James' Line Array speakers

bar.gif

The Line Arrays located in my Home theater/ Listening room.

bullet.gif

You can see my prototype MTM center array experiment with about a 20 degree sweep back of the mids, to reduce combing. This layout worked fine, however I will be going with a TM array , basically 1/2 of one of the mains, for a better match.
These arrays are a based on the Linus Array which is a groundbreaking project by Rick Craig and Jim Griffin demonstrating an affordable line array that can be built by DO IT YOURSELFERS. The cost of the standard Linus Array is about $750.00. This design provides some of the advantages of high end audiophile speakers that cost dozens of times more money.
Basic Description:
Design:
Based on the Linus Array by Jim Griffin and Rick Craig, with changes made for perceived limitations in the original Linus Array design.
Drivers: (Per side)
8 Stryke Audio RTW2 tweeters. (Identical to the Dayton PT2's) $27 each.
9 Vifa PL-14 midwoofers $51 each.
Crossover:
Crossover boards are external to the speaker cabinet.
Tweeter: 4th order LR at 2500hz with amp side resistor for attenuation.
Woofer: 4th order LR at 2500hz with zobel network for impedance compensation
14ga air core coils, Dayton and Solen metallized Poly caps
Cabinet information:
Cabinet is vented. Single 4" port tuned to 50hz.
75" tall, 13" wide, 11" deep
Cabinet is made of 1"mdf, with individual drivers separated by 3/5" mdf for the 1st half.
Back of mid-woofer holes are radiused with a 3/4" router to ease back pressure on driver.
Ribbon opening is backed by 3/4" mdf for rigidity and air tightness, with wire channels cut into it for most of the length.
1 1/2" roundover on each vertical corner beginning only 3/8" from the edge of the drivers.
1/4 hardboard constrained vibration layer attached with flexible chalk.
Veneer is Honduran Mahogany . Stain is Minwax liquid, a mixture of 2 parts Red Oak to 1 part Red Mahogany. Finish is Minwax polyurethane.
Wiring:
No power tapering, even power distribution used. Tweeter array 4ohms. Woofer array 8ohms.
Total cost:
The end cost was approximately $1800.

bullet.gif

Differences from the original Linus Array.
Although the design is based on the Linus Array, I have made a number of modifications to the original design to match my different goals, higher budget, and satisfy my personal preferences and ideas about speaker building. During the same period Jim Griffin was designing the Linus II. We were coresponding during some phases, and much input came from Jim Griffin himself so my array and the Linus II have some strong similarities.
Changes I made to the original Linus Arrays:
Tweeter line is lengthened by 2 tweeters
This is to increase the vertical window of the tweeter line so it extends from below the seating position to above the standing position. The tweeters do not radiate much sound vertically, so if your ear moves above or below the tweeter line, you get a noticeable drop-off of the highs. The highs now sound even in any normal listening position. There is a drop-off if you are laying down or if you place your ear somewhat above 6 feet 2inches, as might be the case for someone 6 feet 8 inches tall. (I am 5 feet 8inches myself.) Jim Griffin also made this exact change in his Linus II design.
Vifa PL14 woofers
This is a personal choice. I feel the MCM-1855 woofers are pretty good woofers, especially for the price. I purchased a pair from Rick Craig, and they are rather nice sounding, and have exceptional bass for a small speaker. They are very tame for an aluminum woofer, but there is still just a bit of harshness to my ear. I am very selective regarding the sound of the midrange, and I find I do not like most midranges out there. I wanted a bit more detail, a more warm, sweet and neutral sound. Several of the members purchased pairs of 5inch midwoofers to audition, building cabinets for each driver. After listening to 7 different 5 inch midwoofers we all concluded we liked the Vifa PL-14. However it was one of the more expensive drivers we looked at. The PL-14 are very low distortion, neutral, and detailed.
I checked out the following drivers all 5"-5 1/4":
GR-130, clean, sweet, runners up, a bit colored, rim too large, $25.
MCM 55-1855, good, very cheap, slight metal sound, not a lot of detail, $12.
Vifa MG-14, bit of harsh breakup, could be tamed. I didn't really like the sound. $45 dollars.
Hi-Vi 5g- Pretty, expensive, rings like crazy, I don't think it could be tamed. Avoid this driver. $90.
Peerless HDS-85488- Not too bad, but not a of detail either $55.
Vifa P-13, an old classic. Pretty good, but inferior to the PL-14 in my honest opinion. NO bass at all. $35.
9 woofers instead of 12
The number of woofers is a bit of a compromise. It is a personal aesthetic choice on my part, to have the tweeter line, and the woofer line of similar length. Also there is a consideration of the end size of the cabinet, since the PL-14s require more volume than the MCMs. However the Vifa's are much more efficient than the MCM's so that equals out. There was also an economic consideration since the price was exceeding my budget.
Jim Griffin has used 10 Vifa TC14 drivers in his Linus II design.
Cost
More tweeters and more expensive woofers, high quality crossover parts, and mahogany veneer. I place the end cost at about $1800.00
Cabinet construction
I tried to work a lot of details in the cabinet construction to try to optimize the effect of the cabinet on the sound.
The cabinet is a single chamber as opposed to 9 separate chambers. This decision was made to make the most efficient use of internal volume, since there is 18 inches or so on the bottom that does not have woofers. I think the ideal would be 1 woofer per chamber, but chose against it due to added size.
The following step were taken to reduce cabinet wall vibration:
------------------------------------------
The cabinet outer walls are 1 inch mdf.
It is very heavily braced with 3/4 inch mdf. With a brace between each woofer.
A 3/4 inch strip of mdf was used as a backing for the ribbon line to provide stiffness, and a better air seal than separate ribbon holes.
There is a 1/4 inch hardboard piece attached to all cabinet wall surfaces with flexible silicon chaulk. This a constrained vibration layer to dampen any wall vibrations.
To reduce standing waves within the cabinet the following steps were taken:
-----------------------------------
Flexible silicon caulk was applied to most internal surfaces to act as damping and an adhesive for the foam.
1/2 inch open celled foam was applied to the interior surfaces. (Ester open celled foam in 2lb density, this is the same type of foam that is used as acoustic foam, but if purchased generically it can be much cheaper. In fact some dealers have 2 price lists, one for regular foam and a SPECIAL price for acoustic applications.)
Miraflex fiberglass insulation was then attached in varying thickness', depending on the amount of clearance available.
Where there were open areas that did not have a direct path to the port, such as the area behind the ribbon line,fiberglass insulation was placed in the area to fill the void.
Other steps that were taken to optimize the cabinet for sound
----------------------------------
To reduce edge diffraction, the vertical edges were rounded over with a 1.5 inch radius.
To reduce any effect of drivers protruding on the baffle face all drivers were flush mounted.
The rear of the driver holes were rounded over with a 3/4 inch radius, to lessen interference with the backwave, and reduce reflection from the area where the driver is mounted.
All internal edges were rounded over, including the braces, and the ribbon backing.
To eliminate any direct interaction of the woofers with the adjacent woofer the brace between woofers was extended from the front wall about 5 inches back.
The cabinets are dead sounding without apparent vibration or resonance. It is like tapping on concrete.
When the speaker is playing loudly if you place your ear flat up against the sides you can just barely make out some sound being transmitted through the wall, this would be inaudible at any distance.
However if you place your ear against the back of the cabinet, you can pretty clearly hear the backwave from the drivers. This is due to an error in my design of the bracing. You will notice I cut a single hole in the back of the brace. This leaves the sides completely braced, however fails to sufficently brace the center of the back wall of the cabinet. If I were to do it again I would cut 2 holes in the brace and leave center bracing from front to back. I don't feel the amount of vibration on the back wall has a detectable influence on the sound at all, since the amount of sound vibration is still very small (my ear is against the back of the cabinet.). However the lesson for me is even with a 1 inch thick wall, center bracing has a significant affect on the vibration being transmitted from the drivers.
The cabinets are covered in a mahogany veneer, extra thick .022mm.
Attached with Weldwood contact cement.
They are stained with a mixture of minwax red oak and red mahogany liquid stain. 2 parts red oak to 1 part red mahogany.

bullet.gif

Some words of thanks
---------------------------------
I wish to acknowledge the fine work and generous nature of Rick Craig and Jim Griffin for so completely documenting the original research and plans of the Linus Array. They were extremely accessible and helpful in discussions, and in answering questions.
The Line Array project is a team effort by a newly founded group which call itself the Bay Area speaker builders. Acknowledgment and thanks to the members of the Bay Area Speaker Builders who spent so much time discussion and working on this first project. Brett Sher, Brenda Make, Warren Harris, Kris Rex.
Also many thanks to members of the Madisound discussion forum, and the Parts Express discussion forum, who helped with answers and advice, especially Wolf, Danny Richie, Alan Ersen, Siegfried Linkwitz, and others that provided suggestions.

links.gif

links.gif

links.gif

Linus Array

Linus White Paper

My Home Theater

bar.jpg

The line arrays use 8 Stryke Audio RTW2 tweeters,(these are the same as the HiVi PT2), and 9 Vifa PL14 midwoofers per-side. To limit the cabinet vibration, I am using a 1/4" hardboard plates as a "constrained vibration layer". These are mounted on flexible silicon caulk to dampen vibration.

This shows the bracing, and partial walls to separate the woofers. The outer walls are 1" thick MDF, the braces are 3/4" MDF. You will notice that even the interior surfaces of the braces are rounded off. (I mean why not?)

I have used a 3/4 inch round-over bit to radius the driver opening for the mid-woofer. For smaller drivers such as the 5" drivers I am using, the size of the magnet is so close to the diameter of the driver, that if you use a thick baffle, it may constrain the airflow of the backwave. This can cause a stiffled sound from the speaker, and some added distortion. It is also possible that some of the sound will reflect back from the wall of the driver hole. If you look below the driver opening your will see that we have chosen to seal the back of the ribbon line with a 3/4 inch board to insure air tightness and provide added rigidness to the front panel. You will notice that we have radiused the edge of the ribbon backing as well.

Here is a closeup of the flush mounting. A lot of milling went into these. As a learning experiece, I made some minor errors along the way, but nothing that wasn't easily repaired, or redone. One major lesson, the better the template, the easier the routering, and the less mistakes. Next time I will spend as much time making templates as is needed. Here you see the tweeter cutout, with a groove for the wires. Both the tweeter and woofers are recessed

Here they are all glued, ready for some final routing, sanding, addition of the base, and finishing. You are able to see some of the damping installed. It uses 8 Stryke Audio RTW2 tweeters,(these are the same as the HiVi PT2), and 9 Vifa PL14 midwoofers per-side. To limit the cabinet vibration, I am using a 1/4" hardboard plates as a "constrained vibration layer". These are mounted on flexible silicon caulk to dampen vibration.

I have always believed in rounding edges to lessen edge diffraction. Of course the more gradual the edge the better. Until recently I thought the 3/4" radius roundover bearing bit at the hardware store was the largest available, and I have used it all over. In a discussion on the Madisound discussion forum the other day, a fellow named Wolf (who I met at a DIY LeftCoast in Stockton, Ca) told me they make a 1.5" radius roundover bearing bit! (I had no idea.) So I HAD to get one of these. I located one at an industrial tool store in town. It was rather expensive. I was extremely scared about whether my router could handle it and how it would cut, since the bit was HUGE, about 3 5/8" in diameter with a 1/2" shank! I had to manufacture a custom base for it since it was about 1" bigger than the current router opening. The bit is actually too large to sit within the router and I had to mount the bit below the router and make the base thick enough to engulf the edge. I used 3/8" clear acrylic. I was very worried about whether it would work, but it cut just fine. It did make the most amazing amount of sawdust I have ever seen, I have a vacuum attachment for the router, but it couldn't EVEN keep up with the 5 foot fountain of sawdust this bit was making. I mean it was pulverizing almost the equivalent of a 1" x 1" on a single pass. Anyway here is a picture of the giant bit, my custom base, the dwarfed router, and the radius that was cut. I love the broad radius, it begins almost at the end of the flush mounted drivers.

All routed. The roundover starts almost at the edge of the flush mounted driver. It is my hope that this reduces the baffle and edge effects somewhat.

Wiring the 8 ribbon tweeters in the tweeter line, with series parallel wiring. The 9 mid-woofers were wired with series parallel 3-3-3 for a system resistance of about 8ohms. No power tapering is used. I tried power tapering, and I didn't like the results. Since the sound is constrained vertically in the array, when power tapering the woofer line, the balance changed between the lines, depending on the height you were listening at. I found this undesirable.

bar.jpg

email.jpg

jmiyake@earthlink.net

HOME | Subwoofers | homeHT | center