I work with underwater vehicles all the time (We do R&D on sonar
systems) and the way we check for leaks is to draw a vacuum inside the
vehicle and let it sit as long as possible and see if the reading
changes. The less obvious advantage of this is that we can back fill
with dry nitrogen and not have to worry about condensation when the
vehicle hits the cold water.
We do have a fancy pressure test facility that tests for structural
integrity, but most of our shells are rated for at least 1200 feet. You
might be surprised how little aluminum it takes in a properly designed
shell to be rated for 1200 feet. I would think that as long as you
design your hull to hold pressure and are just worried about the o-rings
and other seals the vacuum test might work.

-Kris

wrote:
Hi,
I'm building a small underwater tethered / robot / submarine /
doo-hickey and I need a cheap way to pressure test the individual
components to make sure they don't leak. Max diving depth will be 70'
which is around 32psi and each component will be no larger than 6"x6".
I know each component should be tested to 1.5X to 2X its expected
range.

Anyway, my thought was to fill some already pressure resistant device
such as a pressure cooker or autoclave chamber (hello ebay!) with water
and then use shop air to increase the pressure. I don't need a viewing
window as I would basically take it up to pressure, leave it there for
a few hours, relieve the pressure and then check to see if anything
leaked.

Questions:
1) Is this concept of pressure testing sound?
2) Is there a cheaper way to do it?
3) Most importantly, is there an easier way to do this?

Thanks and please resist the urge to point out the 100's of other
failure points that I will be facing in this project such as watertight
bulkhead fittings, ballast systems, propulsion, electronics, stability
control, corrosion, battery systems, faily safe safety systems, etc,
etc...

-M