Building an array with 3" widebands?

St= Z

Member
2002-02-07 5:59 pm
Doing another project (which is comming along well, thanks to everyone) I was pointed to these little wide range drivers

http://tb-speaker.com (under tweeters the 319'S)

I always wanted to make an array but never thought I had the space or could afford too. Well these babies are darn cheap and theyre just 3" so the array would be like a narrow little column.

Anyway theres very very little info about building arrays on the net. So I'm in the dark about calculations and measurements. Would these drivers even work in an array? how many do I need for an 8 foot tall array? how do I figure how to space them? Wired series or parrallel?

Alot of questions but if I could get some basic pointers about arrays it would be much appreciated.

Thank you
 

SteveG

Account Disabled
2002-01-07 7:20 pm
Newton Falls, Ohio
arrays

OK, so I'm gonna open up a big old can of worms here, but here goes:
I am not convinced that arrays aren't inherently flawed designs, at least for home use. The technology got it's start in pro audio, which uses carefully designed (read limited dispertion) horns that were meant for a specific radiation angle to cover an extremely large area at high SPL. I do not understand how this can be applied successfully to home audio use. The problem to me seems to be that there are too many point sources which create delays in the signal from speaker to ear. With each of these distances being different, I don't think it is possible to have a coherent sound with a smooth frequency response. I believe that a lot of people building these systems are overlooking the obvious, because they look really impressive and have great sensitivity. It's the "maxell speaker/guy in a chair effect":) In the array designs the wiring is connected so that the speakers that are further off the vertical axis are producing less volume. It seems that they are trying to minimize the inherent problems of the design with this approach. What is the point? I just don't get it. I could be totally wrong!
If you are considering this type of a design, you need to cross the midwoofers over at a somewhat low frequency (There is a formula for this) and go to a closely-spaced group of on axis ribbon type drivers for the highs. I think that this would negate the point of using the full range drivers for this type of system. The wavelengths at the higher frequencies are so small that you have to have the high frequencies radiate very closely together, or it will turn into a total mess.
I was actually considering the same thing with the same drivers, until I thought it through and concluded that it didn't make any sense. In thinking about it, I considered using less drivers closer together, and putting them on a curved baffle to adjust for timing differences. Maybe even use a series of crossovers (1st order) with a bunch of the same drivers to make a muli-way system where each driver has a very limited frequency range. I may try it some day. For now, I have other areas that need more attention. If you do try it, let me know how it turns out. This is obviously just my 2 cents worth.
Steve
 

St= Z

Member
2002-02-07 5:59 pm
Well, what you are saying makes sense, but I really liked the sound of the array I once heard, so I wont question it just try to enjoy it , this is coming from a guy whos owned quality box speaker, Magnepans, and now have a Apogee Centaurus' (the best of both worlds ;) ) , after being through all these I've come to terms that there is a certain amount of theory that really goes out the window in the real world.

So while a respect and understand what your saying, theres no substitute for the real world test, so what the hey I wanna line up a hella of a lot of wide ranges and see how it sounds, Hell that maxell guy looks like hes having fun! At the same time I want to make it the best I can so I would like to apply some generaly regarded theories to it (line array).

If its bad, I was thinking I could take the drivers and make my sister a sort of "Bose cube" system clone (please no comments on Bose).

With the different volumes at both ends isnt that a "Bessel array"?
 

SteveG

Account Disabled
2002-01-07 7:20 pm
Newton Falls, Ohio
Hey, if it blows your hair back, then that's all that matters, right? ;) just like the maxell guy.
Seriously, I don't get the concept, but if it sounds good to you, then cool. I'm really not an audio snob. I just can't imagine a system like that imaging very well.
Let us know how it sounds!
I would at least cross over to some centrally mounted tweeters, otherwise you will get serious comb filter effects with hot and dead spots. From what I read, this happens pretty early in the frequency range, and most designs I have seen cross over at or under 2Khz. It has something to do with the size of the woofers too. I really did look into building one for a while. The idea of 12 woofers thumping away sounds awfully enticing!
As far as Bose goes, well, I won't go there. Did someone say Better Off with Something Else?
 
And i posted this link in a thread earlier in the week (Line Arrays). This is a paper on line sources and graphically shows the comb effects.

http://homepage.mac.com/hd3dstudios/line-source.pdf

Line arrarys fascinate me as well -- although the only ones i've had any experience with are a set of acoustat 1+1s which, excepting a 2" chunk in the middle is one 8' high radiator.

The advantages of a LS might well outweigh disadvantages in your situation... certainly the PipeDreams usually get positive press.

dave
 

SteveG

Account Disabled
2002-01-07 7:20 pm
Newton Falls, Ohio
We have now come full circle

ST=Z,
It seems you have found my initial source of interest in these drivers... funny how that works, isn't it?
Here's an idea for you-

four tb's vertical with a dome tweeter between the middle two. Make the enclosure big enough to mount an 8" woofer internally in a double-vented bandpass configuration (make sure tb's have their own enclosure of course). Cross the tb's on the low end to blend with the acoustic filtering of the bandpass. Use a 12db/oct on the top to the tweeter to avoid comb filtering effects. Sounds like fun. I might have to try it!
 

St= Z

Member
2002-02-07 5:59 pm
Yes thats pretty much along the lines of what I was begining to think and then I realized I would just be building totaly conventional 3-way speakers, which sort of voids much of the fun and uniqeness of the who'll thing.

My current thought is building mini cubes, two per unit (much like the "B" word), actually I was thinking more like mini tubes. My thought was I could get some 3" PVC piping, and cut them down to 4" sections, pop the driver in one end, pop a 3" round terminal cup (thank you MCM as luck has it) in the other, Voila!! I mean does it get any easier than that.

Now heres the problem, which I'm not sure is a problem, does pvc piping come by size of hole or outer width? if outer width do you think they make a 3 1/4" or something, do the 319's need exactly 3"?

My other thought would be to put 3" pipe peice inside a 4" pipe peice, then pour concrete between them to form a little concrete tube, maybe incorporating some steel mesh inside the mold for strangth. pop in the simple 3" driver and round terminal cup.

Any thoughts on these ideas?
 
PVC rules!

PVC pipe is dimensioned by the inner diameter.
I have also been tempted by stacking modules of lengths of
pipe with a driver at one end and a PVC cap on the other-even to just make a D'Apollito config. The 3" terminal cup is a pretty cool idea, but it must be 3" or less on the outside.

The shape certainly is inherently a good one!
Rmember that at a much greater cost, PVC is available in 6" diameter.

I recently made some floorstanders by taking a 3' piece of 3"
black sewer PVC and adding a 90 degree elbow on top.

The 3' length is vertical as a stand and the 3" coax car speaker was glued with silicone into the opening of the elbow. I stuffed the pipe with pillow filling. I then added a 3" to 4" coupling to the bottom to act like a base, added a carboard disc and filled the coupling with cement-mainly for weight.

Looks like a periscope, my wife hates them. I have 'em in my office.

The coax Boston Acoustics sound pretty darned good-no bass though

A 3" pipe has a pretty small internal volume. I seem to remember that the over 3' length of the assembly only had an internal volume of a 6" cube.
 

St= Z

Member
2002-02-07 5:59 pm
Yes yes, this can all be done with such ease, the terminal cups are supposed to be 3" so I should be able to just pop them in like the drivers.

what about about adding some more elbows and curving it up and down to create a transmission line?

I don't understand the math with these, the TL should be 1/4 the wavelength of the drivers lowest rolloff?? I dont understand how that computes really anybody have any idea what this would be for the 3" tangbands?
 

St= Z

Member
2002-02-07 5:59 pm
how important is tapering in TL's? would a notched taper be of any use, or does the lack of angles nullify any benefit?

Do you guys think a TL will help squeeze some bass out of these puppies? Any opinions on the aluminum TB's vrs Paper vs Polypropalien (Alnico magnet)?
 
PVC Pipe TLs

St= Z said:
how important is tapering in TL's? would a notched taper be of any use, or does the lack of angles nullify any benefit?

Do you guys think a TL will help squeeze some bass out of these puppies? Any opinions on the aluminum TB's vrs Paper vs Polypropalien (Alnico magnet)?

Hey guys...

This thread leads right into a set of project pages that aren't "live" yet. Here is a preview of Laying Pipe with Audio Lego (please report any QC issues to me -- in particular the gamma on the pictures & whether or not the degrees symbol is showing up on Windoz boxes). Rob has come up with a very elegant solution to tapering the line and as a side benefit providing a good place to mount feet.

With the restrictive nature of the PVC, any opportunity to reduce the magnet's shawdow is a good idea -- what is the number of the TB with the alnico magnet?

dave
 

St= Z

Member
2002-02-07 5:59 pm
I'm sure this isnt the first time this has been done but great minds do think alike (or so I would like to think). Heres a totaly unscientific render of what I was thinking.

[IMGDEAD]http://www.stefanziff.com/tbtlspeaker.jpg[/IMGDEAD]

It seems like a cool project because it can modified and changed so easily, like legos.

as best I could figure the three models are..

W3-315SA - Alum
W3-316SA - Paper
W3-532S - Neodyminium

Found in tweeters at http://tb-speaker.com

Why is PVC restricive? it seems that the depth of the magnet is fairly workable no? Anyway I dont have a very good understanding of speaker specs yet, but the paper one looks the best to me.
 
St= Z said:
Why is PVC restricive? it seems that the depth of the magnet is fairly workable no?

"babysnakes" -- another quiet allusion to the late great Frank Zappa

If you are not careful, a big magnet on a small speaker can almost completely block the entrance to the line since the magnet diameter could approach the inner diameter of the pipe.

dave
 

SteveG

Account Disabled
2002-01-07 7:20 pm
Newton Falls, Ohio
Peeing on your campfire again

Not to throw a wet blanket on this whole thing, but why would you do a TL with a little tiny driver that has very limited excursion and no bass response?
It is a very cool idea though, and would be really neat if you could find some large diameter pipe (sonotube?) and use a bigger driver- maybe one with a concentric tweeter... now that sounds tempting.
 

St= Z

Member
2002-02-07 5:59 pm
Well, why not !?

For me its toying with the idea of getting the one driver to deliver as wide a range as possible, in this case that means trying to extend the bottom end. I have a sub to fill the rest of the bottom end, infact I have a pair of apogees to fill the top and the middle, the point is to have a little fun with a pair of drivers that cost 7 bucks.

Considering how much youve considered these drivers I think you should stop over rationalizing everything so much, drop $20 and take the plunge, what do you have to lose?