2206H between 2245H and 2123H or not?

[NiToNi]

Good morning,

I have plans to build an active 3-way using 2245H + 2123H + SEOS-12 with a Beyma CD10Nd. Estimated crossover frequencies are 180-250Hz and ~1400-1600Hz.

Reading comments from other members however, I have become wary of the woofer-to-midrange transition, as some say that the 2245H may not fair that well up to 200 Hz and likewise the 2123H down to 200 Hz. This despite the fact that my modelling of the 2123H in a sealed box using the measured TSP of my pair of drivers at-hand shows no excursion problem up to 500W with a LR4 @ 200 Hz:

http://i63.tinypic.com/2ec0wlk.jpg

Note that besides the LR4 @ 200 Hz, this FR has also been shaped with a LWT with F0=132 Hz, Q0=0.52, Fp=50 Hz, Qp=0.707.

http://i63.tinypic.com/ff3hty.jpg

; but experience beats theory every time, right, of which I have none when it comes to these drivers…

So I am contemplating whether I should use a pair of 2206H drivers that I also have at-hand for mid-bass duty, say 75-300Hz and make it a 4-way system. But I that would increase cost and complexity quite significantly so I would only be prepared to do that if there were clear benefits from doing so. Also, such a 4-way would physically be too large (with too high a tweeter height) so in reality it would have to be made a 3-way system (2206H + 2123H + SEOS-12) with a sub, which of course is better anyway for several reasons, but if I’m going down that route I would probably want to use a more modern 18” driver design with greater Xmax and low-end extension.

I would really value some opinions from experienced JBL users, especially from those who think they can talk me out of going 4-way :D

[bubbleboy76]

"Crossovers suck in general and the fewer and further between, the better." - G.T

[JeffW]

Doesn't the 2245 in the 4345 cross over to a 10" driver at 290Hz?

[1audiohack]

Yup.

[quindecima]

The M2 may sound good but it looks like s--t. When I sit in front of my speakers I like to see a little eye candy not a mound of plastic. No panache

[dprice]

Someone called it a gorilla butt :blink:

(apologies to the OP)

[Lee in Montreal]

What about using the 2245 from 20Hz ---> 50/80Hz
Then 2206 - 50/80Hz----> 800Hz
And Beyma 800Hz ---> top

[1audiohack]

What about using the 2245 from 20Hz ---> 50/80Hz
Then 2206 - 50/80Hz----> 800Hz
And Beyma 800Hz ---> top

That would be the direction I would take as well.

Barry.

[JeffW]

And you could probably run the 2206 up a few hundred Hz more and be OK. Maybe not all the way up to 1400~1600, but 1000~1200 would probably be OK and it seems like the H dispersion of the 12" driver at 1200 is 90°, should work well with the SEOS 12.

[Lee in Montreal]

And you could probably run the 2206 up a few hundred Hz more and be OK. Maybe not all the way up to 1400~1600, but 1000~1200 would probably be OK and it seems like the H dispersion of the 12" driver at 1200 is 90°, should work well with the SEOS 12.

Indeed. And with an active crossover, it will take little experimentation to find the sweet spot between 800 and 1200Hz. This would be a nice 2.5 system. Find the proper cabinet volume to help the 2206 reach as low as it can (w/ BR) and perhaps use a single 2245. ;-)

BTW That SEOS 12 seems "smallish". How low can it go?

[NiToNi]

"Crossovers suck in general and the fewer and further between, the better." - G.T


I agree but a 4-way is hardly unheard of. In this case, the extra crossover would also be low in frequency, where the ear's sensitivity is lower and other factors (i.e. room) may influence the sound to a greater extent. Moreover, an active filtering solution holds the potential for better integration versus passive.

[NiToNi]

Doesn't the 2245 in the 4345 cross over to a 10" driver at 290Hz?

Yes, to a 2122H, second order (12dB/octave).

http://www.jblpro.com/pub/obsolete/4345.pdf

But I can't help thinking that other concerns may have influenced this, e.g. baffle step, floor bounce, cost/practicality of passive filter components at any lower fc, protection/headroom for studio (ab)use etc.

Based on simulations, I cannot see anything that would prevent the 2123H from being used down to 200Hz.

[Lee in Montreal]

Looking at studio monitors (I mean real monitors found in real studios), it can be realized they are often 2-way, sometimes 3-way, rarely 4 ways. I agree that the fewer the drivers and crossover points, the better.

Also, the 2206 is a great driver. But it was also designed for sound reinforcement. Maybe JBL has another 12" driver that is more subtle with a flatter response curve. I never heard one, but if its output is similar to its bigger brother, the 2226, then it could be quite generous in upper bass and lower mid.

http://www.premierstudiosny.com/images/studio%20b%20control%202.jpg

http://www.in2guitar.com/stereo/ureiad39p.jpg

[NiToNi]

What about using the 2245 from 20Hz ---> 50/80Hz
Then 2206 - 50/80Hz----> 800Hz
And Beyma 800Hz ---> top


And you could probably run the 2206 up a few hundred Hz more and be OK. Maybe not all the way up to 1400~1600, but 1000~1200 would probably be OK and it seems like the H dispersion of the 12" driver at 1200 is 90°, should work well with the SEOS 12.


Indeed. And with an active crossover, it will take little experimentation to find the sweet spot between 800 and 1200Hz. This would be a nice 2.5 system. Find the proper cabinet volume to help the 2206 reach as low as it can (w/ BR) and perhaps use a single 2245. ;-)

BTW That SEOS 12 seems "smallish". How low can it go?

Yes that's of course an alternative - and indeed the SEOS-12 is typically paired with a 12" woofer, like in the Econowave-style designs - but I really wanted to center this build around the 2123H since I have it and is not as versatile and useful as the 2206H to use elsewhere. Besides, not entirely convinced I like the 2206H all the way up to meet the horn (1.2-1.3kHz).

800Hz is too low for the SEOS-12 (not holding directivit below 1kHz, rec. min xo 1.2kHz):

http://libinst.com/SEOS/Horiz%20SEOS12.jpg

http://libinst.com/junk/Oct29Meas/SEOS12%20Horiz%20Map.jpg

Vertically it starts losing its pattern even earlier of course so a higher crossover point than 1.2kHz doesn't hurt (up to a point, limited by the vertical null as set by the CtC distance).

Nor does the CD10Nd compression driver go that low (Fs 700Hz, rec. min xo 1.2kHz) - and of course is likely to sound better if crossed a bit higher. Directivity-wise, a good theoretical xo (-6dB @ ±45°) is (0.7 x c)/(sin(θ) x D) = (0.7 x 343) / (sin(45) x 0.2) = 1,669Hz, which reasonably seems to agree with reality:

http://techtalk.parts-express.com/filedata/fetch?id=1128782

But I digress - this was about the addition of 'filler' midbass or not :o: I think I have made up my mind though and will go for a 3-way, aiming to cross the 2123H lower than JBL did the 2122H in the 4345.

Next decision will be whether to go for the standard 10 cu.ft. coffin or try something different just for kicks. I'm thinking making it a corner speaker, with a triangular foot print of about 55 (side) x 55 (side) x 78 cm (45° baffle) and as tall as the ceiling (240 cm), sealed with EQ. Crazy? Well, here is a simulation of one of my 2245H in 120 liters, 600W (peak program) and an LTW (F0=63Hz, Q0=0.65; Fp=27Hz. Qp=0.707):

70114

F3=27Hz, F6=21Hz, F10=16Hz. Pe=600W produces 110dB @ 1m but that is likely to be quite a bit higher with corner loading (but OTOH excluding power compression losses). The grey line is max SPL before hitting Xmech of 12.5mm at each frequency. Here is the excursion graph.

70115

Xmax of 9.5mm is only exceeded below 25Hz. If restricting power to max 500W peak the driver should be safe and cone excursion acceptable (within Xmax) above F3, i.e. for most musical material, with useful output (albeit at higher distortion) down to 20Hz. Crossing actively, a HP would be trivial if further protection is wanted. If SPL is too low in real use, Fp in the LTW can be increased at the expense of extension (F3 35Hz yields another 5dB over 27Hz).

To me it looks like a viable concept worth trying. Not only would this come with all the advantages of a sealed system (IR, GD, transients, easier integration with room and less risk for boom etc), but it would also blend such large speakers into the room, especially if painted in the same color as the walls. 78 cm baffle leaves about 15 cm room (diagonally) on each side of the 2245H, allowing a 4" thick slab of OC703 on each wall either side of the speaker, good for absorbing down to 200Hz, i.e. below Schroeder and the whole passband for the 2123H.

Still crazy?

[Lee in Montreal]

A 2245 in less than 5cft isn't a good use of a 2245 ;-)
If you want something compact, then get a W15GTi

[NiToNi]

A 2245 in less than 5cft isn't a good use of a 2245 ;-)
If you want something compact, then get a W15GTi

Hi Lee,

Why is that? I'm not being cantankerous or questioning you experience, just genuinely curious on what you base that particular number of 5 cu ft on?

For bass reflex I'd agree but for sealed cannot see any technical reasons for why that would be the case. Nor do a simulation. Non-EQ'd, the Qtc of a 120 litre sealed enclosure is 0.65, which is hardly indicative of too small a Vb. With EQ I can shape the response to whatever I feel like, within the woofers excursion and thermal limits. For the response I had simmed and desire (Qtc=0.707, F3=27Hz) 120 litres is the smallest Vb that allows the driver to make full use of its mechanical capacity while not exceeding its thermal capacity of 600W (AES program).

Increasing Vb will increase the system's sensitivity, allowing - or in this case, demanding - a reduction in power to make the very same full use of the driver's excursion capability - but the response and max SPL would still be the same (disregarding that power compression may now be lower than at a full 600W). It's just the usual classic trade-off, right?

If space is not an issue, it would make sense to build a box with the volume required for a vented design (e.g. 8-12 cu ft), in case of a change of heart, and limit applied power to whatever it takes to achieve the very same response and maximum SPL. But the sealed bass system's acoustic Q would be the same whether Vb 4 cu ft or Vb 12 cu ft.

[Ian Mackenzie]

Subjectively a vented 2345 dynamically is far superior to a sealed 2245.

Suggest you try the 2345 and the 2123 crossover 320 Hz actively.

Alternately if you a stressed and you think it would sound better another way which it won't vent the 2123 and it will go well below 200 Hz.

I recommend the 4344mk11 crossover filter on the 2123 as a starting point for passive x over to the SEOS. The EQ for the 2123 is fairly specific and it needs the R/L inductor to flatten the notional impedance below the crossover point.

The SEOS will require some EQ as its CD. Some guys on the SEOS forums have some good networks.

I absolutely no doubt the 2123 will be a winner either the SEOS. Shame we are not neighbours, l could measure it and model in LEAP

Perhaps some local for unites can offer assistance.

Have fun