Center channel designs for home theater

[Don Mascali]

Hello i have one 2446j 2" driver + 2382a horn and 2 2226h 15" in a horisontal MTM config. The box is 2,4 meter wide . Crossover is 700 and 40 hz.

2 closed chambers 140 liter each.

regards mats

Damn, thats what I call a center channel!

[slxrti]

Well 70Hz seems to stick in my mind but obviously fullrange would be the best option. Have you looked at the 2118 spec sheet?? Run them in .5 cubic Ft tuned to 70Hz. With that tuning the knee is down but at 70hz and with room gain it may work out OK.

http://www.jblpro.com/pub/obsolete/2118.pdf

Are you stuck on these drivers??? I would look real hard at what your mains will be and use that to determine the driver set for the center channel. Do you really need 2217 and 2123??? Seem's the 2217 could cover the gap just fine between the 1500 and the 375. Don't make it any harder than you have too. Using the same drivers in the same ranges as the L/R is a good thing.

http://www.jblproservice.com/pdf/Dis...ls/2217HPL.pdf


Have Fun

Rob

Thanks for the info on the 2218, I should have checked the data sheet first.
The issue with the 2123, is that it may not extend low enough for my application.

I was surprise to find that most center channels are rated as full range speakers. Driver complement can range from 3" to 8" drivers with some type of tweeter. As a ref KEF’s were rated there F3 between 50 to 60hz using ~6.5" drivers.

Base on the above, an F3@50hz should have more than enough extension.
I decided on the 2123 for the mains, because after I compared it to the other drivers in my inventory it sounded the best period, clean, dynamic, smooth and authoritative. The 2235 makes it to 1.6k, but the 2123 as a midrange can bring it to the next level. The 2217 will cover what the 2123 can’t.


If the 4430’s were designed with a 2123 as a mid range driver, it could have become a statement speaker. Without it, it’s better than most current designs, but not as good as the best of today.

slxrti

[slxrti]

Hello i have one 2446j 2" driver + 2382a horn and 2 2226h 15" in a horisontal MTM config. The box is 2,4 meter wide . Crossover is 700 and 40 hz.

2 closed chambers 140 liter each.

regards mats

WOW , where the PIC"S. How do you like it?

btw, that won't hang on my wall.

Slxrti

[duaneage]

I would focus more on the crossover design than the low frequency cutoff. OF primary concern are the lobe patterns from the two mid-bass drivers and the tweeter in the middle (if that is what you decide on doing). The crossover point will need to be carefully selected so that it is seemless between the drivers. laying a MTM on it's side changes the pattern altogether. I would opt for 18db/oct crossovers at a point determined by the distance between the midbass drivers where you have 1/2 f wavelength or less. I think you will end up tweaking the crossover a bit so mounting them externally at first will make changes easier.


Testing should really be done at different points in the seating area with white noise and sine sweeps followed by spoken word tests. It is worth noting that dual midbass drivers are not necessary for a center channel. Vertical arrangement would work fine and even a horizontal 2 way with 2 drivers only works if the drivers are close together. A coaxial driver would work well too.

The best part of a center channel lies in the crossover design in my humble opinion.

[Hamilton]

Messing with center speaker crossover, you mean like this?

[toddalin]

My center channel:



The center channel shares a cabinet with one of my two subwoofers, in a separate chamber. Components include a W15GTI (sub), 2235 (center woofer), LE175, 075. New crossovers are under construction. The cabinet has a total volume of 10 cu ft.

Originally the cabinet was designed to hold my 45" table top Mitsubishi, so all alnico components were used to reduce color aberations. We now use an LCOS front projector system, and my Stewart Firehawk sits atop the cabinet on a custom made stand that I had a metal artist weld up for me.

[Zilch]

Messing with center speaker crossover, you mean like this?

TWEAKER!!

[andresohc]

I agree, the crossover is like a good sauce. Its a combination of craft, skill preparation, knowledge, theory, intuition and execution. It makes the meal. I can make a good sauce but not a good crossover. Ian was referring to my attempt at a center channel methinks. I am using the MTM with dual 2122Hs in order to try and push the output down to the top end of the SUB1500s which I hope to biamp. I started out with perfectly good 2123s but had them reconed by OCS to 2122Hs to get that little bit of lower extension. I am aware of the lobbing problem and will experiment a bit. If I pull this off I will have a system with similar SPLs as the 4345 and about one third the size package. I will see how bad the hole is between the subs and the 2122hs and if the elements can handle eq and power. If I dont like it I have other options for the sub.

I would focus more on the crossover design than the low frequency cutoff. OF primary concern are the lobe patterns from the two mid-bass drivers and the tweeter in the middle (if that is what you decide on doing). The crossover point will need to be carefully selected so that it is seemless between the drivers. laying a MTM on it's side changes the pattern altogether. I would opt for 18db/oct crossovers at a point determined by the distance between the midbass drivers where you have 1/2 f wavelength or less. I think you will end up tweaking the crossover a bit so mounting them externally at first will make changes easier.


Testing should really be done at different points in the seating area with white noise and sine sweeps followed by spoken word tests. It is worth noting that dual midbass drivers are not necessary for a center channel. Vertical arrangement would work fine and even a horizontal 2 way with 2 drivers only works if the drivers are close together. A coaxial driver would work well too.

The best part of a center channel lies in the crossover design in my humble opinion.

[duaneage]

I use this:

http://www.vidsonix.com/vx3/vx_aud4ind.html

Really flexible and you can use two together for a 4 way network. It doesn't have to be a science project, you can dial up some values and go with it. Once your happy with the results buy the 12 dollar caps and the 25 dollar inductors and make it permament. I like the external feature that lets you use any value you want.

I bought mine a few years ago and have used it a lot. Not that cheap but cheaper than buying a ton of crossover parts just to test with.

And no I don't work for them or get a cut. Just thought it would be interesting to show it.

[slxrti]

I agree, the crossover is like a good sauce. Its a combination of craft, skill preparation, knowledge, theory, intuition and execution. It makes the meal. I can make a good sauce but not a good crossover. Ian was referring to my attempt at a center channel methinks. I am using the MTM with dual 2122Hs in order to try and push the output down to the top end of the SUB1500s which I hope to biamp. I started out with perfectly good 2123s but had them reconed by OCS to 2122Hs to get that little bit of lower extension. I am aware of the lobbing problem and will experiment a bit. If I pull this off I will have a system with similar SPLs as the 4345 and about one third the size package. I will see how bad the hole is between the subs and the 2122hs and if the elements can handle eq and power. If I dont like it I have other options for the sub.

Hi, it seem that we have similar projects, If you don't mind me asking, which drivers are you using, what are your crossover freq.’s and the distance between each driver?

If I understand correctly the main cause of loping is the over lapping response between driver’s two drivers. Since two drivers share the same pass band loping will be inherent in this topology, Is this the case?

Can I assume the crossover will minimize the problem, but not eliminate it?


thanks slxrti

[duaneage]

Lobe problems are caused by a few things. Directivity of the drivers at certain frequencies. Crossover action and pass charactoristics. Diffraction at the edge of the cabinet and against any protruding object.

In general you want to use a steeper slope to minimize the overlap between the two drivers ( I will call the dual midbass a single driver for this example) so that sound comes from one source as much as possible. The crossover can "work" with the driver arrangement to compliment or cancel lobe patterns as needed. Sophisticated software like LEAP can calculate much of this but at the end of the day your going to have to test it.

Try and crossover as high as you can. Reversing the phase of the tweeter will "tilt" the response and in some cases remedy a problem.

It is tempting to use a crossover network someone else has or that JBL produced in the faith that it is correct, tested and approved but really at the end of the day your going to have to breadboard it sooner or later.

[Ian Mackenzie]

Because the 2212H will be horizontal in this arrangement there will tend to be a lobe with increasing frenquency b/n the two drivers. To what extent will depend on the drivers centre distances and the highest frequency in the passband of the 2122H.

I doubt if this will be too much of an issue but some practical assessment will be needed. The crossover slopes are 18 db.

The advantage of horizontal placement is however that there should be few lobes in the vertical plane other than the transition over to the 2307/2308 and its vertical polar index.

The baffle is sloped as I recall so this shoudl not be too much of an issue if you are seated watching a movie

[duaneage]

I doubt if this will be too much of an issue but some practical assessment will be needed. The crossover slopes are 18 db.


A kind way of saying your going to have to start with 18db/oct slopes and tweak some things. Not just the crossover but driver position on the baffle, distance between them ( try and butt the woofers against each other and put the tweeter above them), maybe even slope the two midbass drivers outward at an angle to change the radiation pattern. Measure the distances for these tests and record them When changing the crossover network values record at what distance between drivers they were used. If moving the drivers keep the crossover values the same. If you change too many things looking for the quick answer your going to be disapointed with the results.

Center channels are Glorious Mono. Since stereo came along the idea of making a single speaker is strange today.

[slxrti]

I agree, the crossover is like a good sauce. Its a combination of craft, skill preparation, knowledge, theory, intuition and execution. It makes the meal. I can make a good sauce but not a good crossover. Ian was referring to my attempt at a center channel methinks. I am using the MTM with dual 2122Hs in order to try and push the output down to the top end of the SUB1500s which I hope to biamp. I started out with perfectly good 2123s but had them reconed by OCS to 2122Hs to get that little bit of lower extension. I am aware of the lobbing problem and will experiment a bit. If I pull this off I will have a system with similar SPLs as the 4345 and about one third the size package. I will see how bad the hole is between the subs and the 2122hs and if the elements can handle eq and power. If I dont like it I have other options for the sub.

w

Which midrange driver are you using? What is the spacing between the drivers? Have you had any luck with the crossovers?

slxrti

sl

[Ian Mackenzie]

I appreciate your thought but are you asking or telling me what I should be doing?

Initally the system will be triamped.

As far as a passive crossover goes the starting point is the mid range aspect of the JBL 3145 network of the 4345 monitor and it will evolve from there. The more fundermental issues are match of the drvier sentivities and getting the 2122's to work properly down to the desire crossover point to the subs and get that crossover point working properly.

The rest is fine tuning. If necessary the baffle as to be modified we can screw around and get a nice peachy response curve latter on.

I doubt if this will be too much of an issue but some practical assessment will be needed. The crossover slopes are 18 db.


A kind way of saying your going to have to start with 18db/oct slopes and tweak some things. Not just the crossover but driver position on the baffle, distance between them ( try and butt the woofers against each other and put the tweeter above them), maybe even slope the two midbass drivers outward at an angle to change the radiation pattern. Measure the distances for these tests and record them When changing the crossover network values record at what distance between drivers they were used. If moving the drivers keep the crossover values the same. If you change too many things looking for the quick answer your going to be disapointed with the results.

Center channels are Glorious Mono. Since stereo came along the idea of making a single speaker is strange today.