Please help design an MTM center

[Don Mascali]

I would like suggestions and opinions on the box size, driver spacing and Xover design of a new horizontal MTM center channel speaker for HT/Music use. The components on hand are two 2123J drivers that I got new on EBAY and for the HF I have a 2416H that will go on a 2342 horn. I also have a 2425J and 2344 horn I could use too.

My current fronts are actively crossed over and are as follows per side;
2404-5khz and up
2426H/2344-1200hz and up
2-E120s-150 to 1200hz
B380 clone 150hz and down
4645C 150hz and down

I really am not concerned with deep bass or extended VHF performance. We are after voice and fill for the center channel, midrange clarity is most important and voicing compatable with the existing fronts.

My first thoughts are to Xover at 1k to 2k and possibly a sealed box.

I am not at the skill level of many of our members and will be gratefull for all comments.

Thank you very much,
Don M

[Zilch]

"Classic" design calls for woofs centers to be each spaced one wavelength of crossover frequency out from center of tweeter.

At 1 kHz, that'd be 13.56". I'd turn the biradial horn sideways so the "cheeks" don't scoop sound from the woofs.

Drawing posted several times elsewhere on the forum shows patterns achieved with various crossover alignments, but that's for dome tweeters. Biradial's gonna be different, probably....

[Earl K]

Hmmmmm,

I would like suggestions and opinions on the box size, driver spacing and Xover design of a new horizontal MTM center channel speaker for HT/Music use.

It would be a lot better if you were working with a woofer having a larger Xmax value - like twice as much ( such as the 2012h, 2251j ) but ,,,,,

The components on hand are two 2123J drivers that I got new on EBAY and for the HF I have a 2416H that will go on a 2342 horn. I also have a 2425J and 2344 horn I could use too.

- Yes, I would "tune" this big box to recapture some of the midbass rolloff ( that starts occuring below 250-300 hz ) and create some useful response in the 125hz area.

- The 2123h ( IME ) is rugged enough to a point ) to survive a system tuning ( Qtc ) in the .425 to .411 range. Personally I'm always nervous going below .5 Q(tc) since Xmax is invariably compromised . .

- .411 Q(tc) in my software equates to @ .65 cu' for each 2123 . That's fairly oversized . For reference ,my preferred .5 Qtc gives an enclosure size of @ .35 cu' .

- a .65 cu' enclosure tuned to 110hz wants a "tuning" port/hole of 9 sq inches, when cut directly into a 5/8" thick baffle board ( or 9.6 sq inches in a 3/4" thick baffle ) . You'll need to limit input power to each 2123j to 175 watts max ( or the woofer cone will over-excursion ) . This "limit" should not be a problem in a home environment. If you tune to 130hz, you can just about use the full rated input power.

- I'd use the smaller of the horns you have ( the 2342 ) because it allows tighter packing .

- That critical distance between the woofer centers ( for a straight ahead MTM with zero lobing ) is unknown to me at this time . Give me some time to do some handson experimenting and I'll get back to you. ( I own 2123s in @ .65cu' enclosures )

- In the meantime how high off the floor will these boxes find themselves ?

- How soon do you want to get moving on this project ?
( The reason I ask is because within 2 months I'm going to changeout the stock surround on a 2123 for a foam type . The idea is to see if this will offerup a little more natural midbass from this driver - somewhat like a shorter coil 2122 )



PS; I'll post some SIMs ( different tunings ) a little later .

[Earl K]

Well quickly ,

The smoothest power response was obtained with 2 woofers side by each - with the baffle board split in two down the center and angled outwards at an offset angle of 20 to 30° . This is a fairly typical look for a mtm center fill

Even side by each woofers display some midrange nulling when used together on a straight unbroken baffle - though very good response is obtained up to maybe 40 - 45° off-axis / then the nulls become obvious. That gives a good 90° "projection window" .

Tomorrow; I'll jury-rig a 2344 in between my test boxes and see what gives .

[Mr. Widget]

I really am not concerned with deep bass or extended VHF performance. We are after voice and fill for the center channel, midrange clarity is most important and voicing compatable with the existing fronts.

So why go to all the trouble? Get an LE8T or H and plop it in a 1 cu ft sealed box and you will have a very nice center channel. The LE8T is even shielded.

Widget

[Mr. Widget]

Don't be so sensitive.

It will be on it's own amp. Unless he wants to play stupidly loud the 80Hz filter will allow plenty of midrange out of an LE8T.

Keep it simple or pedal to the metal. Why waste your time with all the half ass sh!t in the middle.

Widget

[boputnam]

Well , Okay , I'm laughing !

but

I think its you who's being sensitive here . I think it's just that you now hate MTM and want a return to the simple tranquil life - of no hulking boxes filled with 1500AL(s) . You're just projecting those wishful thoughts onto poor Don and his project.

Weird...

[GordonW]

Well, my calculations, using a 1.2 cubic foot box (.6 cubic foot for each 2123) with fiberglass wall lining, using an 80 Hz port tuning, gives an f3 of 85 Hz and a max acoustic power rating of around 115 dB (@ the biggest excursion frequency, which will be about 100 Hz), before the drivers run out of X-max. Mind you, they will NOT bottom out or anything... not until MUCH higher excursions... so it should be able to reach PRODIGIOUS output levels without damage.

I'd think the 2123s should work fine in such a setup...

Regards,
Gordon.

[Mr. Widget]

Gordon, The LE8T will only require 1 cu ft.

...and Earl, while I do think an LE8T would make a great dialog speaker I was being tongue in cheek.


As for MTM wait for the comments after our listening session this Saturday!

Widget

[Don Mascali]

Thank you to everyone for the quick replies.

This box will sit on a shelf 40" from the floor just below my screen. I use a projector so shielding isn't an issue. I have attached a pic of the setup.

GordonW:
The 1.2 cu ft box and 80hz tuning sounds perfect. I run the bass drivers to 150hz so 85hz f3 is as low as we need to go. The center can be set to "small" and crossed at 100hz.

EarlK:
I would like to stick with a flat baffle. My stuff images very well and for the dialoge I'm not worried about a full pattern. I sit 14' from the screen and the room is 20' wide so off axis response shouldn't really be an issue. This will be driven by my Yamaha A/V receiver w/100w and excursion limiting isn't a problem there. I'm sure my ears would give up long before. I don't know if I have the space to mount the horn above the drivers. I'll stick it in between, even if that is a compromise.

Zilch:
Agreed on the sideways horn. It may look strange, but it's needed.

Thanks again,
Don M

[Earl K]

Well, my calculations, using a 1.2 cubic foot box (.6 cubic foot for each 2123) with fiberglass wall lining, using an 80 Hz port tuning, gives an f3 of 85 Hz and a max acoustic power rating of around 115 dB (@ the biggest excursion frequency, which will be about 100 Hz),

Hmmmmm,

Gordon, my software ( MacSpeakerz ) doesn't agree with yours at all .

For Instance :

- My software shows a F3 plateau ( droop ) that begins around @ 150hz while endng at about 100hz before diving to a F6 point at about 77hz.

- I get maximum acoustic power of about 118db ( extrapolating backwards to a "J" instead of the "H" model seen here ).

- Biggest excursion occurs at around 125hz .

Can you post a jpg of a screen capture - that would show what you are looking at ?

I'll do the same - sometime today .

[Earl K]

I would like to stick with a flat baffle. My stuff images very well and for the dialoge I'm not worried about a full pattern.

Well ,, you've been warned .

- I just did a mockup with flat baffle vs outward angled baffles / using a 2425/2344 combo placed in between the boxes .

- Angled baffle wins on all counts . ( It's really best to not combine the uppers mids of the 2123j(s) - but do combine the lower midrange - this gives the flattest / smoothest passband ( 100hz - 1000hz ) over a broad 180° horizontal listening range )

- The flat/planar baffle has significant lobing issues that color the sound as one rotates ( walks ) in front. This is quite hearable and measurable using pink noise as a source .

- I strongly recommend you repeat my experimental mockups by first building (2) test boxes, each .6 cu' (net volume) . Then repeat this R&D before finalizing your design.

[Don Mascali]

How much of an angle would you recommend? would 15 deg on each side be enough?

I see by your graphs that a higher tuning makes sense.

Thanks for all of your input!

[Mr. Widget]

After seeing your setup, I would strongly suggest you consider building a version of your left and right speakers without the upper woofer. You have the mid horn and driver available. You would have to buy one woofer and 2404, but it would be worth it. Even though you don't think you need the "full" range here, there are some films that put much more than dialog on the center channel and if you watch a concert film you ideally want all five speakers to be identical. More importantly the sound quality as a soundtrack is paned across the screen works best if all of your front speakers are identical. You certainly have the room for a single woofer version of your L/R speakers.


Widget

[4313B]

I'd skip the Junk But Loud and toss a Tannoy dual-concentric in there.