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

Can two loudspeakers one next to other, damage them self?

  1. May 22, 2007 #1
    Can two loudspeakers one next to other, damage them self? I think the question is very clear, and if you know please answer. Thank you.
  2. jcsd
  3. May 22, 2007 #2


    User Avatar
    Gold Member

    The only way that I can think of damage occurring is if one very powerful speaker is face-to-face with a second one. In that case, the air pressure from it might tear the cone of the other. Even that isn't likely.
  4. May 22, 2007 #3
    hmm, if the two speakers oscillate with some matching sound waves, could there be a catastrophic resonance?
    Last edited: May 22, 2007
  5. May 22, 2007 #4
    No, they can't :smile:
  6. May 22, 2007 #5
    Not even if they generate the same (high amplitude) signal but in phase opposition?
  7. May 22, 2007 #6
    No. Why do you think they would damage each other in this case? The only thing that would happen in this case is that you would hear less sound from both of them then you would hear from one alone.
  8. May 22, 2007 #7


    User Avatar
    Science Advisor

    The only way is if on happened to produce a resonance frequency that corresponds to the natural frequency of the cabinet or some other component. They would have to provide a decent amount of energy though.
  9. May 22, 2007 #8
    There is nothing unique about having two speakers, other than the overall power you are dealing with is doubled (if they have the same power).

    This could potentially be a problem only if:
    1. The cabinet is sealed
    2. The speakers are operating near their maximum power
    3. The cones are placed towards eachother close enough to nearly form a seal to keep air out

    At least that's all I can think of. As for having two speakers, even if you have complete constructive interference at some point, the power at that point will only be twice the average power anyhow. Usually that won't damage a speaker.
  10. May 22, 2007 #9


    User Avatar
    Science Advisor
    Homework Helper

    A loudspeaker in a properly designed cabinet does not have any undamped resonant frequencies. If it did, they would distort the sound it was producing.

    Also speakers are very inefficient at transferring energy into the air. Good quality speakers are often only 1% efficient.

    So the amount of energy the air could put back into the speaker is small compared with the amout of electrical energy driving the speaker, and "resonance" would not do any damage.

    However, if you take a speaker which is NOT mounted in a cabinet, it does have a natural resonance at a low frequency (lower than its operating frequency range). It's very easy to destroy a speaker in free air, by applying a sine wave signal at much less than its rated maximum power, at the self-resonant frequency. That has nothing to do with two speakers interacting with each other, though.
    Last edited: May 22, 2007
  11. May 23, 2007 #10


    User Avatar
    Science Advisor

    I was really reaching with the resonance angle. I didn't say it was likely to happen, but the possibility is always there as with any structure.
  12. May 24, 2007 #11


    User Avatar
    Science Advisor
    Homework Helper

    KingNothing sent me a PM with this question, and I'll try and answer it here since others might be interested:

    To be specific, let's talk about a fairly large (say 12in diameter) moving coil speaker mounted in a cabinet which is a sealed box. (And I'll mention ported or "bass reflex" cabinets at the end)

    Mechanically, the speaker is basically a mass (the centre part that is driven electrically) on a spring. The mass-and-spring system does have a resonant frequency, but there are two sources of damping which stop it resonating freely.

    One damping effect is the speaker cone, which transfers energy to the air as it moves.

    The other (usually larger) source of damping is electrical. You can think of a speaker as an electric motor. Like any motor, if you try to drive it in reverse, it acts as a generator. If the speaker cone tries to move (for example oscillating at its resonant frequency) a current is generated in the coil. This current flows "backwards" through the amplifier, takes energy out of the speaker, and damps the motion.

    Note that if you experiment with the mechanical resonance of a speaker (e.g. get a 12in speaker, tap the center of the cone with your finger, and watch and listen to what happens) you won't get the electrical damping effect unless you short-circuit the speaker terminals together with a bit of wire.

    The function of a sealed speaker cabinet is to absorb all the sound that would be radiated from the back of the speaker cone, otherwise the sounds from the front and back of the cone would interfere with each other and cancel out. The actual box is made of stiff material (e.g. thick wood or plastic) so it doesn't vibrate and transmit any sound from the inside to the outside of the box. The air inside the box can resonate, the same as in lab experiments showing resonance in an air column, but the cabinet contains absorbing material to damp out that resonance, and the electrical damping described above stops the air from moving the speaker cone easily.

    So to summarize, the idea is to reduce all the resonances from the system as much as possible.

    If a speaker system did have its own resonance, the sound quality would be poor. To reproduce a single pulse of sound (e.g. something vaguely like a bass drum) you want the speaker cone to move in a single pulse and then stop, like the red graph in the attachment. If the speaker had its own natural resonance, it would move more like the blue graph.

    Finally, ported or reflex speakers. These are very common as sub-woofers on PC and home cinema systems. They are not sealed boxes. There is a large hole visible in the front or the side of the box. This is a clever design to increase the volume of low frequency sound from a small speaker.

    The hole is at the end of a tube, inside the box. This tube acts like a pipe. As in lab experiments with resonance at the lowest resonant frequency of a pipe, the air is moving in opposite directions at each end of the tube. The clever bit is that the "tube" works as if it was U-shaped, so the air coming out of the hole at the front has the opposite motion the the air at the back of the speaker cone, i.e. the same motion as the air at the front of the speaker cone. The result is you get twice as much sound energy compared with a sealed box. The audio quality isn't very good, but that doesn't matter much for reproducing explosions in video games etc.

    Attached Files:

    Last edited: May 24, 2007
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook