Why low sensitivity speakers?

You can pick any two of these three options:
1. good bass
2. small cabinet
3. high sensitivity
This is an immutable law of physics. Since many people want the first two due to decorating and other domestic considerations, the third is the one that gets left behind. And since watts are relatively cheap, it is not a problem. High-sensitivity speakers have their own virtues, but nothing will ever change the above law.

First consider radiating area. Typical loudspeakers are only about 1% efficient in an overall sense (that is, for 100 watts of electrical input power, you get an output of 1 watt of acoustical power). The reason for this is that there is a bad impedance mismatch between the surface of the driver and the surrounding air. Think of air as a medium, like a fluid only much lighter and more compressible. The driver only "sees" a very small, light load on its diaphragm, and as such is unable to impart much force against it, because the air moves out of the way so easily. The driver can exert much more force electromechanically than the air can accept acoustically, thus the "impedance mismatch". The reason that horns have such fantastic efficiency is that they gradually expand, allowing the driver to "see" a much larger surface area of air. That is, the driver ends up loaded by the area of the horn opening rather than its own diaphragm area. This makes the air appear much "stiffer" to the driver and results in a much better impedance match. The other way to achieve better impedance matching is to use a lot of direct-radiator surface area. Doubling the radiating area gives you 3dB of efficiency all by itself, just because of the improved coupling to the air. So a 12" driver is inherently four times more efficient at coupling to the air than a 6" driver is, and generally will require four times the enclosure volume as well. (The inside of the box "sees" four times as much air being pushed into it, so for the same compliance, needs four times the volume. It's the same as if there were four 6" drivers in the box.) The reason that 12" drivers aren't vastly more efficient than 6" drivers is that they have a much higher moving mass, see below...

For every additional octave of bass extension, you have to move four times the volume of air. This leads immediately to the necessity for large drivers with large excursions, which in turn requires large enclosures to support them. There is a tradeoff that can be made, though, if you can live with a lower output level capability.

Driver efficiency is primarily a function of magnet force and moving mass. (Think back to F=ma. Sound is nothing but the acceleration of air molecules back and forth. The higher the acceleration, the louder the sound.) So if you increase the mass, you get a lower efficiency but also a lower resonant frequency (better bass extension). Thus you can take a 6" driver which would normally have a resonant frequency of 60Hz, and by doubling the mass and the suspension compliance, get the resonant frequency (and thus the extension) down one octave to 30 Hz. You lose 6dB of efficiency and 12dB of output capability in the process! (Remember the four times air volume principle? This is where it comes back to bite you.) But in many cases, a tradeoff like this is made in order to get good bass at limited output levels out of a small driver.

Hope this helps.

To Unsound: to use my analogy above, the point to a horn is that the air is unable to "escape" away from the radiating surface of the driver, that is, to just move to the side and out of the way of the driver, thus unloading it. In a horn, the gradual expansion forces the driver to load a much greater volume of air directly in front of it, and doesn't "release" the driver from this loading until the horn mouth opens into the room, at which point the surface area that the air is pushing against is immense. True horns are so efficient at loading their drivers that they require what are called compression drivers, which are designed to deliver much higher than normal force at much lower than normal excursions.

To Zaikesman and Pmwoodward: a stiff vs. flexible cone mostly affects the fidelity of the signal, not the inherent efficiency of the driver. That, as I said, is mostly dependent on magnet force and moving mass. Of course, enough flex will result in energy being dissipated as heat within the cone itself, which lowers the efficiency, but very few drivers flex enough to do this within their designed passbands. They will do it as they start to break up at the top of their range, but hopefully the crossover has taken over by then.

As usual, I agree with Twl. The problem has to do with two different materials properties: STIFFNESS and DAMPING. These are two entirely different things, but you would be amazed how much confusion there is about them. I have seen plenty of references to how well-damped aluminum is, for example in the Stereophile review of the Krell LAT-1. For anyone who thinks that aluminum is well-damped, do a simple experiment: go to your local musician supply store, and pick up a tuning fork. Knock it against your skull, then keep it next to your ear until you can no longer hear it ringing. Multiply the 10 seconds (or whatever) by the 420 Hz frequency (or whatever), to get the number of cycles it took to die out of audibility. Now ask yourself one simple question: "What's it made of?" Chances are, it's ALUMINUM, one of the most poorly damped materials known to man.

Sorry, but had to get that off my chest. To get back to the issue, different materials have very different combinations of stiffness and damping. Aluminum is very stiff and very poorly damped. Kevlar is still quite stiff, better damped, and lower in density as well. Plastics are generally quite flexible, and usually better damped still, but have a wide range of variability in both stiffness and damping depending on the formulation and the fillers used. Paper is typically stiffer than most plastics, and by itself is not as well damped, but when coated with the correct polymer coating achieves a very good compromise between stiffness and damping.

What you take from this is that there is no perfect material for making cones. What you would like is something with infinite stiffness and infinite internal damping, in addition to zero mass. This combination does not exist in the real world. Aluminum's advantages are very high stiffness, good formability, and relatively low cost. The price you pay is a GIANT resonant peak when the thing finally breaks up. There are two solutions: one, do what Joseph does and use a very high crossover slope to get it out of audibility, or two, do what Thiel does and push the resonant peak high enough in frequency that even a shallow-slope crossover can do a good job of removing it. (Notice the large voice coil on the 1.6? It's there to get the cantilevered length of aluminum down to a smaller value, to drive the resonant frequency up.) Kevlar is probably a better material in that its stiffness is still very good but it has much better internal damping. I believe B&W has a patent on single-layer Kevlar cones, which may be why they are "attached" to that lately. (Like a lot of patents, I'm not sure it would hold up in court, but they have enough lawyers to discourage anyone from trying.) There are many plastic formulations, some of which are quite good, usually using some type of mineral filler to add stiffness, often magnesium based in order to keep it light. The French manufacturers Audax and JMLab/Focal have experimented with all kinds of cone materials, including all kinds of exotic polymers and fiberglass and kevlar sandwich cones, but none of them have impressed me much. Eton made their living on fiberglass/honeycomb sandwich cones, which are very stiff but also have a nasty breakup peak. The only exotic construction I've seen that had some good sense behind it was the Ensemble driver from Switzerland, a sandwich of two thin layers of aluminum separated by a thick layer of EPS foam. Stiff, light, and well damped. Also very expensive to make and even harder to get good quality control due to the variability in EPS foams.

After many years of experience, I personally have come down in the camp of "make it as stiff and light as possible while still keeping very good internal damping". This, believe it or not, means coated paper cones or a very few select plastic-cone drivers.

Cdc, to answer your specific questions: Yes, ribbons eliminate this type of breakup, but they have resonant modes of their own of a different kind. Re cone "breakup", there are two separate issues. You have to distinguish between "piston mode", which is defined as the frequency range in which the cone still behaves as a piston (a flat inflexible surface), and "breakup mode", where the accelerations have gotten so high that the cone itself begins to flex and resonate in response to the drive signal. Even in "piston mode", the sound waves travel outward from the voice coil to the edge of the cone, and then reflect backwards down the cone. Note that sound waves travel MUCH faster in solids than in air, and that they will dissipate as heat if the material has good internal damping. The reflection at the edge of the cone is also highly dependent on the type of surround used; some surrounds do a much better job of absorbing this wave than others. B&W's claim in this area is that since the bidirectional Kevlar cloth they use for their cones is a non-homogeneous material (different stiffnesses in different directions across the cloth), it will do a better job of breaking up this wave. This may or may not be of much significance; I personally would much rather see a cone material with very high internal damping. In "breakup mode", where the cone is literally going crazy with internal resonances, again I would prefer a material with very high internal damping. Because you're right, you can hear cone resonance very easily, and it's not a pretty thing.