Close phased pair for recording guitar - why no obvious phase issues?

All about the tools and techniques involved in capturing sound, in the studio or on location.

Moderator: Moderators

Close phased pair for recording guitar - why no obvious phase issues?

Post by forumuser915213 »

I haven't played much acoustic guitar for years, but recently decided to buy a small-bodied one that tuck away in a corner of the living room and doesn't need to be put away.

The experiment has worked - I'm really enjoying playing it on a whim, and it helps that it's a sweet sounding instrument.

Naturally, my mind turned to recording it. Up to now I've simply aimed an ldc or sdc at the 12th or 14th fret of my old dreadnought and been quite happy with the result, but this time I decided to read and view around the topic.

In doing so, I found this video. https://youtu.be/qggDfSVkU0o

Around 5m50s, there's a demonstration with two TLM107s in figure of 8, just 20cm apart, are set up so the singer-guitarist's voice is in the null (he's singing into a third TLM107).

The great sounding video demo was obviously done with an upmarket guitar and performer and posh Neumann mics, but it sounded good to me, so I tried the same trick using a pair of CAD M179s in roughly the same position - and was delighted with the results. I may do a demo; folks here may or may not agree.

But I don't understand why it works. With two cardioid mics 20cm apart in clear contravention of the 3 to 1 rule, why doesn't this set up produce very audible comb filtering phase problems? Does anyone know? Or are they there, but perhaps don't matter for some reason?

I notice that this arrangement is very like the Faulkner array, though I believe that is intended for micing from one end of a long room, which is clearly a very different proposition.

Gavin
forumuser915213
Regular
Posts: 143 Joined: Fri Jan 26, 2018 1:08 pm Location: Kent

Re: Close phased pair for recording guitar - why no obvious phase issues?

Post by Wonks »

The 3-1 rule is for when you are miking different instruments and getting spill between mics.

Here you have one instrument, and the mics are the same distance from the instrument. As a result, the lower frequencies will be pretty well phase-matched. It’s the higher frequencies where small distance differences will give bigger phase shifts, especially as like most instruments, a guitar produces sound from all over the body, it’s not a point source. But it’s those phase and level differences that give you the wider stereo image. You are hearing what comb-filtering there is, but that is why you get the stereo soundstage. If you picked up exactly the same sound on identical mics with everything in phase, there wouldn’t be any point in using two mics.
User avatar
Wonks
Jedi Poster
Posts: 17003 Joined: Thu May 29, 2003 12:00 am Location: Reading, UK
Reliably fallible.

Re: Close phased pair for recording guitar - why no obvious phase issues?

Post by Tim Gillett »

Also the timing differences dont cause much comb filtering to our ears, because the two guitar signals remain separated. If the two guitar mic signals were hard summed to mono, audible cancellation would be more likely. To be truly mono compatible maybe the two mics rather than three would be the better option.

For me the great takeaway is the use of the nulls from the two bipolar mics for
effective separation of voice and guitar.
Tim Gillett
Frequent Poster (Level2)
Posts: 2625 Joined: Wed Jan 30, 2013 12:00 am Location: Perth, Western Australia

Re: Close phased pair for recording guitar - why no obvious phase issues?

Post by Hugh Robjohns »

forumuser915213 wrote: Tue Aug 03, 2021 11:33 pm But I don't understand why it works. With two cardioid mics 20cm apart in clear contravention of the 3 to 1 rule, why doesn't this set up produce very audible comb filtering phase problems? Does anyone know? Or are they there, but perhaps don't matter for some reason?

Firstly the '3:1 Rule' is really a guide rather than a 'rule'. ;-)

Secondly, it is a guide to minimising unwanted spill between mics. The mic you want to pick up a source is one distance from it, the other mic that you don't want to pick up that source has to be three times further away so that what it does pick up is so much lower in level that it's not going to cause a problem in the mix.

In your stereo guitar situation, though, you want both mics to hear the same guitar, so the 3:1 guide doesn't apply at all.

What Neumann have demonstrated here is a 'near-spaced stereo pair' -- just one of a great many variations on the theme. They are using the fig-8 pattern here because they are trying to minimise the vocal pickup, but for a solo instrument ('cello, for example) it would be more typical to use omnis in a similar configuration.

With fig-8s on a 20cm spacing, the stereo recording angle (SRA) is potentially enormous -- close to 180 degrees -- and it generates a combination of both amplitude difference and time of arrival difference between the two channels, which is what provides a stereo image, albeit one with rather vague and ill-defined imaging.

The parallel fig-8 array with 20cm spacing is attributed to Tony Faulkner, as you say, who resorted to it in a chapel with particularly troublesome side reflections. He called it a 'phased array' after a superficially similar arrangement of parallel aerials used in some directional radio systems, but the name is really a misnomer and misleading to many who hear the 'phased' bit and think of that familiar whoosing effect!

Any situation involving spaced mics can cause audible phasing -- or more correctly, comb-filtering coloration -- issues, where some frequencies add and some cancel when the two channels are summed together to mono but the critical aspect here is the scale of the time of arrival differences: the greater the difference, the lower the first comb filter notch.

In this case, the breadth and proximity of the sound source in front of the mics, combined with the relatively narrow spacing of the mics, means that there isn't a sufficiently strong time of arrival difference to cause audible problems, and what phase cancellations there might be will only tend to be at very high frequencies anyway.

At the end of the day, if you like the sound it gives in stereo AND mono -- and they will be slightly different of course -- that's fine. Near-spaced pairs like this (usually with different polar patterns) are very commonly employed for solo instrument recordings.
User avatar
Hugh Robjohns
Moderator
Posts: 38835 Joined: Fri Jul 25, 2003 12:00 am Location: Worcestershire, UK
Technical Editor, Sound On Sound...
(But generally posting my own personal views and not necessarily those of SOS, the company or the magazine!)
In my world, things get less strange when I read the manual... 

Re: Close phased pair for recording guitar - why no obvious phase issues?

Post by forumuser915213 »

Thanks folks! I'll use it with more confidence in future!

I have my channels hard left and right, but with the volumes roughly adjusted to make the image appear to be centred.

A concern in my amateur mind, as you'll probably realise, is the thought that in recording something (a project, say) in what seems to be an unusual way (judging from what I see in articles, YouTubes etc), I'm in danger of making a recording that sounds good to me now but which I'll come to dislike later.

Experience teaches that this could be for some reason I'm not yet aware of but could become keenly aware of later... Oh, the embarassment...

Meanwhile, here's the demo I mentioned. https://soundcloud.com/gmatkin/m179-pha ... uitar-demo

One thing I do get from this is the importance of staying still!

Gavin
forumuser915213
Regular
Posts: 143 Joined: Fri Jan 26, 2018 1:08 pm Location: Kent

Re: Close phased pair for recording guitar - why no obvious phase issues?

Post by forumuser915213 »

I've also just heard something striking... I centred both channels and got something that sounds just slightly odd.

Back on hard left-right and it's nice on headphones (I think), and still pleasant on my speakers. Do we think much about mono much these days?

I wonder what would happen if the two mics were turned inward to the same target on the guitar? (I've just tried: the answer is that it's horrible!)

Gavin
forumuser915213
Regular
Posts: 143 Joined: Fri Jan 26, 2018 1:08 pm Location: Kent

Re: Close phased pair for recording guitar - why no obvious phase issues?

Post by Hugh Robjohns »

forumuser915213 wrote: Wed Aug 04, 2021 10:50 am I've also just heard something striking... I centred both channels and got something that sounds just slightly odd.

Er... yes... that's the comb-filtering coloration we spoke of...

Do we think much about mono much these days?

Well, I do...

I guess it really all depends on your audience and how you expect them to hear your music.

I'd like to assume that most people would listen on a half-decent stereo with the speakers placed 60 degrees apart, or with stereo headphones or earbuds, or in a car...

But the reality is that the world's best selling digital radio has a single mono speaker. And rapidly increasing numbers of people now have a single Alexa (other voice-controlled speakers are available) speaker in each room through which they listen to Spotify etc... in mono!

Moreover, if the FM radio reception is poor in a car the radio switches automatically to mono, and a surprising number of night clubs and stage PA systems are run in mono...

So while I'd certainly optimise my recordings for stereo listeners, I'd still want to make absolutely sure that it still sounds acceptable in mono -- recognising that mono will always sound different to stereo.

I wonder what would happen if the two mics were turned inward to the same target on the guitar?

Try it and see...

I suspect it will change the tonality and stereo imaging slightly, but the 'slight oddness' is a factor of the spacing between the capsules. If, by turning the mics inwards, you also bring the capsules closer together you will reduce the comb-filtering slightly, as well as the stereo imaging.

The only way to guarantee fully mono-compatible stereo is to use a coincident mic array of some form, rather than any kind of spaced array.
User avatar
Hugh Robjohns
Moderator
Posts: 38835 Joined: Fri Jul 25, 2003 12:00 am Location: Worcestershire, UK
Technical Editor, Sound On Sound...
(But generally posting my own personal views and not necessarily those of SOS, the company or the magazine!)
In my world, things get less strange when I read the manual... 

Re: Close phased pair for recording guitar - why no obvious phase issues?

Post by forumuser915213 »

Good points Hugh.

To be fair, in general I don't expect very many people to listen to what I record and I don't have much idea who they would be or what they would use. I suspect they're mostly friends and family, but I'd still rather avoid future embarrassments ;-)

I'm thinking that while YouTubes are normally played on some stereo medium, recordings without video may be listened to in mono.

If so, I should probably use a different approach to guitar depending on the medium. Maybe the phased pair for YouTubes and single M179 in figure of 8 for other media.

(I'll ignore the tiny speakers in phones and tablets for now, since they generally sound pretty poor, whatever the recording quality.)

Gavin
forumuser915213
Regular
Posts: 143 Joined: Fri Jan 26, 2018 1:08 pm Location: Kent

Re: Close phased pair for recording guitar - why no obvious phase issues?

Post by Drew Stephenson »

If you're in the mood for experimenting I'd suggest having a play with a mid-side combo. It's generally my starting point for guitar now and has the advantage that it's very mono compatible (you just get the mid signal).
User avatar
Drew Stephenson
Jedi Poster
Posts: 24450 Joined: Sun Jul 05, 2015 12:00 am Location: York
(The forumuser formerly known as Blinddrew)
Ignore the post count, I still have no idea what I'm doing...
https://drewstephenson.bandcamp.com/

Re: Close phased pair for recording guitar - why no obvious phase issues?

Post by forumuser915213 »

I've liked M/S when I've used it, but I'm not sure I could use it recording together with a singer. You may know better! Does it work well with a voice close by?

Most of my acoustic guitar playing is with either my voice or my wife's, and neither of us would be comfortable recording separate tracks.

Buckets of either voice typically spills into my guitar mic unless it's very close, which for obvious reasons means fierce equing with the cardioid mics I've got. It's true that the final result of that equing isn't necessarily bad, but it isn't something I feel good about doing. The best single mic position I've found so far is pointed slightly downwards at the upper bout of my guitar, just above my knee when playing - which places it as far from either singer as possible.

Since I think the CAD M179 is quite nice on the guitar in figure 8 (clock my sample above), I think one option is to try that in different positions.

Gavin
forumuser915213
Regular
Posts: 143 Joined: Fri Jan 26, 2018 1:08 pm Location: Kent

Re: Close phased pair for recording guitar - why no obvious phase issues?

Post by Hugh Robjohns »

forumuser915213 wrote: Wed Aug 04, 2021 12:52 pm I've liked M/S when I've used it, but I'm not sure I could use it recording together with a singer. You may know better!

It can be done!

https://www.soundonsound.com/techniques ... ides-array

You need a fig-8 mic facing sideways as the Sides element, and then two other mics for the voice and guitar. If they are also fig-8s you an get amazing separate of voice and guitar, rigging them in the same way as that Neumann video.

If you right the voice and guitars mics at right-angles per that video, then place the Side mic in the V notch behind them, as close as you can get it to both mic's capsules.
User avatar
Hugh Robjohns
Moderator
Posts: 38835 Joined: Fri Jul 25, 2003 12:00 am Location: Worcestershire, UK
Technical Editor, Sound On Sound...
(But generally posting my own personal views and not necessarily those of SOS, the company or the magazine!)
In my world, things get less strange when I read the manual... 

Re: Close phased pair for recording guitar - why no obvious phase issues?

Post by forumuser915213 »

Thanks again. I guess you realise that my concern would be that if accompanying a singer, the 'sides' mic could be pointing at them.

I'll give it a go, once I've got my current writing project done.

I suppose it doesn't matter if one of the three figure-8 mics is different? I'd guess the way to go might be need to use the same mics on the vox and guitar, and something different on the sides. I've got two M179s and two sE4400as.

Gavin
forumuser915213
Regular
Posts: 143 Joined: Fri Jan 26, 2018 1:08 pm Location: Kent
Post Reply