- or hypo-cardioids all combine proportions of both fundamental principles of operation to arrive at the desired polar pattern, and are therefore 'second order' microphones.

H

[The Korff]

Cheers Hugh!

[Sam Inglis]

I guess with XY you can get away with putting the mics nearer the sound hole because they're not pointing directly at it?

That's an interesting perspective -- not one I'd considered. My instinct is that this would be more dependent on the XY array's mutual angle or the MS array's mid-side level than on the array itself. Even assuming matched acceptance angles, though, I'm not sure it would make that big a difference: it's the HF that always seems the first thing to suffer off-axis, rather than the LF. The greater proximity effect bass boost of the MS rig's fig-8 mic might be an issue, though.

Hmmm...

Hmmm, yes, now you mention it I'm having my doubts. I can't remember the last time I tried crossed cardioids on acoustic guitar in any case. I generally use one mic at the neck/body join and one behind the bridge. If I use a stereo mic it's usually in MS, because then you have the option of canning the Sides component altogether and treating it as a mono recording.

[The Elf]

Cardioids are the work of the devil!

That's why the devil has all the best mic's!

I don't find much use for fig 8 beyond M/S, but I use omnis constantly. I can strongly recommend an MXL 603 with omni cap for melodica!

The big advantage for me is being able to get very close in without having to tame the bass bump you get with a cardioid. With solo acoustic guitar this is often imprtant to capture the feeling of intimacy.

When I do use XY cardioids on guitar I'm typically aiming the mic's at the top and bottom of the body. After that it's just about finding a height that balances out the resonance with the details - often I end up pointing up at the guitar. XY can work nicely, since it plays down the pick sound a little. I'll usually pull the panning in a little to help solidify the centre. This can work really well, and certainly gets a nice sound with my K184s.

[Sam Inglis]

I don't find much use for fig 8 beyond M/S, but I use omnis constantly. I can strongly recommend an MXL 603 with omni cap for melodica!

Funnily enough I think fig-8s are perfect for melodica because it benefits from some bass boost. Try a ribbon mic underneath it -- that way you get a nice rich sound and minimal clacking from the keys.