Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Physically Invert Sound

  1. Feb 11, 2012 #1
    First off, I'm new here so I'll do a brief introduction:

    I've never taken a physics class, but am taking AP Physics next year (didn't know they offered it at my high school). I'm also a musician and wannabe engineer, so I'm really interested in sound properties, acoustics, etc. So that's me. Anyways . . .

    If I wanted to physically invert a sound wave, how would I do that? I saw someone said to just connect a speaker in reverse (+ to - and - to +), but couldn't find any confirmation on that. Also, how would one go about finding the distance required to create deconstructive interference?

    I'm basically seeing if I can capture any given sound or melody and mute it (or reduce its volume) by playing it back with an inverted sound wave to create deconstructive interference.
  2. jcsd
  3. Feb 11, 2012 #2
  4. Feb 11, 2012 #3
    Is there an easier way to do it though? I feel like my project isn't quite that complex to require a special circuit (but I'm the one with no experience so =P). It's just a microphone capturing sounds, and sending those sounds back out through a speaker, but inverted.

    Taken from another forum:

    ". . . It really doesn't matter the polarity of the wire as long as you hook up the correct terminal of your receiver to the correct terminal of your speaker (hot to hot cold to cold) if the (-) wire is connecting the hot terminals that's fine. If you end up hooking your speaker up backwards you are basically inverting the phase of the sound wave so it will be 180 degrees off (speaker moves "in" when it should move "out")."

    I'm assuming, given this is true, that while you can't hear a difference if it's by itself, you can use this inverted wave to cancel out the original one
    Last edited: Feb 11, 2012
  5. Feb 11, 2012 #4


    User Avatar
    Gold Member

    The bolded part is totally wrong. Do not believe everything you read, and trust in competent advice like that above already given to you. You must use circuitry to "invert" sound, for example given by gnurf in "noise cancellation".
  6. Feb 11, 2012 #5
    That's why I came here- I couldn't confirm it. So is there a "simple" way to do it then with circuits (simple in quotes because I know it won't be)? Any direction I can be thrown in that's a bit more specific than Active Control (just due to time and school)?
  7. Feb 11, 2012 #6


    User Avatar
    Gold Member

    Bluestribute, there is no “simple” solution to active noise control, or the destructive interference of sound waves. We must invert the original wave, amplify it, and then transmit it into the original sound field in order for them to mix and cancel one another.

    As gnurf suggested in post #2, and I repeat, you must study this theory: http://en.wikipedia.org/wiki/Active_noise_control

    Another page for you to study is: http://en.wikipedia.org/wiki/Interference_(wave_propagation [Broken])

    For a good article on noise cancellation see this: http://doctord.dyndns.org/Pubs/POTENT.htm
    Last edited by a moderator: May 5, 2017
  8. Feb 11, 2012 #7
    I know this won't be simple . . . I'm feeling this might be a summer project O.O Do it right while learning enough to know how and why I'm doing it.
  9. Feb 11, 2012 #8

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    """ you can use this inverted wave to cancel out the original one""

    i believe that is correct. you'll need to match the one you're trying to cancel...

    Ever notice the "snout" on modern ship hulls, just under water ahead of the bow?
    It's called a "bulbous bow" and is there to make a wave that cancels out the bow wave.


    waves is waves.

    With you interests, peruse thrift shops and used bookstores for old engineering texts.
    There's an industry classic, "Acoustics", from about 1940 that is a treasure.

    as you get into electronics you'll like this site too.
    http://www.tubebooks.org/technical_books_online.htm#Audio (hi-fi, amplifiers, speakers...)
  10. Feb 11, 2012 #9


    User Avatar
    Gold Member

    Old Jim, """" you can use this inverted wave to cancel out the original one""
    i believe that is correct. you'll need to match the one you're trying to cancel...""""

    Sorry, not wholly correct. Trouble with this is: the swapping of the speaker wires does NOT produce an inverted wave.

    IF the swapping of speaker wires would invert the sound wave that would sure simplify things....

    Thank you for those classic references! Those huge vacuum tube amps kicked butt!
    Last edited: Feb 11, 2012
  11. Feb 12, 2012 #10


    User Avatar
    Science Advisor

    When you apply a DC voltage to a speaker, one connection of the positive lead (usually marked as "+" on the speaker) will cause the speaker cone to move forward.

    If you reverse the speaker leads, the speaker cone will move backwards or towards the magnet.

    One connection gives a compression and the other gives a rarefaction for the same input.

    So, yes, reversing the speaker leads does invert the audio out of the speaker.
  12. Feb 12, 2012 #11


    User Avatar
    Gold Member

    vk6kro, Bluestribute, and Old Jim: Oops! Sorry for my mistake! Thank you for pointing out my error about speaker wire reversal and phase inversion.

    Every microphone, cable, mixer electronic path, outboard gear, recorder, electronic path, speaker cables, speakers and headphones are wired in such a way that keeps phase integrity intact (absolute phase). In the most basic wiring path, there is a “hot wire” (positive) and a ground wire (negative). Reversing this anywhere in the signal path will lead to inverse phase (inverse polarity).
  13. Feb 12, 2012 #12
    I would like to confirm that reversing speaker leads does 'invert' the sound.
    It can be demonstrated in an interference experiment using 2 speakers connected in series with a signal generator. With a frequency of about 1500Hz the wavelength is about 0.2m so speakers placed about 0.4 to 0.5m apart facing outwards produces an interference (2 slits) pattern of max and min. Students walking across the space in front of the speakers can easily hear the max and min. If the connections to one of the speakers is reversed the positions of max and min are reversed.
    This works especially well in a hall rather than a classroom to reduce the effect of reflections from walls.
  14. Feb 12, 2012 #13

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    No Sweat Bobbywhy, a large percentage of communication should be prefixed with mis-,,,.

    indeed the ear is absolutely insensitive to phase. Think about it, hearing a chemical reaction in the nerves of the cochlea. No way that can sense phase.

    To cance a sound wave one must first measure the wave he wishes to cancel, then produce one that will arrive at the point where you want it cancelled with matching amplitude and opposite phase. The two waves add to zero so dont move an eardrum in that location.
    That takes more than reversing a speaker wire.

    The technique was used for radar masking aircraft in 1950's - a B52 of the day could absorb an incoming radar wave and return one that looked like Cessna 150 going the other way.
  15. Feb 12, 2012 #14


    User Avatar
    Science Advisor
    Gold Member

  16. Feb 12, 2012 #15
    Wait wait wait . . . so reversing the wires does in fact invert the waves??

    Once I can get the incoming signal reversed, this is my next task- seeing just how difficult placement is, and if there's anything I could do to make it easier.
  17. Feb 17, 2012 #16
    I should be getting a spare amplifier next week so I'll let you know how it goes . . .
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook