Make your own silicone couplers?
#1
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
Has anyone explored making custom silicone couplers? I could replace 6 parts with one part if I could make a 90-degree reducing elbow to go from 5/16" to 1/2". But, like in most areas, I'm so far out in front of the rest of the universe that I can't find anything remotely close to that. So I thought it might be cool to make it. Anyone tried something like that before?
Without it:
Hose clamp
5/16" hose
hose clamp
5/16" barb to 1/4" NPT male
1/4" NPT female to 1/2" barb
hose clamp
90-degree bent 1/2" hose
hose clamp
With it:
hose clamp
5/16" to 1/2" reducer elbow hose
hose clamp
How about a tube of silicone caulk (I said caulk!) and some open-weave ribbon/tape? Make a mandrel out of aluminum on a lathe, bend the 5/16" leg, coat it with 6 layers of PTFE release agent and start wrapping. Hmmmm...now I gotta try it.
Without it:
Hose clamp
5/16" hose
hose clamp
5/16" barb to 1/4" NPT male
1/4" NPT female to 1/2" barb
hose clamp
90-degree bent 1/2" hose
hose clamp
With it:
hose clamp
5/16" to 1/2" reducer elbow hose
hose clamp
How about a tube of silicone caulk (I said caulk!) and some open-weave ribbon/tape? Make a mandrel out of aluminum on a lathe, bend the 5/16" leg, coat it with 6 layers of PTFE release agent and start wrapping. Hmmmm...now I gotta try it.
#5
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
Found it:
REDUCING ELBOW 90 DEGREE SILICONE HOSE 8mm-16mm Roose
I've done a ton of Googling. Didn't think of Ebay for anything but generic crap.
For paNX2K&SE-R, thanks!
For magnamx-5, GFY. I've made **** that's been on the cover of Popular Science.
REDUCING ELBOW 90 DEGREE SILICONE HOSE 8mm-16mm Roose
I've done a ton of Googling. Didn't think of Ebay for anything but generic crap.
For paNX2K&SE-R, thanks!
For magnamx-5, GFY. I've made **** that's been on the cover of Popular Science.
#6
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
Found it:
REDUCING ELBOW 90 DEGREE SILICONE HOSE 8mm-16mm Roose
I've done a ton of Googling. Didn't think of Ebay for anything but generic crap.
For paNX2K&SE-R, thanks!
For magnamx-5, GFY. I've made **** that's been on the cover of Popular Science.
REDUCING ELBOW 90 DEGREE SILICONE HOSE 8mm-16mm Roose
I've done a ton of Googling. Didn't think of Ebay for anything but generic crap.
For paNX2K&SE-R, thanks!
For magnamx-5, GFY. I've made **** that's been on the cover of Popular Science.
#10
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
Sweet find! No, there are things I won't do for a living. This is undoubtedly one of them.
I think the DSM guys are screwing up with the tape selection. The material they are using is too thick, and you can see on the cut ends that it's not being fully impregnated. These tapes
STANDARD E-GLASS TAPES from Aircraft Spruce
are meant to have liquid composite matrix material impregnate them. You should be able to get similar stuff at any fiberglass/composites supplier.
I would also experiment with peel ply rather than the self-fusing silicone. Another layer of silicone that's not doing anything is pretty pointless. If the peel ply works with the silicone, it should compact the wraps, allowing excess silicone to fuse through it, leaving a uniform, OE-like texture. The non-smooth surface might be harder to clean, but if you're making couplers out of anything other than black silicone, you're ghey.
PEEL PLY from Aircraft Spruce
I would want the hoses relatively thinner for weight and packaging reasons, letting the strength come from the fibers and the sealing come from the silicone matrix.
Or you could just wrap it with inside-out electrical tape. Use the cheap stuff.
Just occurred to me that if you used the actual bead-rolled tubes you are connecting as part of your jig, you would end up with a neat ring that would snap onto the bead, locating everything perfectly. If you used mold release on the tube and inserted your main mold between the tube ends, you would end up with a coupler ID that yielded a perfect, no-ridge interface with the tubing. You could similarly eliminate the step on a compressor inlet or discharge.
And you can make any angle/size/length you want. No pieceing it together with straights, 45s and 90s. You can have arbitrary changes in cross section, or additonal fittings coming in at any angle. For example, I'm thinking of a reducing elbow with an integrated BOV port that has to be oriented very specifically. This would replace a pipe welded to the charge pipe joined to the BOV inlet with a silicone coupler and two hose clamps. The "old way" would force the BOV position to be farther from the centerline of the pipe, which you might not want.
More permanent mandrels could be carved from wood and refined with Bondo if you wanted to have multiple goes at a shape.
It looks like it's exactly as easy as I thought it would be. A bit messy, but massive elegance can ensue. This is exciting!
Jeez, I'm a dork.
I think the DSM guys are screwing up with the tape selection. The material they are using is too thick, and you can see on the cut ends that it's not being fully impregnated. These tapes
STANDARD E-GLASS TAPES from Aircraft Spruce
are meant to have liquid composite matrix material impregnate them. You should be able to get similar stuff at any fiberglass/composites supplier.
I would also experiment with peel ply rather than the self-fusing silicone. Another layer of silicone that's not doing anything is pretty pointless. If the peel ply works with the silicone, it should compact the wraps, allowing excess silicone to fuse through it, leaving a uniform, OE-like texture. The non-smooth surface might be harder to clean, but if you're making couplers out of anything other than black silicone, you're ghey.
PEEL PLY from Aircraft Spruce
I would want the hoses relatively thinner for weight and packaging reasons, letting the strength come from the fibers and the sealing come from the silicone matrix.
Or you could just wrap it with inside-out electrical tape. Use the cheap stuff.
Just occurred to me that if you used the actual bead-rolled tubes you are connecting as part of your jig, you would end up with a neat ring that would snap onto the bead, locating everything perfectly. If you used mold release on the tube and inserted your main mold between the tube ends, you would end up with a coupler ID that yielded a perfect, no-ridge interface with the tubing. You could similarly eliminate the step on a compressor inlet or discharge.
And you can make any angle/size/length you want. No pieceing it together with straights, 45s and 90s. You can have arbitrary changes in cross section, or additonal fittings coming in at any angle. For example, I'm thinking of a reducing elbow with an integrated BOV port that has to be oriented very specifically. This would replace a pipe welded to the charge pipe joined to the BOV inlet with a silicone coupler and two hose clamps. The "old way" would force the BOV position to be farther from the centerline of the pipe, which you might not want.
More permanent mandrels could be carved from wood and refined with Bondo if you wanted to have multiple goes at a shape.
It looks like it's exactly as easy as I thought it would be. A bit messy, but massive elegance can ensue. This is exciting!
Jeez, I'm a dork.
Last edited by SolarYellow510; 04-26-2009 at 11:50 AM.
#12
Elite Member
iTrader: (15)
Join Date: Dec 2007
Location: San Antonio, Texas
Posts: 4,847
Total Cats: 27
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
I like the idea as well. I have an engine bay full of back-to-back silicone couplers and t-bolt clamps. If it were possible to fuse the back to back silicone couplers together and eliminate some t-bolts that would be nice.
I'll have to do some reading and see if it is possible to overwrap a set of already cured couplings with new silicone and bond them together.
I'll have to do some reading and see if it is possible to overwrap a set of already cured couplings with new silicone and bond them together.
#13
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
I like the idea as well. I have an engine bay full of back-to-back silicone couplers and t-bolt clamps. If it were possible to fuse the back to back silicone couplers together and eliminate some t-bolts that would be nice.
I'll have to do some reading and see if it is possible to overwrap a set of already cured couplings with new silicone and bond them together.
I'll have to do some reading and see if it is possible to overwrap a set of already cured couplings with new silicone and bond them together.
#15
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
Doods, I'm geeking out. Out of control like Barney with the non-alcoholic champagne. The design possibilities are endless. You can vary the reinforcing material. Say, use cotton gauze on intakes that are only going to see atmospheric pressure. Play with thickness. Add strategic reinforcements or customize the direction of the layup. Add some wire mesh for abrasion resistance. Use fiberglass or thin spring steel wire to prevent collapsing the cross section on really thin sections that need to be flexible. Embed larger metal reinforcements if you want something to be axially stiff, like able to hang a filter out in space. Ooh, ooh - imagine the multiplication of ricer attraction if you had carbon fiber reinforcements in your silicone intake!
![The Brain](https://www.miataturbo.net/images/smilies/the_brain.gif)
#16
Cpt. Slow
![](/mt2011/images_td/ranks/supporter1.gif)
iTrader: (25)
Join Date: Oct 2005
Location: Oregon City, OR
Posts: 14,360
Total Cats: 1,184
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
why not use a 1/2" 90* elbow, with a piece of 5/16" hose slipped into the end, and clamp down on your 5/16" hose barb. That would eliminate the 1/4" NPT BS, and eliminate the wasting of money on all the materials necessary to make your own.
#18
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
lol atleast you have some fight in you. Seems more of a clusterfuck than its worth to me though. Solar you might have made **** on popular science, but do you print GRM (Grassroots Motorsports Magazine), Nature (Journal home : Nature) I did :P
#20
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
I have in my possession:
fiberglass cloth in 2, 4, 6 and 8 ounce weaves
light nylon cloth
light polyester cloth
13 ounces of black rtv silicone
I'm thinking the first thing to be tried should be test samples. Peel strength to see if the silicone sticks especially poorly or well to any of them, then tensile strength.
I've learned so much about hose today just by obsessing about laying up pipe![Giggle](https://www.miataturbo.net/images/smilies/gay.gif)
For example:
Having the reinforcement plies at 90-degrees and zero degrees will prevent stretching and ballooning, but make it hard to slide over barbs and beads. You could alleviate that by having some extra silicone on the ID of the hose in places you know may need a little extra cushion. Having the plies at 45 degrees will allow the hose to expand to a larger diameter while shrinking in length, making it easier to get over barbs and beads but potentially ballooning and screwing up the geometry of your system by getting shorter when pressurized.
The DSM guy screwed up in more than his selection of cars and fiberglass tape. Wrapping the tape in a spiral means for a layup of n plies, you have a spiral through the whole length of your component with only n-1 plies. If you were to test the tensile strength of the tube, it would unzip like flypaper along that joint. His saving grace is that he went back the other direction diagonally relative to his first one. But where he has invested the time, materials and finished weight to have a "four layer" hose, he still has a "fishnet" of weakness where there are only three layers, and only two layers where the three-layer spirals intersect.
It would be much better to wrap wide sheets, the length of the tube if geometry permitted. However, that's only easy on straight sections, which are cheap enough there's no point to doing it yourself. Complex geometries, which are what's exciting about this whole thing, will prevent that. Designing a hose becomes like designing any other composite structure, where the best results will be achieved with carefully detailed joints, and great care must be taken to prevent accidentally building in some non-obvious flaw. Or you can just throw a lot of extra material at it and suck.
fiberglass cloth in 2, 4, 6 and 8 ounce weaves
light nylon cloth
light polyester cloth
13 ounces of black rtv silicone
I'm thinking the first thing to be tried should be test samples. Peel strength to see if the silicone sticks especially poorly or well to any of them, then tensile strength.
I've learned so much about hose today just by obsessing about laying up pipe
![Giggle](https://www.miataturbo.net/images/smilies/gay.gif)
For example:
Having the reinforcement plies at 90-degrees and zero degrees will prevent stretching and ballooning, but make it hard to slide over barbs and beads. You could alleviate that by having some extra silicone on the ID of the hose in places you know may need a little extra cushion. Having the plies at 45 degrees will allow the hose to expand to a larger diameter while shrinking in length, making it easier to get over barbs and beads but potentially ballooning and screwing up the geometry of your system by getting shorter when pressurized.
The DSM guy screwed up in more than his selection of cars and fiberglass tape. Wrapping the tape in a spiral means for a layup of n plies, you have a spiral through the whole length of your component with only n-1 plies. If you were to test the tensile strength of the tube, it would unzip like flypaper along that joint. His saving grace is that he went back the other direction diagonally relative to his first one. But where he has invested the time, materials and finished weight to have a "four layer" hose, he still has a "fishnet" of weakness where there are only three layers, and only two layers where the three-layer spirals intersect.
It would be much better to wrap wide sheets, the length of the tube if geometry permitted. However, that's only easy on straight sections, which are cheap enough there's no point to doing it yourself. Complex geometries, which are what's exciting about this whole thing, will prevent that. Designing a hose becomes like designing any other composite structure, where the best results will be achieved with carefully detailed joints, and great care must be taken to prevent accidentally building in some non-obvious flaw. Or you can just throw a lot of extra material at it and suck.