Greets!
I don't build anymore, so don't keep up with what's available and I'm sure there's many others, but the MAW-15, MJ-18 have been popular lately and at the quantity you need they may make customs at a reasonable price: http://www.mach5audio.com/index-2.html
Perennial favs:
http://www.partsexpress.com/pe/showdetl.cfm?&Partnumber=295-130
http://www.partsexpress.com/pe/showdetl.cfm?&Partnumber=295-325
GM
I don't build anymore, so don't keep up with what's available and I'm sure there's many others, but the MAW-15, MJ-18 have been popular lately and at the quantity you need they may make customs at a reasonable price: http://www.mach5audio.com/index-2.html
Perennial favs:
http://www.partsexpress.com/pe/showdetl.cfm?&Partnumber=295-130
http://www.partsexpress.com/pe/showdetl.cfm?&Partnumber=295-325
GM
OK so I have the 604's in the large open baffles (60" high x 49" wide) and it is a monster. Actually two monsters. I haven't gotten the XO's hooked up yet but hopefully tomorrow. I spent most of the afternoon and evening making the panels and I am beat. Moving these things around takes some work. They sure look impressive. Lets hope they sound as good as the model looked.
I checked out Mach5Audio and they have an 18" sub that has an Fs of 15.4 Hz, a Qts of .36, and Xmax=22mm! It models great in a ~6 ft^3 BR cabinet (-3db@15Hz). I want to use four of them to get the level to match the OB. Nothing fancy. A closed box rolls of sooner and is down 6db from BR at 20Hz. As with the 604, I would like to maximize the performance of this sub.
Any ideas?
I checked out Mach5Audio and they have an 18" sub that has an Fs of 15.4 Hz, a Qts of .36, and Xmax=22mm! It models great in a ~6 ft^3 BR cabinet (-3db@15Hz). I want to use four of them to get the level to match the OB. Nothing fancy. A closed box rolls of sooner and is down 6db from BR at 20Hz. As with the 604, I would like to maximize the performance of this sub.
Any ideas?
Greets!
Yeah, ~50 OBs are large, but it's relative. I mean W.E.'s original theater horn driven by a single 1" exit compression driver had a folded up ~15 ft path-length terminating into a 40^2 ft mouth to get to 50 Hz, though obviously with much > efficiency/directivity.
Hmm, I thought you were going to make OB subs also.
Regardless, this is a true sub driver with only a ~85.6 HF mass corner, so won't have the high SQ BW required to blend to a low Q OB, ergo a wide BW woofer system ideally is required. With that in mind, like most drivers, an optimum BLH is the theoretical ideal alignment (stupid large in this case), with IB or TL next, etc. WRT trading gain BW for maintaining high SQ.
GM
Yeah, ~50 OBs are large, but it's relative. I mean W.E.'s original theater horn driven by a single 1" exit compression driver had a folded up ~15 ft path-length terminating into a 40^2 ft mouth to get to 50 Hz, though obviously with much > efficiency/directivity.
Hmm, I thought you were going to make OB subs also.
Regardless, this is a true sub driver with only a ~85.6 HF mass corner, so won't have the high SQ BW required to blend to a low Q OB, ergo a wide BW woofer system ideally is required. With that in mind, like most drivers, an optimum BLH is the theoretical ideal alignment (stupid large in this case), with IB or TL next, etc. WRT trading gain BW for maintaining high SQ.
GM
OB subs aren't out of the question, but I agree with you on the efficiency (lack of) and respective power/eq requirements. I wanted to see what the driver does in a closed and vented box.
To design the OB sub requires knowing what maximum (linear) excursion is required for a particular driver with the -6db oct rolloff compensation applied to achieve the desired levels. I suppose by clustering enough subs with enough power, excursion isn't a problem as much as cluster/amp count. I would still engineer the system as to know the operating conditions and limits. As they say, measure twice and cutting once.
So I need a sub driver that will go low enough for real bass and high enough to smoothly transition to the OB panels. Got it.
I'll keep looking. It's too bad we can't make our own drivers to get the exact Qts needed for a particular design.
To design the OB sub requires knowing what maximum (linear) excursion is required for a particular driver with the -6db oct rolloff compensation applied to achieve the desired levels. I suppose by clustering enough subs with enough power, excursion isn't a problem as much as cluster/amp count. I would still engineer the system as to know the operating conditions and limits. As they say, measure twice and cutting once.
So I need a sub driver that will go low enough for real bass and high enough to smoothly transition to the OB panels. Got it.
I'll keep looking. It's too bad we can't make our own drivers to get the exact Qts needed for a particular design.
Well I did not know that. I will check it out.
BTW, I cut down the OB panels a little. My driver placement was causing about a 9db notch around 300Hz. Oops. I measured this with a RTA using pink noise. I then modeled the baffle using a piece of software called EDGE, and lo and behold, it pretty closely modeled the resposne I was measuring, though not quite as deep. I put the driver at IEC offsets and sized the panel at 2000 in^2 like you suggested previously. The bad notch turned in to a little wobble. Much better. The MathCAD worksheet is a little off I guess.
So I bit the bullet and trimmed the panels, reinstalled drivers, and measured again with RTA and pink noise and sure enough, the dip was gone. Computers are great, but it takes a little sweat and trial and error to really cut through the fog. I'll keep EDGE in my toolkit.
BTW, I cut down the OB panels a little. My driver placement was causing about a 9db notch around 300Hz. Oops. I measured this with a RTA using pink noise. I then modeled the baffle using a piece of software called EDGE, and lo and behold, it pretty closely modeled the resposne I was measuring, though not quite as deep. I put the driver at IEC offsets and sized the panel at 2000 in^2 like you suggested previously. The bad notch turned in to a little wobble. Much better. The MathCAD worksheet is a little off I guess.
So I bit the bullet and trimmed the panels, reinstalled drivers, and measured again with RTA and pink noise and sure enough, the dip was gone. Computers are great, but it takes a little sweat and trial and error to really cut through the fog. I'll keep EDGE in my toolkit.
Greets!
Oops! That should read: "After all, the no-brainer matching low Fs woofer is the 515-8C........" since the actual one would be the higher Fs 515-8G.
FYI, found this today while browsing for something else, though it's got to be several years out-of-date based on the Adire driver listings: http://www3.sympatico.ca/bonaz/IB-Sub/IBdrv-01.htm
Anyway, nice to know the software agrees with my experience.
GM
Oops! That should read: "After all, the no-brainer matching low Fs woofer is the 515-8C........" since the actual one would be the higher Fs 515-8G.
FYI, found this today while browsing for something else, though it's got to be several years out-of-date based on the Adire driver listings: http://www3.sympatico.ca/bonaz/IB-Sub/IBdrv-01.htm
Anyway, nice to know the software agrees with my experience.
GM
Thanks for the listing.
The Dayton 295-130 (15") driver you put me on to earlier looks somewhat promising. Have you used this driver yourself? I like the transient response of a closed box, but even in a vented box would the differences (exclusing low end roll off) be that audible?
I haven't heard back from GPA on the 515 woofer. That would work nicely. You mentioned optimum alignment in a BLH, so I take it the BIB would fall in that category? The BIB seems to have a pretty shallow flare rate.
I gather the cone mass of the Mach5 IXL-18 is too high (almost 1 pound) to produce the upper range I need. You mentioned:
"Regardless, this is a true sub driver with only a ~85.6 HF mass corner, so won't have the high SQ BW required to blend to a low Q OB, ergo a wide BW woofer system ideally is required."
How did you calculate the 85.6 Hz 'mass corner'? And mass corner I guess means the point at which the HF rolloff begins, -3db down? Also, does SQ mean 'system Q'?
OK, I think I'm hearing low Qts woofer driver (acoustically match 604 rolloff characteristics and enough bandwidth) in a BLH. Back to the alignment question of total system Q: .5? .56? .7? What is the ideal system Q for this application? It seems the .5 value would have a soother rollof but down real low, at some point don't we want some cliff edge drop? What is the generally accepted operating practice? Secondly, does MJK's BLH worksheet model this system correctly? It missed the open baffle config. What can I use to accurately design a BLH for a particular driver?
Thanks again for your feedback. I really am learning something.
The Dayton 295-130 (15") driver you put me on to earlier looks somewhat promising. Have you used this driver yourself? I like the transient response of a closed box, but even in a vented box would the differences (exclusing low end roll off) be that audible?
I haven't heard back from GPA on the 515 woofer. That would work nicely. You mentioned optimum alignment in a BLH, so I take it the BIB would fall in that category? The BIB seems to have a pretty shallow flare rate.
I gather the cone mass of the Mach5 IXL-18 is too high (almost 1 pound) to produce the upper range I need. You mentioned:
"Regardless, this is a true sub driver with only a ~85.6 HF mass corner, so won't have the high SQ BW required to blend to a low Q OB, ergo a wide BW woofer system ideally is required."
How did you calculate the 85.6 Hz 'mass corner'? And mass corner I guess means the point at which the HF rolloff begins, -3db down? Also, does SQ mean 'system Q'?
OK, I think I'm hearing low Qts woofer driver (acoustically match 604 rolloff characteristics and enough bandwidth) in a BLH. Back to the alignment question of total system Q: .5? .56? .7? What is the ideal system Q for this application? It seems the .5 value would have a soother rollof but down real low, at some point don't we want some cliff edge drop? What is the generally accepted operating practice? Secondly, does MJK's BLH worksheet model this system correctly? It missed the open baffle config. What can I use to accurately design a BLH for a particular driver?
Thanks again for your feedback. I really am learning something.
Hi Martin
It missed on driver placement on the OB, I think. Here is what happened:
I recently purchased access to your MathCAD worksheets and was using various models to explore the implementation of the GPA 604 driver. I decided to do an open baffle as it was the easiest and quickest to allow me to start listening to the speakers. I used your OB worksheet and came up with a design. I would be happy to send you PDF's of the worksheets.
After cutting the panels and mounting drivers, I started listening. Something wasn't quite right but I couldn't precisely diagnose it with my ear. I ran pink noise and measured with calibrated mic via RTA each panel individually and each had approx 9db notch around 280-300Hz. These panels are simple 13ply BB panels standing on the carpeted floor.
I have some freeware called EDGE and I used it to model what I had built and it indicated a large notch and overall response closely resembling what I was seeing with the RTA. I decided to alter the driver placement and in my case, had to cut the panel size down as well. The new panel is 49H x 40w and driver placement is 31.1" from bottom edge and 24.4 from left edge. This is a scaled down IEC standard baffle layout. Anyway, before cutting I ran new numbers via EDGE and notch turns in to a little wobble.
After cutting and remeasuring, the RTA response showed the notch was gone and there was just a little wobble in place. They sounded better when listening to some music as well.
Your worksheet is sensitive to driver placement as I used it initially and tweeked driver position to eliminate any bad peaking or notching. I remember that a small move of the driver placement (3-4") would create a notch in the response. The original panel was 60"H x 49"w and the driver was 32" from bottom and 26" from left side. Your worksheet indicated a very flat response before rollof using the original dimensions. The panel as built had the notch.
It is entirely possible I am doing something incorrectly, but I have checked and double checked and that is what is happening.
Does that help? I would be happy to provide whatever information you would like.
It missed on driver placement on the OB, I think. Here is what happened:
I recently purchased access to your MathCAD worksheets and was using various models to explore the implementation of the GPA 604 driver. I decided to do an open baffle as it was the easiest and quickest to allow me to start listening to the speakers. I used your OB worksheet and came up with a design. I would be happy to send you PDF's of the worksheets.
After cutting the panels and mounting drivers, I started listening. Something wasn't quite right but I couldn't precisely diagnose it with my ear. I ran pink noise and measured with calibrated mic via RTA each panel individually and each had approx 9db notch around 280-300Hz. These panels are simple 13ply BB panels standing on the carpeted floor.
I have some freeware called EDGE and I used it to model what I had built and it indicated a large notch and overall response closely resembling what I was seeing with the RTA. I decided to alter the driver placement and in my case, had to cut the panel size down as well. The new panel is 49H x 40w and driver placement is 31.1" from bottom edge and 24.4 from left edge. This is a scaled down IEC standard baffle layout. Anyway, before cutting I ran new numbers via EDGE and notch turns in to a little wobble.
After cutting and remeasuring, the RTA response showed the notch was gone and there was just a little wobble in place. They sounded better when listening to some music as well.
Your worksheet is sensitive to driver placement as I used it initially and tweeked driver position to eliminate any bad peaking or notching. I remember that a small move of the driver placement (3-4") would create a notch in the response. The original panel was 60"H x 49"w and the driver was 32" from bottom and 26" from left side. Your worksheet indicated a very flat response before rollof using the original dimensions. The panel as built had the notch.
It is entirely possible I am doing something incorrectly, but I have checked and double checked and that is what is happening.
Does that help? I would be happy to provide whatever information you would like.
I would be very interested in seeing your simulation results, the EDGE results, and the measured results if possible. A pdf of the MathCad worksheet would be great. If you e-mail them directly I will look at the problem this week. I am a little surprised since I used the EDGE to calibrate and double check my baffle modeling and it seemed to agree very closely.
Thanks in advance,
Thanks in advance,
cesout said:
Greets!
You're welcome!
No, just read the praises of a number of satisfied customers.
Depends on the alignment. Max flat sealed Vs max flat vented will sound different, with sealed sounding subjectively better to many discerning listeners. With low tunings, a vented alignment can have as good or better group delay/transient response than a 0.5 critically damped sealed in the BW that matters WRT our hearing acuity plus some extra acoustic efficiency to boot and why I prefer them. Short of a <10 Hz Fc compression horn, a multiple extremely low Fs, Qts, HE driver IB is the goal IMO, but few audio buffs have the requisite resources for either.
A 515 optimum BLH alignment has a near DC cut-off, so not a practical solution. A much higher Fs driver is required for such an alignment down to a practical Fc.
A BIB is anything but optimum from a technical POV since it's a too short linear expanding horn that is a high gain TL over much of its usable gain BW, so has a nasty 3rd harmonic dip to deal with.
Cone mass per se doesn't limit a driver's HF BW, its Le does. That said, it's a true sub driver, so I imagine it has a limited HF BW by design.
The mass corner is the beginning of the driver's flat BW, which is falling at 3 dB/octave and is calc'd: Fh = 2*Fs/Qts, though I see the driver's specs have changed again in the 18.4 version, so it's bumped up to ~97 Hz.
SQ = sound quality
sysQ = system Q
There are no vented Q alignments other than 0.707 max flat, but TLs can closely track the driver's IB response down to Fs. In theory the room gain is 2nd order, so a max flat sealed that sums ~flat with the room response is the theoretical ideal from some folk's POV, while a low Fs transient perfect 0.5 is the goal and the room response should be damped to suit. With such a large room though, damping it would be very costly and the room gain will be mostly non-existant except for its reverberant field, so I personally would go for 0.5 if sealed/TL or low tuned EBS (aka LLT) if vented and use as many drivers as required to get the desired low end acoustic efficiency. If you decide to do a large BLH, then it will be very highly damped with a Qtc somewhat < the driver's.
As horn programs go, MJK's is accurate enough within its limitations, but you'll need to use Akabak for more design flexibility, though lately DMcB has been adding some nice features to his Hornresp program.
GM
cesout said:
from GPA-
http://cgi.ebay.com/Altec-Lansing-5...ryZ50597QQssPageNameZWDVWQQrdZ1QQcmdZViewItem
http://cgi.ebay.com/Altec-Lansing-5...ryZ50597QQssPageNameZWDVWQQrdZ1QQcmdZViewItem
these might be the 2ed best thing-
http://cgi.ebay.com/Altec-Lansing-G...ryZ50597QQssPageNameZWDVWQQrdZ1QQcmdZViewItem
it's my understanding that a phone call ,
has better results than e-mail-
405.789.0221
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
862
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.