Welding Tips and Tricks

Hey Everyone,

I need to fix a aluminum gas tank. Whats the best way to do this without going Boom? Is purging with argon or similar nonflammable gas good enough or what else can I do?

Thanks!

Hey,

I bet there is about a million ways to do it, This is the i have read. (never done though) First you must wash the tank out with a detergent. I have heard out steaming them out but obviously you need a suppy of steam, as in from a boiler. When you have given it a very good wash out.
Fill with water to just below the repair area then purge the rest of the space with your non flammable gas, im not sure if you can use co2 so argon is prob a good choice. Filling with water reduces the poosible area for vapour build up and also reduces the amount of purge gas required.

Im sure some one out there will have done this before so please take my advice with a grain of salt.

PS Welcome to the forum.

Mick

There's heaps of threads around on welding fuel containers... I've never done it, but I've seen enough dire warnings to know that a bit of solid research is definitely the way forward. A couple of examples/ideas:
http://www.woodworkforums.com/f65/petro ... um-159029/
http://www.gmh-torana.com.au/forums/top ... r-welding/
What's the configuration of the tank? does it have seams or just welded edges? Apart from the porosity of the aluminium itself, are there areas where residual fuel can lurk and form vapours? Apart from detergent washing and purge gas, which I would regard as no-brainers, it seems a good safety measure is to use heat (hot water, steam, etc.) as part of your cleaning process.
I also remember reading somewhere that aluminium fuel tanks (in boats) have an average life span of 5 to 7 years. Can't remember the reasoning or the source. It does raise the question, depending on the condition of the tank, of whether a new replacement tank is an option

I have repaired lots of fuel tanks and have only had two go boom. For steel tanks I take them to my local rad shop and have them hot tanked for at least 24 hours emphasis on I that way you know it was actually done. You cannot do this with aluminium as the caustic soda in the hot tank will turn the aluminium black so you need to flush it with as hot of water as possible for several hours then put an argon purge on it. It doesn't take much I usually run about 10 cfh and stick the purge hose as far in as possible, make sure you purge it for at least 10 minutes before you weld on it to ensure you have an inert environment inside the tank. The two booms were a result of not following the rules. If it is a customers tank make sure they understand if the tank goes boom you are not responsible to replace it. Motorcycle tanks turn into footballs. It takes very little fuel for an abrupt event to happen literally a few drops in a clean tank is enough. That was my first boom I welded a tank and stepped away helper put fuel sending unit back in I came back and took out fuel sending unit and noticed just a few drops of gas I should have let it evaporate but didn't and lit up on it and shat my pants. I always purge with argon now no matter how clean I think it is. Also make sure openings in tank are facing away from you to ensure you don't get a face full of fire.
Hope this helps and I am not responsible if you blow your self up or wreck your tank.

Lots of good advice on this!

When I deal with a baffled tank, once it's as drained as I can get it, I like set it in the sun in the AM with the hose of my shop vac (on exhaust/blower side!!!) as far in the fill neck as it will go. At the end of the day, I'll do the detergent wash. I have the advantage of access to BlueGold, used for removing organics for "oxygen clean", but Dawn works just fine. Then, absolutely, purge. Nitrogen is fine, and cheap. You already have argon, though. CO2 is fine for preventing "BOOM", but is an active gas and may not give the best weld on aluminum. Purge slow and patient for an hour, so you know there's no O2 hidden in a baffle somewhere (and follow pipe-purge protocol... All exits taped up, with several small holes adequate to let the purge gas out, but no air mixing). The advice about keeping the ports/openings away from you is wise.

If you will be welding close to your purge vent, pipe the vent gas away from you, just for insurance.

Steve S

I charge enough to make building a new one a better option. Why? Because of the hazmat clause with our liability insurance and workman's comp. If they found out we were welding gasoline tanks, our rates would go thru the roof no matter how diligent we were with the cleaning. I weld a lot of open gasoline pipe and fittings and diesel tanks though. My situation is somewhat different. Most of the marine tanks we come across, by the time they need any welding, they are iffy anyway. People will still try to cheap out and fix a pitted tank but I find it is just not worth it, as most of the cost is in removing it and re-installing it. Having to repeat that process a year down the road because pits started seeping etc, just makes it false economy to bother with it.

Also, if you can get away with it, it's often better to weld a patch on the exterior of the tank so that you don't have to introduce any flame or sparks to the interior.

Many years ago, I owned a 500cc Matchless Motorcycle. It developed a fuel leak from a visible crack on one corner of the fuel tank where a threaded plate held one corner down. I took the tank to a local welding shop and asked to have it fixed. The owner who was also a weldor put a hose from his car exhaust pipe into the filler opening on the tank and ran the motor for about 15 minutes. (Fuel was $.029 a gallon at the time).

After that, he filled it with water with the fuel filler hole open and placed it on an angle so that he could access the leak. He was going to gas weld the crack. He no sooner touched the tank with the torch flame than there was a sharp explosion that exited with a flame out of the fuel filler opening and went up about 20 feet. No harm was done, except to our ear drums.

After that, he finished the weld with no further problem.

Since then, I've TIG welded several motorcycle tanks. I would clean them out with phosphoric acid to remove rust and that seemed to be enough, although I did fill the tanks up with water up to the cracks. Perhaps the TIG process uses less heat than a gas torch. I never feel comfortable welding fuel tanks, however.

Hey,

I have heard about running vehicle exhaust into tanks before welding, but what if the vehicle were running rich? Could this introduce more possible flammable vapours into the area?..

Mick

Not really, Mick...

If it were rich enough to be an explosive mixture, It'd take out the muffler!

Also if it's running rich at least you know all the oxygen has been consumed from the exhaust. If it was running lean, there's a chance that oxygen is still getting through and mixing with any fuel vapours

Hey,

I may be showing my extreme lack of knowledge of cars. What is the condition where un burnt fuel might come out the exhaust?

Mick

A given quantity of oxygen can burn only an exact amount of fuel. A carburator gives a relatively fixed amount of fuel to air, regardless of what is ideal for temperature and humidity, and even engine load. That's why racers fiddle with the jets during warm-ups.

Modern electronic fuel injection is a big improvement, but no system can perfectly meter fuel/air all the time.

Unburned hydrocarbons is one of the things they "sniff" for in an emissions inspection as required in some States here, and they aren't looking for "zero", but some low number I don't know as I'm not subject to them.

Hope that helps clear it up,

Steve S

Thank You Everyone!

All went well. No Boom! Washed out with warm water and dish soap about 5 times and purged with argon for a few minutes before welding. Welded it with the argon purging at about 10 cfh. Worked great.

Thanks Again!!

highstringer wrote:Thank You Everyone!

All went well. No Boom! Washed out with warm water and dish soap about 5 times and purged with argon for a few minutes before welding. Welded it with the argon purging at about 10 cfh. Worked great.

Thanks Again!!

GREAT!

I figured you'd gotten enough information to stay out of the burn unit...

You never asked about leak-testing the finished product, so I assume you had that part figured out.

Steve S