Truths and myths about clipping

A work in progress.

Moderators: Bob Lee, Cameron Shoffner, Thanh Nguyen, Christian Cook, Kirk Fyvie, Martin Barbour, Brad Sambrano, Fred Thomke, Tracey Homan, Gary Evans_QSC

Fri Dec 26, 2008 4:22 pm

  • Truths and myths about clipping

    NOTE: For an explanation of what clipping is, please see the What is clipping? Why is it bad? thread.

    Can you tell which of these statements are true and which are myths?

    Clipping is bad for loudspeakers because it really heats up the voice coils.
    Myth. Dynamic loudspeakers are notoriously inefficient; they turn well over 90% of the audio power put into them into waste heat instead of acoustical energy. A clipped signal does produce heat in the voice coil, but so does an unclipped signal, too.

    A variation of this myth is that extra heating occurs because the voice coil and cone stop moving during the clipped portion of the audio waveform. This also is untrue; even if the instantaneous signal voltage stays the same for some short period of time, like a millisecond or so, the cone stays in motion because at frequencies above the loudspeaker’s resonance the voltage is an accelerative force.

    Clipping at any power level is capable of damaging your loudspeakers.
    Myth. Some may tell you that a clipped signal from even a very low-power amp will blow out a high-power loudspeaker driver, but that’s untrue. Picture a 50-watt amp driving a 500-watt (continuous) loudspeaker driver. Even if the amp is driven into very severe clipping, it will still put out less than 100 watts, and therefore will not be a threat to the loudspeaker (it probably will not sound good, though).

    Try that with an amp rated at, say, 400 watts or higher, though, and the results may be very different. This is because an amp, when it clips, can usually put out much more power than it is rated for. Prolonged, sustained clipping may cause that amp to put more than 500 watts into the loudspeaker for some significant time and cause a thermal failure (melting) in the voice coil. On the other hand, in a system where the amplifier and loudspeaker are well matched (i.e., amplifier power is roughly equal to the loudspeaker's program power rating), very brief and occasional clipped peaks are generally harmless.

    Clipping is bad for loudspeakers because it is DC.
    Myth. A clipped signal is not DC. Even if the clipping is so severe that the waveform shape approaches that of a square wave, it’s still AC.

    Even so, DC by itself is not necessarily dangerous for loudspeakers. In fact, a good way to check polarity of a woofer driver is to connect a small battery across it and see which way the cone moves for a given DC polarity; the amount of power dissipated is minuscule. However, a significant DC offset on an amplifier output is undesirable for several reasons: because it will dissipate power in the voice coil without producing any acoustical output; because it may cause the voice coil to travel out of the magnet gap, which would reduce the cooling; and because it shifts the driver’s at-rest position and thus makes the available excursion asymmetrical, reducing the overall safe excursion limits of the driver.

    Clipping is bad for loudspeakers because the sharp corners rip up speaker drivers.
    Myth. No.

    Clipping is bad for loudspeakers because the amp may put out more power than you expect.
    True. The real danger in clipping is that it could overpower the loudspeaker. Even an amp that may seem rated safely below the loudspeaker's power handling capacity might put out dangerous power levels if you are careless with allowing it to run into clipping.

    In addition, severe clipping reduces the dynamic range because it quashes the peaks even as the average power in the output signal is elevated by excessive gain boost. Thus, there is often a temptation among inexperienced sound system operators to try to recapture some of the lost "punch" by pushing the levels even higher, which only makes the problem worse.
    Bob Lee
    Technical Communications Developer
    QSC, LLC
    Fellow, Audio Engineering Society
    Image
    "If it sounds good, it is good." —Duke Ellington
    User avatar
    Bob Lee
    QSC Audio Products
     
    Posts: 2518
    Joined: Thu Mar 06, 2003 2:01 am
    Location: Costa Mesa, California


Return to Amplifier FAQ




Information
  • Users browsing this forum: No registered users and 1 guest