Mini line array computer speakers

[Mazeroth]

I picked up some Hi-Vi 3" buyouts from PE that have a very flat frequency response up to 8khz that I wanted to put to use in some computer speakers. I usually like to go a little overboard and was thinking of a design like this:

This is a near perfect scale and setup of my computer desk, including the HK receiver and my 22" widescreen LCD. I also have my copy of SoundEasy sitting next to me!





They're six ohm drivers so wiring each set of 3 in series then paralleling the sets should yield close to a 9 ohm load. I'd like to put a tweeter in the center (as in the pictures) and cross it somewhere in the neighborhood of 3000-4000hz. I was thinking about the $5 3/4 Neos from PE but figured I'll already have around $70 invested per speaker with drivers, wood and crossover parts that I might as well step it up another $10 - $20 and get a better tweeter. I was looking at the Tang Band neos that PE sells and also the small flange Vifa XT25 neo. Do you guys have any thoughts on this or any recommendations I should do before I build a pair?

As far as box construction goes they're roughly 26h x 5.5w x 8d. Ported in a 0.3 ft3 cab to 80hz with 2 chambers each having a 1.5" x 4" port. Sealed they cut off waaay too early (in 0.3ft3 F3 = 163hz and F10 = 77hz) so I think ported will be best. The sub is a single Dayton DVC8 with dual 10" Dayton passives.

As always, any input is greatly appreciated!

[nerd of nerds]

I picked up some Hi-Vi 3" buyouts from PE that have a very flat frequency response up to 8khz that I wanted to put to use in some computer speakers. I usually like to go a little overboard and was thinking of a design like this:

This is a near perfect scale and setup of my computer desk, including the HK receiver and my 22" widescreen LCD. I also have my copy of SoundEasy sitting next to me!





They're six ohm drivers so wiring each set of 3 in series then paralleling the sets should yield close to a 9 ohm load. I'd like to put a tweeter in the center (as in the pictures) and cross it somewhere in the neighborhood of 3000-4000hz. I was thinking about the $5 3/4 Neos from PE but figured I'll already have around $70 invested per speaker with drivers, wood and crossover parts that I might as well step it up another $10 - $20 and get a better tweeter. I was looking at the Tang Band neos that PE sells and also the small flange Vifa XT25 neo. Do you guys have any thoughts on this or any recommendations I should do before I build a pair?

As far as box construction goes they're roughly 26h x 5.5w x 8d. Ported in a 0.3 ft3 cab to 80hz with 2 chambers each having a 1.5" x 4" port. Sealed they cut off waaay too early (in 0.3ft3 F3 = 163hz and F10 = 77hz) so I think ported will be best. The sub is a single Dayton DVC8 with dual 10" Dayton passives.

As always, any input is greatly appreciated!

Did you actually measure the drivers or is there a spec sheet on partsexpress?

It looks like its got the same or a very similar motor to the B3S and I was planning a similar project using the B3S and the vifa D26 driver in a mini waveguide but if these are pretty similar and are cheap...

[cjd]

2.5.5 way may net better results that close...

[Mazeroth]

Did you actually measure the drivers or is there a spec sheet on partsexpress?

It looks like its got the same or a very similar motor to the B3S and I was planning a similar project using the B3S and the vifa D26 driver in a mini waveguide but if these are pretty similar and are cheap...

I haven't measured them and there isn't a posted response graph but I just guessed they'd have pretty clean response up to 8khz, seeing how every one of their other 3" drivers hits 8khz with ease. But, you never know. These could suck!

2.5.5 way may net better results that close...

I could see using a 2.5.5 way if this were going to be a horizontal center but can you elaborate on using them in the intended application?

Also, I got bored and was curious what an 8 driver array would look like. I kept the tweeter near ear height and even added a base, you know, for a little stability! :B

[Brandon B]

You forgot the computer.

(j/k, I realize it probably under the desk or something).

BB

[cjd]

Eh. Go look at the ripple tanks.

But, you say you'd do it for a center.. now, why is that? Think about how close you are - how much vertical movement puts you 15 degrees off-axis? How much movement front to back? Unless I'm completely forgetting stuff (and I should add, that's entirely probable right now, given work stresses) line arrays don't work as well in real nearfield.

C

[Mazeroth]

Eh. Go look at the ripple tanks.

But, you say you'd do it for a center.. now, why is that? Think about how close you are - how much vertical movement puts you 15 degrees off-axis? How much movement front to back? Unless I'm completely forgetting stuff (and I should add, that's entirely probable right now, given work stresses) line arrays don't work as well in real nearfield.

C

I've always invisioned a line array producing a continuous cylindrical wavefront that pretty much builds up on itself for how many drivers are in it. That would be different to having multiple point source drivers producing a spherical wavefront, like a conventional MTM does. Here's an illustration from Jim Griffin's white paper:



From Jim's paper I get the idea that the speakers I'm thinking of building will produce a continuous cylindrical wavefront and won't have any cancellation effects. However, I have read that a line array is useful at farfield distances, where the arrays are still in the nearfield, in that where some drivers are cancelling and lobing the others fill in the gaps? I'm not a phsycist by any means so I'm not quite sure if that adds up or not? I'm sure there's also something to do with the line extending nearly from the floor to ceiling, where in my case they won't, and that also limits the amount of floor and ceiling reflections that cause cancellations and what-not.