Stupid Cheap Line Array

[gmcalabria]

All,

I have another dumb idea and I'd like to know why it wouldn't work. Or it might work, but something tells me this is too easy...

https://www.madisound.com/store/prod...oducts_id=8988
Case of 30 for $60...

I'm thinking, for my intro into line arrays, that this would be a near-perfect driver. It's not one of those nice clean 3" drivers, but it has more bass, and it's still pretty small. It looks like it'll go low enough to cleanly mix with a sub. The harmonic distortion curves seem ok, but I also don't really have a reference as to what's good and what's bad.

So, the design is that driver, along with the (in)famous Goldwood GT-1005. Actually, I was planning on using the GT-1001 since I think the flush mounting looks awesome. Of course, I thought, for some reason, I had a frequency response graph for these tweeters, but without one I think I may have to pass..

The plan is to wire the 5.25"ers either 3 or 4 in series, those sets in parallel (16/side). If I ran this as a free-standing speaker, I'd run one set of 4 drivers as a 0.5 way for BSC, but I'm also thinking about making these into a Murphy corner-line array design. As sensitive as this tweeter is, I was thinking about running a single driver it in the middle of the array, a sort of a MMMMMMMMTMMMMMMMM design. Is this problematic?

I haven't thought much about the crossover yet, it doesn't seem like it would be very tough. I've been experimenting with 1st/2nd order low/high and 2nd/2nd. I'll probably need some other woofer correction, especially if I cross it pretty high, but it seems like these drivers are pretty damn usable for the price.

Let me know what you think.

Greg

[weltersys]

For a line array to work as a line array at high frequencies requires drivers to be no more (preferably less) than a wavelength apart, that is about 5/8 of an inch for 16K.

Line arrays generally use special horns that emulate the response of a ribbon driver for frequencies above 1000 to 2000 Hz.

The 5.25 would be OK perhaps to 2000 Hz from a line array standpoint.

Above that, they are quite ragged in response, then roll off anyway.

[gmcalabria]

Originally Posted by weltersys

For a line array to work as a line array at high frequencies requires drivers to be no more (preferably less) than a wavelength apart, that is about 5/8 of an inch for 16K.

Line arrays generally use special horns that emulate the response of a ribbon driver for frequencies above 1000 to 2000 Hz.

Less than a wavelength from diaphram to diaphragm or center to center?

I'll be searching, but any leads on building one of those special horns? Why do so many people (including some old Infinitys) use dome mids and the polycell tweeters considerably farther apart than that? The idea makes sense to me, but it seems like lots of designers seem to skip that rule.

What about putting one very high sensitivity tweeter in the middle of the array? Theoretically, So long as the 5.25" speakers are less than 6.78" apart (2000hz), then the line array shouldn't be "broken", right?

Greg

[Don Hills]

Originally Posted by gmcalabria

Less than a wavelength from diaphram to diaphragm or center to center?

Centre to centre.

Originally Posted by gmcalabria

... Why do so many people (including some old Infinitys) use dome mids and the polycell tweeters considerably farther apart than that? ...

The spacing is only critical for multiple drivers (for example, two mids) which are reproducing the same frequencies. The mid-to-tweeter spacing is not critical because they reproduce different frequency ranges, so there won't be any interference between them. (However, the mid-to-tweeter spacing is important for other reasons related to crossover design etc.)

Originally Posted by gmcalabria

What about putting one very high sensitivity tweeter in the middle of the array? Theoretically, So long as the 5.25" speakers are less than 6.78" apart (2000hz), then the line array shouldn't be "broken", right?

In theory, it could be made to work. You'll need to design the crossover's phase characteristics to properly integrate the tweeter and mids, as you would for a conventional mid-tweeter layout. Personally, I'd try to keep the driver spacing to 1/4 wavelength or less at the maximum frequency to be fed to the array. You may get away with more depending on the performance of the drivers.

[DrDyna]

Those drivers look a bit like they were designed to be installed in a console television O_o.

If you actually order and build a pair of these, you'll have to let us know how it works out. I might be interested in a few cases of them at those prices as well.

[gmcalabria]

Originally Posted by DrDyna

Those drivers look a bit like they were designed to be installed in a console television O_o.

I agree, but the price and frequency response curve seem to suggest that they're a half-step better than that. Plus the distortion curve seems to be at-or-below the going rate for that price range.

Originally Posted by DrDyna

The spacing is only critical for multiple drivers (for example, two mids) which are reproducing the same frequencies. The mid-to-tweeter spacing is not critical because they reproduce different frequency ranges, so there won't be any interference between them. (However, the mid-to-tweeter spacing is important for other reasons related to crossover design etc.)

What I meant was, there's an old Infinity line array system that used a line of 6 mid-domes and 6 or 7 polycell tweeters, and within each line, the spacing was quite far apart. I wasn't describing the spacing between the mid line and tweeter line.

If I did go with the MMMMMTMMMMM design, I'd be worried about having awful vertical dispersion. If I went with a line of tweeters, I'd be worried about finding a dome (or ribbon) cheap enough to mount 5/8" C-C that could go down to 2khz. Would I be right in assuming that if tweets are spaced 10/8" C-C that I would have comb filtering problems at 10khz and up?

[weltersys]

Originally Posted by gmcalabria

Less than a wavelength from diaphram to diaphragm or center to center?

I'll be searching, but any leads on building one of those special horns? Why do so many people (including some old Infinitys) use dome mids and the polycell tweeters considerably farther apart than that? The idea makes sense to me, but it seems like lots of designers seem to skip that rule.

What about putting one very high sensitivity tweeter in the middle of the array? Theoretically, So long as the 5.25" speakers are less than 6.78" apart (2000hz), then the line array shouldn't be "broken", right?

Greg

Greg,

Spacing of drivers on "regular" speakers may be arranged to avoid diffraction problems, or make the dispersion better in the crossover region. Sometimes an odd arrangement is chosen that is a compromise allowing the speakers to be used horizontally or vertically.

Your math seems OK, 6.78" center to center sounds right for 2K.

I have built "special horns" for line array use, based on Tom Danley's Paraline design used in his Genisis and Jericho horn and VTC's line array.

The VTC Paraline would be too narrow in vertical dispersion for home use.
The design could be modified to have more vertical dispersion, but a regular conical horn would be far easier for you to build.

Having tried line arrays for home use, I would not advise using them.

A Unity or Synergy horn design will give you pattern control with out all the comb filtering the line will have.

That said, I'm using a simple 2 way with subs for my home stereo, though I'd like a Synergy design, the cabinet depth and difficult crossover have kept them out of my living room.

[zobsky]

I norder to keep costs low, use a vertical wide disperson piezo array for treble if you want to cross over low. Something like this : Goldwood GT-1016 2" x 5" Wide Dispersion Piezo Horn Tweeter

[kipman725]

The Murphy Corner-Line-Array Home Page
this is an alternate view on the comb filtering. As the troughs are very high Q they should not actually be very noticeable as long as the general high frequency attenuation trend is equalized out. In a real room you get lots of cancellation causing similar comb filtering anyway. You want to use a smaller driver though as the horizontal dispersion of such a large driver won't be good, don't go >3" dia of the radiating surface (some drivers have large surrounds like mark audio).

*After messing around with peizos my conclusion was that they are completly not workable as tweeters due to very uneven frequancy response that would require digital filtering to correct.

[18Hurts]

Sounds like a project I'm working on in the garage

I used those Sony 5" neo woofers and literally overlapped the frames so they are 5.58" center-to-center. The thin sheet metal frames help although I did cover them with plumbers putty to damp frame ring. The crossover at 4050 Hz at 12dB slope to those Audax 10mm tweeters for 50 cents a pop--all 48 of them. To protect the little tweets, I've used a 24dB/Oct filter at 4,200 Hz and spend my days waiting for the shipment to arrive.

My woofer line is 5 feet 7 inches tall and the tweeter line is 5 feet 8 inches tall. My woofer line hits the one wavelength at 2,420 Hz so I get some beaming there until 4050 Hz when it starts rolling off (Linkwitz-Riley -6dB XO type) I'll also get beaming at 10KHz+ from the tweeter line but such is life.

How do they sound? Well, I don't have the crossover parts yet but 48 tweeters running a test tone at 4,200 Hz will wake up the dead! I ran the 12 five inchers "full range" to break them in and check for leaks (sealed box) After a day or two, they seemed to sound better although they have no high treble. They are loud! They don't change their sound if I'm sitting or standing and have a really "big" sound to them.

Once I get the crossover parts in, I can build the filters and test the 6 foot tall monsters out. Their life is a "garage speaker" and I want to see if the vertical line array overpowers reflections from the cement floor and metal roof.

I went with the Sony 5 inchers since they have rubber surrounds, polypropylene coated paper cones and full metal grills to protect them from garage involved flying liquids, parts etc. The boxes are "PA style" with corner edge hardware, handles in the sides and 1.5" tall rubber bumpers at the corners so not to cave in the grills when/if it falls over. Eventually, a tapped horn sub will take over at around 70Hz to get the party going.

The object of my garage speaker mess is to see if they work decently enough in the acoustic hell of a cement floor/metal roof garage. I'll stack them on top of each other floor-to-ceiling, run them in mono and play a guitar through them. Infinite line array theory anyone? If I actually like the sound of them and would feel the need to upgrade in a few years, I'd go with 3 to 4" woofers and keep the same tweets. The Dayton Audio 3.5" full range looks like a decent driver to take the place of the cheap 5" Sonys eventually.

After all, stuff a 100K speaker in a garage with a metal roof and it won't sound very good...maybe the line array will work, maybe it won't. Either way I'll have something very durable, takes up the floor space of a sheet of paper, very, very LOUD and the locals think it is the coolest speaker they've ever seen. There is something about a pair of speakers running 24 woofers and 96 tweeters that entertains people.

I'll let the world know how and if it works. One of the DIY'ers built something like it and said it sounded like a "velvet hammer". That free shipping from Parts Express takes a looooong time...