DIY TC Sounds LMS-5400 18" + 2x18" PR 200L

[Ilkka]

Manufacturer: DIY
Provided by: Ilkka
Manufacturer link: -
Price: -

Please contact if you would like to have the same kind of subwoofer (with better cabinet and numerous finish options) being built for you.

Manufacturer specs:
• Active woofer: TC Sounds LMS-5400 18” (DVC 2x2 ohm)
• Passive radiators: 2x TC Sounds VMP-TC 18”
• Amplifier: Crown CE4000 (~3600 W @ 4 ohm bridged per amp)
• Enclosure: 200 liters, 21mm (15-ply) baltic birch all around, 42mm front baffle, well braced, opposed passive radiators, total weight with woofers around 90 kg / 200 lbs
• Signal processing: Behringer DCX2496

Configuration as tested:
Phase: 0 degrees
Crossover: Bypassed
EQ: +3dB, 20 Hz, Q=1.0
Tuning frequency: 16 Hz
Subsonic filter: ~15 Hz (effective), 2nd order Butterworth
Orientation: Active woofer facing the mic

[Ilkka]

+/- 3 dB points: 18.3 Hz - >200 Hz

[Ilkka]

[Ilkka]

[rolyasm]

Curious on the cost to build this massive beast? I am thinking of using it as my next car.

[Ilkka]

rolyasm wrote:

Curious on the cost to build this massive beast? I am thinking of using it as my next car.

It is somewhat misleading to list the prices since TC Sounds/Audiopulse has since changed the pricing pretty dramatically.

But here are the prices back then:

LMS-5400 18" - $875
TC-VMP 18" - $250 (subwoofer has two of these)

The enclosure and the amplifier were around $800 total, but naturally their cost can vary a lot.

[robertcharles123]

I will be more than glad to start the questioning for you. Just a little bit about me- I have 4 18"5400's and 8 18" passives. Enough about me, lets get to the 5400's. From your experience, did you find that the 5400's were not only quite capable at high sound pressure and high sound quality, but were they also built "heavy enough" for that kind of constant punishment?

Also, can you explain not only with numbers, but with some personal words about how the 5400's perform compared to any other high perfromance woofer you have tested.

From your new avatar, I seem to be getting some personal ideas right now.

[robertcharles123]

Can you explain a little about the harmonic distortion curves and why they go up so drastically towards the low end compared to some of the other woofers tested. It seems that since the woofers had a 15 hz high pass, the distortion should have been somewhat flattened in my mind. Can you explain how the distortion increases so rapidly.

[robertcharles123]

My system will be consisting of 2 opposing woofers with 4 opposing passives- three drivers on each side of two towers. I used winisd to model the speakers. Basically, I have 7.5 cf per driver for a total of 28 feet. 7.5cf is roughly what you had for your one driver, but my model seem to model the pr's best at 2000 grams added to the passives. How did you come by 1470ish for the passives. I would love to get the best perfromance from these woofers and passives and I feel that you have the "best" first hand knowledge on this subject. Will the heavier pr's add to the extension of the system only, or will they help or hinder more parameters.

Thanks,
Robert

[robertcharles123]

Did you find that the woofers were ok mouted like you had them with no magnet mount?

Did you notice that the panels were thick enough at 3/4". Mine will have three layer on the woofer baffles and the other panels will be 1.5" with extensive internal braces.

Also, with woofers of this magnitude, does it matter how big the hole in the braces are cut- do they act like internal ports and possibly cause problems?

[Ilkka]

Thank you for the questions, Robert. I'll address them tomorrow as it is getting pretty late around here.

[Ilkka]

I moved your questions here so that people looking for info about the LMS-5400 could find them more easily.

robertcharles123 wrote:

I will be more than glad to start the questioning for you. Just a little bit about me- I have 4 18"5400's and 8 18" passives. Enough about me, lets get to the 5400's. From your experience, did you find that the 5400's were not only quite capable at high sound pressure and high sound quality, but were they also built "heavy enough" for that kind of constant punishment?

IMO the LMS-5400 is easily heavy enough for constant punishment, of course slightly depending what you mean by that? Of course continuous near Xmax excursion can cause problems, but that applies to every woofer out there. But when compared to any other woofer I've seen, the LMS-5400 doesn't have to shame anything.

Also, can you explain not only with numbers, but with some personal words about how the 5400's perform compared to any other high perfromance woofer you have tested.

The LMS-5400 is the highest performing and the best woofer I have ever tested. It has the flattest frequency response (=very low inductance and its variation with excursion and current), the lowest distortion (all types) (=very linear Bl and compliance) and highest output at every frequency. Of course it sounds extremely good too.

robertcharles123 wrote:

Can you explain a little about the harmonic distortion curves and why they go up so drastically towards the low end compared to some of the other woofers tested. It seems that since the woofers had a 15 hz high pass, the distortion should have been somewhat flattened in my mind. Can you explain how the distortion increases so rapidly.

Distortion of a ported or a PR subwoofer will always shoot the skies below the tuning frequency. Not even a 2nd order HP filter will prevent that. Of course because the measuring system doesn't have an infinite S/N ratio, the distortion is slightly exaggerated when the output of the subwoofer goes very low. And in case of a ported or a PR subwoofer, it goes very low below the tuning frequency of the subwoofer, especially if a HP filter is being used. But the distortion below 15 Hz is the least thing you have to worry about - trust me.

robertcharles123 wrote:

My system will be consisting of 2 opposing woofers with 4 opposing passives- three drivers on each side of two towers. I used winisd to model the speakers. Basically, I have 7.5 cf per driver for a total of 28 feet. 7.5cf is roughly what you had for your one driver, but my model seem to model the pr's best at 2000 grams added to the passives. How did you come by 1470ish for the passives. I would love to get the best perfromance from these woofers and passives and I feel that you have the "best" first hand knowledge on this subject. Will the heavier pr's add to the extension of the system only, or will they help or hinder more parameters.

Here's the natural (i.e. no HP filter or EQ) of the DIY TC Sounds LMS-5400 18" + 2x18" PR 200L. It indicates that the woofer could use a little bit enclosure volume when using such a low tuning frequency. It goes slightly against any simulation, but hey, that's why I take these measurements. So if you can, I would go with full 10 cu ft per woofer. That will boost up the area around the tuning frequency and give you more output down there.



The 1470g I used means the _additional_ weight. The cone, spider and the bolt that holds the weights weigh around 300 grams. So the total weight of the PR I used and what you should enter into the Unibox was around 1770g. I decided to use that amount because it gave me the 16 Hz tuning frequency I wanted.

robertcharles123 wrote:

Did you find that the woofers were ok mouted like you had them with no magnet mount?

Perfectly ok. The frame is strong enough, but of course the front baffle has to be strong too. It's _really_ difficult to make that kind of magnet mount so that it works as it should and truly helps the frame.

Did you notice that the panels were thick enough at 3/4". Mine will have three layer on the woofer baffles and the other panels will be 1.5" with extensive internal braces.

My panels are 21 mm (15-ply BB) which is closer to 7/8". The front baffle has two of these glued on each other. I also have some bracing, but I wouldn't call it extensive. Whether something is "enough" is always a difficult question. Usually one has to do some compromises between the rigidity and the weight of the enclosure. I decided to use 15-ply BB because I had to move that subwoofer a lot and wouldn't want it to weigh over 200 lbs. I doubt one would be able to hear the difference should I had used 1.5" or 2" panels all around. Your cabinets are of course much larger so 1.5" is well justified.

Also, with woofers of this magnitude, does it matter how big the hole in the braces are cut- do they act like internal ports and possibly cause problems?

It doesn't matter which kind of woofer, the hole(s) in the braces should always at least match the area of the cone. I usually tend to use slightly larger holes.

[Ilkka]

Here's a 3D drawing of the enclosure I made.

[robertcharles123]

I can't really go bigger than 7.5 cf per woofer. I basically have no more space left, or everything is going to start being really cluttered. I currently have six jbl 2241 as my low end for the system. The new subs will be augmenting the jbl's which will be crossed over at about 40-50hz and up- on the main channels leaving the new subs for the lfe channel. I hope I will realy be upgrading what I currently have in place.
Have you ever tested any of the jbl's?

Thanks,
Robert

[robertcharles123]

What did you use for the 15 hertz crossover? Did you have other options?

[Ilkka]

robertcharles123 wrote:

I can't really go bigger than 7.5 cf per woofer. I basically have no more space left, or everything is going to start being really cluttered. I currently have six jbl 2241 as my low end for the system. The new subs will be augmenting the jbl's which will be crossed over at about 40-50hz and up- on the main channels leaving the new subs for the lfe channel. I hope I will realy be upgrading what I currently have in place.
Have you ever tested any of the jbl's?

Thanks,
Robert

Ok. 7.5 cu ft will do just fine. You can use some EQ if you need to bump up the low end but probably your room gain will do that just fine.

[Ilkka]

robertcharles123 wrote:

What did you use for the 15 hertz crossover? Did you have other options?

I used the Behringer DCX2496, but it actually doesn't have a 15 Hz HP filter. I used its lowest, 20 Hz (12 dB/octave Butterworth), setting instead but used a wide boost (bandpass filter) around 20 Hz, so effectively the filter was around 15 Hz. I couldn't have gone much lower with the DCX2496.

[Warpdrv]

Robert, I see your still working on this... Look forward to seeing how they turned out... Thanks Ilkka for all your hard work... That LMS is a wicked sub.
Thanks for posting the measurements.. beautiful...

[Mark Seaton]

Ilkka wrote:

I used the Behringer DCX2496, but it actually doesn't have a 15 Hz HP filter. I used its lowest, 20 Hz (12 dB/octave Butterworth), setting instead but used a wide boost (bandpass filter) around 20 Hz, so effectively the filter was around 15 Hz. I couldn't have gone much lower with the DCX2496.

Hi Ilkka,

Awesome results on this sub and of course the data on all the others. FYI, you can use the DCX to high pass lower than 20Hz. There are two means to get an effective lower frequency high pass filter. If you look at both 12 & 18dB/octave Bessel filters, you will see the sharper roll off is in fact a good bit lower than the nominal Fc. Addition of 1-2 PEQ's makes a lower effective XO pretty easy. The other, more straight forward option is to use a Butterworth or LR with a simple 12dB/Oct Low Shelving filter (would have to check the "Q" of the shelf to see if Butt or LR matches best). In simple terms, you are basically applying a Linkwitz Transform to the high pass filter. Works just fine within reason. There is probably some point where the cut and boost exceed the working range of the DSP math, but I've done this in practice a few times with good results.

Another detail you might want to investigate is where the Fp of the PR's ended up. It looks like that might be accounting for a somewhat faster roll off near or below tuning. Not a big deal, as obviously it performs exceptionally well, but good to note in the case that someone wants to tune lower. The natural frequency response shown actually looks to be a decent match for many rooms as is.

Thanks for giving me a great point of measurement comparison for the BMF.

[Ilkka]

Mark Seaton wrote:

Hi Ilkka,

Awesome results on this sub and of course the data on all the others. FYI, you can use the DCX to high pass lower than 20Hz. There are two means to get an effective lower frequency high pass filter. If you look at both 12 & 18dB/octave Bessel filters, you will see the sharper roll off is in fact a good bit lower than the nominal Fc. Addition of 1-2 PEQ's makes a lower effective XO pretty easy. The other, more straight forward option is to use a Butterworth or LR with a simple 12dB/Oct Low Shelving filter (would have to check the "Q" of the shelf to see if Butt or LR matches best). In simple terms, you are basically applying a Linkwitz Transform to the high pass filter. Works just fine within reason. There is probably some point where the cut and boost exceed the working range of the DSP math, but I've done this in practice a few times with good results.

Thanks, Mark!

Yes, that's exactly what I did. I used a few bandpass filters to lower the effective -3 dB frequency of the HP filter to 15 Hz.

Another detail you might want to investigate is where the Fp of the PR's ended up. It looks like that might be accounting for a somewhat faster roll off near or below tuning. Not a big deal, as obviously it performs exceptionally well, but good to note in the case that someone wants to tune lower. The natural frequency response shown actually looks to be a decent match for many rooms as is.

Thanks for giving me a great point of measurement comparison for the BMF.

I did measure the tuning frequency of the system; it was exactly 16 Hz (distortion sweeps show it too). But as I explained earlies, it seems that the LMS-5400 could use a little bit more volume than the simulations suggest. Around 10-12 cu ft might be the sweet spot when using such a low tuning.

[Mark Seaton]

Ilkka wrote:

Thanks, Mark!

Yes, that's exactly what I did. I used a few bandpass filters to lower the effective -3 dB frequency of the HP filter to 15 Hz.

This works well. Depending on what you are after and what the system response is, the shelving approach gives a little wider range to work and allows you do create a softer corner if that is needed.

I did measure the tuning frequency of the system; it was exactly 16 Hz (distortion sweeps show it too). But as I explained earlies, it seems that the LMS-5400 could use a little bit more volume than the simulations suggest. Around 10-12 cu ft might be the sweet spot when using such a low tuning.

I caught your measurement of the tuning frequency. What I was referring to was the natural resonance fo the PRs. This is dependent on the suspension compliance of the PR and the moving mass, which sets the natural resonance of the PR's mass and spring itself. The tuning in box is separate from this and it is not affected by box volume. I don't recall what TC had posted for the range that Fp would be vs. mass added. You find this by measuring the system output well below resonance, obviously without a high pass. At some low frequency, ideally low single digits, sometimes in the teens, you will see a frequency where the PR and driver output cancel, forming a notch below the low corner of the subwoofer. Depending how close the notch is to the corner, this can affect the response around the low corner. This is also one of the reasons we want PRs to have very soft suspensions. I suspect you know this but read my question of asking about the tuning of the PRs in box.

[Ilkka]

Mark Seaton wrote:

I caught your measurement of the tuning frequency. What I was referring to was the natural resonance fo the PRs. This is dependent on the suspension compliance of the PR and the moving mass, which sets the natural resonance of the PR's mass and spring itself. The tuning in box is separate from this and it is not affected by box volume. I don't recall what TC had posted for the range that Fp would be vs. mass added. You find this by measuring the system output well below resonance, obviously without a high pass. At some low frequency, ideally low single digits, sometimes in the teens, you will see a frequency where the PR and driver output cancel, forming a notch below the low corner of the subwoofer. Depending how close the notch is to the corner, this can affect the response around the low corner. This is also one of the reasons we want PRs to have very soft suspensions. I suspect you know this but read my question of asking about the tuning of the PRs in box.

Sorry, Mark. I was in a bit of a hurry while responding to your question, and I didn't catch that you meant the Fsp instead of Fb. Unfortunately the measurements I took won't clearly show it so I can't give you a definitive answer. Also if the Qmp is lower than predicted (as I'm suspecting), the notch won't be very visible. I can take an impedance measurement for some further investigations.

[robertcharles123]

Ilkka, how is your assessment of the in room response of the 5400 compared to any of your other test woofers. What I am getting at is, can you, with your eyes closed and not knowing which system is playing, tell that the 5400 pr system is being played, or, are some of the test systems up to snuff with it.

Thanks,
Robert.

[Ilkka]

robertcharles123 wrote:

Ilkka, how is your assessment of the in room response of the 5400 compared to any of your other test woofers. What I am getting at is, can you, with your eyes closed and not knowing which system is playing, tell that the 5400 pr system is being played, or, are some of the test systems up to snuff with it.

Thanks,
Robert.

It mostly depends on the setup. If one doesn't carefully match the frequency responses of the subwoofers he's about to compare (in-room), the differences are quite audible as most subwoofers have quite different frequency responses (if not equalized). I would say it is the biggest reason why different subs sound different. Small differences in maximum output or extension aren't as audible in many situations. But depending on the material and SPL being used, and of course the actual absolute difference between the subs being compared, also these variables will contribute more and more.

But probably you are more interested in hearing whether the LMS-5400 PR system is the best thing since sliced bread. Well let's say that if I was about to fill a large room with quality bass (if money nor enclosure size wouldn't matter), this subwoofer (or multiples of it) would be my choice of the every subwoofer I have tested so far. But if I had a small room and wouldn't want to use any external EQ, there are better choices since the low end would get boosted too much due to room gain. Also when thinking about the subwoofers of this caliber, one should make sure that also the other speakers are up to the task. High performance subwoofer with traditional low performance speakers can lead out to an unbalanced system which doesn't sound much better than a lower performance, but a well balanced system.

[robertcharles123]

I started building my boxes recently. If there is any interest in my design and build here with the 4 lms 5400's and 8 passives, I will post info as I get it.

http://www.hometheatershack.com/foru...ves-oh-my.html

Thanks,

Robert