I'm trying to design a linear array of woofers to mate with an electrostatic midrange/tweeter assembly (not yet constructed). I envision crossing over from the dynamic drivers to the ESLs relatively high--perhaps 500 Hz or so. I'm considering mounting the woofer array in an open baffle (OB) configuration so that I end up with dipolar radiation patterns throughout the operating range except for the bottom octave or two, where a pair of NHT 1259 woofers in sealed boxes will help pressurize the room a bit.
Given the SPL losses inherent in an open baffle system I figured I'd better use at least 8" woofers instead of smaller drivers. Also, my inclination is to use a larger number of less expensive woofers rather than investing in a small number of very expensive drivers capable of large excursion. My reasons for this decision are:
(1) I'd like the height of the woofer array to match the height of the ESL, which will probably be 5 or 6 feet tall. My goal is to minimize the discontinuities in radiation pattern at the crossover from woofer array to ESL.
(2) Given the excursion demands inherent in an OB woofer system I'd rather not make matters worse by asking a small number of woofers to move all the air. I don't have complete confidence that all manufacturers are being honest when reporting the amount of excursion their drivers are capable of while maintaining low nonlinear distortion. I looked at ZaphAudio for some independent measurements but couldn't find results for 8" woofers.
So, I'm looking for a bit of advice. First, I'm looking for suggestions for suitable woofers. If the woofers in the array are fairly closely spaced then I'll probably need somewhere between six and nine woofers per channel. That number, combined with my budget, means ScanSpeak is out. I was intrigued by the smoothness of the frequency response of the Peerless Nomex drivers but Madisound says they're discontinued and eighteen would break my bank anyway. I *might* be able to swing twelve but would have to wince while doing it. The Dayton Reference Series woofers are priced more palatably, but their response is certainly more ragged. Those high frequency break-up modes are pretty huge! I'm assuming their effect would be minimized by crossing over several octaves before they kick in, but even at lower frequency the Daytons are less smooth than the Peerless woofers. Then there's always the question of which important driver characteristics are not revealed by frequency response measurements. I should probably add that I'll be driving the woofer arrays with a 100 watt/channel amplifier and active crossovers.
The second bit of advice I'm seeking concerns open baffles versus sealed boxes. I'd love to hear from anyone who has tried mating vertical arrays of woofers to ESLs. I've seen one or two reports from happy designers of the OB type, but wondered if those with horror stories just haven't chimed in yet. Anybody with good or bad experiences trying to mate woofer arrays to ESLs please share your experience.
Thanks,
Few
P.S. I wasn't sure whether to post this here or in the "Loudspeakers" list; sorry if I should have made the other choice.
Given the SPL losses inherent in an open baffle system I figured I'd better use at least 8" woofers instead of smaller drivers. Also, my inclination is to use a larger number of less expensive woofers rather than investing in a small number of very expensive drivers capable of large excursion. My reasons for this decision are:
(1) I'd like the height of the woofer array to match the height of the ESL, which will probably be 5 or 6 feet tall. My goal is to minimize the discontinuities in radiation pattern at the crossover from woofer array to ESL.
(2) Given the excursion demands inherent in an OB woofer system I'd rather not make matters worse by asking a small number of woofers to move all the air. I don't have complete confidence that all manufacturers are being honest when reporting the amount of excursion their drivers are capable of while maintaining low nonlinear distortion. I looked at ZaphAudio for some independent measurements but couldn't find results for 8" woofers.
So, I'm looking for a bit of advice. First, I'm looking for suggestions for suitable woofers. If the woofers in the array are fairly closely spaced then I'll probably need somewhere between six and nine woofers per channel. That number, combined with my budget, means ScanSpeak is out. I was intrigued by the smoothness of the frequency response of the Peerless Nomex drivers but Madisound says they're discontinued and eighteen would break my bank anyway. I *might* be able to swing twelve but would have to wince while doing it. The Dayton Reference Series woofers are priced more palatably, but their response is certainly more ragged. Those high frequency break-up modes are pretty huge! I'm assuming their effect would be minimized by crossing over several octaves before they kick in, but even at lower frequency the Daytons are less smooth than the Peerless woofers. Then there's always the question of which important driver characteristics are not revealed by frequency response measurements. I should probably add that I'll be driving the woofer arrays with a 100 watt/channel amplifier and active crossovers.
The second bit of advice I'm seeking concerns open baffles versus sealed boxes. I'd love to hear from anyone who has tried mating vertical arrays of woofers to ESLs. I've seen one or two reports from happy designers of the OB type, but wondered if those with horror stories just haven't chimed in yet. Anybody with good or bad experiences trying to mate woofer arrays to ESLs please share your experience.
Thanks,
Few
P.S. I wasn't sure whether to post this here or in the "Loudspeakers" list; sorry if I should have made the other choice.
a couple of things. As for open baffles, I just finished assembling a test baffle using four monsoon planar 9 ribbon drivers I picked up at a surplus place and the results are astonishing. Even without filters, eq, or low end reinforcement the sound is amazing. My roommate is a professional sound guy doing the big theatre shows in toronto and he was impressed as well. You cant beat the sound of open baffles.
The difficulties come in particularly in the lower end where you get cancellations and some other idiosyncracies I dont fully understand. Look at www.linkwitzlab.com for extensive info, if you havn't already.
As for drivers, I've looked around and the Dayton 8" has the most excursion I could find without going into prohibitively expensive drivers. Its distortion measurements outperform the seas excel equivelents and its cheap. For the frequency range your going to use I dont think you have to worry about the cone break up. Frankly, I cant think why you would want to use anything else. If you find a better choice, please let us know.
You might want to look at my thread "inexpensive open baffles" at htguide. I got some good suggestions there. Check out AJs suggestion for an open baffle design. It might give you some ideas.
The difficulties come in particularly in the lower end where you get cancellations and some other idiosyncracies I dont fully understand. Look at www.linkwitzlab.com for extensive info, if you havn't already.
As for drivers, I've looked around and the Dayton 8" has the most excursion I could find without going into prohibitively expensive drivers. Its distortion measurements outperform the seas excel equivelents and its cheap. For the frequency range your going to use I dont think you have to worry about the cone break up. Frankly, I cant think why you would want to use anything else. If you find a better choice, please let us know.
You might want to look at my thread "inexpensive open baffles" at htguide. I got some good suggestions there. Check out AJs suggestion for an open baffle design. It might give you some ideas.
I suspect you havn't seen this test of the dayton along with many other well known woofers:
http://206.13.113.199/ncdiyaudio/mark/Testing/Woofer2/8 inch woofer test.htm
Also, read Thomas' thread on mating magnawhatever esl panels to some kind of woofer at www.htguide.com . He's one of the gurus over there and has professed his love for esl's.
http://206.13.113.199/ncdiyaudio/mark/Testing/Woofer2/8 inch woofer test.htm
Also, read Thomas' thread on mating magnawhatever esl panels to some kind of woofer at www.htguide.com . He's one of the gurus over there and has professed his love for esl's.
The bass modules on the Martin Logan Statement 2 are sixteen 12" bipolar sealed woofers.
The midrange dipole line array uses eight 7" drivers per side crossed at 150Hz
10"
http://www.partsexpress.com/pe/pshowdetl.cfm?&DID=7&Partnumber=295-368
8"
http://www.partsexpress.com/pe/showdetl.cfm?&DID=7&Partnumber=297-310
The midrange dipole line array uses eight 7" drivers per side crossed at 150Hz
10"
http://www.partsexpress.com/pe/pshowdetl.cfm?&DID=7&Partnumber=295-368
8"
http://www.partsexpress.com/pe/showdetl.cfm?&DID=7&Partnumber=297-310
Attachments
You won`t have beeming problems with a 10" driver if x-over is below 800Hz.
http://www.teleprodottistore.it/shop/ciare/cw250.pdf
http://www.teleprodottistore.it/shop/ciare/cw250.pdf
Hi,
currently we are designing a comparable system for a customer.
1. ESL-panels run from 200 Hz
2. adjecent to the panel is an array of 10 woofers (Tangband W69-1042, oval shape)). Ten of those mounted vertically on top of each other results in a height of about 7'.
3. The width of the open baffle for the woofers is 1,1', which is rather small.
4. We do not extend the baffle at the rear, but limiting it just to the depth of the woofers magnet, in order to avoid any cavity resonance, which we found always audible, even when electrically damped.
5. The W69 corresponds to an 8'' woofer and in combination with 7mm excursion 10 pieces each side move a lot of air.
6. We measured distortion at 95dB sound level and are astonished about the very low K3 and K2 of this array.
7. Building such an array requires just modest equalization. We established +5 dB at 35 Hz with a Q=4, whcih results in flat response down to 32 Hz. trying to go lower makes no sense, since nornmal living room sizes will not provide the required room modes.
What shall i say, i think this is one of the best performing subsystems, we ever build. 20 tangbands are affordable and in comparison to other woofers we tried, it makes no noise even athigh excursions.
If it is setup in the room correctly, the result is breathtaking.
Go on with your idea !! The result is a perfect cylindrical wave radiator with perfect homogenity between esl-panel and woofer.
Regards, Capaciti
currently we are designing a comparable system for a customer.
1. ESL-panels run from 200 Hz
2. adjecent to the panel is an array of 10 woofers (Tangband W69-1042, oval shape)). Ten of those mounted vertically on top of each other results in a height of about 7'.
3. The width of the open baffle for the woofers is 1,1', which is rather small.
4. We do not extend the baffle at the rear, but limiting it just to the depth of the woofers magnet, in order to avoid any cavity resonance, which we found always audible, even when electrically damped.
5. The W69 corresponds to an 8'' woofer and in combination with 7mm excursion 10 pieces each side move a lot of air.
6. We measured distortion at 95dB sound level and are astonished about the very low K3 and K2 of this array.
7. Building such an array requires just modest equalization. We established +5 dB at 35 Hz with a Q=4, whcih results in flat response down to 32 Hz. trying to go lower makes no sense, since nornmal living room sizes will not provide the required room modes.
What shall i say, i think this is one of the best performing subsystems, we ever build. 20 tangbands are affordable and in comparison to other woofers we tried, it makes no noise even athigh excursions.
If it is setup in the room correctly, the result is breathtaking.
Go on with your idea !! The result is a perfect cylindrical wave radiator with perfect homogenity between esl-panel and woofer.
Regards, Capaciti
kifeinthesink said:
I heard a p21 line array crossed to an ESL and thought the paper cone integrated very well. Martin Logan also used paper cone midbass on its Statement2. On a modest width baffle, six 8" woofers can exceed 115db SPL @1m at 100 Hz with 2mm Xmax, which would work in a 3-way. For a 2-way, a line array of a few 10" woofers might save the need for a sub. That was my reasoning. The Martin Loagn Statement 1 was a 2-way with large dipole ESL and monopole line array of 12" woofers.
The Peerless Exclusive 8" woofer 830884 Nomex fiber cone also sounds like it would integrate well with an ESL
Attachments
Thanks, everyone, for very helpful and encouraging replies. I remain interested in the array of dipole woofers idea, although I can't stop mentally multiplying the cost per woofer and the number of woofers. Even relatively inexpensive drivers sure add up if you use sixteen of them! One of the pleasures of the diy audio game is getting lots of bang from fewer bucks---arrays of woofers are tough to fit into that formula. I guess I'll try to take some solace from the reduced cost and complexity of the cabinet (although mdf isn't *that* expensive).
Here's a question that comes to mind frequently when choosing suitable woofers. I realize I'm risking triggering a firestorm by asking it, but here it goes anyway: If woofers are operated only at frequencies at which the cones behave pistonically, what difference does it make what the cones are made of? I'm a firm dis-believer in the meaningfulness of "fast bass" (that alone may ruffle some feathers) so that explanation of the need for exotic cone materials hasn't yet convinced me. I guess the real question is just how pistonically the various (paper, polypropylene, aluminum, carbon, Nomex,...) 8" woofer cones behave below, say, 1 KHz. If two woofers could have cones made of different materials, but be otherwise identical, wouldn't the woofers sound identical when operated within a frequency range where both cones behave pistonically? (By stipulating otherwise identical drivers I'm trying to keep motor and suspension non-linearities out of the question.) If the answer is yes, then the obvious question is how pistonically 8" woofers behave below ~1 KHz? They don't all exhibit identical frequency responses, even in the several hundred Hz range, so perhaps that demonstrates that the cones break up even at fairly low frequencies---or are those frequency response glitches caused by something else?
I have to admit I prefer the driver-to-driver reproducibility that I (justifiably?) assume comes from using materials other than paper, and my geek side probably just prefers "cooler" materials. But do metal cones necessarily outperform polypropylene (or paper) at 500 Hz, assuming both cones are competently designed? Perhaps this is just excessive philosophizing as I try to work up the courage to pay for twelve or sixteen woofers....
Few
Here's a question that comes to mind frequently when choosing suitable woofers. I realize I'm risking triggering a firestorm by asking it, but here it goes anyway: If woofers are operated only at frequencies at which the cones behave pistonically, what difference does it make what the cones are made of? I'm a firm dis-believer in the meaningfulness of "fast bass" (that alone may ruffle some feathers) so that explanation of the need for exotic cone materials hasn't yet convinced me. I guess the real question is just how pistonically the various (paper, polypropylene, aluminum, carbon, Nomex,...) 8" woofer cones behave below, say, 1 KHz. If two woofers could have cones made of different materials, but be otherwise identical, wouldn't the woofers sound identical when operated within a frequency range where both cones behave pistonically? (By stipulating otherwise identical drivers I'm trying to keep motor and suspension non-linearities out of the question.) If the answer is yes, then the obvious question is how pistonically 8" woofers behave below ~1 KHz? They don't all exhibit identical frequency responses, even in the several hundred Hz range, so perhaps that demonstrates that the cones break up even at fairly low frequencies---or are those frequency response glitches caused by something else?
I have to admit I prefer the driver-to-driver reproducibility that I (justifiably?) assume comes from using materials other than paper, and my geek side probably just prefers "cooler" materials. But do metal cones necessarily outperform polypropylene (or paper) at 500 Hz, assuming both cones are competently designed? Perhaps this is just excessive philosophizing as I try to work up the courage to pay for twelve or sixteen woofers....
Few
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.