Center Channel considerations...criticism wanted!

[cobbpa]

I believe I have selected the speakers I would like to use for the center channel I will eventually build (in the money-saving phase right now..). I would like to use this woofer: Peerless 8" Nomex 830869,
[IMG] [/IMG]

the Seas 27TDFC/TV Shielded Textile dome tweeter,
[IMG] [/IMG]
and copy this B&W design:


It will be going underneath my TV in the entertainment center, so the magnet for the woofer should be almost a foot away from my tube; I'm not worrying about its unshieldedness. I don't have testing equipment & realize this poses a problem...but Zaphaudio has a design using this tweeter at an Xover point & slope that seems appropriate (LR4, 1700 hz) for this design. That should put the woofer's 3 khz peak attenuated > 12 db, right? I picked the woofer because it's efficiency matches that of the tweeter. I would like to seal the box, just based on wanting best quality sound out of my center channel & that sealed is supposedly the better alternative for that, but would that cost me efficiency? I would rather port it & have a simpler (non-attenuated tweet) crossover if that's what it takes. Also, I am afraid a ported box may drop it below the minimum 6 ohm impedance the graphs shows. I realize, without testing equipment for my specific (probably unique) box, it's not going to be at it's full potential, but I'm guessing that for <$130, can compete with most store models under $200. Maybe? Or if anyone in western Ohio / Indiana / even Chicagoland wants to do some testing.. :T The graphs are both fairly flat in the regions I will be using them in, I think that helps keep it simple. Eventually, I'd like to build ported MT or MTM mains with this mid/tweet combo. Basically, I'm posting this to be highly critiqued. Look at the numbers, the graphs, the designs, & let me know all that I am missing. Please, I love learning about this stuff, and learning by doing

edit to put graphs in..also just started reading the Modula MTM 1 thread with nearly the same tweet and it has crossover schematics...hmm

[Dotay]

Sorry I can't help with this design but I just wanted to welcome a fellow Hoosier to the board. It seems there is a disproportionate number of people from Indiana on this forum. Where exactly do you live?

[cjd]

Well...

Just glancing at the graph of the Peerless, I would be shooting for a crossover of around 1200Hz, maybe up to 1400, and 8th order slopes. The tweeter you've picked can handle that. Crossover may be more pricey than you anticipate in this case, but I think you will find it has the potential to compete with more expensive commercial speakers as well. You will lose some of the sensitivity of the mid-woofer perhaps due to baffle step, though it gets more complicated given your placement. Putting this under your TV will significantly change how this reacts and what kind of baffle step will be needed vs. "anechoic" and vs. even in-wall. It may introduce other issues if there is a "hollow" around it - better results may be reached by actually making the whole stand be your baffle.

I'm north of Chicago and would be up for doing some measurements. But that's a long haul for just that. If you happen to hit the 6 flags or something, that's right up the street. Otherwise, I don't know.

When it comes to this kind of thing, if you are not going to be doing your own measurements it is almost always most sensible to pick an existing design. And, my target for breakup is usually a minimum of -30dB, I prefer 40dB+.

Oh yeah... 1 foot away, I *would* worry about shielding.

At the ~$130 pricepoint, the RS150 based MTM I've put together may be a better option. There are plenty of other great designs in this range as well, but being the type that just does my own thing I have this bad habit of not remembering them. :|

C

[cobbpa]

Sorry I can't help with this design but I just wanted to welcome a fellow Hoosier to the board. It seems there is a disproportionate number of people from Indiana on this forum. Where exactly do you live?

Actually...i'm a Buckeye :P I'm west of Dayton in Ohio, a small town called Eaton. It's about 10 minutes from Indiana & I make it up to Chicago a few times a year. Sorry to disappoint!

[Dotay]

Actually...i'm a Buckeye :P I'm west of Dayton in Ohio, a small town called Eaton. It's about 10 minutes from Indiana & I make it up to Chicago a few times a year. Sorry to disappoint!

Doh! I guess that's what I get for making assumptions...strike my previous comment from the record. ops:

[cobbpa]

Putting this under your TV will significantly change how this reacts and what kind of baffle step will be needed vs. "anechoic" and vs. even in-wall. It may introduce other issues if there is a "hollow" around it - better results may be reached by actually making the whole stand be your baffle.

I'm north of Chicago and would be up for doing some measurements. But that's a long haul for just that. If you happen to hit the 6 flags or something, that's right up the street. Otherwise, I don't know.



Oh yeah... 1 foot away, I *would* worry about shielding.



C

I had actually thought of this, since it is going into a hollow, of making an oversized baffle so the sound doesn't "leak" into that hole..good to see someone else suggesting a similar idea.
As I said in another post, I'm in Chicago a few times a year..so this isn't out of the question, but it wouldn't be for a while. Maybe expect a PM a month or two down the road!
Shielding...I have a couple of other speakers w/large magnets laying around. I should see what they do at this distance, as it's something that may totally call for me to alter my design. You noted that the baffle could affect efficiency; how? Is this at all preventable, by, say, a narrower or wider baffle?
Also, interesting on wanting the breakup -40db...sounds like a pretty lofty goal, but not unattainable with these speakers. I should start checking prices & such for the required crossover components, be sure I can still keep this in budget with this steep of a slope. Any comments on sealed vs. ported center channels? I'll check out that design you mentioned, too.

thanks to you both for a nice welcome to the forum :T

[cjd]

Well, sensitivity will be changed for any driver on a baffle due to baffle step. The wider the baffle, the lower this point occurs, but the net result is that it's going to lower sensitivity. You'll probably not see the full -6dB loss because it will be a large effective baffle (the TV will become part of it) and because it is (presumably) fairly close to a wall. May only be a couple dB loss.

It happens because driver radiation switches from 2pi (hemispherical) to 4pi (spherical) space relative to the baffle width and where wavelengths coincide. Baffle step. There's some confusing but good info out there if you google for it, and I'm not so good at explaining it.

And, I'm not so worried about "leaking" into the hole as I am about what kind of resonances or odd reinforcement or nulls it could create. Which is I guess sort-of the same thing.


C

[cobbpa]

I getcha...I'm going to try & read more about BSD, I've picked up bits & pieces from different sources, never dove right in though. I'm wondering, could it be made large enough (the baffle, and the TV effect you mention) that it could happen as low as the speaker's roll off frequency? Thus, I suppose, just making roll off steeper? I have no idea if this is possible, just throwing it out on. Maybe I'll find it in some reading..

[cjd]

Diffraction is related, but different.

This is just baffle step.

Um. You could make it wide enough, yes. Infinite baffle (in-wallish) is the "ultimate" in this regard.

It doesn't make *that* much difference - you deal with it in the crossover regardless.

C

[cobbpa]

Understood. I also read some...using calculations, and I assume the formulae are just best estimates, the 13" height my baffle is likely to take the form of means an f3 at 350 hz, rise starting at 130, and ends at 2086 hz. But, thinking about how you said I probably won't get a full 6 db loss makes sense. Here's a pic:



The speaker will be going where the Xbox currently. The front baffle will most likely fill the total width of that opening. Your theory about the TV seems to make sense. Also, the additional flat surfaces to either side could help reducing the loss. Or am I reading everything wrong here?
Notice this is in a dorm room, and next year will be in an apartment; listening distance will only be 6-10 feet in any instance. I don't know exactly what that means, but I know it will eventually have to be considered

Anyone else wanna give comments, conccerns, shortfalls of design, etc.?

[cjd]

The speaker will acoustically "see" the TV. In your case, it will also "see" the sub and everything next to that... at least to some degree.

All needs to include it when calculating baffle step. Which is why it's simply much better to measure in-place to really understand what is going on.

And, just putting the CC in the space under the TV is a recipe for unpredictable results.

How about building yourself a new TV stand that IS your center channel?

C

[cobbpa]

The smaller measurement will be more like 3 feet then. Interesting.

Yes, I have considered building the enclosure in such a way to place it on that shelf, so it slides in perfectly to meet up with the cabinet door & shelf the TV is on.
Measuring in place...I guess I will have to see what I would need to do this. Sounds fun. Maybe I should be reading more about bucking magnets & how that will affect the woofer...that might be a simpler option, to place it on the tv then.

[cjd]

Bucking magnets, sure.

Placing it on the TV still means you have to consider that when looking at baffle step.

At any rate, you can't avoid it given where you're looking at placing this. It still will likely be less than 6dB but it could still prefer 2-4dB of BSC which brings overall sensitivity to 86-88dB.

C

[cobbpa]

This points attention to my another of my questions...sealed vs. ported. I can assume greater sensitivity with ported, but will that hurt the voice reproduction of movies & such? If it was just a music speaker I'd have no problem porting it, but having an 8 reproducing voices seems like sealed may be smart. Should I be concerned, or go ahead & port like the oroginal design I'm trying to clone?

CJD, thanks for all the help & such so far..I'm hoping someone else may chime in soon, so it doesn't seem like i'm asking all this only of you. Freel free to keep contributing though

[cjd]

Ported vs. sealed will not change the sensitivity at all.

It may change the maximum power handling, it may not. It supplements the bottom end of the system, though often requires a bigger box. It increases group delay vs. sealed. Many things. Whether it is important to port would depend in part on where you cross to your sub, and whether the crossover (particularly if it's the built-in one in a receiver) expects the speaker to be flat below crossover, or relies on the slope imposed by the box and driver interaction.

A balanced design will not have difficulty with vocals or any of the rest of the sound that comes from a center channel - indeed, a center is as important as your main L&R (and I would suggest considering mains with the same drivers...)

C

[cobbpa]

What would be the end result of testing for baffle diffraction? Attenuating the tweeter? Crossing over at a different frequency? Completely throwing away the design? :

[cjd]

Diffraction? Or baffle step?

In either case, you simply work it into the crossover. It will be there no matter what you do. The only reason it's gone this far is your assumption that your drivers were the same sensitivity. They won't be once baffle step is taken into account. In fact, were you building this for "ideal" in the middle of the room placement, TWO of those woofers would be a perfect match for a single tweeter since baffle step would reach a full =6dB at some point.

You can help a lot with diffraction (not step) with the application of wool felt.

You can model baffle step (and diffraction) if you have Excel on Windows by downloading the FRD Consortium's Baffle Diffraction Simulator. It's a bit of a learning curve to really understand what it's doing.

C

[cobbpa]

Interesting. I'll download & play If I can't get testing equipment, I'm guessing that basing a crossover on calculated frequencies & numbers is better than nothing, right?

[Kingdaddy]

I like baffle-less designs, especially for center channel, it seems to make the off axis response and detail better then other designs I've tried.

[cjd]

Interesting. I'll download & play If I can't get testing equipment, I'm guessing that basing a crossover on calculated frequencies & numbers is better than nothing, right?

Depends. Getting acoustic offset right can be critical. I've been fairly successful doing this with 2-ways crossing higher, but the 3-way attempt was a joke (fortunately, I was concerned enough that I measured and modeled before purchasing crossover components...)

It doesn't handle more complex interactions, just basic. And there is no way with this tool to say "this is only going to be a foot away from the wall".

C