"massive phase errors"

austin · Post by **austin** » Fri May 27, 2011 9:00 am

I just read Larry's Welcome column in the new issue (#83), and I'm trying to figure out how/why this "180 degrees out of phase" issue would happen so often. I'm sure it's just a failure of imagination on my part, but I can't think of a situation in which I would accidentally have the exact same signal in both speakers, 180 degrees out of phase... What are people trying to do when they accidentally make this happen? (Honest question... I'm now paranoid that whatever it is, I'm going to do it!)

farview · Post by **farview** » Fri May 27, 2011 2:30 pm

IF you take a stereo widener plugin and set it as wide as it will go, the two sides will be out of phase with each other.

All kind of effects, like stereo choruses have the left and right out of phase with each other.

The stereo-ness of some keyboard patches is the same thing.

There are lots of things like that as well as poor mic placement when using multiple mics on the same source.

Scodiddly · Post by **Scodiddly** » Fri May 27, 2011 4:31 pm

There can be a variety of causes - miswired cables lying around (somebody fixed one end, didn't bother to check that they got the color code right), different gear from different eras of "pin x is hot", not checking the channel strip polarity button when tracking, etc.

Brett Siler · Post by **Brett Siler** » Fri May 27, 2011 5:56 pm

I have a pair of Oktava 319's and they are both wired the opposite of each other. When I put them up, say as a stereo pair on overheads, one is 180 degree outta phase.

Theo_Karon · Post by **Theo_Karon** » Sun May 29, 2011 12:47 am

The key distinction to keep in mind here is whether you're summing the phase incoherence in your mix or not, and to what degree. In other words: are the asynchronous elements in your mix panned, and how far? Checking the mix in mono consistently and repeatedly will reveal significant phase incoherence between panned signals and allow you to correct them before they become problematic. The issue discussed here arises when multiple inputs receiving information from the same event are out of phase, or put more simply, when that information is received later by one input than the others(s), and these input sources are panned to different positions in the stereo field- or, when, yahweh forbid, a source is panned opposite to its duplicate with polarity reversed. (Exception: M/S! And that only used wisely!)

Multiple mics on one source will, unless perfectly equidistant from the source, always produce waveforms that cancel and reinforce each other at different points across the spectrum, dependent entirely on the frequencies produced by the source. If the signal from all mics is panned dead center, the cancellations and reinforcements are occurring equally in both channels of your stereo mix, as both L and R are in effect a mono summation of all input signals. in essence you are doing nothing more than shaping the overall tone by correcting phase discrepancies in this situation. It is perfectly acceptable to sum the signals of two mics on one source and then move the mics until a pleasing tonal balance is achieved.

Problems arise when phase-incoherent signals are panned apart, as the sum at each output, left and right, will not reveal cancellation and reinforcement occurring between two panned signals. This can lead to significant loss of information when the stereo mix is summed to mono- extreme cancellations can cause disappearances when summed to mono, or cause the groove cut in a lacquer master for vinly to almost completely close on itself, producing very ugly results.

Another thing to keep in mind is the often misunderstood difference between reversing the polarity of a signal and turning a signal 180 degrees out of phase with another signal. The simplest way to think of it is this: polarity reversal refers to the AMPLITUDE of a signal. Polarity reversal means that peaks become troughs, and troughs become peaks; speaker/mic diaphragm excursions become incursions, and incursions become excursions. Polarity reversal is not at all frequency-dependent- any two identical waveforms of equal amplitude will sum to silence if the polarity of one signal is reversed.

PHASE refers to the time differential between two signals, and is entirely frequency-dependent. One cannot 'reverse the phase' of a single signal, as phase refers to the relationship between two (or more) inputs receiving sound from the same source. Phase coherence can be achieved by ensuring that two microphones are perfectly equidistant from a single source- which is, of course, impossible to implement perfectly in the real world. Polarity reversal switches present us with two options regarding the interaction of different waveforms representing one source, one of which is often preferable to the other; don't make the mistake of thinking the better-sounding option is entirely "in phase," though! That it most certainly isn't! Great scott!

You need a Little Labs IBP for that.

Don't worry, I can't justify the expense right now either.

Theo_Karon · Post by **Theo_Karon** » Sun May 29, 2011 1:02 am

^
drunk post.

Buffalo Trace is good bourbon.

tonewoods · Post by **tonewoods** » Sun May 29, 2011 9:47 am

Keep drinkin'...

Nice post....

rhythm ranch · Post by **rhythm ranch** » Sun May 29, 2011 12:09 pm

Theo_Karon wrote:^
drunk post.

Damn. You must br really good when you're sober!

Theo_Karon · Post by **Theo_Karon** » Sun May 29, 2011 11:29 pm

Oh jesus. I have no recollection of writing either of those posts.

The Scum · Post by **The Scum** » Sun May 29, 2011 11:46 pm

Check out the guitars on the new Mike Watt record. Awesome work by Tom Watson.

They sound super wide. And they disappear in mono.

cgarges · Post by **cgarges** » Mon May 30, 2011 6:05 am

Theo_Karon wrote:Buffalo Trace is good bourbon.

Always nice to learn something new on here!

Chris Garges
Charlotte, NC

lionaudio · Post by **lionaudio** » Mon May 30, 2011 9:19 pm

Theo_Karon wrote:^
drunk post.

Buffalo Trace is good bourbon.

I'm taking the tour of the Buffalo Trace Distillery later this year.. and yeah.. it's hella good. Have you tried the Buffalo Trace Bourbon Cream yet? I'm not sure if they have released it outside of KY yet, but it kicks the crap out of Bailey's

lionaudio · Post by **lionaudio** » Mon May 30, 2011 9:20 pm

Oh and, phasey, phase, things in phase, and stuff

jgimbel · Post by **jgimbel** » Mon May 30, 2011 11:31 pm

"Massive Phase Errors" would make a good band name, but I wouldn't be excited to record them.

jnTracks · Post by **jnTracks** » Tue May 31, 2011 3:44 am

jgimbel wrote:"Massive Phase Errors" would make a good band name, but I wouldn't be excited to record them.

/approving nod...

Message Board

"massive phase errors"

"massive phase errors"

Who is online