New Home for the BG Ribbons - Part 27

[chasw98]

Here is an initial Room EQ Wizard sweep fom 20 to 20,000 Hz. 6 dB boost at 61 Hz with a 12 dB/octave shelving filter, low to mid at 400 Hz LR12, mid to high at 3,000 Hz LR12. 1/3 octave smoothing. This is 3 way active using a DCX2496. Lots of promise here.



5 db increments

and logrithmic

[JonMarsh]

Here is an initial Room EQ Wizard sweep fom 20 to 20,000 Hz. 6 dB boost at 61 Hz with a 12 dB/octave shelving filter, low to mid at 400 Hz LR12, mid to high at 3,000 Hz LR12. 1/3 octave smoothing. This is 3 way active using a DCX2496. Lots of promise here.



5 db increments

and logrithmic

Hows' that look with more conventional logarithmic scaling for frequency response? Does REW support that?

[TacoD]

Hows' that look with more conventional logarithmic scaling for frequency response? Does REW support that?

I am not the only one wondering why a linear scale was used .

[chasw98]

I am not the only one wondering why a linear scale was used .

Add me to the list. It is very hot today....... I just changed it, sorry.

[TacoD]

Add me to the list. It is very hot today....... I just changed it, sorry.

Don't be . It looks promising. The lower mid (200 - 400 Hz) is recessed (you already boosted the low frequencies). Can you apply some EQ to this region (I am not familiar with these electronic filters).

[chasw98]

Here is another graph using LspCAD taken at the listening position. 400 Hz LR24, 3,000 Hz LR24, 1/3 octave smoothing. A filter has been applied with +6 dB gain at 1,250 Hz, and a Q of 2. Discount anything below 200 Hz.



Taco - The lower mid response was changed by moving the array further out from the wall!

[---k---]

Chuck,
At least now I know what the proper attire is around your place. That will save me packing an extra shirt.

[chasw98]

ALL PICTURES HAVE BEEN CHECKED FOR IMPROPER REFLECTIONS!

The arrays assembled and operating.....

[CraigJ]

....and how do they sound? Gotta sound fantastic after you remove what-ever is under the blue tarp!

Craig

[augerpro]

Cool stuff Chuck!

So have you taken some off axis measurements? Reason I ask is that you mention adding some boost around 1250hz for what I assume is the main dipole null after the peak. Boosting nulls is not usually advised and I wonder if in this circumstance it fills in as you move off axis anyway? Just a thought...

How does the response look with say a 500ms gate? Reading Toole's last book it also occured to me to look at an 80ms gate as this jives a bit better with how the ear works. Haven't done it yet though. This would be with the speakers in the intended position and mic at the listening area of course.

BTW are you running any of the planars open back?

[ThomasW]

BTW are you running any of the planars open back?

The Neo8's are dipole.

[chasw98]

....and how do they sound? Gotta sound fantastic after you remove what-ever is under the blue tarp!

Craig

Right now they roughly sound great. I measured on axis FR and went through a series of crossover points. (They are currently setup using a DCX2496 and 6 channels of amp).I went from 300 to 800 Hz on the low/mid point in 50 Hz increments, measuring and listening, on and on. 400 Hz seemed best to me as a good starting point. I did the same thing with the mid/hi point starting at 2,000 Hz going up to 4,000 Hz in 500 Hz increments. 3,000 Hz came out as the sweet spot. I have now built a 4th order 3 Khz LR passive xover to stick in one channel of the array and compare it to the active channel. I will not be able to get measurements until the weekend when I have time to setup but it (the passive xover) sounds promising.

The blue tarp has christmas stuff, TV sets, boxes, hurricane supplies, etc. The grand plan is to move it all out to a hurricane approved shed in the back yard and close in the porch! I have a contractor arm wrestling with the city fathers over permits right now. Turns out I can call it a storage area and close it in without having to pour a new concrete floor or put electrical in the outside wall. Or I can call it a spare room and I have to pour new footers, concrete floor, electrical along the outside wall, etc. It is going to be the nicest storage area you have ever seen!

So have you taken some off axis measurements? Reason I ask is that you mention adding some boost around 1250hz for what I assume is the main dipole null after the peak. Boosting nulls is not usually advised and I wonder if in this circumstance it fills in as you move off axis anyway? Just a thought...

How does the response look with say a 500ms gate? Reading Toole's last book it also occured to me to look at an 80ms gate as this jives a bit better with how the ear works. Haven't done it yet though. This would be with the speakers in the intended position and mic at the listening area of course.

I have not gotten to the off axis measurments at all yet. I am still at the starting gate figuring out the best way to measure these and then trying to get some individual driver measurements so I can use software to help me get some components together. It is hard to get a decent measurement when an individual driver consists of 16 elements all lined up! I have been taking measurements of an individual panel with the mic at a listening position of 250 cm at ear height. I have only set a 10 ms window so far. Further investigation will be done this coming weekend. I didn't have enough time to get anything going but some rough measurements and get them running last weekend. Then Tom calls and gives me some more suggestions of things to build onto the panels. Turns out he and Jon have had this idea (for an array) running around their heads for a long, long time and now someone is putting the pen to the paper so to speak. I feel like Mikey in the cereal ads! OTOH, they have thought through a lot of the design and thrown out the bad parts so I am starting ahead of the game even though one of these has never been built.

The Neo8's are dipole.

He's right. The Neo 8's are dipole and I have a spare Neo 3 that I am going to cut the back off of and see what that sounds like. So I might possibly run the Neo 3's dipole also!

[---k---]

So Chuck and his lovely wife Debbie hosted me for dinner and some listening. The food was great. Anyone in the area should swing by and take them up on their hospitality.

I was looking forward to getting to hear some real line arrays and planers. Never heard either before.

I was impressed. They aren't done yet, and they need some work. Chuck knows this, so it isn't fair to give a full review. But some things that struck me was how big of a wall of sound it was. It was very enveloping - and this is with the back wall an open porch screen. I probably turn mine up a bit more to get the same envelopment. I was somewhat expecting them to have no imaging, based upon what I've heard in the past. I was wrong. It wasn't pin-point with huge depth, but I think that will come with additional work on the crossover. The bass was very good. Tight. I think this is what Chuck is happy with and won't be changing. The great bass with the huge sound stage made a very realistic presentation. Probably more realistic, like you're actually in the room, than I've had before. Chuck kept saying the sound was effortless - yep. Dynamics were naturally big too.

I also got to listen (and see) to Chuck's home theater system. Very nice also. (And in an air conditioned room). Chuck is one lucky guy with all his toys and a wife that lets him have them.

[ssabripo]

dayum, nice work chuck!

[chasw98]

Thank you Ryan and Sherv for the kind words. Just in case you were wondering if I was slacking off, Hank, I have been learning a new piece of software that will allow me to measure the array properly. Now it was a pain in the ... because it runs on a Mac and I don't have one at home! So I grabbed an old G4 from work (under the lend/lease program) and finally got it to load OSX 10.5 and installed Fuzzmeasure along with the M-Audio drivers. It is working and stable now barely. So now I take the measurements in Fuzzmeasure and export them out as FRD, then bring them back in to LspCAD where I can mix and match components to get a decent crossover start and then go back to Fuzz to measure the results and do it all over again. Whew! But it seems to be working just fine and I am starting to see the results of using Fuzzmeasure for getting raw data. If you have a Mac, it is well worth the investment. Here are some pics of the 'measurement rig' as it is for the moment.

These picture were taken at 4:30 PM as a rainstorm was starting up. You can see how dark it got with the clouds. It was tough to get a measurement today between rainclouds and thunder making noise or yard appliances making noise.

[JonMarsh]

Yeah, Jon's been beating up on Chuck, but I think he's beginning to see the light a bit, as soon as those clouds pass!

That is a real antique you've got there; my PPC tower is a G5. :W At least you're not trying to do it on a G3!

Hmmm, I recognize that file!

[chasw98]

That is a real antique you've got there; my PPC tower is a G5. :W At least you're not trying to do it on a G3!

Hmmm, I recognize that file!

Hey now, it has a 1.2 Ghz uprade processor and 2 gb ram! You still wait about 2 minutes for it to process a 400 ms response, but it works!

Let's see if you recognize this file.



This is my first attempt at measuring according the help I have been given by the Dark Lord. He is looking kindly these days. It is of the Neo8's in the array at 100 cm and a 400 ms window.

[---k---]

Jon,
I've read Jon Atkinson uses Fuzzmeasure also. What are the advantages of it that the other tools people use to measure speakers?

[JonMarsh]

Easy to setup and use (faster & nice UI than Praxis), any measurement can be used for both SPL and distortion, can plot overlays very quickly of multiple measurements, can average measurements easily, uses sine sweep chirps like Praxis, can generate minimum phase copies of data in the same file, can store a whole series of measurements in one file, has plug-ins for impedance, reverberation time analysis, CSD waterfalls, clipping latched on level meter, synchronous averaging for measurements in a noisy environment, includes device correction option, modes for room analysis, and best of all, runs on a Mac. :W

Of course, it can generate FRD files, import or export impulse responses, export a graph as an image, export a graph as a CSV data file, export a stimulus signal to use for remote stimulus and capture for a field recording without Fuzzmeasure connected for input (Make a CD recording to use in a Car for the stimulus signal for measurement in a car cabin), use microphone calibration file, can use multiple cal files applied to multiple microphones when doing more complex room measurements, and has the nicest graphing tools I've ever used in a measurement application (but I've only used SoundEasy, CLIO, and Praxis- no experience with LMS or some of the other more expensive packages.

Oh, and it's pretty inexpensive- $150 is something of a bargain in my opinion. Did I mention it runs on Mac's? :W

I'm not claiming it's "better"; I will say I prefer it to the other tools I have at my disposal. I find myself more productive getting things done, and spending much less tine trying to remember how to use the tool, or how to get something done. This program does have a very good UI design, even as Mac programs go- winning an Apple Design Award in 2006 in the scientific/computing category.

[cjd]

Downside to Fuzzmeasure: you've gotta pay through the nose for hardware - first a Mac (a Mini would be plenty), and then an external soundcard of adequate quality since you can't upgrade the garbage Intel ships on their chipset.

What sound input device are you using these days Jon? Still the M-Audio? or something else?

[JonMarsh]

Downside to Fuzzmeasure: you've gotta pay through the nose for hardware - first a Mac (a Mini would be plenty), and then an external soundcard of adequate quality since you can't upgrade the garbage Intel ships on their chipset.

What sound input device are you using these days Jon? Still the M-Audio? or something else?

well, for my Praxis setup to work well I had to buy an internal M-Audio 24/192 for a couple hundred dollars, so there's no difference there. The Macbooks and Mac Mini use an internal Realtek setup that's not terrible, but not good either. With Fuzzmeasure's loop back you can compensate for the lack of flatness, but not the lack of bandwidth. :W

Computing hardware costs for the same performance and hardware specs are pretty much a wash- sometimes Apple has the advantage there- except that you can't buy a Mac notebook as crappy as my HP I bought on sale last year- it is very marginal running Vista, whereas my Mac Mini or Mac Book run Vista or Win7 with decent performance. :W


There are many possible hardware solutions for audio; of course, USB tends to be more popular on the PC side, but a lot of PC laptops and the majority of Mac laptops support Firewire- Apple uses the larger Firewire ports, which output power, so for me, my basic measurement solution is an M-Audio Firewire Solo, which works with up to 96 kHz sample rates, so I use it at 88.2 (Mac's like multiples of CD rate, 44.1, not the Windows KMixer rate, 48 kHz), works fine with my microphone that rolls off in the upper 40's.

Image not available

The standard Firewire 400 and 800 connectors on the Mac is the larger connector that includes power feed, so the Firewire Solo runs right off that, doesn't need an external PS.

Firewire has a much more robust and powerful command set in general and a better interface for audio than USB. Since I use the Mac Mini music server for measurements, too, I'm thinking about upgrading the Firewire Solo to an RME Fireface 800 (the 400 is not that much cheaper and due to the half size rack has too many compromises in controls, interface connections, etc).



The Fireface 800 generates it's own ultra low jitter word clock, uses that for Firewire audio transfers and can distribute it to other gear in your "studio". It also outputs a very high quality SPDIF signal (including AES/EBU as well as SPDIF), and has very good DACs built in, so I'll try this for the audio feed from my music server to the preamp. The AES/EBU/SPDIF can be sent to an outboard DAC, of course. Firewire is the standard for Pro audio these days, has been for some time. RME is a leader from what I can tell in Firewire and is reputed to have very good driver support on both the Mac and PC side.

I'm not here to proselytize for the Mac or for Fuzzmeasure to you guys; I need every advantage I can get.... Y'all just keep using Speaker Workshop or Just MLS... that's fine with me! :B

[cjd]

Jon: if you're in the market for a computer, a Mac isn't that much more expensive (if you actually need all the features they're priced competitively) - but if you're just looking to get gear for doing measurements and want Fuzzmeasure but only have non-Mac hardware, it gets expensive. That's all I was saying - I wasn't trying to get into a "Macs cost more" discussion, because I know too well that the cost is relative - Apple doesn't even make a laptop that meets my needs any more (15" 8bit display and an ExpressCard slot)

Firewire is not being supported nearly as well on the newer Macs either, so for folks looking at one of the newer units they would probably need external power.

[JonMarsh]

Firewire is not being supported nearly as well on the newer Macs either, so for folks looking at one of the newer units they would probably need external power.

Actually, that's sort of another "myth" because the Macbook introduction last fall dropped the fire wire on the new Aluminum body MacBook. But the refresh in June this year when the 13" aluminum model was officially instated as a Mac Book Pro, along with a lower price, AND adding back the Firewire- so now the only portable that doesn't have Firewire remains the MBA. Even if you only have Firewire 400, it just requires a cable conversion to connect to Firewire 800- I do that all the time. That's how the Mini works with the Firewire Solo, or the MBP.

Unlike a lot of folks, I am a fan of the Express card slot, so I'm happy that my 15 MBP is one of the original models that does have that feature. I just usually use it to put a fast 32GB flash drive in, but it's there for other options now and then, like an external SATA connection. OTOH, Firewire 800 works so well for external drives that the SATA is superfluous unless you're connecting multiple drives.

No, I wouldn't make the case that one should convert from PC to Mac just to run Fuzzmeasure. But as a relatively recent switcher (2005), I could never go back, and right now the only things I do on the PC side of life are LspCAD, Altium Design (schematics and PCB), and legacy MathCAD (slowly moving things over to Maple). iWorks is such a breath of fresh air next to Office it's not even funny, though I do have a copy of Office also. I don't originate documents or presentations in it- even my IEEE papers are done in Pages, and presentations and AV training in Keynote. 3D Cad in Shark, project management in Merlin, and general illustration in EasyDraw (I have CS4, and these days rarely use anything in it, except Photoshop; EasyDraw has replaced Illustrator because of some of the technical and mathematics drawing aids).

[cjd]

Guess we need a speaker update in this thread!

I don't know Firewire as well - never actually had something that required it so... thanks for clearing that up. My opinion of Apple is that they're no better than average - their service can be good or it can be downright insulting (two of three times...), and even for a desktop they fail to offer anything at all - a mid-priced workstation, say a quad-core with 8GB RAM, solid video card, somewhere around $1200 give or take... I can do an i7 920 or AMD X4 965 for that price... No Vista though, please! W7 has fixed a lot of the issues but... sigh. OSX isn't any better though and is only now available 64bit... wheee.

Where were we?

The Firewire device'll work on any of my machines including the Mini my wife has, so I do have something here I could Fuzzmeasure with.

[JonMarsh]

Uh, actually, the current "standard" Leopard support 64 bit programs, as did 10.4; Snow Leopard combines both 32bit and 64 bit in one kernel; none of this either or stuff with incompatible drivers that MS saddled us with for Win XP-64, Vista 64. And don't even get me started about all the different SKU's that MS offers with the steadily escalating prices. One $49 Family Pack upgrade will get me OSX Snow Lepoard Ultimate Business Media Premium edition on five systems- full MS Exchange support, full NT Domains support, all the media stuff, etc. And it's all an in place install, unlike that wacky chart MS sent Walt Mossberg over at WSJ!

But to each his own- whatever works for ya! This is off topic, and we should get back to speakers- a birdie whispering in my ear said Chuck is busy with family things today! The nerve! :W

[cjd]

Family! What's that? Those people that ignore me? :P

My MPB isn't 64bit. *shrug* Mostly I use Ubuntu around here these days. Nothing I've found that comes close for measurement duty though, software-wise. Suppose I could try to learn how and write something. I bet Fuzzmeasure could also work perfectly with a tiny bit of work.

[chasw98]

But to each his own- whatever works for ya! This is off topic, and we should get back to speakers- a birdie whispering in my ear said Chuck is busy with family things today! The nerve! :W

Family! What's that? Those people that ignore me? :P

Umm, wrong birdie. Next weekend is family time!

As they say, so easy a caveman can do it. With a little coaching from Jon I am producing decent data to work with. But like most things if I could do it over, I would have done it this way...... So I will be refining my method again next weekend.

Just for reference, the Mac hardware I am using can be picked up off of ebay for under $250.00 if you are so inclined and know what you are looking for. Won't do much else and isn't very portable. In the end the Mac computer is just another tool, not better or worse than any other, just does what it should for the intended purpose.

[chasw98]

OK, need to get some work done on these guys. I am going to try and get off axis individual driver measurements today. BUT! Everything appears to be going against me. So far a neighbor is using his chain saw, the trash truck is making the rounds, the 'claw' is picking up bulk trash, and it is set to thunderstorm and rain all afternoon. The background noise level is just not acceptable at the moment.

I did get a turntable system built. Not as nice as Evil Twin's (His is polished black!) but it will work for my purposes. Here are some shots of it and a graph of the background level I am dealing with.







More to come.......

[Hank]

Makin' progress there, Chuck.
The Fuzzmeausure feature description from Jon makes it seem like the product to own, and the UI does look friendly. I wonder how many sales it would take to convince the author to port it over to PC...

[chasw98]

Makin' progress there, Chuck.
The Fuzzmeausure feature description from Jon makes it seem like the product to own, and the UI does look friendly. I wonder how many sales it would take to convince the author to port it over to PC...

The UI is extremely friendly, once you learn the neighborhood, so to speak. It will get you data (good or bad) very quickly and easily, but it will also let you dig deeper and use functions that to me appear to be very advanced. Having a guide like Jon makes it even easier.

I took measurements at 0 degrees, 15 left & right, 30 left & right, and 45 left & right. Here are some shots of the neo8 and neo3 array at 0, 45L, and 45R.

Neo8 - blue is 0 degrees



Neo3 - orange is 0 degrees

[Hank]

Remarkable how close 45 off-axis is to head-on. Bring 'em over to my house when they're finished :W

[chasw98]

Here is the best I have come up with so far. I have included the frequency response, impedance graph, and crossover schematic of the latest iteration for the arrays. Crossover points at 800 and 4000 Hz, all LR24. (1/6 smoothing). I have hooked it up and tested it under my 'lab' conditions and it seems to be worthy to take it to the next step by using it with the good gear as soon as it is built and ready. More on the 'secret sauce' active analog crossover later. I am testing with a Behringer DCX2496 for the moment.

Schematic



Impedance



Frequency Response

[JonMarsh]

Looks like you're making some good progress, Chuck. Too bad you can't bring these as carry-on to RMAF!

[Bear]

Neo8 - blue is 0 degrees

It looks like this plot also includes the woofers based on the system response?

[Johnloudb]

It looks like this plot also includes the woofers based on the system response?

Yep, the red plot is the woofer response.

Chuck,

Looks like you imported measured data of the drivers into LSP CAD and then applied different filters to get the response you wanted? Does LSP CAD let you design active filters or are you going to use spice for that? Just curious ... I've never used LSP CAD. It's looking really nice there.

Hate to interrupt here, I just have a quick question about my dipole design. Does this look alright. The center to center spacing between the mid/tweet and the woofers is 14 inches. Is that too far? I won't be crossing high, just 300 - 400 Hz.

The middle opening is just there to the let the mid/tweet have it's own baffle. Per Jon's suggestion and that's what B-G suggests and for the baffle width for the RD50 and I don't really want to experiment with something else. Maybe not really necessary though.

[JonMarsh]

LspCAD has support for conventional active filters and also supports some specific boxes, like the DCX2496. It's a great package, IMO, but then I'm a long time user. Probably not objective.

[chasw98]

Looks like you imported measured data of the drivers into LSP CAD and then applied different filters to get the response you wanted? Does LSP CAD let you design active filters or are you going to use spice for that? Just curious ... I've never used LSP CAD. It's looking really nice there.

Yes, the frequency response data came as exported FRD files from Fuzzmeasure. I used the justMLS portion of LspCAD to do impedance measurements and then put them all together. I set the filters up using the 2496 in LspCAD because it is quick and dirty. To start with they are all just convential slopes at fixed frequencies. I could also add numerous parametric and shelving filters but I don't want to 'muck' up the sound. That is why you see 1 inductor in the woofer circuit to adjust some response in that region. I am being very judicious in my use of filters as I build this. I really want to solve the problems acoustically before manipulating them electrically. So far, this scenario gets me in the ballpark. I want to mock up the filters in the crossover I am actualy eventually going to use to see how they sound and then go from there. I have been using spice to design some filters and get a feel for the limitations of how narrow I can get in an analog circuit.

Hate to interrupt here, I just have a quick question about my dipole design. Does this look alright. The center to center spacing between the mid/tweet and the woofers is 14 inches. Is that too far? I won't be crossing high, just 300 - 400 Hz.

It looks OK from tweeter to mid and is probably OK from mid to woofer, but I would check the physical length of a 400 Hz wavelength. Also, never say never. I just ended up using an 800 Hz crossover point and never ever thought I would go that high.

The middle opening is just there to the let the mid/tweet have it's own baffle. Per Jon's suggestion and that's what B-G suggests and for the baffle width for the RD50 and I don't really want to experiment with something else. Maybe not really necessary though.

I may be mistaken.... Is this a 3 way you are doing or is the center 2 1/2" opening just a blank opening? Disrgard my last suggestion if I misinterpreted you.

[Johnloudb]

I may be mistaken.... Is this a 3 way you are doing or is the center 2 1/2" opening just a blank opening? Disrgard my last suggestion if I misinterpreted you.

Yeah, that's just an opening between the woofers and the RD50 driver. I hope 14" is close enough, and I read something somewhere, maybe on this board, that the Usher 8" doesn't like to be played too high due to a resonance. But, like you say, you never know.

Should the wavelength be twice the distance between the drivers? I can't remember what the rule is.

P.S. - Okay I think It's cool. 0.356m distance and the wavelength at 400Hz is 0.853m in air. :T

[PhilDSP]

Hate to interrupt here, I just have a quick question about my dipole design. Does this look alright. The center to center spacing between the mid/tweet and the woofers is 14 inches. Is that too far? I won't be crossing high, just 300 - 400 Hz.

That should work well according to my calculations. The spacing should be within one wavelength if I understand the general rule. 14 inches equals .359 meters. The speed of sound is about 344 meters per second so the upper frequency for your configuration would be 344/.359 or 958 Hz. You're within a half wavelength - no problem.

[Johnloudb]

Thanks Phil, that's good to know!

[Johnloudb]

Well, I'm about to buy wood for my speaker. And, I think I'm just going to use either 3/4" baltic birch laminated to 3/4" of MDF. Or, I might use two pieces of BB laminated together. My cousin is a cabinet builder so he's is going cut the large wood pieces, but the routing and putting together is up to me.

Is there any problem with applying veneer to BB? I may not even use veneer and just put a finnish on the BB. I'd like to do 3/4" or maybe 1" round overs on the edges. Can this be done on the BB without a mess?

I'm a novice and haven't ever used a router, but I do have a ROTO ZIP which I use on occasion. I'm going to get a router. Also, where can I get decent wood working table or can I just get a few saw horses?

I've decided to build an aluminum frame to support the speaker.

[---k---]

I have several projects with a stained BB ply. It can look very nice. It depends on if you like the look of the stained edges. I got a 1.25" roundover on my sub.

Just a couple of things to be wary of. When using a router on the edges, the plywood can tear. So, use a sharp bit, the right speed, and a little practice to avoid. Taping the edge with painters tape may also help.

The veneer layer is very thin, so go easy on your sanding. Shouldn't really need much.

Also, I think the edges of the 13 layer plywood looks a lot nicer than the 7 layer.

These came out nicely and are just BB ply:

Image not available

[Johnloudb]

Yeah, those look really nice. I guess I'll do an all BB ply baffle and if I should screw up (hope not ... fingers crossed) I can always put a veneer over it. I'll see if I can get the 13 ply BB.

[stinems]

I took measurements at 0 degrees, 15 left & right, 30 left & right, and 45 left & right. Here are some shots of the neo8 and neo3 array at 0, 45L, and 45R.

Neo8 - blue is 0 degrees

Whoah....where did that 12khz +10dB hump go that I'm accustomed to seeing on the NEO8's spec sheets? Does it disappear when employed in a line array???
Or does that graph reflect BOTH the neo8 and neo3 array summed responses after crossover?

Sam

[chasw98]

Whoah....where did that 12khz +10dB hump go that I'm accustomed to seeing on the NEO8's spec sheets? Does it disappear when employed in a line array???
Or does that graph reflect BOTH the neo8 and neo3 array summed responses after crossover?

Sam

Nope, those are the Neo8's all by themselves in an array measured at 2 meters in a room with an 8 foot ceiling. So the array effect should be in force.