ScottG said:
Most of the time I do listen at 80-90dB but when the mood takes me and I let it go the v2's can't handle that in a manner that I'd call low distortion - I can hear them struggling so god knows what they'd measure like - the absolute main problems is the bass range upto the low mids - look at the ATC distortion plot for 96dB(90dB/2m) and consider that I'm crossing at 400hz with STEEP filters and you'll see why its muddy when played really loud. It just wants to make me turn the whole thing down because its a strain. Moving the XO up higher between the drivers is a no-no too because it simply doesn't sound as 'correct' and that's no matter how much tweaking you do in the crossover.
I used to think "yeah but I won't listen at 100dB because its just daft". And that's largely true BUT when you do listen at those levels you want it to be every bit as good as when your cruising along at 80dB. Only a handful of speakers that I've heard will do massive dynamics at large SPL's and without a trace of harshness or audible disotrion from any of the frequency range. The Dali line arrays were one, Dynaudio Studio Monitor (think it was the M3 looking at the Dyn. Website), Large 3-way Genelec Studio Monitors installed at PJ Hifi's home cinema demo room(incredible sound) and that's about it. The rest are typical hifi chaff like the Peceive v2's 😀
ScottG said:
The MBLs sound good for certain music, but not so good on others. I think one of two things are involved.
1. Room and placement: These types of design really need to be in a very absorbing room and placed as far way from walls as practicle, normally like how most pillars would be placed if you needed additional two.
2. Driver limitations. The larger the bribrating mebrane, the more of it's own modes are going to exist, and the more difficult it is to control how the membrane flexes thoughout the music.
Thy/Scott
Line array suggestions for $6k please?
This will round out the idea's and I can start to close it down.
Line array suggestions for $6k please?
This will round out the idea's and I can start to close it down.
soongsc said:
1. not really
2. not really
3. not really, except tweeters because the market doesn't offer
alot of choices. Metal drivers would be the worse choice really.
4. a blind man with no hearing can design a better speaker
than commerical. lol ...
😎
ShinOBIWAN said:
Tweeter choice is the problem.
Planars:
Dayton PT2 - $25, this isn't a very good driver standalone, but
in a low impedance array design it will give any commerical line
array costing $20,000 a run for the money. This isn't good enough
for a person with a $6k budget who wants to make a $100,000
line array 😉
What about the $50 - $100 planars? I'm not convinced that those
will be a big upgrade from the $25 planars. Some of them are
just name brands with inflated price tags.
Pro planar - best choice IMO, but this is too much money and
it breaks your budget.
True Ribbon
Aurum Cantus [G series], Raven, Fountek would be candidates.
Fountek being the bang for buck brand.
AC G3(si) is $200 each {16 drivers - $3200, 20 drivers - $4000}
[This gives you $2k for midwoofers]
Raven 2 is $350 each. {16 drivers = $5600}
[Raven = overpriced, IMO]
Fountek NeoCd2.0M - $126 {16 drivers = $2016, 20 drivers = $2520, 24 drivers = $3024}
The Fountek gives you flexibility in design. You can make a monster
12 driver per tower line array and still have $3k budget for midwoofer choice. 12 Drivers is about a 78" tall line which is sweet
and the extra drivers allow you to squeeze out more performance.
2.3 ohm wiring and 12 drivers = calculated 101 dB sensitivity.
If 0.6 ohm wiring, calculated to 110dB sensitivity.
I would make an impedance toggle switch to select both options.
Midwoofers
With a $3k budget, I can throw a dart and not lose. What do you like in the 4" - 6" size?
For 6" size I wouldn't hesitate and get a PHL1340 driver.
I would custom order from my favorite vendor and tell them
to special order a 1341 with phase plug. The number 1 means
rear cone treatment, default is just the front of the cone.
I don't know the cost, but if you bought 24 pieces maybe you can
get them for $3k 🤔
The standard PHL1240 {16 ohm version of 1340 cost $126 here;}
http://www.zalytron.com/
The PHL frame design will offer a problematic install 🙂
The midwoofer line sensitivity is no issue as the tweeters
will be the limiter. You will wire the midwoofer array accordingly.
I haven't auditioned under 6" size mids so you'd have to
get some samples and pick a winner. Look at Zaphs tests,
the link I posted. The Seas Excel 5.5" looks interesting.
I bet I can use a simple Dayton RS for $30 and make an
outstanding array with 12 - 16 of them per tower.
Midwoofer choice is easier, the tweeter choice will give you
problems. I'd place more money into the tweeters over mids,
but there isn't much out there to choose from. I have no faith
in the BG brand. The Newform ribbons would be cool for a mellow
system, but not a system that cranks the tunes hard.
I don't know if you will gain much by buying a $300 Fountek
ribbon over a $126 one. The 3x cost mostly likely won't yield
3x performance rise, probably a small % increase. The bang
for buck isn't there.
Even if you chose a $126 Fountek, use 10 - 12 per tower
to boost performance, that will be a great system and I can't think
of any commercial loudspeaker that will beat it for under $50k cost.
These designs using Fountek have been done already in the
DIY world. Somewhere on my hard drive, I have some cyberspace
line array pics you can look at to get ideas. You have to email me for those.
Shin and everyone else, I've really enjoyed this thread!
Here are my 2cents on this subject.
There are many issues when considering a speaker of this magnitude. First, I will list goals that I would set if I was designing this project in no particular order.
1. Low distortion/high fidelity
2. Fantastic imaging
3. Full range (in room response to 15Hz)
4. Low spl and higher spl ability (sounds good soft and loud)
So, what factors need to be considered with each point??
What are some possible solutions?
1. Low dist.... High quality drivers
2. Fantastic imaging- Well designed, low difraction cabinet AND perfect placement in the listening room.
3. Full range... Tough to do and pull off number 2 above.
4. Awsome x-over design- I prefer passive designs
How can these points be addressed?? One of the biggest problems that I have experienced with BIG speakers has been either placement in the room to get the best imaging and low end(if the subs are integrated with the mids/highs) and then, if the subs are separate, getting the integration perfect between the subs and the mids/highs.
I personally really like the line array ideas that have been thrown around with the stipulation that the arrays need to go down to at least 50Hz and then crossed over to a pair of line source subs. Distributing the bass vertically in the room has many benefits when trying to tame room modes.....
So for $6k here it is!
B&G NEO 3 (17 per side = 51inch line)
32ea x $43.66=$1397.12
Peerless Nomex 832873 - 5.25" (10 per side)
20ea x $50.25 =1005
Peerless XLS 830500 12” XLS Subwoofer (3 per side in a vertical line array)
6ea x $191.70= $1150.20
1397.12+1005+1150.20=$3552.32 + 15% shipping
=$4085
You will prbably need a couple good amps for the subs..
+$1200
=$5285
Which leaves you with $715 for a bunch of high quality caps, coils resistors and wire. The great thing about a Line array is that you will need the minimmum of x-over components as there is only one x-over point.
Just a thought.....
Nate
Here are my 2cents on this subject.
There are many issues when considering a speaker of this magnitude. First, I will list goals that I would set if I was designing this project in no particular order.
1. Low distortion/high fidelity
2. Fantastic imaging
3. Full range (in room response to 15Hz)
4. Low spl and higher spl ability (sounds good soft and loud)
So, what factors need to be considered with each point??
What are some possible solutions?
1. Low dist.... High quality drivers
2. Fantastic imaging- Well designed, low difraction cabinet AND perfect placement in the listening room.
3. Full range... Tough to do and pull off number 2 above.
4. Awsome x-over design- I prefer passive designs
How can these points be addressed?? One of the biggest problems that I have experienced with BIG speakers has been either placement in the room to get the best imaging and low end(if the subs are integrated with the mids/highs) and then, if the subs are separate, getting the integration perfect between the subs and the mids/highs.
I personally really like the line array ideas that have been thrown around with the stipulation that the arrays need to go down to at least 50Hz and then crossed over to a pair of line source subs. Distributing the bass vertically in the room has many benefits when trying to tame room modes.....
So for $6k here it is!
B&G NEO 3 (17 per side = 51inch line)
32ea x $43.66=$1397.12
Peerless Nomex 832873 - 5.25" (10 per side)
20ea x $50.25 =1005
Peerless XLS 830500 12” XLS Subwoofer (3 per side in a vertical line array)
6ea x $191.70= $1150.20
1397.12+1005+1150.20=$3552.32 + 15% shipping
=$4085
You will prbably need a couple good amps for the subs..
+$1200
=$5285
Which leaves you with $715 for a bunch of high quality caps, coils resistors and wire. The great thing about a Line array is that you will need the minimmum of x-over components as there is only one x-over point.
Just a thought.....
Nate
Thanks fella's. especially Thy who has email me a TON of idea's.
I know the sound that Scott is proposing with the Supravox and G1 drivers - very special BUT only with more audiophile geared music.
I'm looking for a multi purpose design that does everything very well, nothing is average or bad and is a statement that is more true to my ideal's.
I think a line array is the only way to go after researching and general intuition based on limited listening experience. I know I won't be dissappointed with a line array, I maybe dissapointed with the supravox/G1 when it comes to high energy music, home theater and high SPL's.
Off to look at drivers, implementation and aestheics. Will post back shortly with a proposal to see if I'm heading in the right direction.
I know the sound that Scott is proposing with the Supravox and G1 drivers - very special BUT only with more audiophile geared music.
I'm looking for a multi purpose design that does everything very well, nothing is average or bad and is a statement that is more true to my ideal's.
I think a line array is the only way to go after researching and general intuition based on limited listening experience. I know I won't be dissappointed with a line array, I maybe dissapointed with the supravox/G1 when it comes to high energy music, home theater and high SPL's.
Off to look at drivers, implementation and aestheics. Will post back shortly with a proposal to see if I'm heading in the right direction.
Member
Joined 2003
To keep the creative juices flowing, here are two more approaches...
Conventional approach with commercial drivers:
8x8" Peerless Exclusive bass
75" BG planar mid
Line of commercial ribbon tweeters
Unconventional:
25mm DIY ribbon in a waveguide to establish directivity, flanked by two columns of 9x7" Peerless Exclusive, also to establish directivity and maintain symetrical off-axis response. Ribbon waveguide would be short radius throat, conical flare, larger radius flowing onto the baffle. Physical width of this approach would depend heavily on xo frequency and desired directivity. If used in a large room, dipole. Small room perhaps cardiod (though it may be to large to be practical in a small room).
I would go with separate subs and tall lines but, with your stated objective of integration, shorten the lines and add subs at the top and bottom of the arrays while keeping the center of the line at listening height. Build the subs as modules 'cause you know you'll want to change something later.
Conventional approach with commercial drivers:
8x8" Peerless Exclusive bass
75" BG planar mid
Line of commercial ribbon tweeters
Unconventional:
25mm DIY ribbon in a waveguide to establish directivity, flanked by two columns of 9x7" Peerless Exclusive, also to establish directivity and maintain symetrical off-axis response. Ribbon waveguide would be short radius throat, conical flare, larger radius flowing onto the baffle. Physical width of this approach would depend heavily on xo frequency and desired directivity. If used in a large room, dipole. Small room perhaps cardiod (though it may be to large to be practical in a small room).
I would go with separate subs and tall lines but, with your stated objective of integration, shorten the lines and add subs at the top and bottom of the arrays while keeping the center of the line at listening height. Build the subs as modules 'cause you know you'll want to change something later.
Nothing is going to be perfect for all rooms. You also might considering whether additional sub placements can help cancel out room modes or not.
Generally, less drivers that will give the necessary maximum SPL in the desired listening point is the better way to go. When you start connection more drivers either in parallel or in series, the slight differences in electrical characteristics will start interacting with each other unless you have an amp driving each single driver. To make things worse, more drivers acoustically interact with each other unless the drivers directivity is selected specifically to minimize these effects. I would recommend a good WMTMW Do very good from 20Hz~40KHz, and probably two more on the very low sub end if you really wanted to challenge the extremes.
If you want good dispersion of high frequency, check out the BIO Lab 5 concept. Some modification is necessary to get the right performance, but you can do a TT in the middle to avoid the edge diffraction that the Bio Lab has.
Generally, less drivers that will give the necessary maximum SPL in the desired listening point is the better way to go. When you start connection more drivers either in parallel or in series, the slight differences in electrical characteristics will start interacting with each other unless you have an amp driving each single driver. To make things worse, more drivers acoustically interact with each other unless the drivers directivity is selected specifically to minimize these effects. I would recommend a good WMTMW Do very good from 20Hz~40KHz, and probably two more on the very low sub end if you really wanted to challenge the extremes.
If you want good dispersion of high frequency, check out the BIO Lab 5 concept. Some modification is necessary to get the right performance, but you can do a TT in the middle to avoid the edge diffraction that the Bio Lab has.
Thy what about this:
Tweeter
I've been working through things and its way cheaper to import fountek from the states(please correct me if there's a cheap source in Europe)
The Neo Pro 5i is $305 which works out at £200 including import charges. Cheapest I could find in Europe is 375Euro or £255. An array of 8 per side would be £3200 at the cheapest price. Verrry expensive and that's all my budget gone.
On the flip side is the Neo JP2.0 which is a similar story to the above, US is cheaper at $126 or £83 inc. import. European price is £118 each. Total cost for 10 per side is £1650 - now that is well within budget.
I'm not sure just how much better the 5i is than the Neo JP2 its double the budget for an almost identical ine length. It seems that higer sensitivity, lower distortion and about half an octave more extension are the benefits.
Should I go with the 5i then I'll either have to drastically compromise my choice of mid/bass and the subs would have to be bought at a later date. I'm not willing to that so the 5i would therefor be a long term project whereby I have to collect and steadily build up the required number of drivers.
Consider that I want to crossover at 1.5Khz and the double premium on the 5i vs. the Neo JP2 for a comparable line length. Which is the more sound choice?
Also: Is the only difference between the JP2 and Neo JP2 the Mms? I'm guessing the Neo JP2 is better suited to a 1.5Khz XO point because of the higer durability of the ribbon element.
Tweeter
I've been working through things and its way cheaper to import fountek from the states(please correct me if there's a cheap source in Europe)
The Neo Pro 5i is $305 which works out at £200 including import charges. Cheapest I could find in Europe is 375Euro or £255. An array of 8 per side would be £3200 at the cheapest price. Verrry expensive and that's all my budget gone.
On the flip side is the Neo JP2.0 which is a similar story to the above, US is cheaper at $126 or £83 inc. import. European price is £118 each. Total cost for 10 per side is £1650 - now that is well within budget.
I'm not sure just how much better the 5i is than the Neo JP2 its double the budget for an almost identical ine length. It seems that higer sensitivity, lower distortion and about half an octave more extension are the benefits.
Should I go with the 5i then I'll either have to drastically compromise my choice of mid/bass and the subs would have to be bought at a later date. I'm not willing to that so the 5i would therefor be a long term project whereby I have to collect and steadily build up the required number of drivers.
Consider that I want to crossover at 1.5Khz and the double premium on the 5i vs. the Neo JP2 for a comparable line length. Which is the more sound choice?
Also: Is the only difference between the JP2 and Neo JP2 the Mms? I'm guessing the Neo JP2 is better suited to a 1.5Khz XO point because of the higer durability of the ribbon element.
To clarify. I think the Fountek JP series is aluminum element,
the newer NeoCD is composite ribbon element. If you blow
an element, you can roll your own out of aluminum, but
I don't know if it's compatible with the Neo series. I don't
know what the sonic difference might be, if any betweeen the two. I haven't heard any bad reviews from either series.
http://www.fountek.net/products.htm
Madisound has;
NeoCd2.0M 5" Ribbon Tweeter - Rectangular Metal Flange
- $126
http://www.madisound.com/cgi-bin/index.cgi?cart_id=1539839.2138&pid=1890
*and*
NeoCD2.0 5" Ribbon Tweeter
http://www.madisound.com/neocd2.0.html
- $118
I'm biased towards quality tweeters first, then finding a suitable
midwoofer second. Others place more priority on mids, then
tweeters last. IMO, the tweeter is the 'bottleneck'. You can make
an array of $20 midwoofers and get amazing performance, but
an array of $20 ribbons/planars won't get you there.
My budget line array has $25 tweeters and 50 cent midranges
and the tweeters do all the hard work playing 1700hz and above,
8th order. This steep slope and low crossover point pretty
much eliminates any issues with the midwoofers and any
sound gremlins manifest from the $25 tweeters because they
are doing the hard work -> high frequency reproduction.
The only question is whether or not the midwoofers you choose
can offer the bass you seek. Do some driver modeling to
check performance. Because my NSB mids were cheap, xmax
limited, bass limited, I resorted to a tuned enclosure to solve that
issue. All the bass is due to port tuning - works well, less stress
on the poor ole NSB's. LOL
I would be biased in choosing the NeoPro5i because the tweeter
line sensitivity calculates higher {is it audible?}, the ability to
crossover lower is the real bonus. This gives you flexibility to fine
tune the SQ later with your digital crossover.
So the question is, what midwoofer do you desire, make a list of
candidates that you will be happy with.
Read the reviews;
http://www.zaphaudio.com/5.5test/
http://www.zaphaudio.com/6.5test/
The idea of using a high end tweeter mated to a medium
grade midwoofer is a hard pill to swallow, but it can be done with great results. The problem here is that you will need to make
a smaller line length of eight tweeters per tower to reduce costs.
You might want to investigate the idea of using 12 $100 tweeters
to boost performance over the 8 $300 tweeters.
If I were in your shoes, I'd make two plans and then evaluate them in detail. Draw up a normal array with 8 expensive tweeters,
then another plan with 12 of the lower cost tweeter.
When you analyze Zaph's data, don't be dissuaded too much
by the performance data. That is single driver data, place 8 - 12
of in an array and the distortion at the same SPL as the single
driver will be uber lower.
the newer NeoCD is composite ribbon element. If you blow
an element, you can roll your own out of aluminum, but
I don't know if it's compatible with the Neo series. I don't
know what the sonic difference might be, if any betweeen the two. I haven't heard any bad reviews from either series.
http://www.fountek.net/products.htm
Madisound has;
NeoCd2.0M 5" Ribbon Tweeter - Rectangular Metal Flange
- $126
http://www.madisound.com/cgi-bin/index.cgi?cart_id=1539839.2138&pid=1890
*and*
NeoCD2.0 5" Ribbon Tweeter
http://www.madisound.com/neocd2.0.html
- $118
I'm biased towards quality tweeters first, then finding a suitable
midwoofer second. Others place more priority on mids, then
tweeters last. IMO, the tweeter is the 'bottleneck'. You can make
an array of $20 midwoofers and get amazing performance, but
an array of $20 ribbons/planars won't get you there.
My budget line array has $25 tweeters and 50 cent midranges
and the tweeters do all the hard work playing 1700hz and above,
8th order. This steep slope and low crossover point pretty
much eliminates any issues with the midwoofers and any
sound gremlins manifest from the $25 tweeters because they
are doing the hard work -> high frequency reproduction.
The only question is whether or not the midwoofers you choose
can offer the bass you seek. Do some driver modeling to
check performance. Because my NSB mids were cheap, xmax
limited, bass limited, I resorted to a tuned enclosure to solve that
issue. All the bass is due to port tuning - works well, less stress
on the poor ole NSB's. LOL
I would be biased in choosing the NeoPro5i because the tweeter
line sensitivity calculates higher {is it audible?}, the ability to
crossover lower is the real bonus. This gives you flexibility to fine
tune the SQ later with your digital crossover.
So the question is, what midwoofer do you desire, make a list of
candidates that you will be happy with.
Read the reviews;
http://www.zaphaudio.com/5.5test/
http://www.zaphaudio.com/6.5test/
The idea of using a high end tweeter mated to a medium
grade midwoofer is a hard pill to swallow, but it can be done with great results. The problem here is that you will need to make
a smaller line length of eight tweeters per tower to reduce costs.
You might want to investigate the idea of using 12 $100 tweeters
to boost performance over the 8 $300 tweeters.
If I were in your shoes, I'd make two plans and then evaluate them in detail. Draw up a normal array with 8 expensive tweeters,
then another plan with 12 of the lower cost tweeter.
When you analyze Zaph's data, don't be dissuaded too much
by the performance data. That is single driver data, place 8 - 12
of in an array and the distortion at the same SPL as the single
driver will be uber lower.
And to add to the slew of ideas... here's my pet idea for "when I have the money, the time, and the room for it" 🙄
Modular dipole line array, preferrably hanging from side wall of listening room. Modules have one 6.5" to 8" driver each, and one 8" ribbon. Caveat, it's a concept only, I have not built anything like it, but I do have experience with dipoles.
Why dipoles? Well, low coloration in bass and mids, and controlled directivity for less room interaction. Why line array? Not just because of imaging, room filling sound, low distortion, but (especially in the home theater context) less fall-off of SPL with distance, therefore less "off-center" imaging clues when sitting off the center axis. Why modular? to make the line easily adaptable to any room height. Why hanging off the side walls? With dipole figure 8 radiation pattern, you can safely place the drivers close to the side walls, which makes even a tall huge line array relatively unobtrusive. Use plexiglass for baffle optionnally (for transparent sound!
)
Tweeter: From reading many tests and opinions, it seems the BG Neo8 Planars would be a prime choice. Zaph for instance rated the Neo3, essentially the same technology in a shorter package, as the only planar / ribbon in his lineup that had decent distortion figures, and the 8" version has the significant advantage of a possible crossover point below 800 Hz. Additional advantage, can be run as dipole or with included rear cover. At USD 50 a piece they're cheap and even cover quite some length for the price. You could use a tall ribbon instead, but then you deprive yourself of modularity and of the potential for power tapering which seems recommended to avoid "bloom" of the image height at greater listening distances. But at the price, you can use as many 8" drivers as you need for a line with no spaces between tweeters, which is optimal.
Mid/bass: Here the choices are many. I'd think the ScanSpeak 18W8531 or 4531 (sliced paper) would be an excellent choice, or the 8543 (polyethylene) which I use in my dipoles. The sliced paper has been tested by Zaph, the 8543 by Mark K, both extensive tests suggest excellent data. I certainly can vouch for the 8543, extremely clean sound. Peerless exclusive would be cheaper, Seas metal cones would be another option. All these drivers, with suitable EQ / Linkwitz transform, should have enough volume displacement to dispense of subwoofers if used in an array of say, 12/side. In other words, it would be a 2-way.
Possible issues: dipoles need EQ, and line arrays need both power tapering and correction for SPL falloff with frequency. Measuring line arrays seems to be problematic, but I have never tried it so I can't tell how problematic. An immediate workaround would be to build and optimize one pair of modules at first using classic outdoors measurements, for FR / dipole effects etc, then adjust for the line array issues using the full line in-room.
Cost would fit, even using expensive drivers. If using the dimensions in my drawing you'd need 10-12 modules per side for a 9 ft ceiling room. The Scanspeaks cost 150-220 USD/piece, the BG cost 50. It's doable for the price. Seas standard line or Peerless are about 1/3 the price.
The drawing is of course just the bare minimum. THis could look very noce if using say, Perspex panels for the module baffles, visible stainless steel bolts to attach them to a rosewood frame.. chamfering of the modules ... etc. Not to mention that the overall frame can be made of irregular size to help with the acoustics (dipole).
i really wouldn't shy away fro the dipoles, it's not black art really, in fact it's the dramatically easiest way to get rid of box resonances and stiffness problems, without weight or bulk. The rolloff is predictable too. The only parameters that need to be determined experimentally are the start of the rolloff (of course you'd still have to add the rolloff of low Q drivers). You might need a notch filter too but that's common to boxed systems as well. And in a line array the volume displacement needed is not really an issue.
Anyway just my $0.02.
Modular dipole line array, preferrably hanging from side wall of listening room. Modules have one 6.5" to 8" driver each, and one 8" ribbon. Caveat, it's a concept only, I have not built anything like it, but I do have experience with dipoles.
Why dipoles? Well, low coloration in bass and mids, and controlled directivity for less room interaction. Why line array? Not just because of imaging, room filling sound, low distortion, but (especially in the home theater context) less fall-off of SPL with distance, therefore less "off-center" imaging clues when sitting off the center axis. Why modular? to make the line easily adaptable to any room height. Why hanging off the side walls? With dipole figure 8 radiation pattern, you can safely place the drivers close to the side walls, which makes even a tall huge line array relatively unobtrusive. Use plexiglass for baffle optionnally (for transparent sound!
)Tweeter: From reading many tests and opinions, it seems the BG Neo8 Planars would be a prime choice. Zaph for instance rated the Neo3, essentially the same technology in a shorter package, as the only planar / ribbon in his lineup that had decent distortion figures, and the 8" version has the significant advantage of a possible crossover point below 800 Hz. Additional advantage, can be run as dipole or with included rear cover. At USD 50 a piece they're cheap and even cover quite some length for the price. You could use a tall ribbon instead, but then you deprive yourself of modularity and of the potential for power tapering which seems recommended to avoid "bloom" of the image height at greater listening distances. But at the price, you can use as many 8" drivers as you need for a line with no spaces between tweeters, which is optimal.
Mid/bass: Here the choices are many. I'd think the ScanSpeak 18W8531 or 4531 (sliced paper) would be an excellent choice, or the 8543 (polyethylene) which I use in my dipoles. The sliced paper has been tested by Zaph, the 8543 by Mark K, both extensive tests suggest excellent data. I certainly can vouch for the 8543, extremely clean sound. Peerless exclusive would be cheaper, Seas metal cones would be another option. All these drivers, with suitable EQ / Linkwitz transform, should have enough volume displacement to dispense of subwoofers if used in an array of say, 12/side. In other words, it would be a 2-way.
Possible issues: dipoles need EQ, and line arrays need both power tapering and correction for SPL falloff with frequency. Measuring line arrays seems to be problematic, but I have never tried it so I can't tell how problematic. An immediate workaround would be to build and optimize one pair of modules at first using classic outdoors measurements, for FR / dipole effects etc, then adjust for the line array issues using the full line in-room.
Cost would fit, even using expensive drivers. If using the dimensions in my drawing you'd need 10-12 modules per side for a 9 ft ceiling room. The Scanspeaks cost 150-220 USD/piece, the BG cost 50. It's doable for the price. Seas standard line or Peerless are about 1/3 the price.
The drawing is of course just the bare minimum. THis could look very noce if using say, Perspex panels for the module baffles, visible stainless steel bolts to attach them to a rosewood frame.. chamfering of the modules ... etc. Not to mention that the overall frame can be made of irregular size to help with the acoustics (dipole).
i really wouldn't shy away fro the dipoles, it's not black art really, in fact it's the dramatically easiest way to get rid of box resonances and stiffness problems, without weight or bulk. The rolloff is predictable too. The only parameters that need to be determined experimentally are the start of the rolloff (of course you'd still have to add the rolloff of low Q drivers). You might need a notch filter too but that's common to boxed systems as well. And in a line array the volume displacement needed is not really an issue.
Anyway just my $0.02.
Attachments
BG Neo8 Planars
Line array distortion is low, you don't really need to focus on
single driver distortion figures as it's not as critical as other
issues.
One ugly demon shown here;
http://www.partsexpress.com/pdf/264-713.pdf
Second issue with BG Neos.
* 4 ohm drivers
* Neo3 - 90dB sensitivity
* Neo8 - 92dB sensitivity
8 - Neo8 in a 2 ohm array is about 95dB line sensitivity. Very
poor for a line array. You can do better.
18 - Neo3 in a 0.9 ohm array is about 97dB line sensitivity.
8- Fountek $100 ribbon, 7 ohm, 98dB, in an array with 3.5 ohm
wiring, you can get tweeter line sensitivity to 101dB. +6dB better
than Neo8, this is huge, very audible in terms of music dynamics.
Neo3 is probably a better choice over Neo8. Cost for 18?
$720. Cost for 8 Fountek? $1k
If you really want to extract high performance, wire those
8 Founteks for .9 ohm nominal and get a calculated line
sensitivity of 107dB.. That is what I'd do. This is the same
0.9 wiring as Neo3, but cost $300 more to execute, but
you get superior performance with +10dB more sensitivity,
perceived as twice as dynamic 🙂
Line array distortion is low, you don't really need to focus on
single driver distortion figures as it's not as critical as other
issues.
One ugly demon shown here;
http://www.partsexpress.com/pdf/264-713.pdf
Second issue with BG Neos.
* 4 ohm drivers
* Neo3 - 90dB sensitivity
* Neo8 - 92dB sensitivity
8 - Neo8 in a 2 ohm array is about 95dB line sensitivity. Very
poor for a line array. You can do better.
18 - Neo3 in a 0.9 ohm array is about 97dB line sensitivity.
8- Fountek $100 ribbon, 7 ohm, 98dB, in an array with 3.5 ohm
wiring, you can get tweeter line sensitivity to 101dB. +6dB better
than Neo8, this is huge, very audible in terms of music dynamics.
Neo3 is probably a better choice over Neo8. Cost for 18?
$720. Cost for 8 Fountek? $1k
If you really want to extract high performance, wire those
8 Founteks for .9 ohm nominal and get a calculated line
sensitivity of 107dB.. That is what I'd do. This is the same
0.9 wiring as Neo3, but cost $300 more to execute, but
you get superior performance with +10dB more sensitivity,
perceived as twice as dynamic 🙂
Think I'm going to have to go with either the G1 or the NeoPro 5i.
The JP2.0 looks too compromised to be of use at 1.5Khz.
Mid/bass looks to be Scanspeak sliced papers.
And integrate 4 Peerless XLS10 drivers per tower. These will be side firing covering below 40hz.
This is going to be a long term project (2 years+). There's no point compromising when this amount of time, effort and money is involved, build the very best and then there's nowhere else to go from there. Your at the top so another project is either sidewards or down 🙂
I also plan to make this modular. ie. one speaker will consist of two stacks with each containing 4xmidbass, 4xtweeters and 2 subs.
This way I can instantly get up and running with a single stack per side and then add the other two stacks to complete the arrays in the following year. I've also got the option to extend the design from the 2 module setup to 3 and thus increase line length from 1.5m to 2.3m should I feel the need if the 2 module setup doesn't completely satisfy.
I'm still working through the details but it looks like I'm ideally going to want to be crossing at 1.2Khz given the 1/2 wavelength rule between mid/bass drivers which is the point where they cease to act as a line source.
How does that sound?
The JP2.0 looks too compromised to be of use at 1.5Khz.
Mid/bass looks to be Scanspeak sliced papers.
And integrate 4 Peerless XLS10 drivers per tower. These will be side firing covering below 40hz.
This is going to be a long term project (2 years+). There's no point compromising when this amount of time, effort and money is involved, build the very best and then there's nowhere else to go from there. Your at the top so another project is either sidewards or down 🙂
I also plan to make this modular. ie. one speaker will consist of two stacks with each containing 4xmidbass, 4xtweeters and 2 subs.
This way I can instantly get up and running with a single stack per side and then add the other two stacks to complete the arrays in the following year. I've also got the option to extend the design from the 2 module setup to 3 and thus increase line length from 1.5m to 2.3m should I feel the need if the 2 module setup doesn't completely satisfy.
I'm still working through the details but it looks like I'm ideally going to want to be crossing at 1.2Khz given the 1/2 wavelength rule between mid/bass drivers which is the point where they cease to act as a line source.
How does that sound?
MBK: Great mind think alike!
I missed your post whilst typing the one above but it seems our opinion shares much of the same ground. 🙂
I missed your post whilst typing the one above but it seems our opinion shares much of the same ground. 🙂
I am still struggling to grasp the huge fuss about the line arrays. I admit they have advantages but the Fountek line array I heard was not the end all of loudspeakers IMO.
Could you please steer me to a commercially built line array that you are impressed by. I would like to hear it and hear what I may have been missing.
My favorite system to date was a set of Quad ESLs I heard in a high-end stereo shop. I was in utter awe of the imaging I heard. I have yet to hear another speaker come close.
With the money Shin is willing to spend, how come the ideas stopped about horn designs?
What are their flaws?
Could you please steer me to a commercially built line array that you are impressed by. I would like to hear it and hear what I may have been missing.
My favorite system to date was a set of Quad ESLs I heard in a high-end stereo shop. I was in utter awe of the imaging I heard. I have yet to hear another speaker come close.
With the money Shin is willing to spend, how come the ideas stopped about horn designs?
What are their flaws?
Think I'm going to have to go with either the G1 or the NeoPro 5i.
If this is a two year project, then save up the money and re-evalute the market in 2 years to see what new tweeter might emerge. Right now, G1 or NeoPro5i are great choices. If you are
leaning towards G1, then re-evalute SA pricing in 2 years and compare notes.
And integrate 4 Peerless XLS10 drivers per tower. These will be side firing covering below 50hz.
You might have issues with integrated subs. You can manifest
issues. It's better to just make two tower line array with
mids and tweeters, then another two tower bass array,
assumnig you even need one! Don't underestimate the power
of a midwoofer array - you will be shocked to hear amazing bass
from quality 4" - 6" midwoofers. You might even abandon the
midbass array and just add a single monster sub.
I'm still working through the details but it looks like I'm ideally going to want to be crossing at 1.2Khz given the 1/2 wavelength rule between mid/bass drivers which is the point where they cease to act as a line source.
Don't feel like you have to follow the rules, there is nothing written in stone on crossover points. You can read all the theory you want, but in the end you will do crossover frequency sweeps
and hear something you like and it may be a higher crossover point which won't really affect performance.
My DCX has the ability to store settings. I have 12 programs in there to accomodate other variables, listening mood, genre of music, song, bad recording, etc. Each program has a different tweak, it can be crossover frequency, can be gain, can be EQ,
can be delay, whatever. The point is, you don't have to follow
the crowd and build one loudspeaker that generates one sound.
With clever electronics, you can make your loudspeaker do more
sounds with a quick press of a button.
If this is a two year project, then save up the money and re-evalute the market in 2 years to see what new tweeter might emerge. Right now, G1 or NeoPro5i are great choices. If you are
leaning towards G1, then re-evalute SA pricing in 2 years and compare notes.
And integrate 4 Peerless XLS10 drivers per tower. These will be side firing covering below 50hz.
You might have issues with integrated subs. You can manifest
issues. It's better to just make two tower line array with
mids and tweeters, then another two tower bass array,
assumnig you even need one! Don't underestimate the power
of a midwoofer array - you will be shocked to hear amazing bass
from quality 4" - 6" midwoofers. You might even abandon the
midbass array and just add a single monster sub.
I'm still working through the details but it looks like I'm ideally going to want to be crossing at 1.2Khz given the 1/2 wavelength rule between mid/bass drivers which is the point where they cease to act as a line source.
Don't feel like you have to follow the rules, there is nothing written in stone on crossover points. You can read all the theory you want, but in the end you will do crossover frequency sweeps
and hear something you like and it may be a higher crossover point which won't really affect performance.
My DCX has the ability to store settings. I have 12 programs in there to accomodate other variables, listening mood, genre of music, song, bad recording, etc. Each program has a different tweak, it can be crossover frequency, can be gain, can be EQ,
can be delay, whatever. The point is, you don't have to follow
the crowd and build one loudspeaker that generates one sound.
With clever electronics, you can make your loudspeaker do more
sounds with a quick press of a button.
tf1216 said:
Will never consider a horn design because of practical issues in implementing a really good one ie. physical size, room size needed and fabrication complications.
There's also the major stumbling block of never having heard a decent horn setup, about the best was the large JBL K series hybrids and these weren't particularly good. The Peceive 2's sound far better IMO but just don't go as loud.
I am still struggling to grasp the huge fuss about the line arrays. I admit they have advantages but the Fountek line array I heard was not the end all of loudspeakers IMO.
You said "The line array I heard used the same Fountek ribbons you proposed and the 5" Dayton drivers"
Lets solve the mystery why the Fountek / Dayton line array
didn't impress you. See how many questions you can answer
to figure this out.
1. How many Founteks per tower ?
2. How many Dayton midwoofers per tower?
3. What impedance was the Fountek wire too?
4. What impedance was the midwoofers wire too?
5. Was there power tapering in the design?
6. What it an active or passive crossover?
7. What crossover frequencies used?
8. What crossover slope was used?
9. What amplification was used?
10. Were you allowed to fully 'crank up the system' to full potential or was the demo just set to 'normal' listening levels?
11. Estimate SPL listening level.
With the money Shin is willing to spend, how come the ideas stopped about horn designs? What are their flaws?
Horns are primarily chosen to gain high sensitivity because the
people using them fear solid state amplification and prefer 1 watt
tube amps
The down side is they have an interesting personality that some
people love, others hate. Personal preference. I love horns,
I have horns, but their SQ is harder to extract, not to mention
you require a high quality driver.
The line array simplifies the goal of getting audiophile SQ
with high SPL potential that horns give. In other words,
the line array will be easiest to please over the average horn
design.
The horns that I have, there is no way they will compete in SQ
vs. the budget line array using cheap drivers, but the horns
will produce alot of SPL that is great if I want to crank tunes
from 3 blocks away.
Subwoofer horn - You need a horn the size of you house to
make a real sub horn with acoustical gain. You can make
bass horns with gain at 40hz and above. The SQ of bass horns
isn't going to be considered audiophile in sound. It will sound
great as if you went to a dancing club to pick up chicks and
the music hits you hard in the chest Nothing bad,
just different.
You said "The line array I heard used the same Fountek ribbons you proposed and the 5" Dayton drivers"
Lets solve the mystery why the Fountek / Dayton line array
didn't impress you. See how many questions you can answer
to figure this out.
1. How many Founteks per tower ?
2. How many Dayton midwoofers per tower?
3. What impedance was the Fountek wire too?
4. What impedance was the midwoofers wire too?
5. Was there power tapering in the design?
6. What it an active or passive crossover?
7. What crossover frequencies used?
8. What crossover slope was used?
9. What amplification was used?
10. Were you allowed to fully 'crank up the system' to full potential or was the demo just set to 'normal' listening levels?
11. Estimate SPL listening level.
With the money Shin is willing to spend, how come the ideas stopped about horn designs? What are their flaws?
Horns are primarily chosen to gain high sensitivity because the
people using them fear solid state amplification and prefer 1 watt
tube amps
The down side is they have an interesting personality that some
people love, others hate. Personal preference. I love horns,
I have horns, but their SQ is harder to extract, not to mention
you require a high quality driver.
The line array simplifies the goal of getting audiophile SQ
with high SPL potential that horns give. In other words,
the line array will be easiest to please over the average horn
design.
The horns that I have, there is no way they will compete in SQ
vs. the budget line array using cheap drivers, but the horns
will produce alot of SPL that is great if I want to crank tunes
from 3 blocks away.
Subwoofer horn - You need a horn the size of you house to
make a real sub horn with acoustical gain. You can make
bass horns with gain at 40hz and above. The SQ of bass horns
isn't going to be considered audiophile in sound. It will sound
great as if you went to a dancing club to pick up chicks and
the music hits you hard in the chest
Nothing bad,just different.
- Status
- Not open for further replies.