I might admire some bigger speakers, but I definitely limit myself to something I can move around on my own.
Fatparrot, yes mass does inhibit speaker movement in the "for every action, an equal and opposite reaction" way, which is to say that a woofer pumping back and forth won't move a speaker noticeably the other way. But you actually don't need a very heavy speaker to accomplish this task, especially if it's spiked to the floor, as evidenced by the fact that none of us are reporting our speakers 'walking' around the listening room (positioning the woofer low in the cabinet also helps reduce rocking motion, and of course positioning two woofers opposite each other and wired in the same phase can eliminate it). What sheer mass doesn't do is prevent the pressure waves from causing resonance -- think of the Liberty Bell. Some of the techniques and materials that are used to reduce resonance in speaker cabinets do add mass, but it's not necessarily the mass per se that prevents resonance (the more massive the thing that's set to vibrating, the longer it takes to damp), and lighter-weight designs employing clever forms and materials engineering can also be made low in resonance. I always thought it was a neat bit of iconoclastic thinking that Hsu made its subwoofers out of lightweight but rigid and nonresonant (as far as the material goes) cardboard tubes -- inexpensive, too. |
Macrojack: What you're essentially saying is that if you turn down the volume, the cabinet will move less. D'oh! :-)
Seriously though, if you want to achieve the same lower-frequency SPL in-room, you have to move the same amount of air. Can be done through either larger driver area or greater driver excursion. Both have engineering challenges if you want to reduce distortions of the driver (and therefore of the reproduced signal) in several forms. So some designs involve using many LF drivers, which can reduce both area and excursion on a per-driver basis, at the expense of increased cost for the multiple drive units and larger cabinets to house them all (which, to the topic at hand, obviously makes for heavier speakers). But the other trend in the industry has been toward EQ'ed subwoofers with small, inconspicuous cabinets, and these drivers have some amazing levels of peak excursion, at the expense of accepting increased distortion in an area where the ear is less sensitive to it, and targeted primarily toward end users who care more about movie soundtrack FX than music quality. Some of these little monsters can move around the floor under high SPLs if not spiked. |
Onhwy61 wrote: "The easiest way to limit the cabinet's output is to make it massive and less likely to be sympathetically excited. There are other ways to make the cabinet highly rigid and inflexible, but they are expensive (contrained layer, exotic materials, etc.)." I don't agree that it's as simple as saying that making something massive makes it "less likely to be sympathetically excited". What that does, all other things being equal (which they're often not), is lower the frequency at which sympathetic vibration occurs. I don't pretend to have anything like a comprehensive grasp of the subject (and seriously doubt that even many speaker designers do), but do feel it's a lot more complicated than just adding mass to deal with the problem. Shape, size, density, points of contact (for both damping and exciting elements), materials' intrinsic damping and rigidity properties, all go into the mix. If you add mass in the wrong shape or the wrong location or use the wrong material, resonance will worsen instead of improve, only shifted in frequency. Mass in and of itself isn't the answer to anything in this area, it seems to me, and if you can achieve the same intended result while using less mass I think it's always better, and not necessarily more expensive all things factored in (weight has its own costs). |
I guess I look at going from 1/2" to 1" or 3/4" to 1 1/2" MDF a little differently than simply saying it's adding mass. Yes, of course it is adding mass. But more to the point, I think, is that it's changing the ratio of dimensions by doubling the front to back wall thickness (plus it also often entails combining two layers with a an adhesive in between, a form of constrained-layer damping). This is primarily where the extra rigidity comes from. (Think of an I-beam -- it can nearly as rigid as the same dimensions of a solid bar, but is obviously much less massive, which can actually make for a stronger structure for practical purposes.)
And achieving higher rigidity itself is also not always the same thing as reducing resonance. Very non-rigid objects can be the most non-resonant objects -- think of a down pillow -- depending on their form and density. Increasing rigidity raises the resonant frequency (again, all other things being equal, which they likely won't be). Of course higher rigidity in a speaker cabinet is a good thing from a wavelaunch perspective. But without proper attention paid to shapes, sizes, damping, points of contact, etc., higher rigidity in itself might not reduce resonance but only change its character. Overall, to my mind both higher mass and rigidity are probably somewhat overrated compared with self-damping as means to achieve the design goal of a quieter cabinet, and probably more expensive as well, but for the fact that most manufacturers are already set up to deal primarily with MDF.
I thought the Sterophile measurements of the new Mordaunt-Short Performance 6 speaker, that's made of a combination of molded structural foam of varying shapes, thicknesses and densities plus reinforcing metal elements, was fascinating. The result (achieved at the initial cost of a reported 10,000 hours of design time and who what tooling costs) is a 48" tall, 18" deep at the base speaker weighing only 66 lbs., and yielding some very impressive accelerometer results in JA's cabinet resonance tests, with low amplitudes, no dominant frequencies and extremely short die-away. They cost $6,500, not really an outrageous price for an imported, high-tech, nearly full-range 4-driver floorstanding flagship, and the molded-contoured form probably means it has better diffraction properties than most MDF boxes to boot. Of course all that doesn't necessarily mean they sound great, but I do want to hear them if I can. |
Macrojack: Naturally your speaker cabs do resonante -- all speaker cabs resonate. I'm surprised that you and Line say you can't feel any vibrations with your hand placed on the cabinet. This is true with any surface, at reasonably lifelike volumes, playing any music? As for rocking, most speakers won't do this to any detectable degree if well designed and anchored.
Even if they don't obviously vibrate a great amount, the large areas of the cabinet surfaces involved in making a 4 ft. high speaker means there will be sonically significant unwanted contributions from the cabinet. The effect is potentially worst with speaker constucted from flat panels (as most still are) -- speakers with curved cabinet surfaces have a theoretical leg up here, especially if we're also trying to keep weight down to sane levels.
No matter what you do with the bass, cabinet talk will exist if there's a cabinet involved, but obviously speakers that roll out on the bottom sooner than others will have lesss of a problem to deal with, and this I assume is a factor with your speakers. BTW, I agree that 70 lbs, which is what my own Thiels weigh (with their 1" panels, 2" baffle, and 5 internal cross-braces), is at most mid-weight for a 4 ft. floorstander.
I do have to admit to a lack of basic understanding in a technical aspect of what you're alluding to here though. John Atkinson has also commented in a few speaker reviews that higher sensivity should translate into less cabinet talk. I think I understand why this would be true with horns, because they work by concentrating the sound pressure in the direction of the listener, thus reducing the overall driver excursion needed to achieve equivalent perceived volume levels. What I don't understand is why this should hold true for direct-radiators like yours, because the amount of air that's required to be moved to achieve any given volume level is the same, no matter what the efficiency of this type of speaker. In other words, I understand that the required amount of watts of power needed to achieve that air movement will be less with a more efficient speaker, but not why, if it's making the same SPL in-room, there would be less excitation of the cabinet. Anybody enlighten me on this one? |
Another advantage of having the full-range driver mounted to the cabinet top is that the "baffle" is thus made really small, while none of the larger-area cabinet surfaces are excited directly by a driver mounted in the middle of the panel. BTW, I thought Walsh Ohms had fairly low sensitivity, since their sound is radiated equally in all directions. Anyway, you can tell a lot about the cabinet sound of a speaker just by doing the knuckle-rap test at different spots on the various panels and listening critically to the results, no test equipment needed. |
Bartokfan: I assume that what you're referring to in B&W speakers, which is shared with many other speakerers, must be their internal bracing? If so, yes, internal bracing is pretty much mandatory for any speaker that asipres to minimal resonance, but the need for it can be lessened by employing curved outer surfaces, an approach which reduces weight.
I don't know what the percentage range would be for typical audiophile box speakers in terms of cabinet contribution to their total output, but you have to remember that energy reradiated from the cabinet has been shifted in frequency and time -- in other words, it's heavily distorted in relation to the desired goal of the output signal closely matching the input signal. As we know from various kinds of distortion is our electronics, if the effect is musically unconsonant or obscuring, distortion can be audible at very low percentage levels. True, within the signal chain speakers in general have much higher overall distortion percentages than do electronics, but if you want to make a better-performing speaker, you need to address these limitations wherever you can.
I'm in agreement with you that for me, any speaker I can't hump around comfortably by myself is just something that I'm personally not interested in owning, but I do think imposing such a limit probably puts a ceiling on the performance that can attained with typical speaker designs if part of what you like to play is full-range, dynamic music at higher volume levels. |
Bartokfan: I was saying that for those who like to listen at higher volumes, the smaller size that usually goes along with lower weight may put a limit on performance. But I don't agree that cabinet resonance, as its own issue, is a problem only for listening at higher volumes. IMO cabinet talk makes its unwanted contribution at all volume levels, and minmizing it, whether through brute-force construction or something more weight-efficient, pays sonic dividends no matter how soft or loud you like to listen. (I also should mention though, that I'm one of those who feel listening at unnaturally low levels, while sometimes unavoidable, does constitute its own form of distortion that renders reproduction less lifelike. Of course the same thing goes for listening too loud as well.) |
Finding a speaker that sounds 'good' at low levels is notoriously tough, but it is also rarer to find a speaker that can play at truly lifelike levels on large-scale material without seriously faltering than most audiophiles seem to assume. (My own speaker accomplish neither.) Either that, or probably most of us almost never listen at truly lifelike levels, but speakers that aren't large, driven by amps that aren't high-powered -- both of which are common in this hobby -- can't really get that done for the most part. Which is okay, because most listening rooms can't support that kind of volume anyway. Of course, if you mainly listen to a guy singing and playing an acoustic guitar, or string trios, the lifelike volume factor doesn't present as much of a problem. I coined a term I use to describe this theoretical factor (to myself, 'til now), "absolute amplitude distortion". But this concept really only applies to acoustic material, since there is no truly 'correct' volume setting for multitracked studio rock/pop, music which always depends upon electronic sound reinforcement in live performance. Still, no matter what kind of material you're playing, if you listen too low you'll miss musical detail and impact, and if you listen too loud you'll hear non-musical detail that's not an intended part of the performance, and most systems will throw the tonal balance off-kilter in either case. |
I haven't heard a Walsh Ohm in years and don't know much about them, but maybe the unique traveling-wave vs. traditional pistonic principle of driver operation works well at high SPLs? |
