I was reading about pressure testing fuel tanks to test for leaks. The idea is to use a bicycle or other small pump to put 3 lbs of pressure in the tank and then see if it holds the pressure for 24 hours.
Is this a good test and is it commonly employed to check old tanks?
Pressure testing fuel tanks
-
JonnyBoats
- Candidate for Boat-Obsession Medal
- Posts: 372
- Joined: Fri Feb 04, 2005 4:54 pm
- Location: Wiscasset, ME
- Contact:
-
feetup
- Almost a Finish Carpenter
- Posts: 99
- Joined: Tue Jan 16, 2007 11:35 am
- Location: Ladysmith, Vancouver Island
Tank testing.
Way back when I apprenticed as a Fabricator we made fuel tanks for fish boats a lot. Way to many fuel tanks. I hated them, never a square corner on them, and usually to a sketch on the back of a cigarrette package.
We used a soap test to check for leaks, but that was in a shop, on horses where all seams were accessable. We would pressure the tanks to about 3 PSI like you mentioned and using a squirt bottle, spray a strong solution of dish washing liquid and water over all the seams and fittings. The smallest leak would show up as a mound of bubbles and foam. The idea of pressurizing to 3 PSI and holding is valid and the same concept as the hydro and air tests that ASME uses to proof test pressure vessels. It will show you the tank leaks, it just won't show you where.
I would recomend that you plug all fittings and vents, pressurize, soaptest all fittings and wherever you can reach and allow to hold. The pressure hold test has one huge disadvantage, in that a relatively minor change in temperature will result in a fairly major change in pressure, and even a change in barometric pressure will affect the 'indicated' pressure. Try to use a guage where 3 PSI is about the middle of the scale.
A word though from someone who found out the hard way. 3 PSI is MAXIMUM!! I changed a basically coffin shaped tank into a football with a split seam because the regulator we used would slowly climb in pressure (Many of them do) and left it hooked up to shop air while we stopped for coffee. I heard a loud "bonk" and before I could register what it meant
the seam split with a sound that I can not duplicate in print. I KNEW what that sound was. I'm not sure what the ultimate burst pressure was but I never saw that regulator go over 20 PSI. I know that every tank I've ever tested began to show signs of ballooning at 3 to 4 PSI. A pressure vessel 48 inches in diameter made to work at 15 PSI with a flat, bolted lid will require that that lid be between 1 1/2 and 2 1/2 inches thick depending on design, even though the cylindrical shell may be as thin as 3/16 inch. Pressure hates flat sided containers.
We used a soap test to check for leaks, but that was in a shop, on horses where all seams were accessable. We would pressure the tanks to about 3 PSI like you mentioned and using a squirt bottle, spray a strong solution of dish washing liquid and water over all the seams and fittings. The smallest leak would show up as a mound of bubbles and foam. The idea of pressurizing to 3 PSI and holding is valid and the same concept as the hydro and air tests that ASME uses to proof test pressure vessels. It will show you the tank leaks, it just won't show you where.
I would recomend that you plug all fittings and vents, pressurize, soaptest all fittings and wherever you can reach and allow to hold. The pressure hold test has one huge disadvantage, in that a relatively minor change in temperature will result in a fairly major change in pressure, and even a change in barometric pressure will affect the 'indicated' pressure. Try to use a guage where 3 PSI is about the middle of the scale.
A word though from someone who found out the hard way. 3 PSI is MAXIMUM!! I changed a basically coffin shaped tank into a football with a split seam because the regulator we used would slowly climb in pressure (Many of them do) and left it hooked up to shop air while we stopped for coffee. I heard a loud "bonk" and before I could register what it meant
the seam split with a sound that I can not duplicate in print. I KNEW what that sound was. I'm not sure what the ultimate burst pressure was but I never saw that regulator go over 20 PSI. I know that every tank I've ever tested began to show signs of ballooning at 3 to 4 PSI. A pressure vessel 48 inches in diameter made to work at 15 PSI with a flat, bolted lid will require that that lid be between 1 1/2 and 2 1/2 inches thick depending on design, even though the cylindrical shell may be as thin as 3/16 inch. Pressure hates flat sided containers.