Applying Pro Audio Speaker Designs To Make a Better Center Channel?

[Fredygump]

I want to design a center channel that is similar to a Line Array box. I want it to be just a single box, and without the line array horn.

I am interested in learning more about uncommon speakers with driver arrangements, and I want to avoid issues "MTM" style center channel speakers have with horizontal directivity. I also just want to build something unusual, with the caveat that it works well, possibly better than the generic options. My goal is to roughly match the horizontal directivity of my main speakers, which are based on 12" B&C 12FHX76 coaxials. (HF is a 60x40 degree horn, crossed at 1,200hz)

The reason I'm not going straight to using another 12" coaxial for the center channel is because it would make a really tall center!

I have made some sketches of the ideas I have come up with. I think they all have problems, but it is a place to start.

The drivers I'm using in these examples:

Beyma 5CX200FE coaxial
FaitalPRO 6FE100 (2x)
FaitalPRO LTH102 60x50 degree horn
(CD to be determined)

Requirements:

Flat frequency response down to 100hz,
Capable of 110dB+ SPL @1 meter (To achieve THX reference level of 85dB + 20dB headroom at listening position)

0: This is the base line. It is the boring design that is a little better than a plain MTM. The 5" coaxial driver moves the crossover down to around 500hz or below, so I don't get beaming from the woofers.

1: This design is my first impression of a line array box, but it also uses a coaxial driver instead of a horn. This configuration narrows the distance between the 2 woofers, so the horizontal directivity from them should be similar to the 12" B&C.

One major issue it presents is that I can't install the woofers! So I have to make a separate box for the coaxial, and it needs to be removable..to allow installation of the drivers.

2: I've seen a few photos of line array boxes that look a bit like this. But of course they are using a special horn designed for the purpose. As drawn, I'd get the ~70 degrees horizontal directivity the coaxial is rated at, but the high frequencies in the vertical axis are bouncing off the inside of the box, which can't be good. Or does that matter? I don't know!

3: This is sort of a cross between a line array box and a Genelec "The Ones" monitor. I feel like it would work, but I don't know if there would be problems with the horn being in the middle, exposed to the woofers? One theory is that the area between the two drivers is pressurized evenly, so sound is not bouncing around, so not reflecting off the horn, causing distortion, smearing, etc. The other theory is that the sound will bounce all around behind the horn, causing major sonic problems. I'm not sure which one is correct.

If sound reflecting around before it makes it's way past the horn is a problem, I can modify it to be more like design #1, where the output of both woofers is isolated on either side of the high frequency driver.


Any thoughts? What ideas are workable, and what needs to be thrown out?

 

[kiwifi]

Have you considered that the best center channel may be a phantom one? I also use coaxial L/R and found that they create an excellent phantom center, better than any physical speaker that I tried.

 

[Fredygump]

Have you considered that the best center channel may be a phantom one? I also use coaxial L/R and found that they create an excellent phantom center, better than any physical speaker that I tried.

I haven't thought about it like that. I am using a phantom center right now, and I don't have a specific problem with it. I guess I don't know if it is better than using a center channel.

I've been following Dave Rat, and one of this big ideas is that the same sound from multiple sources is unnatural and leads to acoustic problems. So he talks about decorrelating left and right channels by using multiple microphones and making slightly different mixes for left and right. This idea got me thinking that ATMOS really might be worth investing in, because it is sending completely different signals to the different speakers.

So at the back of my mind I'm thinking that if I come up with a design that works, I can use it for surrounds also. Once I build something, then I can really test if I prefer it over a phantom center? If not, it could be one of the surrounds.

I'm not in a hurry to build them all. This project will be a slow burn. At the moment I don't have many muli-channel movies, because for a long time I relied on streaming. But when I play a blueray, I'm impressed with the dynamic range and overall sound quality. So I'm starting to change my ways.

 

[Devnull]

Why not use a B&C 8FCX51?

 

[Fredygump]

Why not use a B&C 8FCX51?

I wasn't looking at that one, because when I was choosing drivers for my main speakers, I decided that I liked the look of the protruding horn on a coaxial. So I was either looking for the same thing in a coaxial for the center, or looking for a horn + woofer combination.

It does appear that the 8FCX51 exceeds my requirements--capable of roughly flat to 80hz @ 120dB.

The minimum box height would be 9", and that is with zero baffle. I'm not sure how it would look in the end. I'd be more interested in using it as a surround.

 

[Fredygump]

Well, I guess I am going to buy some drivers and start experimenting. I was hoping to discuss the physics of cabinet designs and the different speaker configurations that are possible.

The show must go on. But I'm still interested if anyone has knowledge, experience or ideas about some of these possible driver configurations.

 

[Koeitje]

Have you considered that the best center channel may be a phantom one? I also use coaxial L/R and found that they create an excellent phantom center, better than any physical speaker that I tried.

A phantom center only works if you have 1 listening position. If you watch movies alone it works great, but not if you want to watch with more people.

 

[Devnull]

Multi entry horns are definitely a challenge. You should get a copy of hornresp and spend a bunch of time simulating what you're trying to do. It's got a brutal learning curve but worth the time.

 

[Fredygump]

I realized that my biggest problem is the horn directivity, not woofer arrangement. I can't use a standard wave guide, because the directivity of horns is opposite of what I need for this to work!

The problem is the directivity pattern of the horn is proportional to the dimensions of the horn exit. It is wider in the wider dimension, and narrow in the narrow direction. But my woofers will give narrow directivity in the wide axis, and wide directivity in the narrow axis.

But then I found a clever idea on DIY forum. Pictures are my rendition of the idea. It's essentially a cone wave guide, with internal angles so the horizontal directivity (when layed on it's side like a center channel) is 60 degrees, and the vertical directivity is 90 degrees. This roughly matches the expected directivity of the drivers themselves--the horizontal is equivalent to a 12" woofer, and the vertical is equivalent to an 8" woofer. (I read the internal angle determines the directivity pattern, but I'm not sure this so simple? More to learn....)

Pattern flip: I'm learning on the fly, and trying to put the pieces together... What I've read leads me to believe that pattern flip of a horn is similar to directivity of a woofer--the frequency where pattern flip happens is based on the relationship between wave length and horn mouth dimensions

I believe the relationship is similar to the way we determine the radiation pattern of a woofer? So a horn that is 8" wide will start controlling directivity at the same frequency a 8" woofer will start beaming? So if the horn opening and the woofer diameter are the same, the directivity will be fairly even, even if the crossover can't be at the exact right frequency?

The woofer arrangement here would be described as a diffraction horn? I just learned what that is! I've been looking at what Genelec is doing in their "The Ones" series...I realized that the output from their "concealed woofer" design would be directed foreward because if diffraction around the baffle edge, but I didn't know that there is a type of horn that works this way.

Anyway, it's looking like something. This is a first draft, but I think it'd work. I guess I'm just copying Genelec now, except they have a much more clever wave guide.

If I got this completely wrong, partly wrong, or whatever...I'd appreciate any input. It's kind of deviating from the original concept, but I guess that's speaker design?