Psst... Ready to join TalkBass and start posting, make new friends, sell your gear, and more?  Register your free account in 30 seconds.

Lab12 subwoofer closed box sub idea?silly?

Discussion in 'Amps and Cabs [BG]' started by Magman, May 11, 2003.


  1. Someone mentioned the Lab12 eminince sub recently and I put its TS params in winISD and have been fiddling around, wondering if its possible to stick a couple in an avatar cab and tune it real low. Apparently, to tune an average 212 cab down to 20 or even 15hz (which the lab12 will happily do it seems, quite an unusual driver) without lots of port noise, you need impossibly huge ports.

    More esoteric design is needed, which the lab12 was designed for (the 'lab 12 project'), however that beyond my current carpentry skills...

    but..

    ..what about closed box designs? In winISD in a non ported cab, the group delay and freq response was very good. Sensitivity is low, but a closed box Lab12 cab should go to 20hz much louder than most other drivers out there in a similar sized ported cab(with enough power) - thats the whole point, because of the huge xMax of the lab12.

    So according to my theory, buying an empty avatar 212 cab, covering up the port, baffling the inside and sticking in 2 lab12's should give a great 800 watt subwoofer which really can handle the lows (for a small PA or bass rig) and would cost about 450 - very simple project.

    Any obvious problems in this idea? Possibly just too quiet? Maybe I should ask on the lab12 list....

    Jamie
     
  2. I estimate the Avatar 2x12 cabinet to have about 4.5 cubic feet net internal volume after subtracting for bracing and two drivers. This an acoustic volume of 2.25 cubic feet per driver.

    The LAB12 requires 2.85 to 3.6 cubic feet for vented alignments. Sealed alignments are 1.8 to 3.4 cubic feet (B2 and D2 respectively). The sealed box is very forgiving for size changes. As the box size increases beyond the B2 size, the group delay improves, F3 moves higher, and power handling ability gets worse.

    The LAB12 has an efficiency bandwidth product of 56, making it ideal for use in a sealed box. Its actual SPL calculates at only 87.3, making it 2dB worse than my Rockford subwoofers. Mine are terribly power hungry, and I push two of them with the left channel of a PLX 3002 at near-clipping input levels, trying to keep up with my band. The LAB12 will be even worse.

    You will need two LAB12 and 500 watts each to drive it. Eminence does not provide a power handling spec for this driver, so I guessed at 500 watts maximum (thermal) input power. This driver calculates at a 113 SPL maximum at 500 watts (vented) and 105 SPL in the D2 sized sealed box. This isn't very loud at all, but is suitable for home subwoofer use.

    As it turns out, the Avatar 212 box is optimally sized and tuned for a pair of Eminence Delta 12LF drivers, which happen to be the very same drivers Dave installs when he manufactures them. Two Delta 12LF in parallel will give you 97 SPL, which is +7 dB SPL better than a pair of LAB12. The F3 on the Avatar is 48 Hz.
     
  3. ESP-LTD

    ESP-LTD

    Sep 9, 2001
    Idaho
    Bgavin's post showed up while I was editing, but perhaps some of the following is still relevant.


    To minimize chuffing you need ports at least 4" in diameter for a 12" speaker (last rule of thumb I heard). They do get pretty long at low frequencies but you could use PVC and make a few corners.

    I was looking up the specs for the Lab 12 and comparing them to my JBL GT-120 subs. I run the JBL's in 2.4' at 28hz.

    One option for tuning- if you make your tuning frequency below your lowest note, the cab will never get there and since it moves maximum air thru the port at that frequency, you can cheat a little on port diameter without chuffing. I ended up using Parts Express 260-389 3.375" angled port which crosses the inside of the box on a diagonal. It's below optimal diameter, but I'm tuning the box to 28 and using a 5 string, so I never really get close to 28hz anyway.

    The 13mm Xmax of the Lab12 looks cool, but I'm getting along ok with only 12mm (total investment for 2 subs was about $300).
     
  4. Wes Whitmore

    Wes Whitmore Supporting Member

    Mar 10, 2003
    Columbus, OH
    I think the Lab12 will handle 400 watts continuous per driver.
     
  5. geshel

    geshel

    Oct 2, 2001
    Seattle
  6. ESP-LTD

    ESP-LTD

    Sep 9, 2001
    Idaho
    Geshel- That's a mighty serious piece of work!
     
  7. Petebass

    Petebass

    Dec 22, 2002
    QLD Australia
    Life's like that......
     
  8. Thanks for the informative replies.

    I am curious about the apparent lack of sensitivity of a lab configuration vs delta LF's, or a single Kappa 15 LF.

    In winISD, the SPL at 30hz is 'about' the same for each (though sensitivity figures are presumably averages and not exactly indicative of actual sens. at 30hz)

    Now, because of the large xmax of the Lab, presumably it will be able to generate much more 30hz sound before entering distortion? ---- but I guess you still need about 4 drivers to give a 'flat' responce with single 210 taking care of 100hz, and you'll need filters/crossovers to balance it all out anyways....
     
  9. geshel

    geshel

    Oct 2, 2001
    Seattle
    If you could make a driver with the excursion and low resonance that the LAB has, but the sensitivity of the Deltas, you'd make a lot of money! The first two basically come at the expense of the last one.
     
  10. geshel

    geshel

    Oct 2, 2001
    Seattle
    P.S. in WinISD Pro, at least, you can put in a power level and get the real (estimated) SPL - so you could directly compare the output of the two configs at say 100w (at higher powers it's probably less accurate due to power compression however).
     
  11. Petebass

    Petebass

    Dec 22, 2002
    QLD Australia
    sensitivity is actually measured by putting a 1,000Hz sine wave through the speaker at 1 Watt, then measuring the output in Db at 1 meter distance. As Geshel pointed out, you can change these parameters in Win ISD. Trouble with winISd is it doesn't give cone excursion graphs.
     
  12. geshel

    geshel

    Oct 2, 2001
    Seattle
    WinISD pro does. It's got a lot more graphs. It's just in beta but it's still free and everything. Seems to work OK for me.
     
  13. alexclaber

    alexclaber Commercial User

    Jun 19, 2001
    Brighton, UK
    Director - Barefaced Ltd
    Do they really do this for subwoofers? Surely it would make more sense to measure the sensitivity within the usable bandwidth of the speaker?

    Alex
     
  14. Most speakers have a sensitivity graph which shows repsonce across all freqs, the JBL GT120 spec sheet has such graphs for different types of boxes which is cool. Seems to be a good driver as a sub, and not too expensive.

    Given the omnidirectionality of bass freqs, I'm wondering if a sort of "rumble cube" - say an 16 inch sealed square cab loaded with 3 GT120's pointing left/right/forward might work rather nicely, with a 100hz lp filter built in.
    But each driver is 4ohm so thats not going to work...
     
  15. Petebass

    Petebass

    Dec 22, 2002
    QLD Australia
    As far as I know - Yes. 1k isn't that high that a sub can't produce it. It doesn't matter if it doesn't sound "musical" at that frequency because it's only a sine wave after all.

    Remember that sensitivity is only one of the many T/S parameters. None of the T/S parameters tell you much on their own. You've gotta look at them as a whole.
     
  16. Petebass

    Petebass

    Dec 22, 2002
    QLD Australia
    Just curious. What am I looking for in the T/S parameters that suggest a speaker would be good for a folded horn. I'm talking from a PA sub perspective........
     
  17. geshel

    geshel

    Oct 2, 2001
    Seattle
    Actually, it's more likely for a sub to have nasty midrange resonances than other drivers (as they don't care about them in the design). If one of these resonances is at the frequency used to take the sensitivity, it will be misleading. Or even if the output in that range is elevated.

    For instance, someone could build a sub, put a 100Hz low pass crossover on it, and measure -3dB points at 30Hz and 100Hz. Then take off the crossover and measure it at 500Hz and it's 5dB louder than the zero-level used to find the -3dB points. If they publish that number, it's wrong. I doubt, for finished boxes w/ built-in xovers that anyone does this. But, for a raw driver, the output at 1kHz doesn't have a whole lot to do with the output at a lower frequency on an infinte baffle.