Possible Out of Phase issues

[Kenny B]

Recently, I started recording acoustic guitar with three mics. The middle one being panned Center with the outside mics hard left and right respectively. When I add the center mic into the mix, I seem to cause the entire balance to go off by -3 or 4 dB. I need to reduce the right side by that much in order to achieve an even level of sound between the left and right channel. And although the center mic seems to add in some flavor, I'm not sure it is even worth using unless I can someone get rid of what is happening. Any tools to use to see out of phase? And should I just not use the middle mic?

[Rick Cornish]

Hey Kenny B.......
What you're experiencing sounds like it might indeed be a phase issue due to the difference in distance from the different mics and your guitar.

A couple things to try:
• Record your mics to three individual tracks simultaneously, then zoom way into the waveforms in the Sequence Editor and see if the wave traces are going up and down together. If one is backwards from the others, try a different mic cable or see if it's reversed on your input/interface.
• After recording, try muting one of the recorded tracks at a time and see what happens to the level (particularly the low end) and stereo image. Phase issues are usually felt most at the low end where the wavelengths are longer.
• After recording, you could also try reversing the polarity of the center mic (DP has a "Invert Phase" plugin—which really just reverses the polarity) and see what that sounds like.
• Timing can also have an effect. I have been recording my guitar with one mic, plus a mic on an amp, plus a stereo feed from a preamp. All those devices take different amounts of time to get to DP, so sometimes I have to slip them forward or backwards a few samples to get them to line up.

(FWIW: When I hear people talking about something being "out of phase", its often really out of *polarity*—which is an electrical issue, like when something is wired backwards).

Good luck.

[mikehalloran]

You can't be in-phase with that setup except to reverse polarity on either the R or L and pot the C all the way down. This is one of my favorite mic arrays. Learn how to use it to your advantage.

Long before I could afford a figure 8 mic, I was recording M+S using three mics. One pointed straight ahead (the Mid) while the other two were 180° with the two diaphragms lined up and pointed hard R and L (the Sides). You do not need an M+S decoder with this setup. Back in the day, M+S was expensive: you needed a figure 8 and a matrix decoder (most of us built our own but there were issues with that: Passive knocked off -3dB while Active added noise).

If I had reduced these to two tracks, I could use the M+S plug-in, now part of DP's Trim plug, to adjust the L/R balance (or even reverse it) and balance it with the M. Since you have recorded to three tracks, no reason to use the Trim plug since you can do everything with the Volume and Pan controls — or the pots on mixers in the 1960s and '70s in my analog tape days.

My very first DP project in the '90s was to record a live concert in a church across the street from the Anchor brewery. About a half hour into it, the church sounded like it was under arial bombardment. The conductor didn't stop. At intermission, I ran outside expecting to see an industrial accident at the brewery but nope, it was the annual KFOG Radio "Kaboom!" fireworks show that they ran every May from a barge about 200 yards away in the San Francisco Bay — and my AKG mics + DP 2.7 recorded every nuance of every deafening blast.

Afterwards, my friend said he wasn't worried because he figured that I could somehow filter out the noise. My only thought was, "There are no filters than can fix this! I'm so screwed..."

Well, back to my Beige G3 Mac, I opened the project up and panned the three tracks hard L, hard R and C. It sounded just as bad as it did in the church. Yikes! On a whim, I potted the C (M) all the way down and centered the R and L channels and it was amazing: The noise was gone leaving only the music.

Since what sounded was WWII came from outside the church, it entered both S mics equally and cancelled out when I centered those channels. As I told my friend, the result was no longer stereo but it made the fireworks show disappear. Last year when I remastered those tracks for his box set, I used a low pass filter and compressor to add some of the treble back that I mixed in as stereo. This improved the sound while adding almost none of the objectionable noise (that I was to filter again).

Had it a) been a large ensemble with any players or singers in the middle, b) I hadn't recorded M+S, c) the mic array was further from the talent or d) the noise emanated from inside the church, then the results would have been closer to "I'm so screwed...!" I got lucky.

So, my gig was saved, my studio got a boost (word spread about what I did but no one knew how), I met Renée Fladen, one of the singers (the real life inspiration for the song, "Walk Away Renée") — and I got to see the rest of the fireworks show up close.

Had I used a figure 8 microphone for the S instead of a pair of cardioids at 180°, the results would have been the same if I thought to use the M+S plug-in the way that I used three volume controls and pan pots.

If you do the math on a 3-cardioid M+S array, it shouldn't work as well — but it does — and bumblebees aren't supposed to be able to fly.

[Rick Cornish]

M-S stereo is magic.

(OK—going just a little off-topic.)

20 years ago, I got hired by Target to follow their open wheel racing team (Target/Chip Ganassi Racing). (Hardly spent a summer weekend at home in 5 years.) Back then, we were still shooting in BetaCam SP, but my guys built a rig with an Ikegami camera and a mixer and camera shade device. Then we crammed a Sennheiser hypercartiod and figure 8 into the zeppelin and recorded the shotgun into the two analog audio channels (set hot and cold) and the bi-mic into the two digital audio channels. After we decoded the M/S tracks, the sensation of the cars going by was startling. Even the drivers were impressed when they saw the season-ending video.

[bayswater]

A couple things to try:

Thanks for posting this well thought out set of suggestions. It goes into my HowTo folder. I expect the MOTU Precision Delay plugin might also come in handy in these situations.

[Rick Cornish]

A couple things to try:

Thanks for posting this well thought out set of suggestions. It goes into my HowTo folder. I expect the MOTU Precision Delay plugin might also come in handy in these situations.

My pleasure, sir. Some hard-learned lessons there.

(And I agree, Precision delay is a good one.)

[Kenny B]

Rick and Mike, I really appreciate your input on this. It does look like I am out of phase based on what I'm hearing.

When doing an Invert Phase on the Center mic, everything appears louder for sure. Same thing if I invert phase on the Right channel. If I invert phase on both R and C, then the results are the loudest. Not sure how that makes sense. If I listen to what I recorded, I notice that the there is great stereo image, but poor sound quality. Sounds like the mics are too far apart... when I go to invert phase the sound is louder but sounds less like a stereo image, more mono. I know I should be checked the bass response but that is secondary to the volume that I'm hearing with inversed phase.

I also checked my mic cables. I am using prolink studio pro 1000 cable on the L and R, but my center mic is a homemade Neutrik cable that I made years ago. I have really no way of telling if it could the cable polarity causing the problem. I was thinking of just ordering another Prolink cable, they have the studio pro 2000 now. These cables of mine are at least 15 years old so the probably don't make the 1000 any more. I'm still going to dig up some older mic cables to see if one (not homemade) can make a difference.

I could send a small wav snippet of the three channels if anyone has time and inclination to give me their ideal of what is happening. I know there is a plug-in that can be downloaded that shows phase issues, not sure what the name of that is.

My center mic can do figure 8. For awhile I was using that in figure 8 with pointing horizontal to the sound source, picking up L and R sides but not really helping much or making much difference is sound quality.

Mics are Schoeps MK4 L, R. and Charter Oaks E700 for center.

[Rick Cornish]

I also checked my mic cables….

You could always just try switching cables to see if that makes a difference.

My center mic can do figure 8.

Just for fun, you could try a center+figure 8 (right on top of each other) and then decode the figure 8 as M/S stereo. I’m not in my studio today, but I *think* DP has a plug for that.

Good article here:
https://www.uaudio.com/blog/mid-side-mic-recording/


Sent from my iPhone using Tapatalk Pro

[mikehalloran]

Nice mics! The MK 4 a standard go-to M in an M+S array.

As long as that cable is wired correctly, it’s not the problem. 2 and 3 need to match on each connector so pop the shells and look. Don’t buy into the boutique cable marketing nonsense.

You are telling us how you are trying to solve a problem without telling us what that problem is. What are you trying to accomplish?

Any 3-mic array the way you have it is going to have polarity, phase and timing issues that cannot be "solved", only tweaked. A 3-mic M+S array should have the diaphragms lined up for best results. Only a Decca Tree has no polarity issues since all three are pointing the same direction—which leaves timing and phase. If you solved all three, the result would be the same as a single large Omni, so, if that’s what you want, just use one. Polarity and phase "issues" is how cardioid and other pickup patterns work, either by mechanical baffles (single diaphragm) or electronic manipulation of the back plate (dual diaphragm).

Physics — not just a good idea, it’s the Law.

I only use M+S on solo guitar if I was playing in a great sounding room. Being a longtime church music director, I’ve had access to many. If that accomplishes what you want, you’ll be in the market for a MK 8 capsule on Reverb where they show up all the time.

If you want to explore M+S, ditch one of the MK 4s, point the other straight ahead and set the other to Figure 8 and have fun.

You seem to be wedded to an idea that you haven’t shared and want our help to solve the resulting problems. Rick gave you a number of good suggestions for how to tweak your setup. I’ve tried to give a short course on the physics involved.

[mikehalloran]

Just for fun, you could try a center+figure 8 (right on top of each other) and then decode the figure 8 as M/S stereo. I’m not in my studio today, but I *think* DP has a plug for that.

I was using the M+S decoder in DP 2.7. As I posted earlier, it’s now part of the Trim plug-in (since DP 7 or 8?). DP has a separate Plug-ins manual but I’ve found it intuitive.

[Kenny B]

Nice mics! The MK 4 a standard go-to M in an M+S array.

As long as that cable is wired correctly, it’s not the problem. 2 and 3 need to match on each connector so pop the shells and look. Don’t buy into the boutique cable marketing nonsense.

Actually I was thinking of getting another Prolink cable. I dug out a 20 foot mic cord (not homemade) and used that in testing. Results were the same.

You are telling us how you are trying to solve a problem without telling us what that problem is. What are you trying to accomplish?

The problem to me is the sound. I don't like the sound I'm getting. I had previously used 3 mics and didn't have this problem with the right side giving an unbalanced (higher db) once I added in the third mic. Something is off and I don't know what the problem is. I want to figure out how to get the best sound in my studio but not sure how. I could do the M+S setup, haven't tried that in awhile.

Any 3-mic array the way you have it is going to have polarity, phase and timing issues that cannot be "solved", only tweaked. A 3-mic M+S array should have the diaphragms lined up for best results. Only a Decca Tree has no polarity issues since all three are pointing the same direction—which leaves timing and phase. If you solved all three, the result would be the same as a single large Omni, so, if that’s what you want, just use one. Polarity and phase "issues" is how cardioid and other pickup patterns work, either by mechanical baffles (single diaphragm) or electronic manipulation of the back plate (dual diaphragm).

"A Decca Tree setup uses three omnidirectional microphones arranged in a "T" pattern outlining a triangle, often equilateral; the center microphone is mixed with the two spaced microphones to fill the "hole in the middle" in their imaging; it points the sound source."

I would need three Omni mics and I don't have that.

Physics — not just a good idea, it’s the Law.

I only use M+S on solo guitar if I was playing in a great sounding room. Being a longtime church music director, I’ve had access to many. If that accomplishes what you want, you’ll be in the market for a MK 8 capsule on Reverb where they show up all the time.

If you want to explore M+S, ditch one of the MK 4s, point the other straight ahead and set the other to Figure 8 and have fun.

You seem to be wedded to an idea that you haven’t shared and want our help to solve the resulting problems. Rick gave you a number of good suggestions for how to tweak your setup. I’ve tried to give a short course on the physics involved.

I'm wedded to the idea that I really need to get the best sound possible with my setup. I want to keep experimenting to see if how that is possible. Maybe I only need 2 mics? My goal really is to record solo acoustic guitar. Should be simple right?

In the meantime, I have a few recordings that I did recently here in my studio of a guest (friend) who came over to record his guitar. This was recorded using the third middle mic but back further than the Schoeps, maybe a few inches. Wish I would have taken a pic of the setup but next time I will.

https://www.youtube.com/watch?v=kwgcAKVIY7E

https://www.youtube.com/watch?v=a-CSBPLPmBc&t=11s

The first one is a large body guitar and to me it sounds a bit weak in terms of bass. the second is an OM size body.

Better examples I have of the .wav files if anyone has time and inclination to listen to what is going on. I tried my best to record a short snippet using a baritone guitar to get more bass in order to better determine if out of phase is really causing havoc.

[Phil O]

Some have touched upon this, but just to be clear, time displacement and phase are related but NOT the same. Time displacement issues can sometimes be mitigated by inverting the phase but this is not necessarily indicative of an out of phase (i.e., 180 degree) track. One needs to be cognizant of the possible ramifications of time shifting or phase inverting (or both) of any recorded tracks. One needs to also consider time and phase of bleed, reflections, etc. It ain't that simple. As always, use your ears. If it sounds good, it sounds good.

Or as my daughter the chef always says, "Be careful. Hot things are hot."

Your friendly neighborhood Phil

[Kenny B]

I spent this morning recording acoustic guitar with M+S setup. I actually sounds a lot better. One less channel to worry about as well taking up space on the hard drive. I also set the MS Decoder a .83 so it is not full 100% stereo mix. I'm amazed to how much better it sounds.

My next thing would be to try and go back and fix some recordings I did with the three mics with a guitar I know longer have access to but did 3 weeks of recordings with it. It might be nice to hear those better with just playing with either the phase (inverted) or the time. Not sure how I would do that.

[mikehalloran]

In DP, you can invert the waveform of any track using the Invert Phase plug-in, pg 90 of the Plug-ins manual. I would make a copy of the track first so that you have another one track uninverted.

You can also pan your two S tracks hard R/L, then bounce to a single Mono track. Now, pull up the Trim/M+S plug and make that your Side. Since you can accomplish the same thing with Pan and Volume controls if you have three or four tracks to play with, I wouldn't.

There are many ways to record a guitar. An oldie but goodie is to place one cardioid in front and another behind with the polarity reversed (or inverted in post as above)

There's also X/Y and its sibling, near coincident—both popular, ORTF (I wouldn't on a solo guitar) and Blumelein with a pair of Figure 8s (in a nice sounding room). Others will point a mic at the soundhole and another at the neck—also popular.

I'm in the Mono if it's going to be in a multi-track; Stereo if it's going to be a solo guitar school but nothing is sacred. Were that so, I never would have attempted 3-mic M+S with a trio of SONY ECM-32/32Ps in the 1970s. Not only couldn't I afford a Figure 8 but I had to use some weird batteries because my mixer didn't have phantom power.

[Kenny B]

Did a little more research with Phase/Time issues and called on a buddy of mine who records acoustic guitar. Apparently there is a plug-in called Auto-Align 2 that will analyze your tracks and correct Phase and Time issues on the raw tracks before you mix or bounce tracks.

https://www.soundradix.com/products/auto-align/

Just wondering if anyone has experience with this method. I have now downloaded the trial version and applied it to my three mic tracks as well as the M-S tracks I just recorded yesterday. Might be the ticket, like autotune for microphones.