What is clipping?

Discussion in 'Amps and Cabs [BG]' started by fleabass89, Jul 7, 2001.

  1. I hear this all the time but don't know what it means.. I tried the search function but I can't find anything helpful. Can someone please dumb it down for me? I appreciate any replies, thanks.
  2. Munjibunga

    Munjibunga Total Hyper-Elite Member Gold Supporting Member

    May 6, 2000
    San Diego (when not at Groom Lake)
    Independent Contractor to Bass San Diego
    Clipping occurs when you put too much input signal into an amplifier stage, resulting in the peaks of the output waveform being "clipped" off, resulting in a square-wave output. Bad, BAD! Speakers are not designed to handle a signal with sharp little corners on it. Just rips 'em right up.

    For example, if your amplifier is a tube, the input voltage (signal) goes in on the control grid, which modulates the current of electrons flowing from the from the cathode to the plate of the tube. This current develops a higher voltage across a resistor at the plate (unless, of course, it's a cathode follower), resulting in an amplified copy of the input signal. If you apply too high of an input voltage on the grid, the current through the tube "maxes out" at a certain level, and no further amplification is possible. The result is that, for the period of time that the tube is at maximum amplification (while the input wave is above the threshhold), the output wave holds at a constant level, creating a flat top, so to speak, and creates a so-call square wave.

    Bottom line, don't overdrive your amp ... it's death on speakers.
  3. Thanks for the help, guys. How can I avoid it?
  4. Munjibunga

    Munjibunga Total Hyper-Elite Member Gold Supporting Member

    May 6, 2000
    San Diego (when not at Groom Lake)
    Independent Contractor to Bass San Diego
    If you start to hear a really distorted sound out of your amp, turn down your gain (pre-amp) and turn up your volume. It's most likely that clipping will occur in the pre-amp stage.
  5. 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
    "Sharp little corners" on the waveform aren't what destroys speakers. In sustained clipping, the average power in the signal is well above the amp's power rating. In fact, there's twice as much power in a square wave than in a sine wave of the same peak voltage. Power causes heating in a speaker voice coil, and too much power average power causes overheating. You could melt or burn out the coil, or melt the adhesives in the surrounding assembly.

  6. MikeyD


    Sep 9, 2000
    Hi, Bob,
    Welcome to TB. I would assume by your member name you work at QSC. If you are an amplifier expert, that would be really awesome! Good post; I would elaborate though, that the sharp corners represent higher-frequency components of the waveform. If the speaker system is two-way (or higher), the high-frequency power generated by clipping will be routed to the tweeters and may damage them if they are inadequately protected. On the other hand, one-way systems (without crossovers and tweeters) may handle the HF just fine if, as your post suggests, the voice coils are rated to handle all of the power of the clipped waveform.

    Looking forward to more good input from you! :)
    - Mike
  7. Sorry for jumping in on your question.
    But I have one relating to this as well.
    My setup is a 400w@4ohm head going into a 500w@8ohm enclosure and it clips quite a lot. So is that because it is only approx 1/2 the power going into the enclosure so therefore not enough to power it properly. If so whats the best way to look after it?
  8. would a hartke 3500 cab pushing 240 watts @ 8 ohms and a carvin 600W handing 8 ohm cab pose any problems?
  9. Phat Ham

    Phat Ham

    Feb 13, 2000
    As long as you don't drive the amp into clipping it won't cause any problems. You might, however, find that the Hartke isn't pushing the cab hard enough.
  10. and how would that be a problem? Sorry for all the questions..
  11. MikeyD


    Sep 9, 2000
    It's not a problem. Not "pushing hard enough" only means you can't get enough volume from the rig without going into clipping. If you're not clipping and the rig is loud enough, you are just fine. You can push a mountain of speakers with one watt, BTW - they are perfectly happy with too little power, even though *you* might not be.

    PUNK&DONUTS - you'd have to be *sure* that your amp is rated for less than about 250 watts into 8 ohms in order to be "safe" clipping it heavily driving a 500-watt cabinet. Don't make the assumption that the amp will deliver half the power at 8 ohms as it does at 4. This may or may not be the case. A lot of amps do significantly better than half power. Then if you clip badly, you can about double that. So if your amp can put 300 watts (RMS sine) into 8 ohms, then it could pump 600 watts clipped square wave power into that load.

    Anyway, if you are clipping a lot you either need to (a) get a more powerful amp and NOT overdrive it, or (b) get more or more efficient speaker cabinets, or (c) both.
    - Mike
  12. Phat Ham

    Phat Ham

    Feb 13, 2000
    It won't cause any technical problems like MikeyD said, but I come from the school of thought that says you should have more power in your amp than what your speakers are rated for. For example, if I have cab that handles 300W, I would want a 400 or 500 watt amp. That way I wouldn't have to push the amp too hard and have plenty of clean head room.
  13. I'm of the other mindset, where I want the cabinet power rating to be equal or greater than the amp power rating. If your amp is rated higher, then you could be putting out a clean 500W signal with no clipping and blow your 300W rated speaker.

  14. MikeyD


    Sep 9, 2000
    True, if it is putting that kind of power into the speaker continuously. A momentary transient of that (or more) might be fine, depending on the frequency content and duration. But some of the stuff I've read starts to look at this statistically, in which case the term "music program power" does have some merit. I had some correspondence with dudes on the live-sound usenet NG last year. The gist was this: to get an average of 300 watts into that speaker with a 500-watt amp (unclipped), the ratio of peak to average is 500/300 - or 2.2 dB! Imagine playing steadily enough so that your signal did not deviate by that amount. I argued that a long pedal tone (or letting the bass feed back with a lot of compression) might get you there, but for most playing, the peak to average ratio must be quite a bit higher. This includes rests and spaces between the notes. One guy argued that a bass player is very unlikely to see less than a 5 dB power variance in normal playing. This would translate to 3x power ratio, or 158 watts average power for a 500-watt unclipped peak!

    Messy, huh? But this is the reason why some guys get away with having much bigger amps than the speakers are rated to take (i.e., overpowering). What would be really cool is if amps had a power output meter with peak hold and average power, with options for how long the averaging is done. It would be a great way for players to actually see what's getting sent to the speakers. Anyway, if using an amp exceeding the speakers' rating, try to be really careful about not pushing it too hard.
    - Mike
  15. Richard Lindsey

    Richard Lindsey

    Mar 25, 2000
    Metro NYC
    But see, the value of having a big amp in that case has *nothing* to do with the fact that its power rating is higher than the cab's power handling rating. Nothing at all. Given equal efficiency (let's say, 101 dB w/ 1 W/1 m) and equivalent impedance (let's say, 4 ohms), you are not working your amp one bit harder going into a 600 watt cab than you would going into a 300 watt cab. Higher power handling in a cab does not make the amp work harder.

    Efficiency has some relevance, but power handling doesn't. Think of it this way: if you have a cab that handles 300 watts and has an efficiency of 97 dB, it's going to need in the ballpark of twice as much power to produce the same volume as a cab that handles 600 watts and has an efficiency of 101 dB. (This is simplistic, but the principle is basically right.) That is, you're actually going to work the amp *harder* with the 300 watt cab.

    What allows your amp to "loaf" and keeps you from having to push it too hard--which, as you note, is generally undesirable--is *not* having more power than your cab can handle. That's irrelevant in itself. What allows your amp to loaf is *having significantly more power than the job needs*. That is, if you could do your gig well enough with a 200 watt amp, using a 600 watt amp allows that amp to loaf.

    Now, this doesn't mean you *can't* or shouldn't use an amp that puts out more power than the speaker is rated for. As MikeyD explained, that can work. But there is no inherent benefit to this mismatch. It's in no sense better than having a cab with higher power handling, all else being equal. The benefit comes from having a buttload of power available, not from exceeding the cab's power handling capacity.
  16. That last statement is very true, as is almost everything that's being said.

    I'm reading this thread with interest. Don't have time to jump in and respond, however.
  17. MikeyD


    Sep 9, 2000
    Well, just to muddy up the water a bit... I generally don't believe in oversimplifications about underpowering or overpowering right/wrong. Let's say you have a cabinet which is rated to handle 600 watts continuously, and you find that your preferred average program power (the actual average of your playing) is 400 watts continuously. And let's also say you are a really percussive player (e.g., slapper) who doesn't like much compression. Let's say you often see peaks that are easily 10 dB higher than your average. Guess what: that's 4000 watts for the transients. So now let's say you're a hi-fi purist who doesn't want said peaks to be limited or clipped in any way. So you buy an amp capable of 4000 watts short-term, clean power output. Do you go out and buy a set of cabinets that can handle 4000 watts continuously? That might be overkill - if not physically, then maybe financially. This situation might be good reason to have a cabinet that is capable of handling at least the continuous average power output, but have an amp capable of transient power significantly higher. This is a situation that is very difficult to plan appropriately with only two wattage ratings.

    Now that I've thrown that wrench in the works... :)

    - Mike
  18. CaptainWally

    CaptainWally Supporting Member

    Oct 21, 2000
    Sandy Eggo, CA
    Thanks for the informative postings.

    Here is the question I still don't understand. Why
    do manufacturers build amplifiers that are capable
    of clipping? Couldn't you limit the gain (input signal)
    so that the amplifier is never forced to augment
    a signal it doesn't have the juice to faithfully
    produce properly (no clipped wave)?

    Follow-up question (real life case): I used to own
    an Acme B-2 which is known for its inefficiency
    running with an Ampeg SVT3pro (450W). I did
    have the gain up a bit (12 o'clock), but never
    had the volume over 1/2. Would it be possible
    to clip with only 1/2 volume of the amp or does
    efficiency factor in....
  19. MikeyD


    Sep 9, 2000
    Well, part of the problem is the feedback used in a lot of amplifiers. Some of the devices used in an amplifier (transistor, for example) are very nonlinear. Designers use internal feedback to correct the nonlinearities. In some amps, when the feedback circuit detects that the top of a sine waveform is getting clipped, it generates an error signal that forces the amp to compensate by *boosting* gain on that segment in hopes of getting the sine shape back. Problem is, the power supply won't give it any more current, so the clipping actually gets worse. It's sort of an unstable situation in that once such an amp starts to clip a waveform, it tends to clip it a LOT.

    An amp could dynamically adjust the gain to avoid the clipping phenomenon - the effect would be limiting or compression. The amp would have to "look ahead" at the input signal and "ride the gain" if it saw a big peak coming. You'd hear compression or other kind of limiting. That would be a tradeoff, for sure. When bass is played percussively, outboard compression is often employed to prevent these huge transient spikes (which often cause clipping). If you then use a lot of compression and still crank up the amp, it not only clips on the peaks, it clips *constantly* which makes a bad situation even worse - the steady power to the speakers can nearly double in that case.

    So, one way or the other, you can't get "faithful" reproduction unless you stay well below the headroom limitation of the amp (meaning always within its linear range of operation). You either take a chance on clipping, or you alter the signal by compressing the peaks before the amp clips them.
    I don't think the volume setting has a lot of correlation to whether an amp will clip. I've seen a lot of amps driven to clipping at 1/2 or less. It depends on the strength of the signal reaching the volume potentiometer.
    - Mike