Cabinet efficiency vs. power rating

Discussion in 'Amps and Cabs [BG]' started by crdeppe, Dec 4, 2004.

  1. crdeppe

    crdeppe Guest

    Nov 19, 2004
    There seems to be a lot of talk about the pitfalls of using a high-power cabinet with a low-power head. I don't remember when this dicussion came to be, but I'm still not sure I really understand the details. Just because a cabinet can handle 400 watts does not mean it needs 400 watts to work. Isn't the sensitivity rating the more important? For instance, the Ampeg SVT-412 is rated for 400 watts, and has a sensitivity of 98 db. On the other hand, the Aguilar GS-212 is rated for 600 watts and has a sensitivity of 103 db. That says to me that the Aguilar needs less power to drive it, and also handles more power.

    So what am I missing?
  2. Tim__x


    Aug 13, 2002
    Alberta, Canada
    Nothing, alot of people say you can damage a cab by "underpowering" it, but they're just plain wrong. It's seems like you have a fairly good grasp of what's important so just ignore the foolish people.
  3. seems like you get the point, its more about efficiency then it is about power ratings.
  4. Doug Parent

    Doug Parent Supporting Member

    May 31, 2004
    San Diego, Ca.
    Dealer Nordstrand Pickups.
    I politely disagree. Under powering can result in cooked voice coils just as easily as too much heat/power. But I'd like to hear someone with some electrical engineering chime in. What say you David Funk?
  5. Low clean power will never damage a speaker. It's when you overdrive a low power source into clipping that you run into trouble.

    With that said it sounds like you know what's up already. Just wanted to chime in ;)
  6. In the case of cooked voice coils, the culprit is always too much power. Of course, how that power is arrived at is a little more complicated....;)

    It is possible to damage a speaker rated for higher power than the amp powering it if you drive it into clipping, but there's a limit. You might be able to damage a 500W speaker with an amp rated for 400W but it's extremely unlikely that you could damage the same speaker with a 50W amp.

    Amps will produce more than their rated power when clipped (up to a theoretical maximum of twice as much).
  7. nonsqtr

    nonsqtr The emperor has no clothes!

    Aug 29, 2003
    Burbank CA USA

    Here's what happens in the underpowered amp scenario.

    The problem is, that the amp is underpowered "relative to the gig", not "relative to the speaker".

    The fault lies with the player, who turns the underpowered amp up too loud.

    Although, the primary problem is in the amp itself, namely in the electrical properties of the output stage when it is being overdriven.

    In the underpowering scenario, speaker problems only occur with a poorly designed amp. A well designed amp will never fry a speaker for the reason about to be mentioned.

    With a poorly designed amp that is being pushed beyond its limits (ie overdriven for long periods of time), the waveforms "clip", meaning that there is significant high frequency energy being generated even from low-frequency input. Often, the resulting high-frequency energy is "ultrasonic", way beyond the range that any speaker can comfortably reproduce. And, in a speaker coil, any energy that isn't being used to move the coil, goes directly into heat. Hence the possibility of a fried voice coil.

    What typically happens to the signal waveform, is that the percentage of high frequency energy tends to increase, relative to the percentage of low frequency energy. When the amp "clips" (compresses) at high volume, the low frequency fundamental is limited by the maximum current the power supply can provide (or the maximum current the output stage will use), while the high frequencies (which are ordinary much less energetic, relatively speaking), will continue to increase. This results in a much higher percentage of high frequency energy that won't move the coil, but will instead get converted into heat.

    So is it "just as easy" to fry a coil this way, as with too much power? Probably not. You'd need an amp that's underpowered "just enough" relative to the speaker's capacity, and you'd also need an amp that's poorly designed to begin with.

    If you use a good quality amp, and refrain from turning it up beyond the clipping point, then you can use as highly rated a speaker as you want, and there won't be any danger of frying it.

    That's also why it's a good idea to pay attention to those little "clip" LEDs. They put them there for a reason - after all, if your amp could handle a lot of clipping, there'd be no need for the LED, would there? They only put those in there 'cause the solid state power amps don't like to be overdriven, and usually (because of the cost-consciousness in manufacturing) it's risky to do so.
  8. Tim__x


    Aug 13, 2002
    Alberta, Canada
    Not really, do to voice coil inductance a speaker will dissipate slightly less power at high frequency's then at low frequency's. At any frequency, more than 95% of the input power becomes heat anyway. A square wave with a 100w fundamental contain's only 20w of harmonics, so even if the speaker lost all efficiency and converted all input above a certain frequency to heat, a 120w square wave would only cause 1w more dissipation than a 120w sine wave.

    To put it simply there are only two ways to destroy a speaker (with an amp that is :smug: )

    1: Overheat the voice coil. This is caused by too much power, it doesn't matter what frequency, if too much power is dissipated by the coil it will burn up.

    2: Over excursion. This happens when the cone moves to far. It's caused by to much power in the low end.
  9. nonsqtr

    nonsqtr The emperor has no clothes!

    Aug 29, 2003
    Burbank CA USA
    I agree with most of your post.

    Except for the above quote, which needs to be considered in the context of the "compression" that occurs with power amp clipping. That was really the whole point of the scenario I outlined. A 100 watt square wave "isn't", when the fundamental can't drive the full energy through the amp (usually because the power supply can't deliver the needed current). Under those conditions, you won't get a square wave, you'll get a much nastier waveform with very little fundamental and a whole lot of harmonics. It won't be "square", it'll be all pointy and wierd-looking.

    If your input is a sine wave, and then gets "clipped", said clipping may result in something that resembles a square wave. However when the amp is pushed significantly beyond the point where the power supply can deliver the current required to sustain the fundamental, the low frequencies will begin dropping out and the waveform will change from a square wave to something entirely different. Typically the rising and falling edges of the square wave will have large "spikes" associated with them, which is where a lot of the high frequency energy will reside. And that's why solid state clipping sounds so nasty, most of the time. It really doesn't resemble any "normal" waveform, like the nice square wave that might result from overdriving a tube amp.

    So, the only point I'm making is that the 1w/120w ratio is unrealistic in this scenario. In real life it might be closer to 20w/120w, which can be far more significant in terms of what it might do to a speaker. Power is power, and it don't matter what the spectral impedance curve looks like if all your energy is beyond the speaker's useful range. If you're pumping enough power through the coil, and it can't move, the energy still has to find some way to do some useful work.

    Other than that though, it seems like we're saying the same thing. Over-excursion is the far more common speaker-killer, when people use "over-powered" amps. This would be especially true when using loud clean amps at low frequencies. And the other way to kill the coil is by overheating it, and as you say, that would usually be an excess of "sustained" power, as distinct from the "instantaneous" power that would cause an over-excursion.

    The moral of the story is, "get a well designed amp". Many amps have all kinds of built-in protective circuitry that mitigates the undesirable side effects when the amp is "accidentally" overdriven (and also gracefully shuts down the amp, when for instance the transistors get too hot, or the output is "accidentally" plugged back into the input). :D

    It pays to invest in a good quality rig, that's not going to fall apart the first time it's pushed a little harder at a loud gig. :)

    Edit: by the way, most solid state amps will deliver significantly more than the nominal RMS power level, when the amp is driven into the high-THD area. For instance, I was just looking at the manual for my Acoustic 370, which has a pretty typical transistorized output stage. It does 325 watts at 5% THD, and 385 watts at 10% THD. I'm guessing that's pretty typical unless one's power supply has specific limiting. So under those conditions, the extra 60 watts would probably end up "mostly" as heat (in the speaker coil, or maybe in the output transformer if the amp has one). And also as a further aside, the exact same thing happens with output transformers. Output trannies are notorious for burning up when they have to handle lots of distortion. A similar concept applies to power trannies, even the big ones used to drive the AC power grid. Generally speaking it seems, harmonic distortion is a coil-killer. :)
  10. Every amp will deliver more power than rated under overdrive conditions simply because there is more energy in a clipped signal than in an unclipped one. More area under the curve as it were. This is the most significant factor when addressing voice coil damage from 'underpowering.' The extra HF energy isn't really a significant issue in woofers, but can be deadly to tweeters in passively crossed over systems. Like Tim_X said, in a woofer, the coil inductance will increase the driver's impedance with increasing frequency which results in less power dissipated at high frequency. Couple that with the extremely crappy efficiency of speakers in general and it's not very significant at all. Same goes for the so-called 'DC' at the crests of clipped waveforms 'stopping' the cone dead and derating the power handling capability. First, a square wave at X-frequency will move the cone in and out the same number of times per second as a sine wave at X-frequency. Second, the inertia of the cone won't allow it to stop anyway unless the frequency is extremely low (i.e. less than 5Hz). The inertia allows it to follow a nearly regular path. Of course, the amp loses damping control during this... Of course, we're never getting into true square waves no matter how hard the amp is driven anyway...

    Of course there are also factors like whether the coil is coming out of the gap or not, and how much the speaker needs to be derated at certain frequencies due to cabinet tuning. You can fry a voice coil with perfectly clean power much less than rated power if you concentrate at the tuning frequency of the cabinet, because the acoustic suspension almost entirely damps cone movement at that frequency.

    Brick wall clip limiters can also cause problems with overheating voice coils, especially if they're really limiting the peak-average ratio.
  11. Man, you guys are smart! Very interesting points, thanks.
    Hey, Nonsquarter, what about our lil'ol WWU amps? They don't have clips lights, I guess w have to use our ears and common sense... I don't want to blow my little Bergies up...
  12. The Eminence bass speakers seem to be very widely used in many cab designs.
    The common ones seem to have 56 oz. 80 oz. or neo magnets.
    They also seem to have 2.5 & 3 inch voice coils.
    Are they the most efficient speakers out there ?
    I just got a 200w bass amp and since I don't have a lot of power I would like to get a cab that is VERY efficient.
    I'm looking at the Dr. Bass efficient is it ?
    They say on their website that it's 103 db.
    Is that true ?
  13. nonsqtr

    nonsqtr The emperor has no clothes!

    Aug 29, 2003
    Burbank CA USA

    WW's are "well designed" amps.

    They have all kinds of protection circuitry. They're very hard to kill. :)

    With a WWU, any speaker-related problems are much more likely to be related to over-excursion. Those amps are very dynamic.

    Also, you will definitely hear it, if and when your amp starts clipping. It'll be the nastiest, most horrible sound you've ever heard in your entire life. At that point, it's just a matter of heeding the warning. :D
  14. Tim__x


    Aug 13, 2002
    Alberta, Canada
    Nonsqtr, I think we do mostly agree.

    But, I (half) stand by what I said, a standard amp when clipping heavily and overloading the power supply, will output a lot of 120/100hrtz from the rails and there may be a bit of high-frequency ringing, but not much.

    I do, however, have to admit that ther are some VERY badly designed amps out there that will break into oscillation when clipped, and that the power compression you described is the usual 'cause of death' for tweeters.

    P.S. It's nice to see someone else who advocates differential preamps for basses.
  15. BassGreaser


    Aug 22, 2002
    Austin, TX
    I'm guessing ym old bassman 135 is not a well designed amp :eyebrow:
  16. Tim__x


    Aug 13, 2002
    Alberta, Canada
    I'm not familiar with the internals of that amp, but just because something is inexpensive doesn't mean it's badly designed. Do you have reason to believe it is?
  17. BassGreaser


    Aug 22, 2002
    Austin, TX
    it's just an old late 70's all tube head, and doesn't have any fancy stuff to it like alot of todays amps
  18. Tim__x


    Aug 13, 2002
    Alberta, Canada
    Sometimes it's the fancy stuff that causes problems, if it's never caused you trouble before I wouldn't worry about, it's probably very well made and well designed. Think about it this way, anything that survives 20, 30 years of use with no problems has to be fairly high quality.
  19. Jerrold Tiers

    Jerrold Tiers

    Nov 14, 2003
    St Louis
    Under-powering, as noted, isn't the culprit, I agree, the net result has to be too much power, or you won't hurt the speaker power-wise. (over-excursion may occur at way BELOW the rated power, at very low frequencies, like stage rumble, etc)

    As far as the power is concerned, however, there are some details.

    The speaker voice coil can only get rid of significant heat to the magnet structure. It isn't a good radiator by itself, no matter if wound on an aluminum former or not.

    So.....if (case 1) the voice coil is longer than the gap, which isn't unusual on a long-throw speaker, the part that hangs out can overheat faster, IF the speaker isn't moving much. So if you feed it lots of higher frequency power, where the speaker movement isn't as much, it might not tolerate as much power as at a frequency where all parts of the coil are moved regularly into the gap.

    Conversely, a speaker with a shorter coil, about equal to the gap, (case 2) may be able to handle more power if it does NOT move much, because the coil stays in the gap. If it moves a lot, some of the coil is outside the gap more than the rest, and loses dissipation capability.

    A speaker like "case 1" might be easier to kill "underpowering" than "overpowering", because you might not use the excursion, and fail to cool some voice coil area.

    Conversely, a speaker like "case 2" might be easier to kill by overpowering, and more stout if underpowered.

    I have seen speakers with a burnt area on the ends of the coil that looked like it had been done with a pen and ruler, it was so straight. That was the area that wasn't cooled as well.

    You won't know which "case" your speaker is like, so go by the ratings. About double the "RMS" is a good rule if you are likely to play hard. More if you just use "headroom" but you start to risk damage if there is a problem like feedback etc.

    Another and better reason for not under-powering is if the speakers have a "character" that you like. (often more a problem for guitar players than for bass) Typically the speakers need to be "pushed" hard to develop their "character".

    So underpowering may not allow them to "breathe" , "open up", or whatever description you want to apply to the "character" of the speaker coming out.