Dismiss Notice

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

Question for BGavin or Joris - Avoiding driver destruction

Discussion in 'Amps and Cabs [BG]' started by ESP-LTD, Dec 1, 2002.


  1. ESP-LTD

    ESP-LTD

    Sep 9, 2001
    Idaho
    I have built a few boxen and was acoustically satisfied with most of the results. I have blown drivers out of commercial 2x10 boxen using a bolt-on 4 string (hyperexcursion I imagine).

    I have been playing a neck-thru 5 string and am getting pretty spoiled with the extra range. I am considering making some 2x10 boxes based on Eminence BP-102's. I realize that these won't be flat down to 31hz, but I have to sacrifice somewhere because I want small/light boxes for gigs, and I figure I can do 1/2" Baltic Birch on these and they will survive physically.

    Is there a way to assure I won't smoke these drivers with a 5 string? Tuning the cabinet to 30hz? Blowing off low end and going sealed? Or am I just destined to trash these with occasional use of a low B?

    My runner up choices are Dayton 15 (4"vc fs=37) and Eminence Magnum 15LF's but I would have to move up to 3/4" Baltic Birch and a lot more weight.
     
  2. flacko

    flacko

    Dec 6, 2001
    Surrey , UK
    ESP, without wishing to hi-jack your thread, I am also eager to see the reply to this. I am thinking of starting a similar project.

    I am considering the same driver but two 1X10 ported cabs. I'm a 4 string player. What concerns me is the 38 ounce magnet - I thought this looked a bit light (the B102 for example has a heavier magnet). Is this a sign of a slow sounding speaker?

    I intend going with a thinner ply (half or 3 quarter) and placing hardwood bracing where needed.
     
  3. Making a sealed cab will only get you in more trouble, excursion-wise. So that's a no-go.

    If you wanna play it super safe, tune the cab to 31 Hz, that way, the low B will never overexcurse (sp?) the cones, unless your ports are too small.

    I'd do it like this: I'd tune it to 36 Hz (a little above the low B), because decoupling - and thus overexcursion - won't occur immediately below the tuning frequency.

    For lightweight construction, 1/2 inch birch is tricky. You'd have to at least put 2" ribs along every inside wall to stiffen things. My 0.02.

    Regards,
    Joris
     
  4. ESP-LTD

    ESP-LTD

    Sep 9, 2001
    Idaho
    Joris-

    I appreciate your thoughts on it and the tuning.

    I have used 1/2" and 5/8" before but have never tested them for flex on the sidewalls. They survived careful handling but I can see why a manufacturer would use heavier for production jobs. Perhaps I'll use 3/4" front and back and 1/2" on the sides with more bracing.

    Is it of any benefit to use the asphalt-based anti-vibration sheets they use to dampen car sheet metal? It could be that the pressures are just too high.

    I also wonder if anyone has compared corrugated foam sheets (like mattress pads) against loose polyester batting as a fill material? I would think the foam would kill reflected highs better, but I don't know about the apparent volume effects.
     
  5. Joris, say it ain't so! I've been under the assumption that running sealed was a safety measure to protect speakers. You know, using the trapped air as kind of an acoustic suspension shock absorber to limit excursion. Seems like it would always be better or at least equal to a ported cabinet in the amount of resistance it applied to the speaker movement.

    Clue me in.

    Thanks,
    Chris
     
  6. ESP-LTD

    ESP-LTD

    Sep 9, 2001
    Idaho
    I notice that many of the Eminence 10's have the same cone mass but bigger magnets. I did have some cheap aluminum cone 10's years ago ($15 from Mcm) that had 38 oz magnets. I thought they sounded fine (although only 92db sensitivity) until their 70w voice coils warped.

    The Vas is a lot different than the other 10's and it makes the BP-102's pretty softly suspended so the magnets don't have so much to overcome.

    Maybe they trade sensitivity for linearity, by giving the cone a fast but gentle shove against soft suspension? I would guess the soft suspension limits the upper frequency response as well. Perhaps the solution is to use a 3rd 10 of a different model at 94db or so crossed over at 1k?

    (anyone with real knowledge feel free to correct me!)
     
  7. A sealed box requires more excursion than a vented box and this gets worse the lower the frequency goes. A sealed box has less total output because it is more inefficient.

    A vented box offers maximum cone damping at the tuning frequency and the minimum cone movement. This is one of the reasons for finding a driver with an alignment that is optimally tuned at the lowest working frequency (low B).

    Group delay numbers (read: mud) are the worst at the tuning frequency. Some drivers are very good, others are just awful. It depends on the individual driver in question.

    The weight of the magnet is an indicator of driver quality. So is the size of the voice coil. The JBL 2242 has a 368 ounce magnet, a 4-inch voice coil, and a BL (motor force factor) of 23.7. This is a seriously high quality (read: expensive) driver. The Eminence Alpha 15 has a 1.5" voice coil, 25 ounce magnet and a 7.7 BL factor. This is the other end of the spectrum.

    I have included a ratio of BL to MMS (force to cone mass) in my spread sheet. This is the ratio of motor strength to cone weight. More strength per gram is tighter control. This is an abitrary ratio, and compares equally between drivers. It is just a reference point. But I do notice that snappy drivers have a higher ratio than slug drivers.

    Download my spread sheet and sort the DRIVERS tab by column "C" which is "Use". Look for the green colored drivers. The SBB4 alignment group is the one I favor because it offers the best transient (read: "fast") performance, albeit with a larger cabinet size.

    If you don't have Excel, you can download an Excel Viewer from the same directory where my spread sheet is found.
     
  8. I can dig all that. I can see that to get equal volume level, you'd have to use more power and move the speakers farther in a sealed box.

    But I'm coming at it from a different angle. I'm not talking about getting the same output volume level from a ported versus a sealed, I'm talking about keeping input power the same and thinking a sealed would lower the excursion under these circumstances, and like you say, output volume would be lower too since since the sealed is more inefficient.

    Is that correct?

    Chris
     

  9. Link to Eminence Kilomax Pro 18 Excursion Plot


    The Kilomax Pro 18 has an EBP of 57 meaning it is more suited for a sealed box. I have plotted it both Sealed (B2) and Vented (B2).

    In the Sealed plot, note where Xmax is exceeded at 60 Hz with a full 1250 watts input power. This is not the point of self-destruct, as this driver has a 9.8mm Xmax, but an Xmech excursion of 38.1mm. This particular driver runs out of power handling ability before it runs out of cone movement.

    In the Vented plot, the driver excursion peaks near 18mm at roughly 35 Hz. Excursion begins to reduce as the frequency drops closer to the tuning frequency. In the sealed box, excursion continues to rise. Note the excursion slope is virtually identical in both cases. With 1250 watts input power, this driver is not going to over-excurse beyond the Xmech limits, but both will exceed Xmax, and the distortion will go way up.


    Link to Eminence Magnum 18LF Excursion Plot


    This driver shows the advantage of tuning at 31 Hz for a low B application. Note how the sealed excursion continues to rise below 50 Hz, where the vented excursion decreases down to the tuning frequency.

    Xmax is 6.4mm for this driver and it is exceeded below 60 Hz at full power. The vented box remains within the linear range down to the tuning frequency, where the sealed box continues to increase distortion by operating outside the linear Xmax region.
     
  10. Good plots!

    I see what you mean about tuned boxes being better around the tuning frequency and slightly above. Upwards beyond that they track with the sealed box. But what I am referring to is how the vented graphs skyrocket off the page below the tuning frequency, and the sealed boxes level off. That's the effect I was envisioning. It probably doesn't come into play very much since it occurs at such a low frequency, but it still seems to act like a shock absorber at these ultra-low frequencies.

    Plus, I like the phase plot of sealed boxes, even if they don't go as low. :D

    Chris
     
  11. ESP-LTD

    ESP-LTD

    Sep 9, 2001
    Idaho
    I assure you your spreadsheet does not go to waste!

    Thank you for another excellent lesson in cabinet design.
     
  12. geshel

    geshel

    Oct 2, 2001
    Seattle
    Yup. Actually, those low frequencies show up all the time in bass signals, if you play aggressively at all or slap. Slapping a bass string is more or less an impulse, and contains a good deal of energy below 20Hz. Using piezo elements is even worse. Thump the strings, and watch the drivers jump.

    I think, though, you're probably OK for tuning frequencies in the 30-40Hz range like the example Bruce posted. I'd worry about playing a five through a box tuned to 50Hz or so (Eden? :) ).
     
  13. This is the basis of why I don't like playing low notes on high-tuned boxes.

    The "skyrocket" you describe occurs at a rate of 24 to 36 dB per octave. This is a steep slope, and one doesn't have to get much below the tuning frequency before the cone really starts dancing again. Hooking your rig up to a sound card LINE OUT and (gently) driving the speakers with various low notes is quite revealing. The cone motion on a D410XLT is zilch at 44 Hz, and really humping at 31 Hz.

    Sealed boxes have been the audiophile staple forever. Nearly all the high end subwoofer drivers are optimized for sealed boxes because of their tight response. I wonder what the cutoff point is where you can no longer hear the differences between vented and sealed. I've seen 25msec bandied around, but that seems awfully high. I'd think a 10msec or less delay would be pretty great sounding, and it is achieveable with the right drivers and vented alignments.