Page 1 of 1
Gas tank testing
Posted: Fri Sep 09, 2022 8:13 am
by bill goodheart
I have my reconditioned gas tank installed under the front seat in my 1919 touring. A gas line has not been installed yet. I installed the sediment bulb in the tank before putting the tank in the car, I used Permatex Aviation grade sealant on the threads. Now I would like to find out if it leaks any around the sediment bulb threads. I would like to put 3-4 gallons of water in it to check for leaks. If it leaks, I would rather have water on the floor than gasoline. If no leaks I would open the sediment bowl drain valve to let the water out. Will I be able top get all the water out this way, maybe using a heat gun through the fill opening to completely dry it or is the water going to cause long term water in fuel problems?
Bill Goodheart
Re: Gas tank testing
Posted: Fri Sep 09, 2022 8:35 am
by speedytinc
Why not use a few pounds air pressure & spray down the outside & threads with soapy water, like testing tubes for leaks?
I wouldnt be testing with water in the tank.
Re: Gas tank testing
Posted: Fri Sep 09, 2022 8:58 am
by Allan
If you are just testing the outlet, you won't need much gas. A leak will be gradual, so not much loss, if any.. Excess has can be drained off via the peacock in the outlet, if necessary. Forget the water.
Allan from down under.
Re: Gas tank testing
Posted: Fri Sep 09, 2022 9:37 am
by NY John T
Hi Bill,
I used water to test a gas tank for my 42 jeep after I used tank sealer, and it didn't leak. Boy was I surprised when I put gas in. Leaked in places that the water didn't. So water is NOT a good test to tell if a tank leaks. Gas is much thinner than water and it is the real test of leakage. Air would be my choice first. If you already tried water, you can dry it out if you put it in the sun and let nature dry it out or use a hairdryer. Good luck. You and I are at opposite ends of the state.
John
Re: Gas tank testing
Posted: Fri Sep 09, 2022 10:05 am
by TXGOAT2
Gasoline and diesel will often leak from containers that will hold water. A better test would be to apply very low air pressure to the tank while spraying the suspected leak points with soapy water, as suggested above. I would not use a hair dryer or heat gun around a gas tank. If you use compressed air, it's important to keep the pressure in the gas tank VERY LOW. It takes very little pressure to damage or destroy a gas tank. If you want to pressure up the tank, a good way to do it is to cut the stem from an old inner tube, leaving a round flange of rubber about 3" in diameter attached to it. Hold that loosely over the gas tank fill opening and apply pressure through the stem from an air tank or air line while someone sprays the tank fittings and seams with soapy water.
Re: Gas tank testing
Posted: Fri Sep 09, 2022 10:33 am
by TWrenn
Gallon of gas is plenty. My money's on it won't leak. The stupid petcock is more likely to leak than anything.
Re: Gas tank testing
Posted: Fri Sep 09, 2022 3:15 pm
by John kuehn
The sediment bowl has 3 places that can leak.
The on-off handle, filter screen and at the fuel line connection.
Not counting the sediment bowl end that’s threaded into tank.
Most of the time it’s on the sediment bowl in my experience with Model T’s.
Others may have been different expierence’s.
Re: Gas tank testing
Posted: Fri Sep 09, 2022 4:38 pm
by Craig Leach
Be very careful with compressed air in a non pressure vessel! There is about 780 square inches in a Model T gas tank so every pound of pressure will be exerted on each one of them. If you make a mistake and loose control of the flow into the tank it can get real ugly. I use a modified cooling system tester hand pump, it has a gauge and gives you total control over the volume & pressure going into the tank ( will take some time to fill a 10 gallon tank ) I only use a couple of pounds to test for leaks with soapy water.
I just replaced the bottom of a copper gas tank to make repairs on it & clean out the sealer that a previous owner had put in the tank rather than repair the leaks. The tank was make of 24 GA. copper and started to distort @ 2 lb. of air pressure.
If you have already put water in the tank drain out as much water as you can then put some alcohol in the tank swish it around. Drain the alcohol out and blow out with air ( not a hair dryer alcohol will blow up just like gas ) The alcohol will absorb the water and will mix with the gas just fine. If you put sealer in the tank make sure it is compatible with alcohol. The sealer in the copper tank was not and the gas blend we have here turned into mess in the tank, carb & engine.
Craig.
Re: Gas tank testing
Posted: Fri Sep 09, 2022 8:59 pm
by bill goodheart
Thanks for all the great responses. They are all very helpful, every one says don't use water so I won't. I am not concerned about the tank leaking, only the sediment bowl. The tank was sent to a radiator shop in Kingston, Ontario. They are experienced in using the "gas tank renu" process. They picked it up here and brought it back here. Total cost $350, it took about 6 months, but it was through the winter, Nov- Apr which was fine with me. At this point I think I will hook up the gas line, put in a couple gallons of gas and see if there are any leaks.
Re: Gas tank testing
Posted: Fri Sep 09, 2022 9:58 pm
by Scott_Conger
I believe it is quite unlikely that your sealant will fail on the sediment bowl. That said, before putting fuel in, I would remove the tapered valve on the shut off and the drain and schmeer on some "fuel lube" and simply be done with it. Nothing will leak and the valves will operate smoothly.
Re: Gas tank testing
Posted: Sat Sep 10, 2022 5:06 pm
by Professor Fate
Make up one of these. Great way to check stuff out. Just a couple of psi will do it. Minimal just a pound or two to make the valve or cap hiss when you open them. Make sure you plug the vent hole on the cap. I used a pop rivet.
Don't over pressurize!
Re: Gas tank testing
Posted: Sun Sep 11, 2022 12:06 am
by Phoenix88R
I would be extremely careful applying air pressure to a sheet metal gas tank. At 2psi each square foot will have 288 pounds pressure acting on it. One the side of a T gas tank, which might be 3 square feet in area would be 864 pounds. More than enough to split the seams if you are lucky.
On two occasions I have witnessed the result of using compressed air on vessels not designed for it. The first was in Alaska where a pipeline contractor made a glycol storage trailer using a 80 foot section of 48" pipe by welding a 3/4" plate on the end. They routinely used compressed air to transfer the fluid. On one occasion due to operator error it was over pressured and the plate was blown off, completely severing the entire engine compartment off of a pickup truck parked across the end of the tank. Fortunately, no one was injured. The other time compressed air was used to aid in draining water from a square waste oil tank the size of a small bedroom. The tank was fabricated of 3/8 plate with angle iron ribs about every 3 feet. Air was connected and everyone went to lunch, upon returning the new tank was empty but had also been inflated so it resembled a pumpkin. It had to be demolished and rebuilt with all new material.
.
Re: Gas tank testing
Posted: Sun Sep 11, 2022 9:04 am
by TXGOAT2
You do not want flat surfaces or right angles on a pressure vessel. They will FAIL. A vessel designed to hold pressure may be destroyed if "vacuum" is applied. Gasoline tanks and propane tanks are built very differently, and it's not just for looks.