Mass tweeter array assembly
-
- Posts: 2623
- Joined: Sun Jan 28, 2007 11:13 am
- Location: Denver, CO
- Contact:
Mass tweeter array assembly
I'm in the middle of retro-fitting my 4 OT12's with melded arrays. That's a mere 48 tweeters cut and glued, precision cuts and assembly required.
Here's what I found sped up my process.
Identify as many good tweeters as you'll need for your arrays. Out of 50 tweeters delivered I found 3 obviously bad and 6 that lacked the high frequencies of the rest of the tweeters. From now on to get consistent response tweeters I'll plan on 1 in 5 being bad or questionable. Bill says the single tweeters lacking HF response are fine in arrays. For the low cost of 1016 tweeters I figure I'll just plan on ordering an extra tweeter for every 4 I need.
If you're wondering how it all adds up to the 48 I needed, I already had about 20 on hand from my original build of the OT12's when I ordered many more than I actually needed to get my arrays up to snuff after not testing the tweeters. Test your tweeters, order more than you need. It bears repeating.
I'm cutting and assembling the tweeter arrays in 2 phases rather than 1 as the plans show. I don't trust myself to do the 45 degree cuts with perfect precision on individual tweeters so all the parts line up just like they should.
Phase 1:
Step 1: Trim the tweeters down to their final 2 1/8" width. Use a tablesaw with a squared and calibrated fence. If you don't have one, use someone else's. I highly recommend this for safety and accuracy. My panel jig had enough play that my first two cuts weren't quite square so I abandoned it. I would definitely not use a miter saw to cut these edges. Your fingers will be way too close to the blade and most of the saws I've worked with won't support that narrow a material next to the blade.
Set the fence at 2 3/8". Cut one side, cut all your ends, make sure the ends are in pairs facing opposite directions so you have a top and bottom. Set the fence at 2 1/8" and cut the other side of all the middle pieces. Make sure the blade is as high as specified in the plans to trim the bolts and housing. Make especially sure your fingers are out of the blade path. I felt comfortable enough to do it all by hand, you may want to use a pushblock if you don't feel that comfortable. Make sure any pushblock you use holds the tweeter straight up against the fence and down against the table surface.
Step 2: Assemble vertical sets of tweeters. Plans say use model cement, Testors liquid model cement says "contains Methyl Ethyl Ketone". MEK is a solvent you can purchase in larger containers at any hardware store. At Home Depot, as I recall, a can of MEK costs about $7. Model cement in about 1/100 the volume cost $2.50. Use MEK.
I made a squaring jig which was just a 1" strip of 1/2" plywood attached to a larger flat piece of wood. Make it at least as long as the vertical dimension of your array. If you use model cement you have to apply it with a brush. I did this for two sets of tweeters and it only took 10 minutes. Using a cookie sheet and my squaring jig, I poured out a thin layer of MEK in the pan, briefly dipped the cut edges of the tweeters in it and pressed them together in batches of 6 on the jig. Make sure all your wiring posts face the same direction. Using this method I finished the remaining 6 sets of tweeters in an additional 10 minutes.
Glue directions are the same as in the plans or on a bottle of cement. Let it set for a few seconds, press the edges together and hold for at least 20 seconds. Let the arrays set for at least 2 hours while the plastic solidifies.
Make very sure you do this in a well ventilated area unless you want to relive the 70's while building tweeter arrays.
Phase 2:
Step 3: Cut your 45 degree angles. Again using a table saw, since most miter saws can't crosscut the dimensions of an array, set your blade to 45 degrees, measure the length specified in the plans on the fence and cut the 6 element parts in one pass. Make sure you cut them in pairs and cut the 45 degree angles so the wiring posts face each other as specified in the plans. You may want to test the 45 degree cut with a piece of scrap wood. I sometimes find it hard to get the lengths right when cutting angles on a table saw. Measure twice cut once. Piezos are cheap but messing up a set of 6 or two gets a little costly.
Step 4: Assemble the crossfiring arrays. You can use a 90 degree jig of some sort if you want. Repeat the cookie sheet MEK gluing process with the array elements and hold the two parts at 90 degrees for 20 seconds to a minute. I used a sheet of wax paper, a flat surface and held the two array parts together by hand since there was enough surface to "enforce" the 90 degree angle. Using dipped MEK I found the arrays stuck together quite well after only a minute. Leave them on a corner of some sort for 5-10 minutes and handle very carefully. Let them set supported at 90 degrees for at least 2 hours. Goop over the back with black ABS pipe cement to seal the array and let it set for 24 hours.
Install the arrays per the plans.
Hope this helps. I'm very happy with the results of this process and everything looks solid.
Here's what I found sped up my process.
Identify as many good tweeters as you'll need for your arrays. Out of 50 tweeters delivered I found 3 obviously bad and 6 that lacked the high frequencies of the rest of the tweeters. From now on to get consistent response tweeters I'll plan on 1 in 5 being bad or questionable. Bill says the single tweeters lacking HF response are fine in arrays. For the low cost of 1016 tweeters I figure I'll just plan on ordering an extra tweeter for every 4 I need.
If you're wondering how it all adds up to the 48 I needed, I already had about 20 on hand from my original build of the OT12's when I ordered many more than I actually needed to get my arrays up to snuff after not testing the tweeters. Test your tweeters, order more than you need. It bears repeating.
I'm cutting and assembling the tweeter arrays in 2 phases rather than 1 as the plans show. I don't trust myself to do the 45 degree cuts with perfect precision on individual tweeters so all the parts line up just like they should.
Phase 1:
Step 1: Trim the tweeters down to their final 2 1/8" width. Use a tablesaw with a squared and calibrated fence. If you don't have one, use someone else's. I highly recommend this for safety and accuracy. My panel jig had enough play that my first two cuts weren't quite square so I abandoned it. I would definitely not use a miter saw to cut these edges. Your fingers will be way too close to the blade and most of the saws I've worked with won't support that narrow a material next to the blade.
Set the fence at 2 3/8". Cut one side, cut all your ends, make sure the ends are in pairs facing opposite directions so you have a top and bottom. Set the fence at 2 1/8" and cut the other side of all the middle pieces. Make sure the blade is as high as specified in the plans to trim the bolts and housing. Make especially sure your fingers are out of the blade path. I felt comfortable enough to do it all by hand, you may want to use a pushblock if you don't feel that comfortable. Make sure any pushblock you use holds the tweeter straight up against the fence and down against the table surface.
Step 2: Assemble vertical sets of tweeters. Plans say use model cement, Testors liquid model cement says "contains Methyl Ethyl Ketone". MEK is a solvent you can purchase in larger containers at any hardware store. At Home Depot, as I recall, a can of MEK costs about $7. Model cement in about 1/100 the volume cost $2.50. Use MEK.
I made a squaring jig which was just a 1" strip of 1/2" plywood attached to a larger flat piece of wood. Make it at least as long as the vertical dimension of your array. If you use model cement you have to apply it with a brush. I did this for two sets of tweeters and it only took 10 minutes. Using a cookie sheet and my squaring jig, I poured out a thin layer of MEK in the pan, briefly dipped the cut edges of the tweeters in it and pressed them together in batches of 6 on the jig. Make sure all your wiring posts face the same direction. Using this method I finished the remaining 6 sets of tweeters in an additional 10 minutes.
Glue directions are the same as in the plans or on a bottle of cement. Let it set for a few seconds, press the edges together and hold for at least 20 seconds. Let the arrays set for at least 2 hours while the plastic solidifies.
Make very sure you do this in a well ventilated area unless you want to relive the 70's while building tweeter arrays.
Phase 2:
Step 3: Cut your 45 degree angles. Again using a table saw, since most miter saws can't crosscut the dimensions of an array, set your blade to 45 degrees, measure the length specified in the plans on the fence and cut the 6 element parts in one pass. Make sure you cut them in pairs and cut the 45 degree angles so the wiring posts face each other as specified in the plans. You may want to test the 45 degree cut with a piece of scrap wood. I sometimes find it hard to get the lengths right when cutting angles on a table saw. Measure twice cut once. Piezos are cheap but messing up a set of 6 or two gets a little costly.
Step 4: Assemble the crossfiring arrays. You can use a 90 degree jig of some sort if you want. Repeat the cookie sheet MEK gluing process with the array elements and hold the two parts at 90 degrees for 20 seconds to a minute. I used a sheet of wax paper, a flat surface and held the two array parts together by hand since there was enough surface to "enforce" the 90 degree angle. Using dipped MEK I found the arrays stuck together quite well after only a minute. Leave them on a corner of some sort for 5-10 minutes and handle very carefully. Let them set supported at 90 degrees for at least 2 hours. Goop over the back with black ABS pipe cement to seal the array and let it set for 24 hours.
Install the arrays per the plans.
Hope this helps. I'm very happy with the results of this process and everything looks solid.
Last edited by gdougherty on Thu Feb 21, 2008 1:47 pm, edited 1 time in total.
- David Carter
- Posts: 1842
- Joined: Mon Jan 01, 2007 7:20 am
- Location: (East) Tennessee, USA
Great post! Thanks for sharing your process. I'm going to be putting melded arrays in a couple of DR250's so this was very helpful.
One question that's been on my mind ever since I first read the plans for building a melded array... The plans I had when building my O10 earlier this year said to use ABS cement when gluing the cut piezos together. Now the plans for the melded array say model cement and coat the back of the finished array with ABS cement. Why the change? I already have plenty of ABS cement and was wondering what would happen if I just used that for the whole thing...
One question that's been on my mind ever since I first read the plans for building a melded array... The plans I had when building my O10 earlier this year said to use ABS cement when gluing the cut piezos together. Now the plans for the melded array say model cement and coat the back of the finished array with ABS cement. Why the change? I already have plenty of ABS cement and was wondering what would happen if I just used that for the whole thing...
Dave
Built:
- Omni 10 (S2010 + piezo array)
- 2 x DR250 (DL II 2510 + melded array)
- 2 x Titan 39 (BP102 - 14"W)
- 2 x Titan 39 (3012LF - 20"W)
- 4 x DR200 (Delta Pro 8B + melded array)
Built:
- Omni 10 (S2010 + piezo array)
- 2 x DR250 (DL II 2510 + melded array)
- 2 x Titan 39 (BP102 - 14"W)
- 2 x Titan 39 (3012LF - 20"W)
- 4 x DR200 (Delta Pro 8B + melded array)
- Bill Fitzmaurice
- Site Admin
- Posts: 28967
- Joined: Tue May 02, 2006 5:59 pm
Model cement comes in a tube with a small applicator tip. ABS cement comes in a can with a large applicator swab. The stuff inside the containers is the same.David Carter wrote: Now the plans for the melded array say model cement and coat the back of the finished array with ABS cement. Why the change? I already have plenty of ABS cement and was wondering what would happen if I just used that for the whole thing...
- David Carter
- Posts: 1842
- Joined: Mon Jan 01, 2007 7:20 am
- Location: (East) Tennessee, USA
Thanks for the clarification, Bill.Bill Fitzmaurice wrote:Model cement comes in a tube with a small applicator tip. ABS cement comes in a can with a large applicator swab. The stuff inside the containers is the same.David Carter wrote: Now the plans for the melded array say model cement and coat the back of the finished array with ABS cement. Why the change? I already have plenty of ABS cement and was wondering what would happen if I just used that for the whole thing...

Dave
Built:
- Omni 10 (S2010 + piezo array)
- 2 x DR250 (DL II 2510 + melded array)
- 2 x Titan 39 (BP102 - 14"W)
- 2 x Titan 39 (3012LF - 20"W)
- 4 x DR200 (Delta Pro 8B + melded array)
Built:
- Omni 10 (S2010 + piezo array)
- 2 x DR250 (DL II 2510 + melded array)
- 2 x Titan 39 (BP102 - 14"W)
- 2 x Titan 39 (3012LF - 20"W)
- 4 x DR200 (Delta Pro 8B + melded array)
-
- Posts: 2623
- Joined: Sun Jan 28, 2007 11:13 am
- Location: Denver, CO
- Contact:
Ah, so the plans are talking about the tube paste type cement rather than the bottled brush on cement. From my modeling days, either works and if anything the straight MEK worked better than the brush on cement.
I will repeat again, if you use MEK do it only in a well ventilated area. The fumes are very potent. ABS cement seems aromatic and a little "enlightening", MEK feels like it's eating your brain. I also recommend against soaking your hands in it. It disolves paint on contact and I'm sure doesn't do good things to your hands.
I will repeat again, if you use MEK do it only in a well ventilated area. The fumes are very potent. ABS cement seems aromatic and a little "enlightening", MEK feels like it's eating your brain. I also recommend against soaking your hands in it. It disolves paint on contact and I'm sure doesn't do good things to your hands.
- David Carter
- Posts: 1842
- Joined: Mon Jan 01, 2007 7:20 am
- Location: (East) Tennessee, USA
Thanks for the warning. If MEK clears your sinuses even better than the ABS cement, I think I'll stick to the black ABS cement I've already got on hand!gdougherty wrote:I will repeat again, if you use MEK do it only in a well ventilated area. The fumes are very potent. ABS cement seems aromatic and a little "enlightening", MEK feels like it's eating your brain. I also recommend against soaking your hands in it. It disolves paint on contact and I'm sure doesn't do good things to your hands.
Dave
Built:
- Omni 10 (S2010 + piezo array)
- 2 x DR250 (DL II 2510 + melded array)
- 2 x Titan 39 (BP102 - 14"W)
- 2 x Titan 39 (3012LF - 20"W)
- 4 x DR200 (Delta Pro 8B + melded array)
Built:
- Omni 10 (S2010 + piezo array)
- 2 x DR250 (DL II 2510 + melded array)
- 2 x Titan 39 (BP102 - 14"W)
- 2 x Titan 39 (3012LF - 20"W)
- 4 x DR200 (Delta Pro 8B + melded array)
-
- Posts: 2623
- Joined: Sun Jan 28, 2007 11:13 am
- Location: Denver, CO
- Contact:
It's not that bad, just do it somewhere ventilated and don't stand over the cookie sheet of MEK. It evaporates into the air rather quickly. I'm just posting fair warning for the guys who do their builds down in the basement. I'd move the project outside briefly for tweeter assembly.David Carter wrote:Thanks for the warning. If MEK clears your sinuses even better than the ABS cement, I think I'll stick to the black ABS cement I've already got on hand!gdougherty wrote:I will repeat again, if you use MEK do it only in a well ventilated area. The fumes are very potent. ABS cement seems aromatic and a little "enlightening", MEK feels like it's eating your brain. I also recommend against soaking your hands in it. It disolves paint on contact and I'm sure doesn't do good things to your hands.
- kdchandler
- Posts: 133
- Joined: Thu May 18, 2006 12:44 am
- Location: Salem, Oregon
MEK warnings
I agree with the more severe MEK warnings. In the old days aircraft mechanics would use it to wash their hands. By the time I worked for a major airline (in Canada, anyway) we were cautious enough that the entire hangar (DC10, B737/767, etc) would empty when the painters brought out the MEK cans. Even with respirators, we didn't want to be there very long.
-
- Posts: 2623
- Joined: Sun Jan 28, 2007 11:13 am
- Location: Denver, CO
- Contact:
I used it in the kitchen right next to a super high-flow stove exhaust fan setup. Of course I was gooping in the garage last night with the ABS cement and it wasn't too much better. Best reccomendation is to follow the directions on whatever it is you're using, which of course I'm sure we all do all the time.
MEK DANGER!!
Hi all,
Please be aware that MEK is a nasty chemical - when absorbed in your body it goes straight for your kidneys and WILL do you harm!! DO use ventilation to avoid breathing fumes, skin contact also should be seriously avoided. Glue outdoors, a desktop type fan blowing the fumes away would be good way to minimise exposure.
Please be aware that MEK is a nasty chemical - when absorbed in your body it goes straight for your kidneys and WILL do you harm!! DO use ventilation to avoid breathing fumes, skin contact also should be seriously avoided. Glue outdoors, a desktop type fan blowing the fumes away would be good way to minimise exposure.
-
- Posts: 2623
- Joined: Sun Jan 28, 2007 11:13 am
- Location: Denver, CO
- Contact:
Re: MEK DANGER!!
Thanks for the heads up. I think next time I'll take it outside then and wear gloves.GeoffBass wrote:Hi all,
Please be aware that MEK is a nasty chemical - when absorbed in your body it goes straight for your kidneys and WILL do you harm!! DO use ventilation to avoid breathing fumes, skin contact also should be seriously avoided. Glue outdoors, a desktop type fan blowing the fumes away would be good way to minimise exposure.
Jeez, I built models for years as a kid and am still ok. I remember when they came out with the 'safe' stuff, it didn't work worth a shit.
As I recall, plastic model kits are made of polystyrene, while the tweeters are ABS.
Both of these products are 'cements', not glues. That is to say they chemically weld they're respective plastic types together. Done correctly the weld is as strong as the material itself. I have to wonder if the model airplane glue is doing justice to the ABS, or maybe if they're close enough in composition that it doesn't matter?
As I recall, plastic model kits are made of polystyrene, while the tweeters are ABS.
Both of these products are 'cements', not glues. That is to say they chemically weld they're respective plastic types together. Done correctly the weld is as strong as the material itself. I have to wonder if the model airplane glue is doing justice to the ABS, or maybe if they're close enough in composition that it doesn't matter?
- Bill Fitzmaurice
- Site Admin
- Posts: 28967
- Joined: Tue May 02, 2006 5:59 pm
Says who?Tim Ard wrote:Jeez, I built models for years as a kid and am still ok.

The glue is mainly solvent that dissolves the material being bonded, which bonds when the solvent evaporates. The solids in the glue seem to work equally well with ABS, styrene and PVC.I have to wonder if the model airplane glue is doing justice to the ABS, or maybe if they're close enough in composition that it doesn't matter?

-
- Posts: 2623
- Joined: Sun Jan 28, 2007 11:13 am
- Location: Denver, CO
- Contact:
- David Carter
- Posts: 1842
- Joined: Mon Jan 01, 2007 7:20 am
- Location: (East) Tennessee, USA
I've been working on melded arrays for a pair of DR250's using gdougherty's assembly process. I cut them all down to 2 inches (except the ones on the ends), and glued them together in groups of five. Today I made the angle cuts on each group of five. I realized after cutting the first group that my measurement was slightly off, and the length after cutting was about 1/16" short of the specified 3-3/8". The top piezo shown below is straight from the plans, and the bottom one is how mine look.

My question is whether this slight deviation from the specs in the plans is going to cause any performance problems once I get the cabs built? I'm guessing that it won't, but I want to make sure before I spend a lot of time getting the arrays completely assembled.

My question is whether this slight deviation from the specs in the plans is going to cause any performance problems once I get the cabs built? I'm guessing that it won't, but I want to make sure before I spend a lot of time getting the arrays completely assembled.
Dave
Built:
- Omni 10 (S2010 + piezo array)
- 2 x DR250 (DL II 2510 + melded array)
- 2 x Titan 39 (BP102 - 14"W)
- 2 x Titan 39 (3012LF - 20"W)
- 4 x DR200 (Delta Pro 8B + melded array)
Built:
- Omni 10 (S2010 + piezo array)
- 2 x DR250 (DL II 2510 + melded array)
- 2 x Titan 39 (BP102 - 14"W)
- 2 x Titan 39 (3012LF - 20"W)
- 4 x DR200 (Delta Pro 8B + melded array)