Phase question

general questions, comments and ideas about recording, audio, music, etc.
fifthstory
pluggin' in mics
Posts: 32
Joined: Thu Apr 08, 2004 2:56 pm
Location: Greensboro, NC
Contact:

Phase question

Post by fifthstory » Sun Feb 13, 2005 2:10 pm

I understand the idea of phase relationship. The idea that a sound wave caught at two different spots (mic placement) can be destructive depending on how the physical wave is shaped at the two spots.

I was under the impression that when using a DAW you could physically move a track to line up with a second similar track in order to remove the phase issue.

The other day when I was doing this, i tried both moving the wave and inverting the phase on one track. It seems that inverting the phase worked better. Is there a reason for this sort of thing. Is there something happening before the tracks hit the monitors that makes moving the tracks not work.

I do realize that better mic placement is the true answer, but sometimes you just gotta through em up and go.

thanks,
Chris

User avatar
andyg666
re-cappin' neve
Posts: 669
Joined: Tue Jun 17, 2003 9:25 am
Location: Witchtown, MA
Contact:

Re: Phase question

Post by andyg666 » Sun Feb 13, 2005 2:23 pm

i think moving the tracks sometimes sounds weird because you're altering the left/right arrival-time differences that your ear expects to hear from a stereo recording with a spaced pair of microphones. chackin the phase on all open mics and flipping the phase if neccessary before you record (if you have the option) is the best way to keep phase issues out of your hair.

User avatar
gabriel
ass engineer
Posts: 48
Joined: Sat Oct 23, 2004 2:49 pm
Location: San Francisco, CA
Contact:

Re: Phase question

Post by gabriel » Sun Feb 13, 2005 8:28 pm

Also keep in mind that a condensor mic's signal will be 180 degrees out of phase with a dynamic, so if one of your mics is condensor and one dynamic, that would explain why you needed to invert one channel.

-gabriel

User avatar
assfortress
steve albini likes it
Posts: 388
Joined: Sun May 11, 2003 11:45 pm
Location: salt lake city, ut

Re: Phase question

Post by assfortress » Sun Feb 13, 2005 8:31 pm

gabriel wrote:Also keep in mind that a condensor mic's signal will be 180 degrees out of phase with a dynamic, so if one of your mics is condensor and one dynamic, that would explain why you needed to invert one channel.

-gabriel
why is that?
"It?s the consequence you?ll pay, as long as you got the sound of it blowing up on tape."

User avatar
soundguy
ghost haunting audio students
Posts: 3182
Joined: Wed May 07, 2003 12:50 pm
Location: NYC
Contact:

Re: Phase question

Post by soundguy » Sun Feb 13, 2005 8:36 pm

that is so not the case.

dave
http://www.glideonfade.com
one hundred percent discrete transistor recording with style and care.

User avatar
gabriel
ass engineer
Posts: 48
Joined: Sat Oct 23, 2004 2:49 pm
Location: San Francisco, CA
Contact:

Re: Phase question

Post by gabriel » Sun Feb 13, 2005 8:53 pm

My understanding (this is what I was taught) is that a dynamic mic converts the movements of the diaphragm directly into voltage, whereas a condensor mic senses the change in capacitance between the diaphragm and a backplate, creating an inverted signal. I dunno. Could be wrong. Anyone with more specific knowledge care to confirm/deny?

-gabriel

User avatar
Mark Alan Miller
dead but not forgotten
Posts: 2097
Joined: Wed Apr 14, 2004 6:58 pm
Location: Western MA
Contact:

Re: Phase question

Post by Mark Alan Miller » Sun Feb 13, 2005 9:11 pm

Depends on how *any* given mic is wired.
I have three of one mic model, identical except that one of them is pin 3 hot instead of pin 2, and thus is "out of phase" with the others. (It is polarity flipped, to be exact...)
he took a duck in the face at two and hundred fifty knots.

http://www.radio-valkyrie.com/ao/aoindex.htm - download the new record (free is an option!) or get it on CD.

User avatar
soundguy
ghost haunting audio students
Posts: 3182
Joined: Wed May 07, 2003 12:50 pm
Location: NYC
Contact:

Re: Phase question

Post by soundguy » Sun Feb 13, 2005 10:15 pm

most mics use a transformer on the output and that transformer can be wired to accomodate for whatever polarity the designer wants to present to the world.

dave
http://www.glideonfade.com
one hundred percent discrete transistor recording with style and care.

Zoltar
gettin' sounds
Posts: 132
Joined: Thu Jan 13, 2005 10:39 pm
Location: Ontario Canada

Re: Phase question

Post by Zoltar » Sun Feb 13, 2005 10:36 pm

As far as sliding tracks around, this is the same as moving mics farther away from the source. This will help change some phase issues, but if you are trying to correct say, drums in two mics, lining up the snares (which you can do visually) might wack the bass drum out. Also, you should note that doing this might correct phase issues at one frequency, but create phase issues at other frequencies.

Inverting is the same, but different.

What you can do is draw yourself some nice little sign waves. use 13,500 inches per second for the speed of sound at room temp. this gives you a 13.5 inch long sound wave at 1000Hz. so if you have two mics 6.75 inches apart, draw them on your sign wave. They will always cancel each other out on the 1000Hz wave. Now for 2000Hz, you have a 6.75 inch wave length. These same two mics will always add to double the actual signal. I wish I could draw it here, it isn't hard to see, but it is hard to explain.

digdug
pushin' record
Posts: 296
Joined: Thu May 20, 2004 1:27 pm
Location: Dallas, TX

Re: Phase question

Post by digdug » Sun Feb 13, 2005 11:06 pm

Hijacking a bit:

I've always wondered about the 180 flip on the mic UNDER the snare when you've got one on top, too. I understand that in principle, the drum head is pushing towarded one and away from another, but does that really matter if on mic is, let's say, over three time the distance to the head that the other one is? My top mic is considerably closer than the under mic, so does it still make sense to flip the phase?

I know, I know, just listen and see.

Or better yet, just ditch the bottom mic.

Just curious.

User avatar
The Real MC
steve albini likes it
Posts: 399
Joined: Fri Feb 04, 2005 6:50 am
Location: Tranquil secluded country
Contact:

Re: Phase question

Post by The Real MC » Mon Feb 14, 2005 11:06 am

I was under the impression that when using a DAW you could physically move a track to line up with a second similar track in order to remove the phase issue.

The other day when I was doing this, i tried both moving the wave and inverting the phase on one track. It seems that inverting the phase worked better. Is there a reason for this sort of thing. Is there something happening before the tracks hit the monitors that makes moving the tracks not work.
It worked better because of the differences between working in time domain (shifting audio by time) versus phase domain (shift audio by phase).

A phase shift of all the harmonics in the audio is not the same as a time shift of audio. Inverting a 40hz harmonic is a time shift of 12.5 milliseconds, but inverting a 4Khz is a time shift of 125 microseconds. If you time shift both the 40hz and 4Khz harmonics by 125 microseconds then the 4Khz harmonic will be inverted but not the 40hz harmonic.

User avatar
8th_note
buyin' gear
Posts: 524
Joined: Tue Oct 19, 2004 4:58 pm
Location: Vancouver, WA
Contact:

Re: Phase question

Post by 8th_note » Mon Feb 14, 2005 5:30 pm

Good explanation of phase vs. time, MC.

I line up the drum tracks and invert phase (polarity) on my DAW. I line everthing up with the top snare. When I expand out the wave form so that I can move the tracks to match the attack of the top snare I can see if the polarity is inverted. I've wondered the same thing as digdug - why would the bottom snare signal have reversed polarity? But it always does. And it sounds a lot better if I invert the polarity of the bottom snare track.

To further the mystery, sometimes one of my overheads will have reversed polarity. I can see this in the waveform and if I invert it, it has the expected sonic effect. My only explanation for this is that maybe I have a mic cable that is wired wrong because my mixer doesn't have polarity switches. I've checked both mics and they are the same polarity.

Reversing the polarity of a track has a bigger effect if the tracks are lined up in my experience.

Regardless of theory, I can say that lining up the drum tracks and reversing polarity when necessary has a substantial positive impact on the sound. Over the last couple projects I've tended to move the overheads farther away to get a better cymbal sound and lining up the tracks to the snare really helps tighten up the sound. It's like focusing a camera. It's one of the nice advantages of a DAW IMO.

User avatar
NeglectedFred
pushin' record
Posts: 248
Joined: Tue Jan 18, 2005 6:53 pm
Location: Salt Lake City

Re: Phase question

Post by NeglectedFred » Mon Feb 14, 2005 7:33 pm

Usually what I do to avoid phase problems is concentrate each mic on certain frequencies..

For example, lately I've really been enjoying the sound of a 57 inside the kick and a D112 out. I use a HPF on the 57, and roll the hi's off starting at about 5khz, whereas I boost the sub freqs on the D112, scoop the mids until there's practically none, and boost above 5KHz. I get a mid range slap in the face from the 57, while the D112 moves the ground with a nice counter-balance at the top end.

I suppose using my method could still pose some threats with harmonic overtone comb-filtering, but it sounds good to my ears, so why get science overly involved?
I eat glue.

User avatar
Mark Alan Miller
dead but not forgotten
Posts: 2097
Joined: Wed Apr 14, 2004 6:58 pm
Location: Western MA
Contact:

Re: Phase question

Post by Mark Alan Miller » Mon Feb 14, 2005 8:59 pm

NeglectedFred wrote:Usually what I do to avoid phase problems is concentrate each mic on certain frequencies..

For example, lately I've really been enjoying the sound of a 57 inside the kick and a D112 out. I use a HPF on the 57, and roll the hi's off starting at about 5khz, whereas I boost the sub freqs on the D112, scoop the mids until there's practically none, and boost above 5KHz. I get a mid range slap in the face from the 57, while the D112 moves the ground with a nice counter-balance at the top end.

I suppose using my method could still pose some threats with harmonic overtone comb-filtering, but it sounds good to my ears, so why get science overly involved?
Watch out for the phase-havoc that severe boosts/cuts withmost designs of eq can play... Very few eq designs are truely phase-accurate/linear.
But as you say, if it sounds good, it's good. :)
he took a duck in the face at two and hundred fifty knots.

http://www.radio-valkyrie.com/ao/aoindex.htm - download the new record (free is an option!) or get it on CD.

User avatar
NeglectedFred
pushin' record
Posts: 248
Joined: Tue Jan 18, 2005 6:53 pm
Location: Salt Lake City

Re: Phase question

Post by NeglectedFred » Mon Feb 14, 2005 9:18 pm

Not to hijack the direction this thread is currently going, but does anybody ever use phase cancellation as a tool (other than removing vocals)..

There's a Nine Inch Nails song recorded live (no audience, though) where it's just Trent singing, a piano, and an accoustic guitar. It sounds like a fairly small room. Trent begins to get emotionally loud towards the end. As he does, you can hear certain mics picking up the vocals more than others (probably piano mics) while phase problems become pretty hairy -- but I like it. It sounds 3D-ish.. While the center channel becomes a little quieter, you hear more room ambience as he raises his voice..

Most engineers I think would have strived to correct the problems without hesitation (myself included), but I love it more and more every time I hear it.

BTW, in case anyone is interested, it's a nice little version of "Something I can never have".
I eat glue.

Locked

Who is online

Users browsing this forum: No registered users and 30 guests