Fuel Tank Construction

[rodsmith]

I'm ready to start building fuel tanks. I have tank fittings from Spruce, but not sure what the proper way to mount these is. Does the whole fitting get welded on the outside of the tank, or do you cut a hole the size of the central part and most of the fitting is inside the tank. I've enclosed a photo of the fittings. Also, I believe Kent White at one time said the fittings and rivets could be aluminum brazed instead of welded. Has anyone used that method.

Rod

[alaskabearhawk]

Hi Rod...

As far as I know they are welded to the tank like you have them The outer flange is intended for thickness so you to have something to weld to.

[S Lathrop]

The upturned flange around the outer edge of the fittings is for welding. Ideally one would flair the hole outward so the tank and the flange are welded by fusing the 2 upturned edges together.

When you flair the hole, you minimize the problems that come when you heat the aluminum and the welding surfaces warp as you weld.

The next option is to cut a hole so the tank and the edge mate. You can tack the edges. This will help minimize distortion as well.

The other option is to weld the flange to the tank and have a hole in the tank large enough for the fittings to clear. Distortion can be an issue with this approach.

[rodsmith]

Thanks for the info. I don't have a way to flange the holes. I am going to experiment with brazing the fitting on, I have an extra end rib to play with.

[Bearhawk535]

I went a slightly different way on my tank fittings. I turned the fittings out of round stock without the flange. I cut a step on the side against the tank surface that sits in the hole so that it is flush on the inside. Then it was a matter of going around this on the inside with the TIG torch and adding just a little filler. I did not get much distortion on the test pieces I welded and no leaks. I had center drilled the hole for the fitting slightly undersized. After welding the fitting in place, I drilled to the appropriate diameter for the pipe fitting tap. I did the filler neck the same way. You have to weld all of the fittings prior to closing up the tank, but this seems like a good plan. I haven't finished the tanks yet, as I don't need them yet and want to practice my TIG skills just a bit more before welding up the final assembly.

[mike 427]

If you have a finished wing to fit the fuel tanks in you are ready.that includes upper and lower stiffeners installed so you can get proper dimensions. Form a internal bulkheads with outer skin thickness and strap and wear pad considered. Lay the fittings on outside of bulkhead where they will fit,clamp and tig weld in place.Watch your heat to control distortion .I did not rivet internal bulkheads in place and try to weld rivets as there is a cracking problem.I drilled as if you were going to use rivets but used clecos to secure and then tig welded them in place and welded shut the drilled holes.That required making welds on the inside of the tank and i stitch welded the bulkheads all the way around.The outer skin was made from one piece of aluminum with the single seam at the top and front of tank.It made for one less weld and one less possible leak. the end bulkheads were then fitted in, drilled and clecoed as the others were and welded up after all the drains were already installed..by doing it this way I had a minimum of distortion and no problems with leaks and the fit was perfect.The clecos work perfect for holding and welding and you can check dimensions before final welds Keep in mind to give your self room on the length. Its real tight in there when your installing them,you need room for tubing,fittings,wrench and hands.good luck. welding rivets is a disaster waiting to happen.

[bergy]

I have been working on the gas tanks for my Patrol. I now have them riveted together, except for the last end. I plan on leaving that open until the very last so I can make sure it stays as clean-as-possible inside the tank during welding and such.

Different builders have expressed problems with welding the aluminum rivets; leaking welds have been a problem for some. Some builders have asked about the possibility of using aluminum-brazing instead of welding.

I checked out aluminum-brazing online (a rather misnomer, since there is no brass in the principally zinc/aluminum rod) . http://www.weldingtipsandtricks.com/...zing-rods.html

So I thought I would do some experimenting with some rod I bought at Harbor Freight yesterday.

I tested it by trying to seal-up empty-holes, solid-riveted holes, and pop-riveted holes in 0.050 5052 aluminum sheet.

I first cleaned the aluminum with Scotchbright pad.

I had mixed results. It is easy to use either too much heat or too little heat. I used a small pencil-tip propane torch adjusted to have about a 1" inner-cone in the flame. I had the best success by first warming the rod slightly, probably 200 to 300F, then carefully heating the aluminum sheet. I applied heat until a momentary touch of the rod to the sheet revealed that a small speck of the aluminum rod would stick to the surface. I would then apply heat for a couple more seconds, then rub the end of the rod around the rivet or around the hole to be filled. I works surprisingly well ... sometimes.

Examination of the back side will reveal when you use too much heat. Also, heating for too long soon develops a scum of slag on the melted puddle. Getting it done quickly seems to make the prettiest weld.

Also, the puddle of melted rod has a lot of surface-tension. It wants to bead-up on the surface, but by the same token, it will span amazingly large holes in the aluminum-sheet.

After cooling my various welds I stuck the samples in a vise and beat them up with a hammer. Some welds proved to be stronger that the 5052, The melted rod does not fill cracks by capillary action like lead-solder would, but it did stick surprisingly well to the aluminum sheet and rivet heads. It appears that one must use enough rod on each river to completely cover the rivet-head.

Some of the weld-puddles did partially peal-off when threated severely by a ball-peen hammer. The problem, I suspect, is the oxide-coating on the 5052. So I then did another experiment. I cleaned the 5052 sheet with fine sandpaper, but first put a few drops of 10W30 engine oil on the 5052. I sanded lightly, then wiped off the gray gunk formed on the surface with a clean rag, but left behind an oily surface. I reasoned that if, as web sites claim, one can use aluminum-brazing on oily aluminum castings, the oil must burn off and cause no problem. Well, the oil did seem to help reduce the formation of an oxide coating; my weld-puddle now stuck much better to the surface, as attested to by the hammer-test.

The really nice part of using aluminum brazing is that the stuff melts at about 730F, much lower than the melting point of 5052. So if one does not like what he has done he can go back, warm up the surface until the weld-puddle develops a slight sheen, and wipe it all off with a dry rag. It is sort of like a Welder's-Mulligan. Also, when one wipes off a substandard weld one can see just how/where the original weld puddle was bonded to the 5052 and where one might need to go back and clean the surface again.

[Bcone1381]

I found this article in the 2010 Q1 Beartracks issue, and wondered if it might be applicable.


PROPER PROCEDURE FOR WELDING AA 5052 (fuel tanks) -- Bob Smith #727

As bearhawkers we are primarily interested in making sound crack free welds on our gas tanks.
The information in this article is gleamed from the CSA standards as well as ASTM.
The cracking of aluminum alloys is primarily due to the hot shortness of the alloy. Hot shortness is that characteristic of a metal that results in complete loss of ductility while the alloy is solidifying (contracting). Cracking is caused by the chemistry of the weld bead and the shrinkage stress due to cooling.

There is two ways to prevent hot shortness. When they are used together they produce a sound crack free weld ( no pinhole leaks either).

1. Change the alloy
2. Reduce the dilution of the base metal
   
   AA5356 was designed to weld AA5052. By moving the alloy out of the crack sensitive area however you must

make sure that your weld bead consists of at least 50 percent filler metal (5356). This is accomplished in two ways.

1. Weld procedure - full fusion but low penetration and lots of filler metal
2. Joint design - a strait but joint has the most amount of parent metal dilution (85% not good) A V groove equals

45% base metal dilution and will give an acceptable weld. A lap or a T joint is best because there is only 10% parent metal dilution, so this joint design is best. So how do we put this information to use while welding our gas tanks?

1. We use 5356 filler rod
2. All joints are lap joints so use only enough heat to allow full fusion and add lots of filler metal

When I welded my first gas tank I had 6 rivets with small leaks from trying to make small welds with too little filler metal. On the second tank I realized my mistake and added the proper amount of filler metal on each rivet. The second tank had no leaks. I hope this helps. If you need any help or advice with 4130 or 5052 please don't hesitate to contact me at
smith727@drytel.net
Bob Smith #727

Note: Bob Smith is a Red Seal inter provincial Journeyman welder, qualified to weld pressure vessels and boiler tubing in both Stainless steel and mild steel. He has held these tickets for the last 31 years consecutivetly.

[Sir Newton]

Bcone1381 great post.
I hold the same qualifications. Only 25 years experience though.
Also wanted to ask you if your BH is flying?

[Bcone1381]

Not yet. I'm building a Patrol from a kit. I build alone at my home. Education takes time. Building comes after that. Right now I'm covering stuff, specifically, I have the interior covered and am setting up to paint the interior fabric, then will reassemble various control cables, then cover the exterior fuselage. Yesterday I designed and fabricated a paint booth in my workshop to accommodate painting the interior.