Because  water is virtually incompressible, hydrostatic testing is employed
to test  tanks. Compressed air alone is a no-no as it can create a bomb in a
hurry!

The  standard ABS/USCG method for non-pressurized liquid tanks such as fuel
tanks is  to fit a standpipe to the tank and fill with water to a height of
approx 4  feet above the tank top in order to create a test pressure of approx 2
PSI at  the tank top if I recall correctly.

Pressurized  tanks such as starting air tanks are also filled with water and
then  pressurized to something like 1.5 x the working pressure using a
hand-operated plunger pump and a calibrated test gauge. Some tanks also have a  test
pressure stamped on them in which case that is used.

Testing and  construction methods are based on the ASME code. Exact methods
and  pressures used by CG for inspection may be found in Title 46 CFR.

The  only tanks I ever saw tested with air were galvanized lifeboat buoyancy
tanks,  which had soldered seams. CG always had us uncap, then cap  them in
the morning and then put them in the sun. Later in the day, they  would have
bulged a bit and some air would escape when uncapped if no  leaks.

Regards,

John
"Seahorse"

empty tank.
NOT recommended when tank contains fuel.

A vacuum  test would also tend to flex the tank walls and most likely cause
the leak  points to get worse.

Visual inspection is still the most practical  test.
If that does not produce any usable result then empty the tank  completely
before applying any sort of pressure or vacuum  test.

************************************** See what's new at http://www.aol.com

to a

REPLY
Current ABYC standards call for a test pressure of 3 PSI.
But the open standpipe only applies hydrostatic pressure as a function of
depth. In other words only the bottom seam is subjected to the max pressure.
The top of the tank seams only see a partial pressure.
Conversely what about a deep tank  of four or five foot maximum depth plus
the 2 foot stand pipe would have in excess of 3 PSI at bottom seam.

It was my understanding that the hydrostatic test was done with a capped
pipe and then air pressure of 3PSI was applied because this now equalized
the pressure on all surfaces.

An interesting side bar to this concerns the issue of water retention in the
tank after testing.
If you have a large capacity tank with multiple internal baffles, will the
surface of these baffles collect enough water film to amount to any sizable
quantity. In other words how do you ensure the tank is completely dried out
after testing?
I'm assuming the tank is properly designed with adequate limber holes in the
baffles so as not to create any pockets for water to collect. But with the
pickup pipe kept of the bottom by 1/2 inch or so there could be several
gallons of water in the tank from the start.

John, have you seen any specifications on how integral tanks are tested. Are
such tests any different?

regards
Arild