thumbs up! first shakedown!

Okay, just a limp across town to a friend's afternoon barbecue, but still a real enough ride to shake it through some gears and attempt a stoppie or two (that 320mm brembo works). Still, Sarah was happy for me to have reached this point and insisted on some pictures. Well it's been 2,1/2 years. Through open heart surgery, 2 moves, shlepping parts here and there. Countless hours pondering, sketching, rebuilding, pondering more...

...And now that I'm back home, I see so much more to do. 3 miles of loose wiring needing to be trimmed out of the situation, suspension is totally out of whack, I think I ran out of clearance under the fender in back and scuffed the pipe and made an unnatural sound with the back of the license plate on a bump...but still...Face in the helmet says it all.



A little back yard burnout


Spent the morning wiring all the signal lighting gear in place. Man, I went through so many failed crimp connectors. Good thing I bought 3x more than I needed. It was good practice, I'll be revisiting this all at some point for the tacho. Dad turned me onto this coiled plastic stuff that really bundles up your loom nice and you can see it below. Since I had the harness all apart trying to figure out the turn signal logic, I took some time to clean off all that gooey tape that had been resident with the spiders since before polyps were even a twinkle in Ronald Reagan's colon.


Also took some time to affix some adhesive insulation under the tank just for a little more peace of mind

plumbing fuel tank, revisit

OK, finally got the moto specific fuel line from J&P. It's much more flexible than the automotive kind while still having some substance. With the elbow filter, and a Tee connection, I'm able to get both sides of the tank connected to the carb without any kinking. It's draped across the rubber carb/intake manifold adapter there so hopefully not too hot! 8 hoseclamps later:

By the way, is my filter oriented with the flow of fuel correctly? I couldn't find any indicator arrows on the body of the filter saying which direction the flow went. This is a screen type filter so not sure it matters as long as you don't switch after in use

plumbing fuel tank

OK, got back to some progress on the bike. Used the Caswell snot to seal the interior, hope it holds. Been off line because I've got a bit of a conundrum with fuel plumbing here, and it's probably, well, important to get it right to make sure I don't get a lap full of gasoline while straddling this home made beast.

Recall I've got petcocks on both sides of the tank here and I need to run an inline fuel filter. Clearance is going to be tough and my first parts sourcing for the application is a total failure. Here, you see a petcock and the filter attached to the carb. In between these two needs to be a "TEE" connector to receive the fuel line from the other side of the tank, joining this petcock to then lead to the filter:



I got the petcocks from Clarke Mfg (they make the big plastic tanks for motorbikes, primarily). They're pretty large and I did myself no good by attaching them so far aft on the tank. Anyway, the Automotive 1/4" fuel line is way too stiff for bending and the fuel filter I bought from J&P Cycles is too long.

I found some more flexible (tho reinforced) moto-specific fuel line in their catalog with some help from an actual human on their end. Plus a shorter fuel filter that should only take up about 1/2" of length, barbs aside. Hopefully that will work

I'll also need a better "Tee" connector for joining the two sides of the tank first before leading to the fuel filter and then the carb. I'm on the lookout for a sort of 3 dimentional Tee, I think: where the barbs are all oriented 90 degrees off of a central origin, each pointing along the X, Y, Z axes...That might help keep the fuel line from making such a tight bend.

aft mounting bracket II

Here's how it looks now

aft mounting bracket

The forward mounting brackets worked out okay so I reinforced them and sanded them down a bit. Now it is time to get the rear bracket sorted out.

I wanted to use the OEM rubber damper that fit the original tank to it's mounting post that we left intact here, so I'm going to fabricate a tab that attempts to use it. The layup here was as quick as possible: build up a cardboard form with some thickness at the bottom to accommodate the offset of the rubber. I'd then prep it and rest the tank on the glass/epoxy layup while it was still curing so that everything dried in place.


I folded up some newspaper to support the cardboard form right where it would make contact with the tank.


Here you see everything prepped and draped with paper to prevent drips from sticking to the bike. In the upper left you see a wooden block that the tank will rest on during the cure.


Here it now sits. We'll see how it looks tomorrow morning!

forward mounting brackets

Had the hardest time trying to figure out how to place the forward mounting tabs for the tank. The tabs are nothing more than a "C" cross section that I molded around a foam block. Well I eventually cut the tabs to size and then press fit that foam block in where the mounting posts would be, and then tacking the "C" tabs into place. It sort of works, but It was hard to transfer the mounting points exactly onto the tank when it was in position.

I'll lightly test fitment tonight and if it looks okay, reinforce the attachment with more glass and epoxy.

tank assembled

OK, mixed another slury of the milled glass fibers with the slow cure epoxy and laid a bead along the interfaces. Got a lot of squeeze-out once the thing was taped down (especially along the inside when I peered in with a flashlight through the cap hole). Hopefully it'll be alright.

one last check before closing up the tank

OK, I talked to the tank sealer guys at Caswell and they said I ought to scuff up the insides of the tank since I can and it'd help adherence. They said go ahead and glue everything up before dumping their sealer into the tank to coat the interior. So that's what I'll do.

It's been a long process so I'll take a sec to make sure everything's wiped down clean with acetone after sanding and vacuuming all the dust. Nothing but rubber gloves at this point; you don't want any of your finger grime/oil getting onto joint surfaces.

petcock fitment

looks like they'll bolt right on no probs. You can see on the right side how the positioning plates formed the epoxy/paste mixture into a flat surface that the petcocks can interface with all nicey-nice.

more fitting glue-up

OK, Milled glass fibers came in the mail and so now onto mixing into a thick slurry with epoxy to finish off the T-nuts that mount the petcocks, and then the gas cap assembly.

I pressed some PVA-released positioning plates into the mixture as it'll make a flat surface interface to the petcocks.


I took the outer surface of the tank down to 600grit sandpaper to clean up the flash and a few boogers on the outer surface; It'll look okay once I get a coat of wax on there, kind of satin hopefully.

mounting petcock nuts.

OK, managed to epoxy on a few "T" nuts to the bottom piece which will attach the petcock valves. I've positioned them with a few jigs I cut from some old fiberglass layup done on a previous project; just cut some holes the correct distance apart to fit the petcock valves. Once this glue sets, I'll lay down a slurry of epoxy and milled glass fibers to really solidify the assembly. I actually would have done this all in one step but ran out of the milled fibers to make the reinforcement with. Figure it's okay to "tack" the nuts down right now and then reinforce them later.

demolding the top part

I probably should have let the thing cure a bit longer in the mold but it looked solid enough and I just couldn't wait. I'm that way with scabs, too. And like a scab unready for the picking, it took plenty of encouragement to bust the mold loose. First some prying between the flanges, then a bit of drumming on the hull to pop free along the surface, more prying, more pounding, you get the idea. These things stick together quite a bit because of the PVA. It's weird stuff: a thin membrane that starts off applied to the mold but when you take everything apart, it ends up on as a coating on the final part that you just rinse off with water.

This is how it looked immediately out of the mold, so a lot of boogers and some flash to clean up, but the part seems pretty sound!



I ran it out to the garage to confirm fitment with the bottom, and after roughing out the edges a bit they look to be mating pretty well.



Next step will be to double up a few more layers on the top part from the inside where it looks a little thin, then it's on to fitting up the petcocks and gas cap.

An opera in 3 pots.

Never made a gas tank before. It's obvious I'm way over my head here. Anyway, let's begin making the top part.

I chose to use four facets of glass per layer. As luck would have it the tank is reasonably symmetrical, so the pattern consisted of 2 facets, roughly looking like this:



Keep that mildly pissed off woman's expression in mind, because by 8:00pm yesterday, I was right with her, maybe even several lengths past that, ready for a bucket of plonk and a flintlock to point at strangers.

Making the patterns is pretty easy, just hanging some news print off of the plug, and cutting notches around the compound contours. Kind of like so:


(I like how her picture ended up upside down...perhaps that's why she's giving me that scorning look of disapproval -- as if my shenanigans have no purpose to her discussion on philanthropy for the Arts in the post-Bush, pre-economic apocalypse).

Layup for the top was a complex affair. I knew I'd take too long to get it all in one pot of resin before it started to gel. Even with the slow hardener, after 30 minutes resin begins to cure really fast when it's amassed together in the mixing dish. I carried forth the layup in these overtures:

• First a topcoat of black tinted resin right on the mold that I'd let cure to the point of tackiness. This from an idea taken directly out of Tygaboy's advice on Bay Area Riders Forum to achieve a smooth outer surface and facilitate adhesion with the first layer of glass.
  
• Then mix some milled glass fibers with more black resin to fill in the cracks/crannies around the fuel cap recess, as well as the dimples around the shoulder of the tank that are there to: a) clear the fork triple clamp bolts, and b) make it harder for the glass to lay down smoothly.
  
• While the topcoat dried to a tacky consistency, I cut the glass for the main facets. Once cut, I mixed the first pot of tinted resin and went at it. After a couple layers the resin started to get kind of snotty, so I found a stopping point, ran inside and took the turkey thighs I had roasting out of the oven, and mixed a pot of clear resin to finish out the layup.
  
• Vacuum bagging. This exercise was fail from the get-go. First, I didn't cut the peel-ply big enough to extend past the overhang in the tank -- crucial because I wanted to tape it down to the outside of the mold. I then had a few wads of breather cloth jammed into the interior, some bunched up into the shoulder and under the overhang to help apply pressure from the bag. I then stuffed all this into the bag and began pulling air with my bicycle pump. After not long, it was apparent that there was not enough surface area of the bag to reach into the overhangs of the tank *and* all the way into the depths of the shoulder of the tank, even with all the extra wadding I had used to take up these areas. I decided to quickly pull the layup from the bag, remove the wadding, peel-ply and just let it cure on its own.
  

Six pairs of rubber gloves later, and this is what the place looked like last night. Note the brush right in front of the mold, I sawed off the handle for easier maneuverability within the mold:



Yeah, vacuum bagging is a nice to have thing, and a lot of composites geeks use it to consolidate a layup and get the optimal ratio of resin to reinforcement for the lightest possible construction. I've never been as interested in that factor as much as the way vacuum bagging helps the fiberglass to wrap around complex edges. Since I'm laying glass on the interior of the mold, it actually seems to have laid down pretty well. I'll leave it and see how it comes out of the mold in a day or two.

On Tank Lining

From the Caferacer.net board:

Tank lining with composites seems to be a hotly contested subject of late especially with ethanol based fuels. I've read about several treatments. The most noteworthy ones to date are:

• POR-15: This is a paint on substance. I first learned about it from an epic moto tank building thread on Bay Area Riders Forum.
  
• Caswell: A two part epoxy based treatment designed specifically for ethanol fuel resistance. One source on the Bay Area Riders Forum has had a lot of experience with both POR-15 and Caswell from running a painting business and restoring 100's of moto tanks. He swears by the Caswell. I'm inclined to agree because it's epoxy -- hard to beat it for adherence and toughness.
  
• Fuel Safe: Manufacturer of fuel bladders for racing appliations. You send them a copy of your plug and they build you a physical bladder that I guess is just draped into your tank "shell". Probably bucco $$ but also probably the best.
  
• Kreem: This has been around for ever and is widely regarded as garbage by people who've used the other products. I guess it doesn't bond as well and ends up delaminating from the walls of your tank. ick.
  

There's a lot of discussion among the boating guys, too. I picked up on an article below from a post on the ADV rider forum. Frustrating that the study does not give specifics on when it was conducted, what epoxy was used specifically in this test, nor what kind of lining was used:

http://www.boatus.com/seaworthy/fueltest.asp

[...]
Finally, chemical resistance data from a leading epoxy supplier showed that even epoxy can be attacked by ethanol. The test was made using the company's most resistant epoxy and exposing fiberglass lab samples to 10% ethanol gas and regular unleaded gas as well as diesel and aviation gas. The results for the ethanol gas showed a 10% loss in hardness and a 10-15% loss of compressive strength over a 16-week period and it’s likely that the loss of hardness and strength would continue to fall at a similar rate. The unleaded gas, diesel, and aviation gasoline tests, none of which contained ethanol, showed virtually no loss of strength.
[...]
What to Do?
Many boaters have made the decision to replace what were thought to be superior fiberglass gas tanks with aluminum tanks (tank replacement is not covered by insurance). Seaworthy has talked to resin manufacturers who say that there are resins that resist ethanol, but simply coating a tank with one of these resins is not likely to work, since the tank has to be thoroughly cleaned and prepped on the inside and the resins have to be specially cured.

Actually from the article at the top,there's a link out to some boat builder's own empirical test with ethanol vs epoxy using the caswell jizz as a barrier. It is not terribly promising.

But I am dubious of anybody being able to get a good result from a composite tank that has already been in service!:
http://egyptian.net/~raymacke/Cbnskif27.htm

From that same thread on Caferacer, Matt (sign-on 'monkey') who's a very well regarded moto tank fabricator had this to say:

as far as a tank lining... epoxy works better than anything.

the problem with most linings is that people are appling them after the tank has been made or even used... good luck getting whatever you are pouring into the tank to bond with the inside surface and actually last for more than a few months.

once you have both skins of the tank prepped for bonding together...
1. make sure all your fittings are set ( much easier to install things such as caps, inserts for petcocks etc etc before you bond together).

2. sand the glue surface with 80 grit so you get a good mechanical bond... also sand the entire inside surface ( for resin coating)

3. i then apply epoxy resin to the inside faces ( usually 2 coats ) and let it tack up for an hour or so. i also apply the epoxy adhesive at this time. with the flow coat and the adhesive ( on the glue surface) tacky... they stay put when you are handling the tank halves rather than ooze all over the place

4. align and stick together.. use small spring clamps to hold bonding surfaces together. once happy, i then pour a few ounces or catalysed resin into the filler hole to seal any potential leaks.

then the tank is then placed on a level surface to cure.

5. come back after resin is completely cured and machine edges for neatness.

6. cleanliness, clean hands, lots of gloves, masking tape... etc etc

Make sure to get a vented fuel cap or you will get vapor lock.

now this all sounds easy.... but it is not fuel tanks are a pain, if this fails on you while you are riding.... major problem

my father in law made one for himself in the 60's and it split at the start/finish line dumping fuel all over his running bike.
it was my goal to prove to him i could successfully make them and race em. i will never be on the racetrack without one... except my crf 150... leaving it alone. your glass wont be too thick.. i just want my tanks to be significantly lighter than stock... otherwise why bother? my goal is to always make them lighter than 2 lbs.
Just a reminder.. i have been in the composites fabrication business for over 11 years... building everything from custom sailboats, custom offshore racing sailboats, lobster boats, America's Cup parts.... the list goes on and on... i always love when people try to get me to give em a crash course in this stuff and think they are going to make money off of me... sounds big headed.. but it is funny. i started building bike parts when i didn't have the money to go out and buy it... so make it better and suit my needs.

also... if you crash... any tank can potentially crack... steel, alum, glass... if you are concerned about things breaking... run a 1/2" thick steel plate tank... then you'll be fine

from the pics on your post... you are doing perfectly fine.
have fun and good luck

matt

Thanks, Matt!

Cleaning up the top mold

OK, Now I'm cleaning up the edges of the mold and am holding it together with some old 5mm machine screws I had:



I left about an inch of overhang where final top piece will be glued to the bottom. Gonna be hard to reach in there and get the glass applied well. Especially if I want to vacuum bag the layup. I'll probably carve out some foam blocks that will fit under the overhang which will apply some compression once I pull the air out of the bag.



Sarah has grown a little dubious of all this work entrenched in my subconscious.

Snap, Crackle, POP!

OK, well I built up all the other facets much like the first and this morning used a butter knife to pry them apart. It's quite a noisy affair as the PVA releases the fiberglass from the mold popping and cracking. I had to rap on the outside with a wooden dowel to help release some areas. Sarah didn't like that at 5:30AM.

One trick I used to get started was to lay down a few pieces of packing tape on the opposing side of the flange being erected, followed by a bit of PVA. This allows you to tug the pieces apart a bit.

Anyway, here is how they look!



The "Locating Dimples" I cut with the drill bit really help in lining this thing up; glad I took a few extra minutes to do them. That one hole you see was where I actually went all the way through on the drill so I had to cover it up with tape.



This thing will get some bolts to hold it all together during layup. I'll trim the interior overhang down a bit from what you see here, but this is still going to be a very challenging layup to execute. I think I'll have to cut the handle off of the chip-brushes that I use to apply epoxy for a little extra maneuvering room.

Begin layup on top 2

After the glass had cured, I tried pulling the weirs used in it's construction. Sadly the adhesive from the glue gun stuck a little too well to the plug. I guess I got carried away with the glue and used a lot, figuring it'd just peel right off but it didn't!

Ended up tearing some of the skin off the plug in the struggle; so i patched with a bit of smooth packing tape. Next time, I might attempt covering the entire plug in fiberglass beforehand as a way to toughen up the skin.

Anyway, I carried on with the other side, this time just using a few small dollops of glue to hold the weirs onto the plug.

I cut a few depressions with a 3/8" drill bit into the opposing flange there (marked by arrows) so that the two sides of the mold have a means of orienting themselves neatly.

Top mold layup 1 complete

One step at a time! I used a slurry of milled glass fibers and epoxy to get into the nooks and crannies before doing the primary layup using 10oz fabric. Used a bunch of small rectangles of glass in a patchwork so that they'd lay down okay along the weir. Fingers are crossed.

Is this thick enough?

8 layers of 5.6 oz "S" glass ends up being roughly 1/16" thick. Hmmm...I'd like it to be about twice as thick just for peace of mind but just 8 layers feels awfully tough.

I can bend this sample about 1/4" with my bare hands; certainly couldn't fold it without a vice and some pliers. But is it enough?

Addendum A user going by the handle 'monkey' frequents the Caferacer.net forums who does a lot of this stuff professionally. His current race tank build up process just uses 3 layers of 200g Carbon, and thinks this is more than enough, even 'gnarly': http://www.caferacer.net/forums/topic.asp?TOPIC_ID=9756

weirs erected for first section of top mold

So I'll apply glass/resin to the mold and a ways up on the weirs to build some flanges for the mold pieces to bolt together with. This thing has a coat of PVA that is drying for an afternoon layup.

I learned this trick of roughing out a complex curve from a guy who build a gas tank in a similar way on Bay Area Riders Forum. Reminiscent of differential calculus in a way, but there's nothing some old junk mail cardstock and a glue stick can't do



Once the patterns are made, I just transfer to some tougher card board and then cut out and seal with packing tape. Few minutes of hot glue gun action later et voila