Subwoofer Tests - Fall 2007

[ISLAND1000]

It's revealing that four out of the top five subs use TC Sounds drivers, that the top six subs are DIYers, and eight out of the top ten subs are DIYers.
This list contains some of the best commercial subs available and the DIY subs outperfomed them in these tests. We are at a point where DIY knowledge, available software modeling programs, and cooperation on these forums enable the hobbyest to make for themselves a sub that can outperform commercial units and probably do it for less money per performance dollar.
Congratulations all around.

[/quote]

[Sonnie]

Ilkka... weren't you going to test the SoundSplinter RL-p15 driver? I'm wondering if it might be comparable to one of the TC Sounds drivers.

[WillyD]

Ilkka... weren't you going to test the SoundSplinter RL-p15 driver? I'm wondering if it might be comparable to one of the TC Sounds drivers.

It should compare to the TC-2000s that Ilkka tested this spring. They aren't exactly identical, but close enough I'd say.

[Ilkka]

Sonnie wrote:

Ilkka... weren't you going to test the SoundSplinter RL-p15 driver? I'm wondering if it might be comparable to one of the TC Sounds drivers.

Unfortunately no one brought any RL-p based sub in. I would gladly measure one (15" or 18").

But as Willd said, it should be pretty close to TC-2000.

[anidabi]

funky_waves wrote:

When I was talking to Mike at Tc he said the were working on a new rubber for the tall surouds specificaly for the Audiopulse LMS ultra that dosn't cavitate(wrinkle) nearly as much. Would be nice to see measurements of the LMS Ultra to compare. Too bad they are $2999 and anyone who sells them can only sell them in there local area, I know because I am going to be selling them, localy only of course.

Yeah, good news, but with that price I would expect neodymium magnets for that driver.

[anidabi]

Ilkka wrote:

How can you say that? Have you done any testing at different temperatures?

When I said that lower temperature means slightly better results, I meant regular long sine sweep tests. They put a lot of heat into VC and ambient temperature plays some role on how the VC dissipates this heat. But the short burst of the CEA-2010 is vastly different. Because it's only 6.5 cycles long, it doesn't create much heat at all. Therefore ambient temperature doesn't matter that much. 10 C or 20 C will give very close to identical results. That is one more reason why I started to use it: accuracy and especially consistency.

And in any case, the ambient temperature for rounds 4 and 5 was within 2-3 C.

This is interesting. I didn't think about the voice coil heating at all, but like jakeman said, the first thing that came to my mind was the driver surrounds. How does that affect to the results? Especially with LMS drivers wich have rubber surrounds. Maybe the noise at high output was partially from the surround, because of the coldness? Higher distortion levels or something like that?

[Ilkka]

anidabi wrote:

This is interesting. I didn't think about the voice coil heating at all, but like jakeman said, the first thing that came to my mind was the driver surrounds. How does that affect to the results? Especially with LMS drivers wich have rubber surrounds. Maybe the noise at high output was partially from the surround, because of the coldness? Higher distortion levels or something like that?

10 C isn't cold enough to have any real difference. Below 0 C would be a different case though but no one ever tests at those conditions.

The tall profile surround makes the cavitation noise even at normal room temperatures at high excursion (over ~25-30mm one way; think about the SPL at this point! ). But it's not as bad as people may think, it's just shows up in measurements because the LMS has such ultra low "other" distortion. Other drivers have so much other distortion that this kind of surround distortion doesn't present itself.

[ISLAND1000]

Ilkka wrote:

The tall profile surround makes the cavitation noise . . . . . . it's just shows up in measurements because the LMS has such ultra low "other" distortion.

Ilkka, will the other remaining test parameters indicate the presence of the "cavitation noise"?
In my listening tests of the LMS driver I noticed the "beating or slapping noise" IE: cavitation only at high excursions and the lower end of 10-19 Hz sweeps. I was attempting to fiind the "safe" outer limits of the driver. I had originally thought the sounds might have been soft V/C bottoming but after seeing your tests and this discussion I think it was the tall surround.
This cavitation noise would only occurr or correlate to ear splitting spl at higher frequencies. At this extreme level of spl my sub unit would be many times louder than my mains.
If the surround is producing sounds of it's own, at what frequencies does it occurr?

[SteveCallas]

I ignored the SDX15" results the first time looking over this data, looking only at the LMS, but now that I do, things get even more interesting. The 15" SDX, with its smaller voice coil and all, can outdo the TC2k from 25hz on up, and two can outdo the 18" LMS in the same frequency range for less money. That's pretty significant me thinks....I wonder why it doesn't suffer from nearly as much compression? It looks like it would be an excellent fit for a LLT, as it could really benefit from the additional clean headroom down low - that would make for a great sub. It would look to be a notch above a TC2k LLT performance wise.

Ilkka wrote:

Do I see a dual LMS-5400 18" sealed 200L (with two CE4000s) in the future? That would be pretty close to my ideal subwoofer. Talk about some crazy output in relatively small package. And the total should still stay below ~$3500.

I dunno, from this preliminary data, it looks like multiples of the SDX would be the better option, no?

ISLAND1000 wrote:

And I'm guessing the reason you'd choose sealed over ported or PRed is because of the lower distortion numbers and flatter, lower, frequency response possibilities using EQ for the sealed subs

Hehe, no. Ported or reflex will always have less distortion, sometimes considerably less, and they can aways be made to result in a flatter, deeper extending FR with more headroom. The main reason to go sealed is to save space.

[SteveCallas]

Ricci wrote:

It doesn't outdo it by 6db at any one frequency but an average of 4.45db greater across the board in a smaller enclosure isn't too shabby in my book.

No, not too shabby indeed, but by going from a ~$300 15" driver with no exotic low distortion motor technology and a ~$10 6" diameter port to a ~$875 18" driver using LMS technology and two ~$300 (?) PRs, I was expecting more. Two SDX LLTs would look to outperform the LMS even more, while still costing considerably less.

You pay a big premium for a smaller enclosure.

[WillyD]

I ignored the SDX15" results the first time looking over this data, looking only at the LMS, but now that I do, things get even more interesting. The 15" SDX, with its smaller voice coil and all, can outdo the TC2k from 25hz on up, and two can outdo the 18" LMS in the same frequency range for less money. That's pretty significant me thinks....I wonder why it doesn't suffer from nearly as much compression? It looks like it would be an excellent fit for a LLT, as it could really benefit from the additional clean headroom down low - that would make for a great sub. It would look to be a notch above a TC2k LLT performance wise.

Its the <25Hz performance that concerns me with the SDX. And since when have you or anyone else cared so much about >25Hz performance from a sub? Getting that has never been the hard part. Sure the pair of SDXs in 140liters edges out the single 5400 in 100liters from 31.5Hz to 80Hz, but wow, look at the 20Hz and below performance. The 5400 crushes it, with an average of 5.7dB more output; and that is with one CE-4000 vs the pair of amps for the twin SDX sub. So in reality, the pair doesn't outperform the LMS-5400 for less.

And you've seen the power compression magnitude measurements already? I didn't know Ilkka released them yet.

And there are a few differences between the tests in the fall and the ones in the spring. Fall, Ilkka used the CE4000 all around. Superior amp, as you know, the 2000 could've definitely used a more powerful amp in the spring. There is at least a 1000W difference between the CE4000 and the t.amp TA2400 MK-X used with the TC-2k. Something tells me that extra 1000W would bump up the TC-2000s CEA-2010 clean output results from the spring just a tad.

I'll still wait and see the full set of results. The CEA-2010 table isn't enough by itself for me to completely judge a sub. Its neat, but the individual threads tell the real and true story.

[Ricci]

SteveCallas wrote:

No, not too shabby indeed, but by going from a ~$300 15" driver with no exotic low distortion motor technology and a ~$10 6" diameter port to a ~$875 18" driver using LMS technology and two ~$300 (?) PRs, I was expecting more. Two SDX LLTs would look to outperform the LMS even more, while still costing considerably less.

You pay a big premium for a smaller enclosure.

Of course it is much cheaper to go with the SDX's or TC2K's. Not everyone wants to put 600litres worth of subwoofer/s in their room though, and some people will always be willing to pay to get big performance out of a smaller sub. Also how is it that the premium is that BIG? 2 SDX's is about $600 and keep in mind that you need 2 CE4000's at $600 a pop used, (that is actually quite the deal there), that's roughly $1.8k. The LMS was $875, $600 for the CE4000, and $600 for the PR's(I think that this is slightly high, especially when CSS has 18" PR's for $139), for a total of $2075. $275 is not a lot of money when your already close to 2G's invested.

[bsoko2]

Ilkka, is there any way to test the HSU 3.3 Turbo with the HSU MBM-12 together? This is what the MBM-12 was made for, use with the HO or 3.3 Turbo. I would be interesting to see how they would rank in the pack as a single unit.

Bill

[Ilkka]

ISLAND1000 wrote:

Ilkka, will the other remaining test parameters indicate the presence of the "cavitation noise"?

Basically no.

This cavitation noise would only occurr or correlate to ear splitting spl at higher frequencies. At this extreme level of spl my sub unit would be many times louder than my mains.

That's true. It's not that of a problem in real world use.

If the surround is producing sounds of it's own, at what frequencies does it occurr?

It's around 8-10th harmonic.

[Ilkka]

SteveCallas wrote:

I ignored the SDX15" results the first time looking over this data, looking only at the LMS, but now that I do, things get even more interesting. The 15" SDX, with its smaller voice coil and all, can outdo the TC2k from 25hz on up, and two can outdo the 18" LMS in the same frequency range for less money. That's pretty significant me thinks....I wonder why it doesn't suffer from nearly as much compression? It looks like it would be an excellent fit for a LLT, as it could really benefit from the additional clean headroom down low - that would make for a great sub. It would look to be a notch above a TC2k LLT performance wise.

When comparing those results, do remember a few things. The dual 15" TC-2000 had a surround failure. I'm sure it affected the CEA-2010 numbers below ~40-50 Hz. The single 15" TC-2000 had an off-spec driver. I'm sure it affected the CEA-2010 numbers above ~30-40 Hz. It's unfortunate that these things happen, but on the other hand it's the reality, too.

I dunno, from this preliminary data, it looks like multiples of the SDX would be the better option, no?

It depends what do you mean by better? Distortion and enclosure size wise the LMS is much better.

I would definitely wait for the full results before jumping into conclusions.

[Ilkka]

WillyD wrote:

And there are a few differences between the tests in the fall and the ones in the spring. Fall, Ilkka used the CE4000 all around. Superior amp, as you know, the 2000 could've definitely used a more powerful amp in the spring. There is at least a 1000W difference between the CE4000 and the t.amp TA2400 MK-X used with the TC-2k. Something tells me that extra 1000W would bump up the TC-2000s CEA-2010 clean output results from the spring just a tad.

Around 1 - 1.5 dB more as 2400W vs. 3600W should be. Though it should be noted that not all drivers can take 3600W, even peak.

I'll still wait and see the full set of results. The CEA-2010 table isn't enough by itself for me to completely judge a sub. Its neat, but the individual threads tell the real and true story.

Spot on, Willd.

[Ilkka]

bsoko2 wrote:

Ilkka, is there any way to test the HSU 3.3 Turbo with the HSU MBM-12 together? This is what the MBM-12 was made for, use with the HO or 3.3 Turbo. I would be interesting to see how they would rank in the pack as a single unit.

Bill

I don't have them anymore so that's not possible. Nor it would give us any new data. You can simply add their SPLs together to see how they would do as a pair. Of course it also depends on what kind of crossover configuration one is using with them.

[Chris Brunhaver]

Hi Ilkka,

You didn't happen to measure the amplifier output voltage during the distortion test sweeps, did you? While you are already doing a massive amount of work, this would help settle the debate about how much power is required to really push these DIY sub drivers tot their CEA-2010 distortion limits etc.

As you mentioned, most subs aren't going to do much with anything with that 1.76 dB of gain (going from 2400 watts to 3600 watts).

Wild,

It's not like you need 7200 watts of power to run a pair of 15" subs to their limits. While I haven't read everything in detail, I think Ilkka used this much power so that clipping etc. was simply not a factor.

[WillyD]

You didn't happen to measure the amplifier output voltage during the distortion test sweeps, did you?

Good Q. Another reason I am critical of not Ilkka, but comparing the spring DIY results to the Fall. I trust the CE-4000 map more than I trust the t-amp, if you know what I mean.

As you mentioned, most subs aren't going to do much with anything with that 1.76 dB of gain (going from 2400 watts to 3600 watts).

That is, if the difference was really 2400w to 3600W.

Wild,

It's not like you need 7200 watts of power to run a pair of 15" subs to their limits. While I haven't read everything in detail, I think Ilkka used this much power so that clipping etc. was simply not a factor.

Absolutely, but the amount of power available would most certainly make a difference. You even bring up clipping, which was present when the TC-2000 was tested. I am just calling it like I see it. Its not fair to completely compare the spring 2K results to the Fall SDX results and declare a "winner". And even if one wants to, wait for the full results, and look at the whole picture (as in, not just 25Hz+ performance, but <25Hz performance...you know, where subs are supposed to excel.)

The CEA-2010 standard defines a new way for measuring and determining the clean maximum output level for a subwoofer. The signal used is a 6.5 cycle long sine burst which allows a safe measurement of the maximum output level. The max SPL is limited by the stepped distortion threshold (allows less distortion for higher harmonics) or the limitations/limiters of the subwoofer itself, which ever is reached first. The standard defines the max SPL normalized to 1 meter distance (half-space) and peak value of the sine burst, but I use the more familiar 2 meter distance and the RMS value of the sine burst. That way these figures are more comparable to the results measured by using the more common sine wave sweep/tone method.

[SteveCallas]

Willy wrote:

Its the <25Hz performance that concerns me with the SDX

Correct, it's not as good as the TC drivers in low end efficiency, but using it in a LLT would bring up the low end output. The TC2k quasi LLT was cleaner from tuning to 35hz than it was above 35hz due to the port output, so if you start with a driver that trades a bit of cleanliness down low to gain it up top, you should end up with more consistent, low distortion.

And since when have you or anyone else cared so much about >25Hz performance from a sub?

Umm, I do like to hear bass in addition to feeling it . The majority of the output of the sub will come from >25hz. <25hz is definitely very significant nowadays, but you gotta have your ducks in a row and have the frequencies we are more sensitive to clean first.

And there are a few differences between the tests in the fall and the ones in the spring. Fall, Ilkka used the CE4000 all around. Superior amp, as you know, the 2000 could've definitely used a more powerful amp in the spring. There is at least a 1000W difference between the CE4000 and the t.amp TA2400 MK-X used with the TC-2k. Something tells me that extra 1000W would bump up the TC-2000s CEA-2010 clean output results from the spring just a tad.

For sweeps, I would definitely agree. For 6 cycle bursts like were used (I believe?) for the CEA testing, I'm not so sure. Ilkka - was power the limiting factor for the TC2k CEA test above 25hz, or was it distortion?

The CEA-2010 table isn't enough by itself for me to completely judge a sub

I would definitely wait for the full results before jumping into conclusions.

Absolutely - we have to talk about something though. I'm neither completely judging a sub nor jumping to any conclusions, I'm simply making observations based on the data presented.

Ricci wrote:

Also how is it that the premium is that BIG? 2 SDX's is about $600 and keep in mind that you need 2 CE4000's at $600 a pop used, (that is actually quite the deal there), that's roughly $1.8k

I disagree. You may need two such amps to extract the bulk of the performance from a LMS 18, but you definitely don't need that much to get the bulk of performance from a SDX. A pair of EP2500s would do it.

Ilkka wrote:

Distortion and enclosure size wise the LMS is much better.

From sweeps or from bursts? Isn't burst testing more representative of real world material reproduction?