phase position

[vanishbox3]

i'm doing a project for my Musical Acoustics class and i didn't even bother to research this topic to see if it could be done!
please give me any tips you can on getting mics in the WRONG phase position.
my goal is to make whatever i'm recording disappear...
i guess when mics are in a bad phase position, all you record is hiss... so i was going to demonstrate this for my class
how should i set up my mics? i'm using two condensers... should i be using dynamics? i really need some advice here. i feel very stupid at the moment

 

[scrubs]

Getting two mics to be completely out of phase with each other is just about impossible since the capsules are not in the same place and will exhibit variable phase across the frequency spectrum, I believe. Your best bet might be to record something with a figure 8 mic (two diaphragms pointing in opposite directions. Then, copy the track and invert the polarity (phase) of the copied track. Pan one track hard left and one hard right. You'll hear sound. Then, pan them toward the center and the sound will disappear.

 

[DonGraham]

Getting 2 mics perfectly out of phase is probably just as hard as getting 2 mics perfectly in phase............either move the room mic round till it sounds the worst/weakest (preferably using a noise generator) or use your college math to figure out the incorrect distance per frequency per time that you need.........For instance:

If you are recording a 1kHz sine wave.
Use phase velocity=frequency*wavelength where frequency=1/period
Theoretically if you can place the 2 mics at the minima AND the maxima of the waveform the 2 waves will cancel out.
For phase as a function of (distance, time):...use this:

Phi(x,t)=(2*pi*t/T-2*pi*x/lambda+phi(0))...where x=distance, T= period, t=time phi(0)= reference phase, lambda= wavelength....


good luck...

 

[northsiderap]

Testing

If you had a nice anechoic chamber, you could use two closely matched mics to record a sine wave or some white noise.

Move one of the mics around a little bit while recording....

Record the mics to two separate tracks, and use FFT analysis to show that certain frequencies are being added or subtracted based solely on the position of the mic...

Or mix them equally to mono if that's all you have available, then analyze..

You are looking for a freq. response that deviates from flat.

 

[SonicClang]

use your college math

Phi(x,t)=(2*pi*t/T-2*pi*x/lambda+phi(0))...where x=distance, T= period, t=time phi(0)= reference phase, lambda= wavelength....

You FREAK!!!!!! Holy crap man, do you use that kind of math on a regular basis or something?

 

[punkin]

Problem with this idea is that the sound goes everywhere then comes back. We're talking about changes in the audio and delays/reflections. It's all very unpredictable. It might be possible to phase cancel the incidental or predictable waves but after that, it ain't ever gonna happen. Especially if you're messing with complex/compound signals like voice or guitar or just about anything that's not a fixed pure test tone. Forget about room or background noise or hiss and stuff of that nature.

If you're bent on doing this, you'll want to work with a fixed test tone...nice and strong. Close mic the audio source and then use a second mic directly in front of the sound source and start backing off to the first cancellation point. There will be others the further back you go but the effect will be much less noticable. If you can get your hands on a pair of highly focused directional mic this might help as well. The room will need to be as dry as possible.

 

[DonGraham]

You FREAK!!!!!! Holy crap man, do you use that kind of math on a regular basis or something?

Unfortunately i do....

 

[punkin]

I might add that if you're wanting to experiment with this, a low frequency will be a little easier to work with. A long wave length is going to take more movement of the microhpone and should be a little easier to get close to finding that phase cancellation. If you pick someting up at say 2 kHz, a very slight change in mic placement will make a big change.

Just a thought...maybe a bettwer way to work this would be with two audio sources and one mic. This way you can remove some of the room influence. Take two identical speakers fed by the same audio source (hooked in parallel) face the speakers toward each other with the mic in the middle...I'm thinkin out of the box now but put the mic one wave length away from speaker #1...then move the #2 speaker slowly closer and closer to the mic. If you can monitor this, on headphones you should hear the beating and eventually something close to a null.

 

[Robert D]

Phi(x,t)=(2*pi*t/T-2*pi*x/lambda+phi(0))...where x=distance, T= period, t=time phi(0)= reference phase, lambda= wavelength....

I think I did a frat party gig at lamda phi once. Bunch of math freaks.

Don - you rock, pocket protecter and all.