LLT Explained - Discussion Thread

[Zene]

Steve ... That helps a lot. I think I'm on at least one rail of the track.
Price does come into the mix as I need to do a complete HT system from scratch.
Would you be kind enough to list acceptable 15" drivers for LLT so I can choose the optimum for my budget. Is the Dayton 15 Ref in the group or any of the Daytons? I can do multiples if that helps, boxes are cheap for me. Reason for Daytons is I can trade for them and pay no shipping.
Is there an easy way to evaluate whether any two or more woofers would be better than one of better known quality, i.e. two Dayton model X's vs. one RPL-15?
Zene

[SteveCallas]

The Dayton Ref HiFi 15 would work, but you'd probably want to use two of them to ensure plenty of headroom. The only other real options at the moment are the already mentioned TC 2000 15 svc, the SS RLp15 D2, the LMS 4000 15 (need two), the LMS 5400 18, and maybe the new Adire Tumult 15 D2. Cost/performance wise, the SS RLp15 comes in first, with the TC 2000 15 offering slightly more performance for slightly more cost and being a little larger, and then dual Dayton HiFi 15s for a little more and double the size of the TC 2000 enclosure, though it needs the least power.

If you list your budget and size limitations, I can help you reach a solution pretty easily.

[WillyD]

Just thought I'd mention that the LMS 4000 has apparently been discontinued.

[SteveCallas]

Just thought I'd mention that the LMS 4000 has apparently been discontinued

Saw that one coming, just didn't think it would be so soon. I wonder if it even made production, let alone if they sold a single driver...

[WillyD]

Yeah dude, it definitely made production....there was even a build thread on the TC Sounds forums.

But no doubt it wasn't very popular. I am hoping the 4100 Neo will be much more successful.

[Zene]

Steve ... Will figure budget. I have a few more people to talk to about equipment. In the mean time can you tell me what excursion limits limits as a % of Xmax you feel is good headroom. I can work backwards with SPL wanted to find power limits without exceeding mfg power limits. So for Xmax = 20mm, 10mm would be 50%. Then inputting power into UniBox to match that limit. As long as that does not exceed Pe I should be good to go, but I am not positive which mfg power limits to use Pe or max. Hope that makes sense?
Zene

[SteveCallas]

Well I try to amp limit so that max excursion limits can't be reached, and typically this ends up being below or near the power handling limits. Depending on the output levels max power would correlate to, it can be ok to use an amp with more power than the driver can handle. As for the amount of headroom, based on many LLT owner experiences, I would advise that modeling show the sub to be capable of 112db+ levels above tuning for spirited playback. Depending on how loud you listen to, you may need more than that. Overall though, most are very satisfied with that capability.

[Zene]

Steve .. 112dB sounds reasonable. Music ; I understand levels, but HT is another animal. I do not get a clue in demo rooms. Walking around with a RS meter is pointless. Just to suck-up a little more, your work explaining LLT is going to help a lot of old Star War fans (me, me).
Much appreciated. Zene

[SteveCallas]

Thanks. To be honest, I find output levels for home theater much easier to understand than for music. Assuming calibrated systems and capable gear, if I watch a movie at -15 on my system, then watch it at -15 on a friend's system, I should be exeriencing virtually the same output levels. Assuming the movie wasn't recorded hot, I would also expect the potential for each speaker to hit 90db levels and the LFE channel to hit 100db levels....factor in the redirected bass that goes to the sub as well, and the sub may be asked to reproduce ~105db levels on its own. This is because a calibrated master volume level of 0 equates to the potential for each speaker channel to hit 105db peaks and the LFE channel to hit 115db peaks. You simply adjust from there.

When it comes to music, I really don't know if there is a standard or reference level. I don't know what the dynamic range is for PCM, DVD-A, SACD, or other music formats. Differences in compression/dynamic range vary WILDLY from cd to cd.

But anyway, getting back to the sub, if it's capable of 112db+ levels above tuning, that means you should be able to listen to movies on a calibrated system up to a master volume level of about -8 (that's pretty loud ). Of course there are hot movie tracks every now and then (War of the Worlds DTS), but by and large, I've found I typically listen at the same master volume for every movie.

[Zene]

Steve ... Stereo does have it ups and downs (punny); cd vs records,type of music, mood, purpose, wine intake and a hundred other variables as you mentioned. My saving grace is a remote volume control on my preamp. I could not get along without one. I have a small red dot on the knob so I can see it and simply make a mark on a sticker on the cd case or record cover that shows clock position of desired level. Match it with the preamp knob and I'm close.
Sorry for stereo interrupting HT, but may be useful to others.
Zene

[Zene]

P.S. Sub amp is at arms length for quick change, noting the approx positions for any cd or record I particularly want different settings than normal. It's easier than one might think. Practice makes quick. Marks on knobs this time is for flattest response return.
Zene

Steve,

In case the TC2000 15" is on backorder, is the Ficaraudio Q15 suitable for LLT?

Q15 specs
https://ssl.perfora.net/www.ficaraud...iew.shopscript


DUAL 1 | DUAL 2

Fs: 25.9 Hz | 24.9 Hz
Re: 0.7 Ohms/coil | 1.4 Ohms/coil
Qms: 5.64 | 5.72
Qes: .54 | .54
Qts: .49 | .49
Mms: 257g | 279g
Cms: 0.15mm/N | 0.15mm/N
Sd: 810cm^2 | 810cm^2
Vas: 135 l | 135 l
Spl: 88.2dB 1W/1m | 87.7dB 1W/1m
Bl: 10.4 N/A | 15.1 N/A
Xmax: 27mm
Rms: 1000W
Sealed box: 2-2.6 cuft
Ported box: 2.8-4 cuft @ 28-33Hz
Sub OD: 15.625”
Cut ID: 14.125”
Mounting depth: 8.500”
Displacement: 0.19cuft


Regarding the questions for 12" drivers, the Adire Brahma 12" works well in the Sadhara sonotube. I'm looking for tighter, chest pounding with extension below 20Hz.

thanks for your efforts.

Al

[SteveCallas]

Yeah, looks good, I'd probably go with the dual 2.

[WillyD]

On another forum someone built a Q15 LLT , but the Q15 he used was custom. By his accounts, it turned out great.

It was a Dual4 ohm version, I remember.

http://www.avsforum.com/avs-vb/showthread.php?t=728542

[SteveCallas]

Wow, never even caught that thread

soloxp wrote:

Now how does it sound? Well I put in WOTW and all I can say is WOW. coming from 234 Liters to 340 Liters made a big different. The bass is lot cleaner deeper

Preach it brother, preach it

Al, if you can afford $30 more and a bit more space than what you were planning, I'd highly suggest going with the 18".

[WillyD]

It'd be nice if Scott could customize it for him as well. Not sure if he'll do just one driver for someone, but I don't recall if the standard Q15/18 models well.

[lienly]

what a great job with so much info, tks steve!

[DrWho]

I don't understand this issue with "resurgance of the port output"...especially in light of all the non-linear behavior:
http://www.klippel.de/pubs/default.asp
http://forums.klipsch.com/forums/sto...6640/PORTS.pdf

With port compression, voice coil temp rise, Bl and suspension drops, inductance modulation, etc etc..., I think there is a pretty good argument for implementing resurgance over the port's passband. It just requires the cabinet to be much larger...

Btw, your response times for the port build up or whatever seem a bit skewed considering that an 18Hz filter is going to affect the output at 20Hz...I think it would be more fair to show the response where the peaks are both .4, not .3 and .4. I'm not sure if it matters, but I would expect output to increase faster when the air velocity is faster...

[SteveCallas]

DrWho wrote:

I don't understand this issue with "resurgance of the port output"...especially in light of all the non-linear behavior:

If given the choice, would you rather have reasonably flat or non flat in room FR down low?

With port compression, voice coil temp rise, Bl and suspension drops, inductance modulation, etc etc..., I think there is a pretty good argument for implementing resurgance over the port's passband. It just requires the cabinet to be much larger...

Why, what do you gain? Also, port compression is something the design strives to eliminate. A great LLT design simply won't be able to generate speeds high enough to experience port compression based on Collo's work.

Btw, your response times for the port build up or whatever seem a bit skewed considering that an 18Hz filter is going to affect the output at 20Hz

Traditional ported designs use a highpass just under the tuning frequency, and I'm trying to show what can be gained without it, so I'm not sure what you are getting at?

lienly wrote:

what a great job with so much info, tks steve!

Much appreciated.

[DrWho]

If given the choice, would you rather have reasonably flat or non flat in room FR down low?

I prefer to look at behavior that doesn't show up in the frequency response - namely the transient behavior (which involves the time-domain, not the frequency response). I would also prefer that the system sound good at all output levels.

Klippel isn't doing all this research for nothing.

Why, what do you gain? Also, port compression is something the design strives to eliminate. A great LLT design simply won't be able to generate speeds high enough to experience port compression based on Collo's work.

The fact of the matter is that you will always have port compression. There's no way around it. But more importantly, you also have the non-linear behavior of the driver....

The tuning alignment is a degree of freedom within the design that can be used to address the non-linearities of the total system. Minimizing the non-linear behavior of one single variable may not be in the best interests of the whole system.

Traditional ported designs use a highpass just under the tuning frequency, and I'm trying to show what can be gained without it, so I'm not sure what you are getting at?

You are trying to show the time-domain behavior of the system at a specific frequency, but you're also changing the amplitude. If you want to show the effect of a single variable (ie, time), you have to keep all other variables the same (ie, the amplitude). Some (better?) alignments are using a peaking highpass to take advantage of the reduction of cone movement at the tuning frequency - which can be used to keep the output at the tuning frequency the same (or higher), while still controlling the excursion of the system.

I was just commenting that it would be more meaningful if you had used a peaking filter, or merely turned up the volume on the highpassed sub so that the amplitudes were the same between trials. Only then will the time-domain differences be meaningful.

I hope I don't come across negative - I totally dig the writeup and just wanted to throw some other ideas out there. It's hard to present an idea without coming across negatively. , I don't even propose these ideas as fact...

[SteveCallas]

DrWho wrote:

I prefer to look at behavior that doesn't show up in the frequency response - namely the transient behavior (which involves the time-domain, not the frequency response). I would also prefer that the system sound good at all output levels.

Is not the FR a reflection of group delay and impulse response? I think Ilkka's extensive subwoofer testing will show that it is. The flatter and deeper extending the FR, the better the GD. The shallower the rolloff, the better the impulse response down low.

DrWho wrote:

The fact of the matter is that you will always have port compression. There's no way around it. But more importantly, you also have the non-linear behavior of the driver....

To what degree do you speak of? Let's say air speed never exceeds 10m/s in a 10" diameter port - will there be compression, and to what degree? If you are saying like 0.5db, then fine, I'll accept that. I'm still not sure what benefits you are suggesting with a resurgence in port output however.

I was just commenting that it would be more meaningful if you had used a peaking filter, or merely turned up the volume on the highpassed sub so that the amplitudes were the same between trials. Only then will the time-domain differences be meaningful.

Fair enough. In the end though, the graphs really aren't even needed. A sharper rollof, which a highpass will create, results in deteriorated impulse response.

I hope I don't come across negative - I totally dig the writeup and just wanted to throw some other ideas out there. It's hard to present an idea without coming across negatively. , I don't even propose these ideas as fact...

No worries at all, I appreciate the discussion

[Ilkka]

DrWho wrote:

You are trying to show the time-domain behavior of the system at a specific frequency, but you're also changing the amplitude. If you want to show the effect of a single variable (ie, time), you have to keep all other variables the same (ie, the amplitude). Some (better?) alignments are using a peaking highpass to take advantage of the reduction of cone movement at the tuning frequency - which can be used to keep the output at the tuning frequency the same (or higher), while still controlling the excursion of the system.

I was just commenting that it would be more meaningful if you had used a peaking filter, or merely turned up the volume on the highpassed sub so that the amplitudes were the same between trials. Only then will the time-domain differences be meaningful.

Maybe I could better answer to your questions, since I was the one who took those measurements.

The small difference in amplitude doesn't matter. ~0.3 Pa vs. ~0.35 Pa is only around 1.3 dB difference.

But to be honest, those graphs aren't 100% compareble since the signal used was a normal non-windowed sine wave, and the other amp had a fixed 1st order LP at 300 Hz, and the other hadn't. So the difference isn't only due lack of HP filtering below the tuning frequency.

I'll see if I could make more comparable graphs now that I have a sub with a bypassable HP filter.

[DrWho]

Here's a simulation for the TC Sounds TC-2000 SVC 15" driver:

Green = small signal with port resurgance
Yellow = large signal with port resurgance

Light Blue = LLT small signal
Dark Blue = LLT large signal

Room gain was predicted with a linkwitz transform and static gain of 6dB (1/2 space versus 1/8 space). The LLT enclosures are 7 cubic foot tuned to 16Hz with a 6" port (40" long) and then my alignment is 14 cubic foot tuned to 16Hz with two 6" ports (each 40" long). Both systems are using a 12dB/octave highpass filter at 20Hz to control cone-excursion. It could probably be omitted, but does yield more output without fear of bottoming out with low frequency transients.

Non-linear variables modded were a 200 deg F temp rise, .75 Bl and .75 Vas, and 3dB port compression. I don't propose these non-linears as factual, but they demonstrate the trends.

As you can see, there is about 3dB of power compression with both alignments. It's hard to see on the attached graph, but the F3 slides up too while the HF corner goes down too. The LLT is flatter with low signal analysis, but my alignment is flatter at very loud levels. I would probably choose an alignment somewhere inbetween.

Group delay and port velocity are similar between both.

[SteveCallas]

The established TC 2000 SVC LLT design is 320 effective liters with a 6" port that is 27" long and 900 watts with no highpass. There have been about five of these built so far, and they all seem to stay relatively flat to ~13hz in room. The 6" diameter port will have some compression with full 900 watts, but it will be used much more often at lower levels then at full power. A better LLT will have practically no port compression.

[SteveCallas]

UPDATE!

I have tweaked and rewritten my entire LLT explanation with the main area of focus reflecting the change in my guidelines regarding amp limiting. This comes as a result of the hard work by member Ilkka and the real world usage of LLTs by enthusiasts and their informative discussions on this forum.