Originally Posted by ChrisBee
Ilkka's tests show that a small, sealed box is unusable with that 18". Distortion rises exponentially above a certain output level. Play an infrasonic (or near infrasonic) tone and you would hear a false harmonic signature. Each harmonic masks the next lower one because they map the Equal Loudness Curves. A flat frequency response favours the upper harmonics and betrays the fundamental. It is a pyramid built to a false god. Of course a bass guitar sounds realistic. It has no low fundamentals and is rich in delicious harmonics.
Ugh. This again.
One more try;
When the LMS was tested in a 100L box, everyone proclaimed it as being the most linear driver ever, although 100L is technically already too small a Vb
for that driver.
In the ASD discussion that lead to the test, my point was that Vilchur proved air spring linearity vs suspension linearity 1/2 a century ago by removing a then popular IB driver from its huge box, decimating its surround and spider and replacing it with a makeshift, extremely high compliance suspension and reinstalling the driver in a very small Vb
The result was:
Julian Hirsch, who at the time published the Audio League Report, wrote that 'the AR-1 had the lowest electroacoustic efficiency of any loudspeaker on the market - but at 25 Hz and below, it was more efficient that the Klipschorn, which had the highest efficiency of any speaker on the market.' Hirsch also said that the AR-1 'established a new industry standard for low distortion bass.' Eventually, the industry began to realize that the smaller size of the cabinet was only a secondary advantage of the extended bass response."
Instead of Ilkka using the 100L results and comparing them against the same driver/amp in the 200L Vb
(or whatever size he preferred), he claimed that the 100L tests were a year old and, therefore, not comparable to any new test, so he fashioned the smallest box the driver could physically fit into and then added ballast to effectively shrink the Vb
A test that would have been in keeping with the actual discussion would have entailed changing the drivers suspension from its production parameters to a high compliance driver in the same 100L box and comparing the results to the tests in 100L that were already done, or using a high compliance driver with more sensible differences in Vb
That said, Ilkka's test was largely unnecessary, as we had already discussed the ASD formula which could easily have predicted the results and the THD by component numbers closely enough.
ASD Formula: 140 * one-way swept volume/Vb
140 * 4.18L/200L = 2.9% THD
140 * 4.18L/75L = 7.8% THD
A difference of 2.67 times increase predicted by the formula.
Looking at the actual results, taking the average THD of max output sweeps at 10, 15 and 20Hz:
200L = 10%
75L = 23.7%
23.7% is 2.37 times 10%, showing a slight difference from the prediction via formula vs actual, which may be attributable to any number of factors, but would have saved the effort to prove something that was already known but didn't address the actual topic of discussion.
Since we have the 75L and 200L numbers, let's take a look at the numbers that no one seems to be interested in, the CEA2010 numbers, taking the average from 10-80Hz:
200L = 109.43dB
75L = 108.06dB
So, here we have the max clean output per CEA's standard. Does anyone out there actually believe that there will be an audible difference at normal listening levels?
Just for fun, let's compare the 200L CEA results with the existing 100L results, averaged from 12.5-80Hz, because there is no 10Hz number for the 100L results:
200L = 111.26dB
100L = 112.05dB
Interesting enough results to warrant discussion, IMO. A discussion which never took place.
There is an ASD rule of thumb for anyone who wants the smallest Vb
for his chosen driver in a sealed. LT system. Vb
= or > total swept volume/5%.
For the LMS-18: 8.36/5% = 167.2L. That's roughly equivalent to a stuffed 140L box. Take that sub and test it against any sized Vb
you prefer (up to and including IB) and you'll see the same things, proportionally, that I've shown in this comparison of the 100L and 200L versions of the LMS driver:
The larger you go, the worse the performance will be below 30Hz unless you a) Use a smaller amplifier, or b) Use a HP filter to prevent severe over excursion.
Option a will limit performance above the knee and option b defeats the purpose for building a 2nd order sub. Generally, under normal use, there will be no discernible SQ or clean output difference.
Bottom line: CEA 2010 results show the LMS 5400-18" and 3KW in any of these boxes (75L + boost EQ, 100L/no EQ, 200L/no EQ), far from one being 'unusable' vs any other, in a typical room will perform similarly enough as to be indistinguishable from each other in actual use.
Except for the fact that one of them will be a whole helluvalot bigger.