Psst... Ready to join TalkBass and start posting, make new friends, sell your gear, and more?  Register your free account in 30 seconds.

Speaker Motion VS Waveform shape

Discussion in 'Amps and Cabs [BG]' started by DaveMcLain, Jun 23, 2005.


  1. DaveMcLain

    DaveMcLain

    Jun 19, 2005
    Cuba MO
    I'm a bit confused about the relationship between the motion of the speaker cone and the waveform's shape. It's easy to think that the speaker moves in and out in direct relation to the voltage level applied and that if the signal is clipped that the cone moves out, stops for a while and then moves back in, stops for a while etc. Question is, is that REALLY what's happening? OR since the signal voltage level represents the AIR pressure on a transducer(microphone) the speaker really has to move much differently to recreate the sound. If for instance a square wave or close to a square wave is fed into a microphone the diaphram is not pressed back for a certain length of time and then snapped back the other direction for a certain lenght of time to represent that sort of a wave form. It is instead pushed back at the speed of sound at the top and the bottom of the wave and the electrical result is that the voltage swing between the two is what produces the "square wave" at the speaker. In other words, it tries to rareafy the air in one direction at the speed of sound and then compress it in the other all to produce our square wave that was input from the microphone, is this correct? AND is it why amp clipping is much more detrimental to speaker life than other things????
     
  2. alexclaber

    alexclaber Commercial User

    Jun 19, 2001
    Brighton, UK
    Director - Barefaced Ltd
    Ignoring port effects, I believe the motion of the speaker tries to track the shape of the wave as closely as possible.

    Perfect square waves rarely happen, and even when they do the speaker's inertia stops it tracking the wave exactly. Consequently the only danger with square waves is their increased average power to peak power ratio.

    Alex
     
  3. BassIan

    BassIan Supporting Member

    Apr 27, 2003
    Cupertino, California
    The cone of the driver is being asked to follow exactly the voltage signal being applied. If you assume a relatively linear motor system (probably a decent assumption for modern quality drivers), then the displacement of the speaker cone should be proportional to the signal.

    A clipped signal is far more complex than merely chopping the top/bottom off a nice smooth sinusoidal waveform. Amplifiers produce higher frequency harmonic distortion as a result of clipping the original waveform. If the amplifier is asked to reproduce close to a square wave within its dynamic ability, it is still only *close* to an exact square wave. As for the speaker's response, it is slightly different here. Since the cone has mass, and the force being applied to it is electromagnetic (and finite), it cannot stop instantaneously. Even if the voltage applied stops increasing and instantaneously becomes constant, the speaker cone takes a short period of time to come to rest. This is a relatively typical problem when it comes to conversion of electrical signals to mechanical motion. Lower mass cones and stronger motor systems reduce the time it takes for the cone to stop or change direction, but they still must obey the limitations of physics. Keep in mind again that these effects are small, since these drivers can be made to oscillate at relatively high frequencies (2-3kHz, though keeping in mind this is usually during cone breakup, and is with significantly reduced dispersion relative to the lower frequencies).

    Clipping is bad for speaker cabinets for two reasons. First, the harmonics introduced during clipping can be harmful to high frequency drivers, or at least trip their protection circuits. Second, a clipping amplifier produces MORE average power than its rating. Peak power can be double that of RMS power. Also, clipped signals are usually compressed, so they are likely to stay near peak power for a longer period of time than an uncompressed, clean signal. It's the simple fact that clipping amps produce excess power that can cause damage to the speaker driver. Overexcursion can also damage a driver, and this can occur with clean, relatively low power levels if the correct frequency is chosen relative to the driver/cabinet design.

    Hopefully that answers some of your questions.
     
  4. Bob Lee (QSC)

    Bob Lee (QSC) In case you missed it, I work for QSC Audio! Gold Supporting Member Commercial User

    Jul 3, 2001
    Costa Mesa, Calif.
    Technical Communications Developer, QSC Audio
    Think of the signal voltage from the amp as an accelerative force on the speaker driver, not a positional one (except at really low frequencies).

    BassIan gave a very good explanation, but I would clarify his statement about displacement to say that the cone acceleration is proportional to the signal, and as a result the actual excursion decreases as frequency increases with a constant signal level. And that means, of course, that lower frequencies at a given level require larger excursion; that's why it's possible, with a signal low enough in frequency, to make a driver bottom out with much less power than it's rated for.
     
  5. 44me

    44me

    Jun 17, 2002
    Bedford, NH USA
    Actually, speakers generate a velocity proportional to the voltage applied. The position or displacement of the cone is the integral of the velocity (for non math types, think in terms of how far you travel when driving at a particular speed). This behavior is obvious when you stop and think about things for a second - for the same volume level (applied voltage) the cone displacement gets smaller and smaller as the frequency increases.

    Complex repetitive waveforms (square waves being a relatively simple one) can be equated to a series of simple sine waves that are harmonically related. A square wave is a series of odd harmonics (3rd, 5th, 7th, etc). When you apply this to a speaker, it reproduces the fundamental and harmonics that are within it frequency range. Omitting the higher order harmonics softens or rounds the edges. An inability to reproduce the fundamental frequency results in a decay after each edge.

    The square waves produced by amp clipping can cause tweeter damage not because of the waveform shape, but rather that the high frequency components (harmonics) of the waveform can have very high amplitude (much higher than normal musical tones).

    - John
     
  6. alexclaber

    alexclaber Commercial User

    Jun 19, 2001
    Brighton, UK
    Director - Barefaced Ltd
    In retrospect, that is so obvious... Duh!

    Alex
     
  7. BassIan

    BassIan Supporting Member

    Apr 27, 2003
    Cupertino, California
    That's exactly right. Thank you for clarifying that.
     
  8. Glad to see people talking sense on this...:D

    The old 'clipping locks drivers in place' myth really needs to be put to rest.

    On top of what's been said already, even if a single pure square wave could lock a driver in place, it never ever happens in practice. Real signals have a lot more than just a single frequency so even if you are clipping the fundamental you probably aren't clipping the 3rd, 4th, 5th etc harmonic, which will have some effect on cone motion, etc....

    It's a complex relationship and trying to boil it all down to one thing usually results in the creation of these myths....
     
  9. 44me

    44me

    Jun 17, 2002
    Bedford, NH USA
    You really can’t think in terms of clipping the fundamental but not the harmonics. With a complex waveform, you alter and add to the existing harmonics when you clip. See my previous post – this is a fundamental property of complex signals. Once you’re comfortable with the idea it becomes more apparent why wave shape has nothing to do with driver damage.

    Driver damage is a result of exceeding its rms or peak power handling capability, with peak power capability varying significantly with frequency and cab tuning. A driver is most vulnerable to excessive peak power at the bottom of its frequency range. Tweeters get damaged not because of very high frequencies at high amplitude, but rather, large amplitude signals at the bottom of their frequency rangem which can be developed when an amp clips.

    - John
     
  10. Its the upper level harmonics that sum up to "square" the wave. Its sort of a chicken/egg thing. Clipping and the upper harmonics are the same thing, it doesn't quite make sense to say you're clipping the harmonics also. I'm not explaining this very well.

    Square waves are a fundamental sine wave with infinite series of harmonic sine waves added and summed together. So in that sense, you're really not even clipping the fundamental either when you're clipping, you're just adding a bunch of overtones that sum up in such a way as to produce a square-ish version of the fundamental waveform.

    Sound like Zen, or something.... :D
    Randy
     

  11. I am aware of all of this. I spent many long hours studying Fourier analysis. ;)

    What I was getting at (poor choice of words I guess) is that real world signals have complex waveforms and you will never see a pure square waveform at a speaker's terminals, no matter how much you clip an amp. A lot of people seem to think that driving an amp into clipping in actual use somehow produces pure square waves. For one thing, no real amp can actually reproduce a pure square wave because they can't react instananeously, so even with square input, there will be some distortion of that by the amp.

    I've been arguing that speakers don't really care about waveform for years. The 'square wave locking the cones in place' argument comes up a lot....
     
  12. 44me

    44me

    Jun 17, 2002
    Bedford, NH USA
    Mark – Sorry I misunderstood your intent. It seems everyone who has responded to this post has a good understanding of the physics involved.

    - John
     

  13. No worries, my original post was poorly worded. I was ill and tired at the time....:D