[Attila]

As I am still in the planning stages of my new speaker system I am playing around with different ideas. My starting point is that I will go with a line array design of some sort, it will be supplemented with subs (2 already built), and the x-over will be active with plenty of EQ'ing ability.

The question is what speaker design to go for. One idea that has some attractive sides is to build a shallow array 'on' the front wall, looking something like this:



or possibly with subs at the bottom only and the array extending closer towards the ceiling. Drivers in this version are 1 array of BG Neo8 and 2 arrays of Hivi B3S a side. Being a somewhat non-standard approach, I am more than usually uncertain of how well this would work however... Hence, I am hoping for some input on the idea, an in particular:

-What will the on-wall (20-25cm / 8-10") from frontwall to baffle) do to imaging? Will the lack of space lead to a 'flat' image, or...?

-Will the arrayed MTM configuration give a more controlled dispersion pattern, or would a 2-line (MT) array work as well in this respect?

-The screen has to be there, but will it be better to bring it forward, flush with speaker baffles, and fill the space behind it with damping/absorbing material?

Most of all: Is this an idea worth pursuing or time wasted?

Best,
Atle

[Rick Craig]

Have you considered the Fountek ribbons? I would spend the extra $ if you can budget for them. Your design looks good and I would leave the screen where it is. With the enclosures against the wall you will lose some beneficial rear ambience but with the way arrays tend to project an image the sound should still be very good. The symmetrical layout with small drivers will give you the same dispersion on both sides of the speaker and a very stable image.

[JoshK]

I'd be inclined to use the Founteks over the Neo8 as well.  I might also choose the Tangband 4" aluminum driver over the HiVi 3 from what has been written about it.  But either way, its going to be tough to get the 4" or 3" drivers to go low enough to cross over to subs I'd think.  That would be my big concern with this approach. 

[Attila]

Thanks for the feedback - good to hear that there is at the least some confidence in the concept! Much appreciated!

I have been (and still am) considering the Founteks, but the cost is a bit of an issue here. Going with the Founteks would cost about twice as much as the Neo8's, or at least 6-800$ more. Also, the very favorable response the LS6 and LS9 from GR has been receiving seems to indicate that the BG is a pretty good driver in an array. It also crosses lower, not that that necessarily matters much in this case. (Actually, as the c-c of the outer lines will be close to 20cm, crossing low may not be such a bad idea?) I suspect this may be a matter of taste and preference - should have had a listen to arrays based on both ribbon and planar to be able to find my preference in this matter, but that will probably not happen...

As for the 4-inch Tangbands that is a good suggestion, and if space was not an issue in the current room I would most likely have gone for 4 or even 5 - 5 1/2 inchers. Problem is there is just not enough room on the front wall to go above 35cm baffle width - the chimney is protruding about 15 cm from the front wall just to the left of the left front and I can't move the screen much more to the right. The alternative would be to go with 1 mid-line of 6,5" drivers, but that would mean leaving the MTM-style array... Am actually quite happy with the mid-range performance of the B3S from what I have heard so far - they really are very good for their price if treated right!

Given the number of drivers, I think getting the B3S down to around 100hz should work out. I have no problem EQing these quite a lot, and the sd of 40 of these is about 1200cm2, equalling 2 12-inchers a side. That's a lot more than I am running for the same passband now without any big problems. There is also the example of Roger Russell's IDS 25 with about half the sd EQing it flat down to 25hz - not something I would want to do, but it does indicate that you can push these little thing quite a bit if you want to!

Is there any point in angling the fronts for a toe-in BTW? Something like 10-15 degrees would bring the sweetspot pretty much on-axis, will this be an advantage if I am unable to (as per WAF restrictions...) dampen the 1st reflections on the side walls?

-Atle

[gitarretyp]

The LS-6/9 don't use the BG Neo 8. It's a special design that's not available publicly. The Alpha LS line arrays that Danny designed a few years ago used the Neo 8s.

[zobsky]

If you have two useable corners, you could consider something like a corner loaded TLAH, http://www.billfitzmaurice.com/TLAH.html . Corner loading should help with baffle step issues.

I'm not sure about what corner loading does to imaging, .. in any case I'm not sure that line arrays are pin-point imaging champs (if that's indeed what you want).

Lastly, Neo 8's (correct me if I'm mistaken) are more of mid-tweeters rather than tweeters. In other words, you may want to check if they extend high enough for your listening preferences. http://www.thorvinelectronics.net/sound/rps_neo8_specsheet.pdf . If not, ... a single supertweeter crossed over high can easily rectify matters

[Daygloworange]

Is there any point in angling the fronts for a toe-in BTW? Something like 10-15 degrees would bring the sweetspot pretty much on-axis, will this be an advantage if I am unable to (as per WAF restrictions...) dampen the 1st reflections on the side walls?

I would strongly recommend you do a full mock up in your room and tweak the angle of your arrays for toe in before committing to building the finished pair. We played around with placement of the LS-6's on Thursday night  during setup at RMAF, and there is a definite angle where the imaging just snaps right into place.

Also, the Neo8's are a fantastic tweeter. Very detailed, sharp, fast, and clean. They are effortless and transparent.

I have a pair of HiVi B3N's running in sealed test boxes in my home theater in a box setup running full range, and I quite like them. They image and soundstage quite nicely. For an 8$ driver, they are quite nice. 

Good luck with the project. 

Cheers

[Rick Craig]

Thanks for the feedback - good to hear that there is at the least some confidence in the concept! Much appreciated!

I have been (and still am) considering the Founteks, but the cost is a bit of an issue here. Going with the Founteks would cost about twice as much as the Neo8's, or at least 6-800$ more. Also, the very favorable response the LS6 and LS9 from GR has been receiving seems to indicate that the BG is a pretty good driver in an array. It also crosses lower, not that that necessarily matters much in this case. (Actually, as the c-c of the outer lines will be close to 20cm, crossing low may not be such a bad idea?) I suspect this may be a matter of taste and preference - should have had a listen to arrays based on both ribbon and planar to be able to find my preference in this matter, but that will probably not happen...

As for the 4-inch Tangbands that is a good suggestion, and if space was not an issue in the current room I would most likely have gone for 4 or even 5 - 5 1/2 inchers. Problem is there is just not enough room on the front wall to go above 35cm baffle width - the chimney is protruding about 15 cm from the front wall just to the left of the left front and I can't move the screen much more to the right. The alternative would be to go with 1 mid-line of 6,5" drivers, but that would mean leaving the MTM-style array... Am actually quite happy with the mid-range performance of the B3S from what I have heard so far - they really are very good for their price if treated right!

Given the number of drivers, I think getting the B3S down to around 100hz should work out. I have no problem EQing these quite a lot, and the sd of 40 of these is about 1200cm2, equalling 2 12-inchers a side. That's a lot more than I am running for the same passband now without any big problems. There is also the example of Roger Russell's IDS 25 with about half the sd EQing it flat down to 25hz - not something I would want to do, but it does indicate that you can push these little thing quite a bit if you want to!

Is there any point in angling the fronts for a toe-in BTW? Something like 10-15 degrees would bring the sweetspot pretty much on-axis, will this be an advantage if I am unable to (as per WAF restrictions...) dampen the 1st reflections on the side walls?

-Atle

The larger element of the BG will provide a lower crossover point but for your application you won't need that. The Neo8 has poor off-axis response in the top octave and the Fountek is much better in that regard. There may be some less expensive alternatives from HiVi but the physics of a wide radiating element are something you cannot correct with DSP. The Founteks also have better sensitivity so you can maximize the benefits of an array. The B3S should be fine down to 100hz and the active front end will give you plenty of control.

[Attila]

I would strongly recommend you do a full mock up in your room and tweak the angle of your arrays for toe in before committing to building the finished pair. We played around with placement of the LS-6's on Thursday night  during setup at RMAF, and there is a definite angle where the imaging just snaps right into place.

That's interesting. And also a bit worrying - I don't really want to build a full-scale mock-up to be quite honest, these are rather large. Any ideas as to what happens when it snaps into place? Can the optimal (or close to) angle be predicted in some way?

Also, the Neo8's are a fantastic tweeter. Very detailed, sharp, fast, and clean. They are effortless and transparent.

Sounds good! I have noticed that the opinion of these differ when it comes to their ability to run all they way up - it may be a matter of implementation or taste. As I said before, I wish I could listen to arrays with both type of drivers to judge for my self, but that is not very likely to happen.

Rick: I am not quite with you on why sensitivity is an issue with actively crossed setups? Isn't max SPL relative to distortion within passband a better parameter for judging the different drivers? Or are you thinking in terms of SPL per watt in an amplifier-limited setting? (tubes etc.)

Zobsky: Thanks for the links - hadn't seen Bill's pages before - interesting stuff! Corner-loading is not an option in the current room - there is a niche to one side and a door to the other. The super-tweeter is a consideration, but due to the differences in dispersion-pattern relative to the line I am less enthusiastic about that route in this application - it would also require a pretty potent (and expensive) tweeter to keep up with the lines.

One more question: How significant will the benefit of keeping the array symmetrical be? There is always the option of going with a single mid/woofer line, 9-12 6,5" or 8-9 8" drivers would be a lot less work to implement. They would get about 12-16 liters each if I went this route, should be enough for the right driver to attain some serious extension given a lot of power and a fair amount of EQ I should think. I could probably even loose the subs with enough real-estate in the woofer line. The question is if this would cost me some significant benefits from the symmetrical arrangement - anyone care to speculate on the differences between these two alternatives in the suggested application?

-Atle

[Daygloworange]

I would strongly recommend you do a full mock up in your room and tweak the angle of your arrays for toe in before committing to building the finished pair. We played around with placement of the LS-6's on Thursday night  during setup at RMAF, and there is a definite angle where the imaging just snaps right into place.

That's interesting. And also a bit worrying - I don't really want to build a full-scale mock-up to be quite honest, these are rather large. Any ideas as to what happens when it snaps into place? Can the optimal (or close to) angle be predicted in some way?

There is an optimal toe in for a given listening position. I don't know if everybody would agree on what angle that would be. But in any event, when you get it right, you'll know.

Honestly, I know you don't want to build a full mock up just for this, but, I'm afraid that you might jeopardize optimization of your imaging and soundstage if you don't.

With the scale of the project your are undertaking, I wouldn't look to cut corners in this area.

On the subject of tweeters, again, one could argue preferences all day long over ribbons vs planars, but I think the planars are a great choice. I think they have the edge in leading edge detail and transient response. YMMV.

Cheers

[Rick Craig]

I would strongly recommend you do a full mock up in your room and tweak the angle of your arrays for toe in before committing to building the finished pair. We played around with placement of the LS-6's on Thursday night  during setup at RMAF, and there is a definite angle where the imaging just snaps right into place.

That's interesting. And also a bit worrying - I don't really want to build a full-scale mock-up to be quite honest, these are rather large. Any ideas as to what happens when it snaps into place? Can the optimal (or close to) angle be predicted in some way?

Also, the Neo8's are a fantastic tweeter. Very detailed, sharp, fast, and clean. They are effortless and transparent.

Sounds good! I have noticed that the opinion of these differ when it comes to their ability to run all they way up - it may be a matter of implementation or taste. As I said before, I wish I could listen to arrays with both type of drivers to judge for my self, but that is not very likely to happen.

Rick: I am not quite with you on why sensitivity is an issue with actively crossed setups? Isn't max SPL relative to distortion within passband a better parameter for judging the different drivers? Or are you thinking in terms of SPL per watt in an amplifier-limited setting? (tubes etc.)

Zobsky: Thanks for the links - hadn't seen Bill's pages before - interesting stuff! Corner-loading is not an option in the current room - there is a niche to one side and a door to the other. The super-tweeter is a consideration, but due to the differences in dispersion-pattern relative to the line I am less enthusiastic about that route in this application - it would also require a pretty potent (and expensive) tweeter to keep up with the lines.

One more question: How significant will the benefit of keeping the array symmetrical be? There is always the option of going with a single mid/woofer line, 9-12 6,5" or 8-9 8" drivers would be a lot less work to implement. They would get about 12-16 liters each if I went this route, should be enough for the right driver to attain some serious extension given a lot of power and a fair amount of EQ I should think. I could probably even loose the subs with enough real-estate in the woofer line. The question is if this would cost me some significant benefits from the symmetrical arrangement - anyone care to speculate on the differences between these two alternatives in the suggested application?

-Atle

Not sure what you mean concerning the sensitivity issue with an active setup. I think that it's a great approach and the active electronics allow you to mix a wider variety of drivers and wiring configurations. This is true in particular with the BG drivers because their sensitivity is not high enough for many passive array systems to give a balanced response.

For your application the symmetrical layout with the Founteks is ideal because you won't have to worry about toe-in. With a single line of woofers there will always be a lobe to the left or right and if you build into the wall you won't be able to adjust the speaker angle.

[el`Ol]

If you bought a big BG, crossed at 200Hz and left the HiVis away that wouldn`t be much more expensive. If the SPL is not sufficient for you, you could ask for pro versions. The pro speaker shown on their site is an array, but I remember large single planars, maybe they are still in production. But let them show you the polar response plots, could be possible that they look like this:
http://www.sonus.de/download/pdf/datenblatt/ZX_101_DE_2007-07-05.pdf

[Attila]

For your application the symmetrical layout with the Founteks is ideal because you won't have to worry about toe-in. With a single line of woofers there will always be a lobe to the left or right and if you build into the wall you won't be able to adjust the speaker angle.

That is a very good point. Not that they will be built into the wall (it's a 40cm brick wall), but angling will be a bit cumbersome when they are up against (and secured to) the wall. I think that nails it as far as the single vs. double mid issue is concerned!

Rick, I noticed that you commented on a similar setup being in the works in the Selah forum - looking forward to hearing reports from that as it moves forward. Looked like the choice of mids (3" with shallow metal-alloy cones) will be quite similar as well - sound like you have decided on a driver?

This leaves the tweeter-issue. There are 3 different tweeter-options: (Al prices include taxes and estimated freight.)

1) Fountek Neo CD2M, whith 9 a side totalling about $2500. Probably best in terms of horizontal dispersion and top end. Can not cross as low as the Neo8, but not critical in this application. Most expensive by far, but a little bit shorter line with 8 would help somewhat.

2) BG Neo8, with 8 a side totalling about $1200. Can cross lower than the others, moving the x-over to around 1k gets it into a less sensitive area. Top-end disputed, but very low distortion. The modified version used by GR may be obtainable, and should improve on the top-end. Cheapest!

3) BG Neo3, with 16 a side totalling about $1500. Very good top end and better dispersion than Fountek. Can not cross as low as the 8, but clearly low enough for my use. Not very expensive. Chamber included. Good compromise between the two others?

I know that Rick favors the Fountek and Dayglow the Neo8 - anybody else want to chime in with an opinion on what way to go here?

el'Ol: I have been down the route of the RD-series BG's earlier and decided against them. They are very nice drivers, but the feedback I have been getting indicate that they have more of a problem with the top octave than the Neo-series. Also, I am using downward firing subs, and they do not sound good played much over 100 unfortunately.

Again - thanks for the great feedback!

-Atle

[Greggo]

Atle,

Like your project, and since I have very similar curiosities I will share some of my own observations here but qualify them with the fact that all I have done to date is listen and read regarding line array's.... no real experimentation or expertise....

1)  Given your diagram/rendering of the system, I would consider angling the side walls of the cabinet out for a smoother transition to the wall and I would also plan for maybe three variations from straight, to angled in slightly, to angled in strongly.  Build the baffles and wire them up and using some weather seal and quick cut plywood for the varrying sides, protect the baffles and maybe just clamp them in place to do a quick experiment regarding effects on imaging.  I agree with Rick that the system should have a very symetrical dispersion pattern, and I think it might even sound better flush with the wall (building a false wall in front of your bricks to act as baffle and cabinet...), but exactly what kind of sound stage emerges in a given room is still a tricky thing IMHO, so with a project of this caliber I would try to leave some options open.

2)  Looks like we are all thinking about the same driver.  When I read the test on Zaph's site and saw his project, and thought about the square frame, seemed right away like an easy start to line array project.

3)  But with 3 inch drivers do you really want to shoot for a low crossover (one under 1.5kHz), or do you push for a higher one and see what you get with the tight driver spacing?  One of the advantages of the Neo3 is that your c-t-c from one side of HiVi's to the other is ontly 5.5 inches, maybe 5 if you can cheat em in enough, and that may allow for a 3kHz or higher crossover.  I have raised this question before on other line array forums, with most folks sticking to the "lower is better" idea, but I haven't heard anyone say they took drivers like this or smaller and tried both approaches before making a decision.

4)  I hope to some day be in a position to do some of these experiments myself, if things sounded good with drivers paired up like that and running out past 3kHz, might be worth going to a manufacturer and having them build two small cone drivers on each side of a very thin planar that only had to do down to 4 or maybe 3 kHz or some point in between, and keep the horizontal c-t-c under 4.5 inches by mounting all three drivers to a commond, single flange.  One that is rectangular so that they can almost form the front baffle themselves, just screwing each corner of the faceplate into side rails and maybe a rubber gasket between and underneath them or something... anyways, if it could be done for $25 dollar a unit and they wanted a thousand minimum, I could probably put a group together to make it happen...

all just food for thought.  Again, a very cool project and I wish you the best of luck.

P.S. - having met Rick and listened to his arrays and point source designs, I would never pretend to be even half as qualified to comment on such things.  His stuff really rocks, and if I decide not to DIY, I would probably buy from him.

Regards,

Greggo
Asheville, NC

[Daygloworange]

I know that Rick favors the Fountek and Dayglow the Neo8 - anybody else want to chime in with an opinion on what way to go here?

Just for the record Attila, I have no experience with Fountek ribbons specifically. My descriptions regarding the Neo8's are based on my extensive listening to the Neo3's in my own speakers, and after spending a number of hours listening to the Neo 8's at RMAF.

I'm confident in how I described their performance, but not in any way trying to dissuade you from the Fountek's.

Cheers

[gitarretyp]

What speakers used the Neo 8s at RMAF?

[Rick Craig]

For your application the symmetrical layout with the Founteks is ideal because you won't have to worry about toe-in. With a single line of woofers there will always be a lobe to the left or right and if you build into the wall you won't be able to adjust the speaker angle.

That is a very good point. Not that they will be built into the wall (it's a 40cm brick wall), but angling will be a bit cumbersome when they are up against (and secured to) the wall. I think that nails it as far as the single vs. double mid issue is concerned!

Rick, I noticed that you commented on a similar setup being in the works in the Selah forum - looking forward to hearing reports from that as it moves forward. Looked like the choice of mids (3" with shallow metal-alloy cones) will be quite similar as well - sound like you have decided on a driver?

This leaves the tweeter-issue. There are 3 different tweeter-options: (Al prices include taxes and estimated freight.)

1) Fountek Neo CD2M, whith 9 a side totalling about $2500. Probably best in terms of horizontal dispersion and top end. Can not cross as low as the Neo8, but not critical in this application. Most expensive by far, but a little bit shorter line with 8 would help somewhat.

2) BG Neo8, with 8 a side totalling about $1200. Can cross lower than the others, moving the x-over to around 1k gets it into a less sensitive area. Top-end disputed, but very low distortion. The modified version used by GR may be obtainable, and should improve on the top-end. Cheapest!

3) BG Neo3, with 16 a side totalling about $1500. Very good top end and better dispersion than Fountek. Can not cross as low as the 8, but clearly low enough for my use. Not very expensive. Chamber included. Good compromise between the two others?

I know that Rick favors the Fountek and Dayglow the Neo8 - anybody else want to chime in with an opinion on what way to go here?

el'Ol: I have been down the route of the RD-series BG's earlier and decided against them. They are very nice drivers, but the feedback I have been getting indicate that they have more of a problem with the top octave than the Neo-series. Also, I am using downward firing subs, and they do not sound good played much over 100 unfortunately.

Again - thanks for the great feedback!

-Atle

There are two different versions of the BG's - the original and the PDR. The PDR version adds a resistive material over the left and right sides of the element to control the horizontal dispersion. This was done because the original version has poor horizontal off-axis response due to beaming as the frequency increases. I've tested both versions of the Neo3 and the PDR has much better top octave dispersion; however, it still doesn't perform as well as the Fountek.

One problem with the Neo design for arrays is that you'll have more comb filtering in the top octave than the Fountek. The reason is that the Neo's element has virtually no directivity control in the vertical plane other than the effects of the element length. This means that the interference patterns will be closer to that of a dome tweeter and this increases the nulls in the response (comb filtering). The Neo3 will also behave differently than the Neo8 in a line due to the difference in their radiating length.

Yes, I already have the drivers. Now I just need just need to find some extra time. 

[Rick Craig]

Atle,

Like your project, and since I have very similar curiosities I will share some of my own observations here but qualify them with the fact that all I have done to date is listen and read regarding line array's.... no real experimentation or expertise....

1)  Given your diagram/rendering of the system, I would consider angling the side walls of the cabinet out for a smoother transition to the wall and I would also plan for maybe three variations from straight, to angled in slightly, to angled in strongly.  Build the baffles and wire them up and using some weather seal and quick cut plywood for the varrying sides, protect the baffles and maybe just clamp them in place to do a quick experiment regarding effects on imaging.  I agree with Rick that the system should have a very symetrical dispersion pattern, and I think it might even sound better flush with the wall (building a false wall in front of your bricks to act as baffle and cabinet...), but exactly what kind of sound stage emerges in a given room is still a tricky thing IMHO, so with a project of this caliber I would try to leave some options open.

2)  Looks like we are all thinking about the same driver.  When I read the test on Zaph's site and saw his project, and thought about the square frame, seemed right away like an easy start to line array project.

3)  But with 3 inch drivers do you really want to shoot for a low crossover (one under 1.5kHz), or do you push for a higher one and see what you get with the tight driver spacing?  One of the advantages of the Neo3 is that your c-t-c from one side of HiVi's to the other is ontly 5.5 inches, maybe 5 if you can cheat em in enough, and that may allow for a 3kHz or higher crossover.  I have raised this question before on other line array forums, with most folks sticking to the "lower is better" idea, but I haven't heard anyone say they took drivers like this or smaller and tried both approaches before making a decision.

4)  I hope to some day be in a position to do some of these experiments myself, if things sounded good with drivers paired up like that and running out past 3kHz, might be worth going to a manufacturer and having them build two small cone drivers on each side of a very thin planar that only had to do down to 4 or maybe 3 kHz or some point in between, and keep the horizontal c-t-c under 4.5 inches by mounting all three drivers to a commond, single flange.  One that is rectangular so that they can almost form the front baffle themselves, just screwing each corner of the faceplate into side rails and maybe a rubber gasket between and underneath them or something... anyways, if it could be done for $25 dollar a unit and they wanted a thousand minimum, I could probably put a group together to make it happen...

all just food for thought.  Again, a very cool project and I wish you the best of luck.

P.S. - having met Rick and listened to his arrays and point source designs, I would never pretend to be even half as qualified to comment on such things.  His stuff really rocks, and if I decide not to DIY, I would probably buy from him.

Regards,

Greggo
Asheville, NC

Greg,

The problem is that the measurements don't always agree with the theory in terms of what crossover point you can use. This would apply to the line of woofers as well as the horizontal dispersion of two woofer lines. The C-T-C spacing is of course the most critical but the driver design also plays a role. The cone shape can vary from something very shallow (Aurasound,HiVi) to more of a straight sided cone that is deeper in profile or a curvelinear which falls in the middle of the two. I'll also have a kit available once I have finished this design.

[Attila]

1)  Given your diagram/rendering of the system, I would consider angling the side walls of the cabinet out for a smoother transition to the wall and I would also plan for maybe three variations from straight, to angled in slightly, to angled in strongly.  Build the baffles and wire them up and using some weather seal and quick cut plywood for the varrying sides, protect the baffles and maybe just clamp them in place to do a quick experiment regarding effects on imaging.  I agree with Rick that the system should have a very symetrical dispersion pattern, and I think it might even sound better flush with the wall (building a false wall in front of your bricks to act as baffle and cabinet...), but exactly what kind of sound stage emerges in a given room is still a tricky thing IMHO, so with a project of this caliber I would try to leave some options open.

Good ideas, but I am (as the sketch below will show) a bit limited in space in the current room. Angling the sides enough to make a difference will take to much space unfortunately, and a false wall will not go well with the W!

I am thinking of a solution where the speaker is secured to the wall on top in a way that allows it to be angled out quite a bit and fastened at that position. Combined with a mounting on a base larger than the speaker footprint and bolting to this I should get enough stability. A bit of decoupling between floor and base as well as some foam against the wall should take care of unwanted resonances. Have to work on this!

4)  I hope to some day be in a position to do some of these experiments myself, if things sounded good with drivers paired up like that and running out past 3kHz, might be worth going to a manufacturer and having them build two small cone drivers on each side of a very thin planar that only had to do down to 4 or maybe 3 kHz or some point in between, and keep the horizontal c-t-c under 4.5 inches by mounting all three drivers to a commond, single flange.  One that is rectangular so that they can almost form the front baffle themselves, just screwing each corner of the faceplate into side rails and maybe a rubber gasket between and underneath them or something... anyways, if it could be done for $25 dollar a unit and they wanted a thousand minimum, I could probably put a group together to make it happen...

Cool idea, and it would certainly make the baffle-building process a lot easier! Still, I am leaning towards the low (<1500) x-over also for other reasons than c-c. Our hearing is very sensitive in the 2,5 - 4,5 area, and if we can keep the x-over out of this region it can't hurt. The Fletcher-Munson curves indicate that the area around 1,2k is an optimum in this regard (That is unless you can push it down to under 500), but in general well under 2k should be a good idea.

all just food for thought.  Again, a very cool project and I wish you the best of luck.

Thanks!

Rick: Didn't mean to say the Neo3 has better dispersion than the Fountek - should have been 'better dispersion than the Neo8'... I have no doubts the Fountek is the better driver over 10k, what I am trying to figure out is if it is deserving of its price premium for me...

Your argument on increased comb-filtering with the Neo's is a good one. Hadn't actually considered that, but particularly for the Neo3 (which also has a lower radiating % than the Neo8 when arrayed) this is a potential issue. Hmmm...

...Suggest you make some time and build the prototype - some real experience with this setup would be good!

Dayglow: Sorry about that, not my intention to put words in your mouth. A bit sloppy in the wording I see, what I should have said was that I had a suspicion as to what way you were leaning...

Anyway, I made another sketch more in line with what this hopefully will look like:




The two boxes are the subs (already in use). I am thinking of using part of the bottom space for wiring (plenty of that..) and possibly class D amps, getting them out of the rack.

The room is a bit off a mess shape-wise and there are some seriously solid walls around parts of it. There is also a flight of stairs up in the back right corner. I plan to get some absorbing material on the two first-reflections on the side walls - a bit lively without. Seating is about 3,5m (10-11 feet) from the front wall (room about 5m front to back).

-A

[Rick Craig]

One of the HiVi planars might be an option that would fit into your budget. I may be testing one of them in the near future; if so, I'll let you know how it turns out.