Big Blue

[John W]

Here are the plans I have in the works for building a big 4-way system. The design has basically evolved from a desire to put a set of accumulated drivers and parts to good use, and try out Jack’s axially symmetric oblate spheroid CD waveguides.

Initially I was considering building a separate “dog box” for the top 3 drivers and then playing around with a couple different options for the lower bass. In the end I decided to build just a big box, rather than 2 or three separate boxes per side. The footprint is about the same either way, and the current design worked out to be the most economical use of materials.
1 – 5x5 sheet of Baltic Birch
2 – 4x8 sheets of 1in MDF
1 – 4x10 sheet of walnut veneer
With separate boxes I couldn’t come up with a way to provide the same volume, have the boxes veneered and stay on budget.

The drivers I have are:
18in 2241G woofers up to 290hz
12in 2202H midrange driver running from 290hz to 1.2khz,
2in 2445 using the same profile horn described in your initial post from 1.2khz to 10khz,
and a 2403 cats eye from 10khz up.

Also, I still have a pair of AB international tri-amp amplifiers to drive them. The amplifiers put out 850w@4 ohms for the woofers, and have two more 250w channels per side. They are currently set up with an internal crossover it 400hz and 1200hz, with some various low frequency boost options. Eventually I would like to try a digital 3-way crossover, but this may have to wait. To accommodate the tri-amp design I decided to implement a passive crossover between the mid-range horn and tweeter. This was pulled from the 4355 crossover, with an added 20uf cap up front of the mid-range as dc and thump protection.

From my calculations the box will end up about 8 cubic feet after subtracting volume for the dog box and drivers. WinISD suggests a 35hz tuning for the 2241G. I plan on making the ports easy to replace in case I end up swapping out the woofers for 2245s or something else later and have to re-tune the box.

Construction has already begun. The turned horns are finished and I’ve started on the cabinets. Let me know if I need to put the brakes on, otherwise I’m trudging ahead and will keep you posted.

[Earl K]

Construction has already begun. The turned horns are finished and I’ve started on the cabinets. Let me know if I need to put the brakes on, otherwise I’m trudging ahead and will keep you posted.

Hmmmmm,,,, John ,

- Only because I have such great respect for your wood-working abilities , do I suggest something a lot more ambitious,( both sonically and build-wise ) .
- The ***sonic purity of a round horn*** just begs to be used with an MTM type system .

- How about something like this kludge-up ? It's an asymmetrical, MTM-based, "fanciful-thought" project ( for the time being ) .
- It's meant to be a biamped / triamped ( only ) system ( just for the sake of simplicity ).
- The 2245H and 2206 ( or other qualifying 12" ) are run paralleled, with various means applied to restrict/blend their respective response curves.
- ( ie; the 18" "working" in a ported cabinet would have a large inductor on it to roll down its midrange ( earlier than the 12" ) / while the 12" working in a "largish" sealed doghouse slowly rolls out its' LF response ).

- You'd need to sell your 2202 midrangers ( I know, I know this is all a stretch ) and obtain midwoofers instead, like the 2206H, to make this sort of MTM concept workable .
- The midwoofer ( 2206H ? ) should be located as close to the horn as possible / with a center to center spacing ( 2245 to 2206 ) of around 34" as a reasonable goal.
- Another ( perhaps conflicting ) design goal is to use a ( large enough ) round horn that will support a crossover point as low as 600 to 650 hz ( with minimal low-mid "correction" EQ ) .
- Whatever 12" midbass woofer is used , it needs to have a similar efficiency to that of the 2245H. It also needs to offer pretty linear output ( from @ 80/100 hz to 650 hz ) when housed in its' sealed ( @ )1 cu' subenclosure. For Xmax reasons, a long coil winding depth ( > than or = to .75" ) offers the best compromise for this job description . ( Aquaplassed cone, a deeper gap, higher compliance surround and higher BL factors are also desirable attributes / though much tougher to obtain through eBay type purchases ) . The ME120H or 1200Fe do come to mind as my first choice twelves to explore ( when designing with specs. from a page / and when referenced through my own hands-on experience with the ME150H ) .

- I believe you'll need to design & build a small round waveguide for your 2403 driver to maximize the imaging capabilites of a 4way system . This will most likely entail copying the initial "compression-chamber" of the 2404 tweeter to achieve the correct, initial compression characteristic necessary for a bit of diffraction.

- The original image belongs to "pantaNS" and is of his 4-way system that uses a 2245H and a 2206H ( among other things ).



Okay , not yet ? maybe in a couple of years.

[John W]

I like your idea. Alas, maybe next time.

[jack_bouska]

Here are the plans I have in the works for building a big 4-way system. From my calculations the box will end up about 8 cubic feet after subtracting volume for the dog box and drivers. WinISD suggests a 35hz tuning for the 2241G. I plan on making the ports easy to replace in case I end up swapping out the woofers for 2245s or something else later and have to re-tune the box.

Construction has already begun. The turned horns are finished and I’ve started on the cabinets. Let me know if I need to put the brakes on, otherwise I’m trudging ahead and will keep you posted.

John
The cad drawing looks interesting, some quick thoughts in no particular order:

- If you have a digital camera, post a photo or two of the horns

- How do you plan to mount the 2245 on the horn? (They are heavy!)

- remember, the 2445 will not have flat freq. Resp on the waveguide, so you will need to do lots of xover horn eq mods to get it to sound right.

- I have not run the simulations, but if I remember correctly, the 2241 has a higher Fs compared to a 2245 (35Hz vs 20Hz?). If you have eq capability in the amps, then consider tuning lower and using judicious boost after testing Freq Resp n your room

- lastly (and you might hate me for this), what about turning the speakers 90deg, and mounting the 2202, 2445 and 2403 on the skinny 18” wide side of the cabinet, leaving the 2241 on the wide 28” side. This would make the original front the side, and the side into the front. You would have enough room to put the three drivers in line vertically on the skinny side (the 18” wouldn’t be in the way anymore) and the narrow dimension would cause fewer frequency response and imaging problems due to the diffraction effect caused by the wide cabinet. You can place the drivers closer to one side than the other, depending on internal bracing, etc.

The 18” does not need to face forward, as ½ wavelength at 290 Hz (2') is around the size of the cabinet, so the 2241 is pretty much omni directional below the xover, so it doesn’t matter which way it faces.

Keep us posted
Jack Bouska

[John W]

Thanks Jack,
I'll post some pictures of the horns tonight.

- The drivers bolt directly onto the base of the horns, which should be pretty evident when you see the pictures. They are quite heavy, which causes me some concern. Within the larger cabinet they will just clear the top of the inner box housing the 2022 drivers. My plan is to securely brace the 2445s to the top of this box using some sort of cradle and strap, and hopefully prevent them from putting too much strain on the horns.

- I plan on a little roughness in the response on these drivers, taming it will have to follow getting them put together.

- The speakers are going to be pretty large. I hear your concerns about baffle diffraction and driver placement, but I think turning the speakers sideways would bring them too far out into the room to be practical. The boxes will be mirror imaged in keeping with the design of other JBL monitors, like the 4344 and 4345. The 4345s are similar in size and they image quite well. From what I’ve read elsewhere there are advantages to having a horizontal alignment of upper midrange and tweeter, both at ear height, especially since the tweeter designs tend to have a narrower vertical beam width.

- I’ve modeled the 2241 drivers in WinISD and played around with a few different box tunings. The picture below shows three tunings: the green at 35hz the orange at 30hz and the grey at 27hz. I guess I don’t see the advantage to lowering the tuning too much?

[John W]

I have been looking at the Behringer dcx2496 digital crossover for this project. Does anyone have any experience with it?
Does it offer any equalization functions or is it strictly a crossover?

[Robh3606]

The boxes will be mirror imaged in keeping with the design of other JBL monitors, like the 4344 and 4345. The 4345s are similar in size and they image quite well. From what I’ve read elsewhere there are advantages to having a horizontal alignment of upper midrange and tweeter, both at ear height, especially since the tweeter designs tend to have a narrower vertical beam width.

Hello John

The 4344's do image well and there is an advantage because of the very narrow vertical dispertion of both the higher you go.

I’ve modeled the 2241 drivers in WinISD and played around with a few different box tunings. The picture below shows three tunings: the green at 35hz the orange at 30hz and the grey at 27hz. I guess I don’t see the advantage to lowering the tuning too much?

Well that depends on how they work out in your room. I have my XPL clones tuned low and with room gain they come in just right. You have to be careful, you don't want to end up with the last 2 octaves all jacked up so leave yorself some room so you can try different tunings to see what works best and what you like.

I agree with Jack on a narrow profile baffle would be prefereable or have seperate boxes but it's hard to get the volume you need in anything other than the shape you have chosen. Those 18's need a good amount of room to breath. Have Fun!!! We want pictures!!!!

Rob

[Titanium Dome]

(snip)- lastly (and you might hate me for this), what about turning the speakers 90deg, and mounting the 2202, 2445 and 2403 on the skinny 18” wide side of the cabinet, leaving the 2241 on the wide 28” side. This would make the original front the side, and the side into the front. You would have enough room to put the three drivers in line vertically on the skinny side (the 18” wouldn’t be in the way anymore) and the narrow dimension would cause fewer frequency response and imaging problems due to the diffraction effect caused by the wide cabinet. You can place the drivers closer to one side than the other, depending on internal bracing, etc.

Jack Bouska

I was surprised and pleased to see this recommendation. (But then you're by no means a traditionalist. )

[Don Mascali]

IMHO, DON'T.
On paper and in use this is an amazing piece of gear for "Home" use. BUT, I have owned three of them and every one has the habit of making an intermittent "Sizzling Bacon" sound when I think it looses a ground due to piss poor construction or some other reason beyond my capacity to under stand. I guess when you steal someone elses design and built it cheaply you miss a few things. It does have PEQ, Limiter, Auto Time Alignment/Phase setup among other functions.

Zilch has one now, maybe he can shine some light here.

I ended up with a YAMAHA D2040 4 way digital that sounds good. It has no where near the bells and whistles as the DCX 2496 but is clean and solidly built.

[John W]

Here are some pictures of the walnut horns.
I glued them up with 1 7/8 hole running down the middle to make turning easier.
My lathe is older than dirt, but it runs smooth and is quite solid.
I plan on mounting them from the back with some screws on the outside baffle edge.
These are currently unfinished. I will apply the finish when doing the cabinets.

[jack_bouska]

[quote=John W]Here are some pictures of the walnut horns.
quote]

Wow, nice looking waveguides, I’m envious, to say the least.
I hope you think they sound as good as they look in the pictures!

Now for some additional quick comments before off to bed:
- I’m not sure the xover circuit shown in post #1 will work all that well with the oblate-tractrix waveguide
- I’m posting a couple of images which show a comparison between the JBL PT waveguide, and 2381 horn against my oblate-tractrix (on a jbl 2441) – all without EQ applied. The oblate-tractrix matches the PT-waveguide frequency response quite closely in overall shape, so if you can find an example of a xover and eq for that waveguide-driver combination, it should be close to what you need for your system.
- Barring that, I might have a few spare moments to cut-and-try some passive xover and EQ later this month. If I get anything to work, I’ll post what I did.
- Passive xovers depend heavily on the impedance of the drivers: What is the nominal impedance rating for your 2202H and 2445?
- let me know and I’ll try some spice simulations.
- in the mean time, I’ll post a first attempt at a 2nd order xover (similar to Wayne Partham Pi-align xover) in an image below.

Best of luck, and let us know how the system turns out when all the sawdust and solder settle

Jack Bouska

[John W]

Mr. Widget did a great analysis of the ring radiators in this post:

http://audioheritage.org/vbulletin/s...ead.php?t=6368

Here is a copy of the impedance plot for 077 and 2405 tweeters with a couple different diaphragms. It should be pretty similar for the 2403. I think 10 or 11 ohms is normally used.

I also cut a little plot out of the 2245j sheet. These are 16 ohm versions.
I went with the version shown up top basically because the drivers are similar to those in the 4355 and the crossover calculators I have access to suggested values that were about the same. I am not a crossover expert, so things like baffle compensation and differences in horn response were not really considered. I’m all ears if you have a better suggestion.

[John W]

Construction is mostly done, and I thought I'd share some pictures:

[John W]

A couple more.

[John W]

So far listening results are postive.
I listend to the speakers tonight with the active 3-way amps and the passive crossover between the horn and 076. They sound big.
I tuned them as best I could with the little dip switches on the amps that assist in level adjustment, but I don't think I was able to balance them just right. The potential is definitely there.
Next step is to get a digital 3-way crossover and try them with some amps that I am more familiar with.