Subwoofer Tests Round 5, 6th of October 2007, Test Summary ***READ ME***

[jakeman]

Ilkka wrote:

Hi all,

I'm heading out on a short New Year holiday so I won't be online for a few days. But when I'm back, I will tackle all the questions regarding the subwoofers measurements you have made.

Happy New Year to everyone!

Hi Illka

Trust you had a Happy New Year and a relaxing holiday. I was idling away some time today during my ski trip here in the Canadian Rockies and went looking for some spectral contamination charts for your latest sub tests and see you have not posted them. While I agree with you that the cross-section of charts does provide a pretty good indication of performance, I found the spectral contamination charts useful as all encompassing charts of distortion and noise.

As per your definition;

Spectral contamination test will show all non-linear distortions and mechanical ‘self-noises’ (panel vibrations, driver motor noise, port noise etc.). Also the limiting circuits often became very audible during this test.

While many readers may not appreciate its significance, this chart does provide a very useful measurement of real time performance. Generally speaking, many larger subs at higher output can produce significant port or resonance noise and do not fair as well as other subs such as several sealed subs in this area. Accordingly, it would be useful information when comparing the relative performance at high output between these two types of alignment and also among vented subs.

[Ilkka]

allredp wrote:

Which one would you buy for yourself, representing the best cost/performance ratio?

If looking at the commercial subwoofers only, the SVS PB13-Ultra stands out as having a very high cost/performance ratio. But then again it's physically quite large so it doesn't come without expenses. For example the JL Audio Fathom f113 represents another kind of approach: it's very small yet very powerful subwoofer. Then again it costs much more than the PB13-Ultra. Pick your poison, as they say.

[Ilkka]

rubbersoul wrote:

I to am curious as to your personal opinion on the JL Audio Fathom 113....I happened to buy one a month ago and recently as Dec 23rd actually got to calibrate it and use it and I was blown away. It has everything I think a sub should be.
Feel free to sendind a PM.

Merry Christmas to everyone!!!!

Frank

tweakophyte wrote:

Now... was the JLA's distortion that audible? In other areas people have compared that sub to the Ultra... now I am doubting their ears

I'll reply to these in the Fathom f113's own test thread.

http://www.hometheatershack.com/foru...-f113-new.html

[Ilkka]

darrellh44 wrote:

Hi Ilkka,

Do you have the means to measure/record amplifier and sub output distortion at the same time. For your next round of tests, it would be informative to see how much distortion/compression the amp is contributing along side the sub measurements. It seems like the CSS 2xSDX15 had an unfair advantage in the latest round of testing since it used two CE4000s. Since it's impractical to use two amps with every sub, at least knowing a single-amp sub's input distortion would help interpret the overall results.

When it comes to commercial subs, it's not possible since I would have tear them apart to be able to access the speaker wires. DIY subs with external power amplifiers are another case; it would be possible to measure and record the amplifier distortion, and in fact I have some preliminary plans for this.

I wouldn't say the dual CSS SDX15 had an unfair advantage since it used two amplifiers. DIY means that one can use as much power as his wallet allows so I decided to see how far I could push those woofers. And as you can see from the results, the woofers became the limiting factor, not the amplifier power. Also remember that 2x power means only 3 dB higher output.

[Ilkka]

jakeman wrote:

Hi Illka

Trust you had a Happy New Year and a relaxing holiday.

Thank you, John!

While many readers may not appreciate its significance, this chart does provide a very useful measurement of real time performance. Generally speaking, many larger subs at higher output can produce significant port or resonance noise and do not fair as well as other subs such as several sealed subs in this area. Accordingly, it would be useful information when comparing the relative performance at high output between these two types of alignment and also among vented subs.

Both ported and sealed subs can (and usually will) start to sound stressed when pushed high enough. Whether this is port chuffing, amp clipping, driver noise etc, it's always not wanted and usually very audible. But neither me or the results I have measured won't agree that larger (ported) subs would be more prone to these kind of noises than small sealed subs. Instead of being a size or an alignment difference, it's usually noticed because some subwoofers have much aggressive and/or transparent output limiters than other subwoofers. Some subwoofers can be pushed several dB into compression without sounding stressed or outputting excessive distortion. Some others let the amplifier clip and/or the woofer excursion run out, and that will usually sound much worse than heavy limiting. In either case, the subwoofer is being overdriven and it isn't operating in its linear range anymore. If this happens constantly, you should buy one more subwoofer (or more if even two won't give you enough clean output).

[jakeman]

Hi Illka,

I found those spectral contamination charts to be another good indicator of sound quality. Its not all about overdriving the sub which should always be avoided. Extraneous noise can occur during normal levels. Those charts also reveal the amount of self noise, as you describe it, during normal operating levels. 90db multi-frequency stimuli should not cause too much distress yet the chart does reveal alot about non-linear behaviour and other noise. Those charts and the spectral decay chart are quite telling about performance.

Perhaps I just like getting into more of the detail but I believe that data will also add to your already excellent body of work. They would also be another unique reference since you have included them in your earlier tests

[allredp]

Ilkka wrote:

If looking at the commercial subwoofers only, the SVS PB13-Ultra stands out as having a very high cost/performance ratio. But then again it's physically quite large so it doesn't come without expenses. For example the JL Audio Fathom f113 represents another kind of approach: it's very small yet very powerful subwoofer. Then again it costs much more than the PB13-Ultra. Pick your poison, as they say.

Excellent insight--much as I figured considering your test results, as well as, user reviews. Thanks for responding...

BTW, I have an F113 in my home right now--along with my own SVS PB12+/2 for comparison (it is a friends JL that I'll be helping him install next week ).

Interestingly, the +/2 dug a little deeper(the inaudible shudder stuff), but wasn't as impressive across the rest of the hz. I have the +/2 in 20hz mode and they have both been in the same room location. I'm running my Yamaha 1800 in pre/pro with my Sherbourn 5 x 200wpc amp.

My favorite demo scene is chapter 12 of The Island, which has all sorts LFE--from machine guns to explosions to score/sound effects galore. I haven't found a better test for the entire LFE range.

I'm going to boost them both a little more tonight and see how they respond (so far I've had them at -4db with the YPAO set sub gain).

At this stage, due to price, I'm really looking at the 13 Ultra, though the ARO and the size of the JL are really impressive!

What I really need to do is get a 13 Ultra and an F113 in my room at the same time! Hummnnn...

Thanks so much for your incredible work, Ilkka!!! You provide an invaluable service man!

[Ilkka]

allredp wrote:

Excellent insight--much as I figured considering your test results, as well as, user reviews. Thanks for responding...

BTW, I have an F113 in my home right now--along with my own SVS PB12+/2 for comparison (it is a friends JL that I'll be helping him install next week ).

Interestingly, the +/2 dug a little deeper(the inaudible shudder stuff), but wasn't as impressive across the rest of the hz. I have the +/2 in 20hz mode and they have both been in the same room location. I'm running my Yamaha 1800 in pre/pro with my Sherbourn 5 x 200wpc amp.

My favorite demo scene is chapter 12 of The Island, which has all sorts LFE--from machine guns to explosions to score/sound effects galore. I haven't found a better test for the entire LFE range.

I'm going to boost them both a little more tonight and see how they respond (so far I've had them at -4db with the YPAO set sub gain).

At this stage, due to price, I'm really looking at the 13 Ultra, though the ARO and the size of the JL are really impressive!

What I really need to do is get a 13 Ultra and an F113 in my room at the same time! Hummnnn...

Thanks so much for your incredible work, Ilkka!!! You provide an invaluable service man!

The SVS PB12-Plus/2 and the JL Audio f113 have very different frequency responses, so without a doubt they will sound different if not using any EQ. The f113 is much flatter while the Plus/2 has a quite wide and large bump centered at ~40 Hz. A good parametric EQ will help the Plus/2 a lot.

[Warpdrv]

Ilkka wrote:

The SVS PB12-Plus/2 and the JL Audio f113 have very different frequency responses, so without a doubt they will sound different if not using any EQ. The f113 is much flatter while the Plus/2 has a quite wide and large bump centered at ~40 Hz. A good parametric EQ will help the Plus/2 a lot.

Thanks for your hard work here Ilkka....
That is excellent info, I will be adding a second Plus/2 for my huge room, should be here this week, EQing with SMS and new firmware, Im looking forward to see how they measure out together... I'm also into a DIY with twin 15" TC2K's 320L each and can't wait to put them up against the Plus/2's. :raped::raped:

Would it be possible for you to do a measurement of the new firmware on the SMS to see what the results turn out ?

Exciting times to experiment with some really amazing subs these days !!

[allredp]

Ilkka wrote:

The SVS PB12-Plus/2 and the JL Audio f113 have very different frequency responses, so without a doubt they will sound different if not using any EQ. The f113 is much flatter while the Plus/2 has a quite wide and large bump centered at ~40 Hz. A good parametric EQ will help the Plus/2 a lot.

That's a good point; in fact, I've got 4 filters running through a BFD taming the +/2. I ran my informal tests with the SVS and BFD active against the F113 with ARO.

It was very interesting to feel the shudder of the really low stuff with the SVS, but not the JL; while the SQ of the F113 was appreciably better across the board, with the exception of the low stuff it couldn't produce...

[Ilkka]

I am a bit surprised that the latest round of results hasn't caused more discussion? Are the results so clear that no one cares to comment anything or are they too difficult to understand? There was a lot of discussion prior to the measurements about the LMS woofer etc, but now no one says anything about the results they got.

Do I have to start giving random points to the subwoofers in order to get some more traffic here?

[SteveCallas]

Well I think it's like we discussed in prior emails Ilkka - after the first round of testing DIY subwoofers, the cat was already let out of the bag in more ways than one. It confirmed the superior performance to cost ratio of DIY, that a large and low tuned ported subwoofer yields the best peformance, that bottoming in such an alignment is extremely difficult, that having more power is a good thing, and that many high excursion drivers have high Le.

After this latest round of testing, two new things were confirmed - LMS technology works as claimed and SVS has stepped up their game. SVS already has plenty of customers, so not much new to speak of from their results. As for LMS, the driver is very expensive, scarce, and implementation is even more expensive. You mention yourself that it would benefit from a larger enclosure (-5db at 20hz compared to 50hz), and it's not all that small to begin with, so where does a sub like that really fit into the equation? It falls into an odd category in my opinion.

But anyway, that's why I think there hasn't been much discussion.

[Ilkka]

Steve,

So basically what you are saying is that since now the LLT is "proved" to be the best thing since the sliced bread, there's no need to continue with these subwoofer tests anymore? I couldn't disagree more.

And I don't think the LMS-5400 falls into an odd category. Its performance in a 100 liter sealed enclosure in phenomenal. No other driver that I know of can have a similar performance in a such small enclosure. But since you won't agree to understand that not everyone want to have a 600 liter enclosure in their room, you probably don't understand why it's so extraordinary.

[SteveCallas]

Further testing will showcase the differences between drivers, and that is definitely beneficial, but it probably won't create the type of discussion that your previous round of testing did, as more "what if's" where being put to the test then.

We've gone over the size thing about 30 times, and I understand VERY clearly that some want or need to stick with small subwoofers. I am only aware of a very small number of projects using the LMS 18 though, and I can count the ones put into a small sealed enclosure using less than one hand. So does the price justify the ability to have one small enclosure that performs well? Proof is in the pudding.

[Ilkka]

I believe there would be a lot more LMS-5400 projects on these forums if TC Sounds would have marketed it better. Similar measurements to mine would have boosted up the sales dramatically. Though IIRC they weren't able to keep up with the demand even now so it seems that they were selling a lot of them after all. Maybe the ones who bought them aren't the typical audio forum guys?

[Itay]

my views on the results:

1) the 5400 is a very impressive driver. in the passive box it's performance is not
supprising, however the sealed box is most impressive.
2) except for the 5400, all of the diy 15 sealed have unusual high THD between
50-100hz. a bit strange considering the level of those drivers.
3) the sdx-15 is disappointing below 25hz and impressive above 25hz.
4) the JL fathom's THD graphs don't look much at first glance, however it's
performance considering the driver size is very good. also in the cea2010
it gave gr8 numbers.
5) the svs ultra is one amazing subwoofer for the money. even it's sealed mode
gave good results. the 10hz mode is pretty useless except for maybe very
small rooms.
6) the yamaha subwoofer shows just how much marketing "bla bla" is involved
in this industry...
7) to Ilkka.

[Ricci]

I think that the LMS results confirm that it is the real deal. Its performance in the 100L sealed is really impressive and the 200L PR system is more of the same. I don't know of any other drivers that will perform like that in those box sizes. The fate of Audiopulse/TC seems to be in doubt so maybe that has hampered interest. That and the $2000 retail price.

F113 i thought did really well for the size of the box and backed up it's hype that it has had going lately.

The SVS PB13 Ultra definately backed up it's rep as well and was the surprise of the group IMOP. Everyone assumed it would do well, but it seemed to do even better than could be hoped. Impressive.

The sealed SDX 15's did ok, but I was a little underwhelmed, especially after the performance of some of the other units. To me it seemed like the performance below 20hz was poor.

[Ilkka]

Ricci wrote:

The sealed SDX 15's did ok, but I was a little underwhelmed, especially after the performance of some of the other units. To me it seemed like the performance below 20hz was poor.

Do you mean the distortion? Yeah, it's a bit higher than for example with the TC-2000, but IMO distortion below 20 Hz isn't the most important thing in a subwoofer. If everything else is OK, then you might be check that also the VLF is clean as possible. Also the SDX15's distortion is mostly 2nd order harmonic so it's extremely hard to hear.

[Ilkka]

Itay wrote:

2) except for the 5400, all of the diy 15 sealed have unusual high THD between
50-100hz. a bit strange considering the level of those drivers.

It's the inductance doing its evil work there. Most of these high excursion drivers have very high real world inductance (and its variation with current and excursion). But luckily the THD contains mostly low order harmonics so it isn't that easy to pick up.

4) the JL fathom's THD graphs don't look much at first glance, however it's
performance considering the driver size is very good. also in the cea2010
it gave gr8 numbers.

Driver size but also the size of the enclosure. It's MUCH smaller than for example the PB13-Ultra.

5) the svs ultra is one amazing subwoofer for the money. even it's sealed mode
gave good results. the 10hz mode is pretty useless except for maybe very
small rooms.

I would choose the 10 Hz mode over sealed. Though I wouldn't use anything else than the 20 Hz or 15 Hz with it. Those others are just wasting its performance.

6) the yamaha subwoofer shows just how much marketing "bla bla" is involved
in this industry...

The Yamaha didn't quite cut it but IMO it's not THAT bad. It gives high SPL in mid bass range which means it's a pretty good "party subwoofer".

[Itay]

Ahh, I don't like high inductance...

The 10hz on the ultra does look better than the sealed option, and both do waste
It's performance.

Yahama went a bit far stating the 1500 does 16HZ, and does it "well"...

[Ricci]

Ilkka wrote:

Do you mean the distortion? Yeah, it's a bit higher than for example with the TC-2000, but IMO distortion below 20 Hz isn't the most important thing in a subwoofer. If everything else is OK, then you might be check that also the VLF is clean as possible. Also the SDX15's distortion is mostly 2nd order harmonic so it's extremely hard to hear.

I agree about the amount of distortion at 16 and 20hz not being that bad in the real world, but it is still higher than the TC2K. Mostly though it was the output. The dual SDX's in 140L had 2 Ce4000's running to it right? The LMS was in a smaller enclosure with 1 amp and it put up significantly better 20 and 16hz #'s while being cleaner too. Maybe that's not a fair comparison but still...

[Ilkka]

Ricci wrote:

I agree about the amount of distortion at 16 and 20hz not being that bad in the real world, but it is still higher than the TC2K. Mostly though it was the output. The dual SDX's in 140L had 2 Ce4000's running to it right? The LMS was in a smaller enclosure with 1 amp and it put up significantly better 20 and 16hz #'s while being cleaner too. Maybe that's not a fair comparison but still...

Well the LMS-5400 is by far the cleanest and the best driver I have ever measured, so naturally any other "normal" driver will look bad compared to it.

[ISLAND1000]

Ilkka wrote:

I am a bit surprised that the latest round of results hasn't caused more discussion? Are the results so clear that no one cares to comment anything or are they too difficult to understand? There was a lot of discussion prior to the measurements about the LMS woofer etc, but now no one says anything about the results they got.

Do I have to start giving random points to the subwoofers in order to get some more traffic here?

LOL, No . . . . we don't need no steenking random point assignments for the speakers!
Ilkka, I think what happened last year was all of us DIYers were interested in the "ultimate" driver. Everything was pointing to the LMS-5400 18" but we didn't have anyone with the right credentials to do the building and testing. The process that you proposed took more than half a year to bear fruit. Then you kinda let the cat out of the bag when you tested the drivers in "free air" after you had finally received your unit that had shipped over seas. We were all blown away as soon as we saw your results. The LMS-5400 performance was easy to recognize as the top driver. The only final proof needed was to measure it in-box.
You constructed TWO boxes and had the CE 4000 to drive them with. There was going to be NO question at the end of this adventure as to which driver was "the best" and for us sumo-scientific guys we thought we were finally going to have THAT driver and box that would rattle the frequency response all the way down to 10Hz and reference levels with little or no distortion.
Your test procedures and display of results leave little else for us to imagine. The graphs and pictures show and answer almost all of our questions. You ARE good at that you know!!!
The LMS-5400 IS THE BEST sub driver to date . . . . . . however to our amazement it is no longer realistically available to the DIY world. We sit in utter amazement and silence. Our dreams have been answered and yet we sit with head in hands wondering how this could have happened. Yes there are a few people lucky enough to have the LMS driver but way fewer than needed for it to become an ICON.
Don't be dismayed at the lack of discussion, we're still trying to wrestle with our realities. We'll get back at it . . . . . but we will have to replace a discussion about the worlds best driver with . . . . . what . . . . .I dunno.
Phil

[SteveCallas]

I think you're taking this a bit too seriously man

[allredp]

I agree with Phil--DIY took a big hit in the gut when TC went bonkers-retail on us...

I know I'm in a bit of a quandry as to what to do; hence the near silence (except when confirming the the 13 Ultra is really an excellent offering).

I love the SVS, and feel it is really the ideal mix of aesthetics, performance and value, but couldn't help thinking that DIY with the LMS unit would finally put the nail in my upgraditis coffin...