AVS Forum banner

SI 24z build....8 of em'

35K views 678 replies 81 participants last post by  Bassment 
#1 ·
Update 3/3/14...so after many issues in the home construction I am happy to say we are reaching the end and HT build can get started. Long story short, the contractor cost me what the IB space would have cost to construct physically about 2times over. Silver lining....with so many options for good diy subs these days, you can't lose. I have considered options carefully and talked with a few folks that I trust about it.....my 8 SI 24z will be here in a few weeks



That's the starting point for the new project, where if you jump to page 12, I'd love thoughts on implementations.


Any and all comments welcome and as always, thanks to all!
 
See less See more
1
#529 ·

Quote:
Originally Posted by ChopShop1  /t/1501074/si-24z-build-8-of-em/500_100#post_24543733


I think this is the same way I see it....the other thing to keep in mind, which some will say should not be taken consideration because the published specs are the published specs, is that xmech on the 24s is close to double the xmax and the thermal limits lie somewhere beyond double the rated power handling. 3000w isn't unreasonable for the 24 in any case.

Interesting claims without being tested. Will be interested in seeing a data-bass report =] Just because xmech is double xmax does not mean you have that much more usable stroke.

Quote:
Originally Posted by bossobass  /t/1501074/si-24z-build-8-of-em/500_100#post_24544081


Man, c'mon. We have to wade through this old debate again?


What advantage do you need across a one octave slice of BW with an 8 x 24" driver system that warrants 4 times the real estate?



You had excess head room before you increased box volume from 10 cubes to 40 cubes.


If you have difficulty in reproducing 20 Hz with 70L of displacement driven by 56KW peak power then state your new standard for reproduction as a disclaimer. These 20dB hot, +10dBRL threads used to be a bit amusing but they really are just getting silly.

Yes because people are building 8x24 systems to reproduce tracks at reference levels. Ok then. Just because it's silly to you does not mean it isn't serious for others. It's a discussion anyway, not really a debate. Also, went from 6.5 cubic feet boxes to 27 cubic feet. I did not have headroom in that range as I was bottoming out the LMS drivers. I gained so much output there I can run at hotter levels and not worry about it anymore. Pod emergence at +10db over reference without a hint of stress now. Before, I had to deal with the lovely sound of metal cones bottoming out when only a couple db hot.
 
#530 ·
I wouldn't say they are "untested", but if by tested you mean data-bass, etc. then yes they are untested. I know and trust Nick and when you see what he does to these to "test" them after assembly....its amazing. Moreso, there are more in use than folks have seen and there are plenty being pushed with double the rated power and no issues. Watching a few thousand watts hit these things free air and seeing them take it in stride makes my confidence level high. Xmax/Xmech are a whole different issue, and it's not that I'm saying the woofer is producing gobs more after xmax, but that the xmech being so high allows us to make use of all of the xmax without worry and the typical goal of modeling for somewhere under xmax by design to keep it safe. Of coure A week from now everyone can watch videos of us at Brandon's place and see just how it works out. I'll say this, I have no problem buying one extra woofer so that we can test its long term power/thermal and excusrion capabilities....we can measure and see what happens to distortion, both audibly and on paper. I'll let one get completely abused in the interest of science and finding out what the "useable" and flat out drop dead capabilities of the 24s are.
 
#531 ·

Quote:
Originally Posted by ChopShop1  /t/1501074/si-24z-build-8-of-em/500_100#post_24544797


I wouldn't say they are "untested", but if by tested you mean data-bass, etc. then yes they are untested. I know and trust Nick and when you see what he does to these to "test" them after assembly....its amazing. Moreso, there are more in use than folks have seen and there are plenty being pushed with double the rated power and no issues. Watching a few thousand watts hit these things free air and seeing them take it in stride makes my confidence level high. Xmax/Xmech are a whole different issue, and it's not that I'm saying the woofer is producing gobs more after xmax, but that the xmech being so high allows us to make use of all of the xmax without worry and the typical goal of modeling for somewhere under xmax by design to keep it safe. Of coure A week from now everyone can watch videos of us at Brandon's place and see just how it works out. I'll say this, I have no problem buying one extra woofer so that we can test its long term power/thermal and excusrion capabilities....we can measure and see what happens to distortion, both audibly and on paper. I'll let one get completely abused in the interest of science and finding out what the "useable" and flat out drop dead capabilities of the 24s are.

I'm not quite sure where you're getting the double xmech figures from either as the posted ones are xmax 36mm xmech 45mm, are you seeing that is not accurate? I'm not trying to start a debate here, 8x 24 will be such overkill for any sane person. I just wonder how many of us fit into that category
The only point I was really trying to make is that with crazy systems like these, we generally have extra headroom above 30-40hz, but in the lower frequencies this isn't always the case as excursion starts to be a limiting factor.


This is in a high Q 0.937 box, 10 cubes with 4000 watts. You can see how crazy the excursion will be and how quickly it starts ramping up.




Anyway, just my $0.02. I considered switching to 24s or 21s as well before going ported.
 
#532 ·
I don't plan on being nice to the 24's 7 days from now. In fact, I plan on trying to power rape them until they scream out the safe word.


If you want to hang with the big dogs with you gotta earn it... By the end of the GTG the 24's will either earn the respect of all, or die trying.


IMO, if the 24's can match the LMS-U's peak output across the FR while keeping it's distortion profile at or below the LMS-U's, then it's a win/win all around...especially at $799 shipped. Anything above that is a bonus...even priced under $1K (IMO).


I still can't wait for Ricci to get his paws on this monster... If the report comes back favorable, hopefully people will give SI a bit of credit for actually knowing how to design a "serious" driver.


Until then, guess next weekends subjective banter will have to make due.

 
#533 ·

Quote:
Originally Posted by notnyt  /t/1501074/si-24z-build-8-of-em/510#post_24544719


Interesting claims without being tested. Will be interested in seeing a data-bass report =] Just because xmech is double xmax does not mean you have that much more usable stroke.

Interesting that you post models as gospel but question the actual performance of the 24".
Quote:
Yes because people are building 8x24 systems to reproduce tracks at reference levels. Ok then. Just because it's silly to you does not mean it isn't serious for others. It's a discussion anyway, not really a debate. Also, went from 6.5 cubic feet boxes to 27 cubic feet. I did not have headroom in that range as I was bottoming out the LMS drivers. I gained so much output there I can run at hotter levels and not worry about it anymore. Pod emergence at +10db over reference without a hint of stress now. Before, I had to deal with the lovely sound of metal cones bottoming out when only a couple db hot.

When you quote double-digit 'model' increases +3dB as though it's in the bag, many readers take that as verified fact. I don't because it's likely not so. Here's what I know to date regarding the data vs the posted model:

http://www.imagehousing.com/image/1194144


The actual measurements show no such gain insensitivity. They also show no THD at excursion max vs output at the HPF and above tune, they show zero port compression, which is impossible, etc. Saying the subs don't bottom with one scene tells us nothing, let alone 12-15dB or whatever it is supposed to be, headroom at 15 Hz.


Until the actual data is presented in a way that's directly comparable, we don't know what the headroom is.


Basically, I see the discussion not as one of reproduction but rather one of compromise to increase playback level. There's a big difference between the 2 with no reference for where you draw the line. That may seem irrelevant to you and others but the majority are interested in 2 things here; bang-for-buck or accuracy. These SPL quests are neither, IMO.


Bottom line is that the proposed 8 x 24" sealed system is going to give reference level playback to the bottom of the signal chain in any room. Suggesting a compromise in BW for a 'house curve' response to 'x' Hz is fun discussion but never seems to be accompanied by the requisites, especially lately.
 
#534 ·

Quote:
Originally Posted by bossobass  /t/1501074/si-24z-build-8-of-em/500_100#post_24545030


Interesting that you post models as gospel but question the actual performance of the 24".

No, I'm questioning figures tossed out on this thread that vary greatly with what the MFR has on their page
Quote:
Originally Posted by bossobass  /t/1501074/si-24z-build-8-of-em/500_100#post_24545030


When you quote double-digit 'model' increases +3dB as though it's in the bag, many readers take that as verified fact. I don't because it's likely not so. Here's what I know to date regarding the data vs the posted model:


The actual measurements show no such gain insensitivity. They also show no THD at excursion max vs output at the HPF and above tune, they show zero port compression, which is impossible, etc. Saying the subs don't bottom with one scene tells us nothing, let alone 12-15dB or whatever it is supposed to be, headroom at 15 Hz.


Until the actual data is presented in a way that's directly comparable, we don't know what the headroom is.


Basically, I see the discussion not as one of reproduction but rather one of compromise to increase playback level. There's a big difference between the 2 with no reference for where you draw the line. That may seem irrelevant to you and others but the majority are interested in 2 things here; bang-for-buck or accuracy. These SPL quests are neither, IMO.


Bottom line is that the proposed 8 x 24" sealed system is going to give reference level playback to the bottom of the signal chain in any room. Suggesting a compromise in BW for a 'house curve' response to 'x' Hz is fun discussion but never seems to be accompanied by the requisites, especially lately.

You're using graphs from measurements that I posted were from a bad mic. Good job.


Red is bad WM61a, blue is good WM61a. Unfortunately I didn't get to measure groundplane outdoors with a good mic. Feel free to interpolate from these measurements, but too many variables for that to be accurate.




Or you know, you can compare actual back-to-back measurements of the same drivers in the same location outdoors, just going from sealed to ported... Granted, mic is bunk, but response at different levels was at least linear.




Also, compression graphs with linear sweeps were posted here
 
#536 ·
How about a mix of ported and sealed? That's my next plan.



I've read many times that mixing sealed and ported is a no no since the ported sub will be out of phase below tuning, but Dom does it and when I had 4 of my SI's at Archaea's place we were able to mix with his 2 ported Captivators as long as we had a 48db slope crossover right around tuning of the ported sub (maybe just above tuning, can't remember for sure). Either way the results were impressive.
 
#537 ·

Quote:
Originally Posted by notnyt  /t/1501074/si-24z-build-8-of-em/510#post_24544872


I'm not quite sure where you're getting the double xmech figures from either as the posted ones are xmax 36mm xmech 45mm, are you seeing that is not accurate? I'm not trying to start a debate here, 8x 24 will be such overkill for any sane person. I just wonder how many of us fit into that category
The only point I was really trying to make is that with crazy systems like these, we generally have extra headroom above 30-40hz, but in the lower frequencies this isn't always the case as excursion starts to be a limiting factor.


This is in a high Q 0.937 box, 10 cubes with 4000 watts. You can see how crazy the excursion will be and how quickly it starts ramping up.




Anyway, just my $0.02. I considered switching to 24s or 21s as well before going ported.

The only thing I take issue with is to suggest that "8X24z would be overkill for sane people, but who many of us are in that category?"....I would make a high dollar bet that my 8 24s sealed in 15cubes with 3kw a piece would make any one of us nutcases giggle with content...IMO, to suggest otherwise is kind of silly.


Yes, I am suggesting that the xmech of 45mm is inaccurate...Nick will never say that because he doesn't want idiots like us taking his word on that and blasting these things out of the enclosures, but the true xmech is at least 10mm beyond that. I don't doubt that you'd prefer the ported versions, as that's what you've come to love. I hope my expectations are met....from conversations with folks who have used them (folks that I respect the opinions of based on experience) they are something special and bring a real presence with them.
 
#538 ·

Quote:
Originally Posted by ChopShop1  /t/1501074/si-24z-build-8-of-em/500_100#post_24545434


The only thing I take issue with is to suggest that "8X24z would be overkill for sane people, but who many of us are in that category?"....I would make a high dollar bet that my 8 24s sealed in 15cubes with 3kw a piece would make any one of us nutcases giggle with content...IMO, to suggest otherwise is kind of silly.


Yes, I am suggesting that the xmech of 45mm is inaccurate...Nick will never say that because he doesn't want idiots like us taking his word on that and blasting these things out of the enclosures, but the true xmech is at least 10mm beyond that. I don't doubt that you'd prefer the ported versions, as that's what you've come to love. I hope my expectations are met....from conversations with folks who have used them (folks that I respect the opinions of based on experience) they are something special and bring a real presence with them.

Interesting. I'd like to see some actual numbers there, including how bl is actually linear and how much of the throw has motor force behind it... for science! I don't agree with spec skewing for any purpose. In this case, it's insulting to the consumer, like saying "you're too dumb to use real specs, we need to trim these down so you don't do something stupid".


Unfortunately the LMS drivers xmech is a hair above xmax. Fortunately, I'm happy with that situation. Obviously 8 24s in ported enclosures would do better than 8 18s in ported enclosures, I just didn't have the space, or it would be an option
Alternatively, could have sold the LMS, got 4 24s using one per enclosure, and come out with some cash in my pocket. However, that reduces overall power handling by at least a factor of 2, so while you're doing well with less drivers where port efficiency is involved, you're at a net loss above that, and we're back to hoffman's iron law and the amount of space needed.


Again, everyone's situation is different. Mine necessitated ported enclosures to get the efficiency required

On the flip side, they won't have the feel of a ported enclosure, which I definitely missed.
 
#540 ·

Quote:
Originally Posted by notnyt  /t/1501074/si-24z-build-8-of-em/510#post_24545493


Also, while we're on the subject, do you happen to have a measurement of how tall the surround is from the edge of the cone? I couldn't find any mechanical drawings of these.

I don't, but I'll ask Nick when I talk to him at the beginning of the week and see if I can get those for you
 
#541 ·

Quote:
Originally Posted by bossobass  /t/1501074/si-24z-build-8-of-em/510#post_24544081


If you have difficulty in reproducing 20 Hz with 70L of displacement driven by 56KW peak power then state your new standard for reproduction as a disclaimer. These 20dB hot, +10dBRL threads used to be a bit amusing but they really are just getting silly.

He doesn't have 56kW @ 2ohm's, he has 35.2kW@ 4ohm's. That is only possible if you had octal 2-ohm's (like half the coils disconnected).


octal 1 ohm's, octal 4 ohm's or quad 8ohm's. There is no way to get there with all the coils pumping.


By that definition the rig I'm in the process of building has 89kW's @ 2ohm's and 68kW @ 4 ohm's.


Yet I have already proven with a Fluke power analyzer that sealed LMS's bottom out at around 400watts continuous @ 20hz.

The rest of the watts is used for musical program in short bursts, mostly at high frequencies.


It would take 16 LMS's in series-parellel and 4 real Gruppen's to be at 56kW (or use 4 clones, and risk blowing them up I suppose).
 
#544 ·
"The actual measurements show no such gain insensitivity."


just to clarify bb, are you now suggesting that ported don't increase sensitivity as predicted by t/s parameters as i thought this was pretty well worked out science?


josh measured the re-xxx in a sealed cab and a ported cab.




while the sealed cab was small for the driver, so was the ported cab.



in any case, we see the same increase in sensitivity there of about ~12db around the tuning frequency that 'not saw in his build and is what is expected according to t/s parameters.


the sealed has a large peak owing to the qt in the enclosure. maybe it is this effect that is the reason why "high q" subs are the shizzle in the minds of some car audio enthusiasts these days.


as for distortion, i'm not quite sure what to make of it, but at the limit, this is what was reported:


ported sub:




sealed sub:




one thing about the sealed subs in 10 cf is whether or not the motor has enough thermal capability to take advantage of the excursion capability in such small enclosures. the number that i have in my file was 1500 watts, which only gets you to about 20mm excursion or so. maybe the 1500 is an old number.


however it is sliced, this is going to be a whopper of a sub-system and either alignment choice will bring good results. no doubt about that.
 
#545 ·
also, not sure if this is relevant or not, but with redirected bass from all the channels, -0 reference is about 123db on the subwoofers @1m. if they *are* co-located, then 8 of these 24's driven to xmax will give about 126db with no room gain at 15hz. if they are *not* co-located, then the number will be less than that. 126db is only +3db of "heat" on the bass. that's not much.


so as to whether or not sealed vs. ported make sense, the room and its attendant gain is really driving a big part of 'how much is enough'. crazy, but right. my guess is this lack of room gain is what 'not was running into and why the change to ported was of such benefit.
 
#546 ·

Quote:
Originally Posted by LTD02  /t/1501074/si-24z-build-8-of-em/540#post_24547903


"The actual measurements show no such gain insensitivity."


just to clarify bb, are you now suggesting that ported don't increase sensitivity as predicted by t/s parameters as i thought this was pretty well worked out science?

Of course not. I said not's posted data did not show a sensitivity increase, suggesting his data is flawed and therefore any conclusions are as well.


Comparing his previously posted compression sweeps of the sealed system to the new ported system:

http://www.imagehousing.com/image/1194242


[Dotted trace = zero compression reference, magenta trace = actual max compression sweep result (with port compression beginning to show) and green trace = sealed system max compression sweep result]


It looks to me as though he gains 6dB across 1 octave, which is what I would expect to see in the comparison.


The tradeoff is a loss of 10dB
 
#547 ·

Quote:
Originally Posted by BassThatHz  /t/1501074/si-24z-build-8-of-em/540#post_24546087


He doesn't have 56kW @ 2ohm's, he has 35.2kW@ 4ohm's. That is only possible if you had octal 2-ohm's (like half the coils disconnected).


octal 1 ohm's, octal 4 ohm's or quad 8ohm's. There is no way to get there with all the coils pumping.


By that definition the rig I'm in the process of building has 89kW's @ 2ohm's and 68kW @ 4 ohm's.


Yet I have already proven with a Fluke power analyzer that sealed LMS's bottom out at around 400watts continuous @ 20hz.

The rest of the watts is used for musical program in short bursts, mostly at high frequencies.


It would take 16 LMS's in series-parellel and 4 real Gruppen's to be at 56kW (or use 4 clones, and risk blowing them up I suppose).

BTH,


Thanks for the catch. I'm off by 2dB.



Not sure what you've proven or how exactly but both Ilkka and Josh fed 170V into an LMS-U with no bottoming reported.
 
#548 ·

Quote:
Originally Posted by bossobass  /t/1501074/si-24z-build-8-of-em/540#post_24548119


BTH,


Thanks for the catch. I'm off by 2dB.



Not sure what you've proven or how exactly but both Ilkka and Josh fed 170V into an LMS-U with no bottoming reported.

170 volts at what frequency, and what alignment... we don't really have to go 'round on this subject again do we? Warpdrv bottomed 6x LMS-U's and dented the cones with Crest 8002's in dual opposed sealed boxes, after a suggestion was made that he could add in some eq down low, since he had 'ample headroom' and 'no one needs that much output'.


Notyt also had bottoming... he built ported enclosures to gain headroom, and suddenly he can't measure accurately, and he's not your friend in the 'sealed is the cure for cancer camp'?


Hoffman's Iron Law states that the efficiency of a woofer system is directly proportional to its cabinet volume and the cube of its cutoff frequency (the lowest frequency it can usefully reproduce). The obvious implication is that to reduce the cutoff frequency by a factor of two, e.g. from 40 Hz to 20 Hz, while still retaining the same system efficiency, you need to increase the enclosure volume by 23=8 times!
 
#549 ·
So tell me, you think the difference is going to be made by taking one cabinet and stacking it on the other, they are basically in the same location, no?


"I also believe the sealed subs would have shown better low end 'sensitivity' if they had been stacked in the corner as the ported system is, for obvious reasons."


I also believe I will have another cup of coffee, and that the tooth fairy, santa and the fairy godmother is not real.


http://]www.avsforum.com/t/1314884/8x-18-lms-ultra-5400s-in-4-sealed-enclosures[/URL ]






http://www.avsforum.com/t/1516234/8x-18-lms-5400-ultras-sealed-to-27-cube-15hz-ported-build/180


 
#550 ·

Quote:
Originally Posted by michael hurd  /t/1501074/si-24z-build-8-of-em/540#post_24548393


So tell me, you think the difference is going to be made by taking one cabinet and stacking it on the other, they are basically in the same location, no?

Michael,

I too think he's gained significantly with the new quasi baffle-wall like approach. I've posted before I believe the acoustic efficiencies of the current approach create a great peak summation of in phase energy. Whereas the old approach suffered a somewhat less than optimal SBIR laden impulse ... especially up in the region where notnyt first noted the subjective benefits. I like what he's done, he was encountering bottoming concerns, he addressed it...noting the differences and tradeoffs.




Independent of my personal convictions and approach, and independent of any potential bias' that he brings relative to his experiences, I find notnyt's system makeover from sealed to vented, a fascinating study. He's a contributor I'll always take the time to read.


He's typically as precise as anyone, ... whether it's measuring acoustic aspects or measuring electronic signal path, or simply articulating his subjective findings. He rarely if ever uses hyperbole etc, in his posts. All said I find his contributions to be top rate.


Quote:
Originally Posted by bossobass  /t/1501074/si-24z-build-8-of-em/540#post_24548097


I also believe the sealed subs would have shown better low end 'sensitivity' if they had been stacked in the corner as the ported system is, for obvious reasons.

Wrt notnyt's preference for vented, or anyone's subjective preference for any audio related anything, it's consistently been shown that merely a raise in level is sufficient to significantly sway subjective preferences toward the louder item.


Some of his first comments clearly and enthusiastically stated the preference for the vented, as it possessed a more punch or greater visceral component. I suggested the baffle-wall like wave-launch was superior in that the energy possessed greater impulse, greater level of peak summation in phase, less 1/2 space diffraction back toward the boundaries behind and to the sides.


So yeah Bosso, I agree the wave-launch as it existed prior to the changeover to vented, wasn't as optimized as it could have been. And this is especially true in the punch or visceral region. Is it enough to matter? It would matter significantly, in my opinion, but I don't know how much. In the upper range of the previous sealed subs, yes I believe there were acoustic gains left on the table. However, the area of interest for notnyt, ~20hz, not as much as he gained going tuned/reflex.
Quote:
Originally Posted by bossobass  /t/1501074/si-24z-build-8-of-em/540#post_24548097


This is nothing new. Callas (with his LLT) and I hashed this out over a number of years. I'm surprised it's being rehashed.

I don't view it as a re-hash, he recognizes and accepts the differences.



Bosso, I know you've mentioned having several dozen varietal subwoofer drivers over the past several years, have you explored either subjective or measured differences, in your room or any room you know well?




Good stuff, thanks,

Kevin
 
  • Like
Reactions: Scott Simonian
#551 ·

Quote:
Originally Posted by FOH  /t/1501074/si-24z-build-8-of-em/540#post_24548460


Michael,

I too think he's gained significantly with the new quasi baffle-wall like approach. I've posted before I believe the acoustic efficiencies of the current approach create a great peak summation of in phase energy. Whereas the old approach suffered a somewhat less than optimal SBIR laden impulse ... especially up in the region where notnyt first noted the subjective benefits. I like what he's done, he was encountering bottoming concerns, he addressed it...noting the differences and tradeoffs.



Wrt notnyt's preference for vented, or anyone's subjective preference for any audio related anything, it's consistently been shown that merely a raise in level is sufficient to significantly sway subjective preferences toward the louder item.



Good stuff, thanks,

Kevin

With respect, in the frequencies we are talking about in the subwoofer realm, and the polar response of the box, the sound will wrap around those boxes just like the sealed boxes were there prior... this is not like mounting an acoustically small speaker in a baffle wall.


Current through the voice coil is going to be less than with the sealed boxes, leading to less inductance variance with respect to the same output level ( not that the LMS has inductance issues, it's a fantastic driver) , as well as less excursion for the same output level. This is a significant change.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top