Half a croissant, on a plate, with a sign in front of it saying '50c'
h a l f b a k e r y
Open other side.

idea: add, search, annotate, link, view, overview, recent, by name, random

meta: news, help, about, links, report a problem

account: browse anonymously, or get an account and write.



Noise Cancellation for ghostly effects

  (+7, -1)
(+7, -1)
  [vote for,

Connect two identical speakers in Parallel. Place them adjacent to each other (rims touching, not facing each other but facing same direction). Feed them output of one of the channels of a stereo amplifier. Now both speakers should be cancelling each other out due to destructive interference. ( If not, swap two connections of any one of the two speakers).

This will form "dark spots" in the room where (and only where) absolutely no sound from speakers will be heard. Arrange Halloween party in such a room.

VJW, Dec 17 2010

random video of demo with two speakers http://www.youtube....watch?v=_MM1l7AhMac
Even has a math derivation! [sqeaketh the wheel, Dec 18 2010]

Public Noise Cancelation Public_20Noise_20Cancelation
[mouseposture, Dec 19 2010]


       Baked ((wrong - see below)) Physics professors do this all the time in classes on acoustics.
sqeaketh the wheel, Dec 17 2010

       [+] I think you'd get a wide middle dead spot if you connect them up out of phase.
FlyingToaster, Dec 17 2010

       //if you connect them up out of phase//   

       which I think will be achieved by swapping the two connections of one of the two speakers.
VJW, Dec 17 2010

       Could be cool. Phase cancellation and augmentation as you walk through the peaks and valleys can give a weird "sound is changing in my head" effect that's kind of hard to describe.   

       You could also use this in conjuction with a sound absorbing "dead room" that feels like being in a dream or something. You hear your breath only and get to see how noisy a "quiet" room actually is by comparison.
doctorremulac3, Dec 17 2010

       You'd be rather limited in what sounds you could play on these speakers, no? Just line spectra. But it could still be pretty eerie.
mouseposture, Dec 18 2010

       At first I too thought it will work only with a fixed sine wave. But Then again this may not be the case.   

       Somebody should try it.   

       //Physics professors do this all the time...   

       [sqeaketh the wheel] do you have more details ?
VJW, Dec 18 2010

       //But Then again this may not be the case.// Explain that, please, 'cause I thought the same (one or a very few sine waves, anyway).
mouseposture, Dec 18 2010

       In case of a general purpose Noise Cancellation device, I think tricky part is predicting what the next signal is going to be and generate opposite of that.   

       But here, that isn't the case. Or, so I think. Also source and anti-source are as close to each other as possible. Which again takes away the need for simple sine wave.
VJW, Dec 18 2010

       //Baked. Physics professors do this all the time in classes on acoustics//   

       My physics teacher never did that. Has any one seen a live demo of this ?
VJW, Dec 18 2010

       I just ran a speaker test program on my computer speakers. The polarity test featured a pure sine-sounding signal that was supposed to die out if the speakers were hooked up wrong. I don't think the noise-cancelling would work on anything else.   

       The storm sirens that are too close to my house give out a fairly pure tone. When my windows are open when they go off, the sound varies wildly as I run through the house to let the dog in before his ears implode.
baconbrain, Dec 18 2010

       Was same signal fed to both speakers; If it is a stereo signal, then you can never be sure of it.   

       Also I think it is very essential that one of the speaker connections must be swapped. This I think is impossible in standard computer/laptop setup.   

       One will need to rip-open one of the speaker cables and swap two connections.
VJW, Dec 18 2010

       //Physics professors do this all the time...   

       [+] Changed my vote. I was wrong about "all" professors, although it is well known for the case of two sine waves. (link)   

       This even works in a crowded lecture hall, where some students hear loud and others quiet. When the phase of one speaker is flipped (by switching the wires [VJW]), the opposite students hear loud or quiet.   

       For music or white noise, it doesn't really work dramatically, unless the two speakers are facing and near (< 1m) each other, and both speakers are fed by the same signal (not two different stereo channels [VJW]). Then, only certain frequencies will cancel, depending on the speakers' separation [bigsleep].
sqeaketh the wheel, Dec 18 2010

       Thanks for link, [sqeaketh the wheel];   

       How ever I think he is forming a stationary wave in the area between two speakers. Hence he is getting fixed peaks and valleys of sound.
VJW, Dec 19 2010

       Yeah, though I walk through the peaks and valleys of sound, I shall fear no weird sounds...
normzone, Dec 19 2010

       I tried this a few years ago. In an ordinary room, the effect is fairly subtle. You certainly don't get any regions where no sound is heard at all, but you do get a //weird "sound is changing in my head"// effect.
spidermother, Dec 19 2010

       If we try to understand noise cancellation in terms of speaker cone movements then really waveform should not matter, isn't it ?   

       We can say that when one speaker cone pushes ahead the other one retracts back by same amount, thus negating +ve pressure with -ve pressure.   

       So it should work for white noise, music or just pure sine wave.
VJW, Dec 19 2010

       That is correct, as long as the detector is equally distant from the two sources.   

       In the real world, you don't hear silence because   

       1) Reflections from walls take various times to get to you   

       2) Speakers are not point sources (they are of significant size relative to audible wavelengths)   

       3) Ears are not point detectors; you (probably) have two of them, and they also have significant size.
spidermother, Dec 19 2010

       Ear size should not matter because if this thing works perfectly there should not any wave at all anywhere in the room, unlike in the case of a standing wave.
VJW, Dec 19 2010

       //there should not any wave at all anywhere in the room// But there will be, if it's done as you describe. You should actually try this. It's not difficult - just reverse the wires on one speaker and play a mono track, or use a sound editor to create a file where one channel is the exact inverse of the other.   

       The result will be different from what you describe, but interesting anyway.
spidermother, Dec 19 2010

       The "dark spots" referred are not really valleys of a stationary wave. I think they will be caused by "imperfections" of experiment due to reflections on the wall and other objects in the room etc.   

       As a result of this experiment, we are not expecting a stationary wave to be formed. We are expecting absolutely no waves formed at all. But due to imperfections there will be dark spots.   

       We must remember that noise cancellation and stationary wave formation are two different things. I think demos which physics professors show is about stationary waves.
VJW, Dec 19 2010

       [VJW] Ah, now I understand, I think. <link>   

       //there should not any wave at all anywhere in the room//
//This will form "dark spots" in the room//

       Can you clarify this apparent contradiction? Also, the first statement is only true if the speed of sound is infinite, no?
mouseposture, Dec 19 2010

       //there should not any wave at all anywhere in the room// as per theory!   

       //This will form "dark spots" in the room// due to imperfections of the experiments such reflections of walls, non-zero speaker size, etc.   

       We can NEVER achieve Noise Cancellation by forming standing waves. Both are different things.   

       As per theory stationary waves are supposed have peaks/valleys (AKA dark spots). ( this is what most physics demos show)   

       Whereas if Noise meets anti noise both totally cancel out each other. There should be no vibrational energy in the medium ( not even dark and non-dark areas). It should be all-dark in the perfect world.   

       Ok here is the crucial difference between two:   

       Stationary waves are formed when two waves of equal amplitude travelling in *opposite* direction meet each other.   

       In case of "Noise" cancellation, two waves of equal amplitude, opposite phase, travelling in *SAME* direction superimpose each other.
VJW, Dec 19 2010

       //Stationary waves are formed when two waves of equal amplitude travelling in *opposite* direction meet each other
In case of "Noise" cancellation, two waves of equal amplitude, opposite phase, travelling in *SAME* direction superimpose each other.//

       You are assuming a space of only one dimension, then being inconsistent with that assumption by squeezing in two speakers "side by side" on a second dimension. You should have said "different direction" rather than "opposite direction." The waves can only travel in the same direction if the two speakers occupy the same point in space.   

       This is easy to accomplish: by linear superposition, you can just use a *single* speaker, and disconnect it from the amplifier so it makes no sound. In fact, you can accomplish this by having no speaker at all. And, just as you predict, there will be no wave, anywhere in the room.   

       Where the idea becomes nontrivial is in supposing that if the speakers are sufficiently close on the second dimension, you'll get a sufficiently good approximation of that state of affairs.   

       The problem, so it seems to me, is this:   

       Consider what constitutes "sufficiently close." The distance must be much smaller than the smallest wavelength in your audio signal. Otherwise, you don't get "noise cancellation" you get "standing waves." Now, consider that your speakers aren't point sources, that their size is, in fact, on the order of the *largest* wavelength in your audio signal.
mouseposture, Dec 19 2010

       Co ol.   

       [mouseposture] exactly.
VJW, Dec 19 2010


back: main index

business  computer  culture  fashion  food  halfbakery  home  other  product  public  science  sport  vehicle