There are too many other design factors to consider than just cone area, especially in your "limited use" situation. This is why we have Thiel-Smalle parameters as it describes most all of the aspects of the driver. Once you know all of those factors, then you consider cabinet rigidity, volume, bass alignment, etc...
Other than that, driver size has nothing to do with efficiency or cabinet volume. It is quite possible to have a 12" that needs a bigger box and requires more power than an 18" driver. This all depends on the above mentioned factors and the over-all design of the product.
I can't really point you in any specific direction as i stopped looking at subwoofers a while ago. Most are ridiculously priced and under-designed. After all, a passive sub is basically a woofer, a cabinet, some stuffing, binding posts and some feet. How they think that this costs as much as multiple drivers, complex crossover networks, etc... I don't know.
The one thing that i would suggest would be to try and listen to / compare some sealed designs to vented designs. The one advantage that vented designs offer is that they will play louder. Since this is not a major concern of yours, you can take major advantage of the improved transient response, increased damping, lack of overhang and ringing, slower resonance, more controlled output at resonance, etc.... of a sealed box. You'll also get less "bloat", which means you probably won't annoy the neighbors as much. Whether or not you prefer the sonics of this type of design is a personal matter, but i thought i would mention it. Sean
> |
El: Thanks for taking the time to perform this second set of tests. I would only comment that a loudspeaker works much like an inductive electrical motor i.e. voltage is fed into a coil and the resultant fluctuations in magnetic field create motion of varying speeds.
Think about that aspect of operation and what is required to bring the motor up to a linear operating speed and you'll have a better idea of what happens to a speaker too. The fact that a speaker also has to deal with the mechanical / thermal losses of the suspension increases the variables involved and the linearity of operation. On top of that, motors typically don't change direction at a rapid rate of speed and / or change loading characteristics as signal is varied.
There's a lot going on here that may / may not show up on an SPL meter. Sean
> |
Bag End's are designed COMPLETELY different from ANY other subwoofer made. As such, they will respond differently. The one thing that will be consistent with Bag End's regardless of size as compared to other subs is that they will require more power to achieve the same level of volume. This is directly related to the design choices that they made when building these products.
Other than that, all drivers have a "knee in the output curve", just like all semiconductors do. The subs using a servo system can take this into account whereas those without some type of active equalization / monitoring system can't. I'm quite certain that the Velodyne's have this factored in. The Bag End's also vary bandwidth as spl is varied i.e. the lower the spl's, the deeper the sub goes. As spl is raised, the active EQ system reduces the boost applied at the extreme low frequencies. While this occurs naturally with ALL woofers to some extent, the Bag End's do this not only acoustically, but also electrically. Like i said, they are kind of a novel design. Sean
> |
A driver has both a mechanical suspension and thermal losses. Until you can overcome the majority of losses in both of those areas via increased drive levels, the driver itself won't be running in the region where it has the greatest linearity. At the same time, driving the speaker harder will eventually go beyond the sweet spot of operation, causing dynamic compression due to saturation of the mechanical suspension and / or thermal losses.
In effect, a driver is no different than any other type of device. They all have a region where they operate most linearly. Above and below that, they still work, just not as linearly. This is why various speakers sound better at certain spl levels i.e. don't come alive until cranked up or sound worse as they are driven harder. Bass alignment also comes into play here as a higher Q design will sound "fuller" at lower volumes, but becomes bloated and tubby as drive levels are increased. This is why speakers are the most specialized component in the system i.e. they vary the most in terms of room interaction / placement, listening style ( spl ranges, seated listening position, etc.. ) and personal preferences in terms of tonal balance, transient response, etc.. Sean
> |
El: Thanks for taking the time to not only perform the testing that you did, but for sharing the results with us. I look forward to your confirming both your earlier test results and my prior comments in your next post : ) Sean
>
PS... Try a bandwidth limited signal for greater accuracy. The larger / heavier / less efficient the driver, the more apparent the variances in drive levels will be until the threshold or "knee" in the curve is reached. After that, the input / output ratio should remain relatively consistent until driver compression comes into play. As can be seen by your initial testing, this is exactly what took place and what i said would happen. |
If you are going to test a sub, try to get the meter as close as possible to try and eliminate room nodes. Otherwise, the linearity that you measured last time will be thrown out the window. The room itself will have a "knee in the curve" in terms of the excitability of nodes, etc... Sean
> |
El: How is it possible for the motor structure of a loudspeaker, which is fed a non-steady state AC based musical signal, which varies quite drastically in both amplitude and duration, to operate anything like a DC motor?
After thinking about the above question and coming to a logical conclusion, look at your own quote here: "Perhaps you are thinking of AC motors, which do not develop much torque at low rpm."
Perhaps i am. The fact that we are feeding AC into a magnetic motor structure could have something to do with my thinking that. The fact that you acknowledge that such a system doesn't develop a linear torque curve, and is "weaker" at lower drive levels, is exactly what i stated above.
Like i said, the testing that you performed may not have been specific enough to reveal these problems. Making comparisons to different motor designs and modes of operation, which an AC motor or voice coil is to a DC motor, just because it fits a specific set of test results, doesn't make it so. It just means that the tests performed weren't capable of revealing what is really going on. The fact that you already knew the answer pertaining to AC motor operation as demonstrated by the above quote, but adjusted your train of thought to support your later test results and twist the comparison around, is neither consistent or scientific. Sean
> |
DC of any given amplitude will have a constant RMS value. AC does not, especially when using non-symmetrical waveforms like that of music.
DC of any given amplitude does not have a duty cycle. AC does not, especially when using non-symmetrical waveforms like that of music.
I didn't suggest that your test results were based upon false data. I said that the manner in which the tests were conducted may not be revealing enough to explain the criteria that i mentioned. Assuming that such data was all that one needed to formulate theories that were set in stone would be nothing less than incorrect and / or misleading to the general public. As i publicly stated, i appreciated the fact that you were willing to conduct such tests. Sean
> |
El: I never said that i doubted your test results, so why would i bother trying to duplicate them? What i'm saying is that this is probably not the most appropriate manner to try and ascertain the type of information that we are looking for. Sean
> |
