Safety during hydrostatic test Pressure Vessels

Good advice above (particularly making sure ALL air is out of the vessel).

Here is a good safety report courtesy of the British:

www.hse.gov.uk/research/crr_pdf/1998/crr98168.pdf

and if you wanted another reference (albeit for pipelines, but still relevant): API 1110

http://www.techstreet.com/cgi-bin/detail?product_id=1506807

Don't go near. Under any circumstances. The overall energy is generally low, but a tiny jet of pressurized water can slice through flesh like a scapel (please don't ask me how I know this to be true, otherwise I'll sound like a careless idiot ). Use a Lexan viewing window, a CCTV, or (if you're really cheap like me) the ordinary blue paper towels that auto mechanics use - drape them across joints and they turn dark blue when even slightly wet.

You should never go close to a pressure vessel that is being pressurized or under a hydrostatic test. In fact you should be standing behind a safety barrier while at the test pressure. There are methods to observe the weld joints including blue paper towels.

You should only approach the vessel after the test is completed satisfactorily and when the pressure is reduced to the desing pressure.

Chama

You have not informed us what size pressure vessel you are testing. We often submerge the item being tested in a water tank to contain any debris that might result from a sudden rupture. This is done with pressure vessels such as fire extinguishers, SCUBA tanks, etc.

We used to test SCBA (for firefighting) air tanks by placing them (empty) in a container of water that was completely full, then filling. If more than a certain amount of water spilled out, that meant the tank was expanding more than the design amount at pressure and should be condemned.

The container was the same length of high-strength seamless pipe we used as a safety cage when filling the tanks.

We pressure test every day. Up to 15,000 psi. Proof and burst testing.

There is very little risk with the proper apparatus and procedure.

Basically a metal box with a plastic viewing window. Ideally you should have a relief on the system to control overpressure and a safety on the door so it can't open when pressurized. A flow control is a bonus to prevent a surge in pressure.

The most important point is to bleed all air from the system before pressurizing.

I would suggest you install a safety relief valve set to just a few PSI over your desired test pressure. I do this on all of the vessels I test/inspect. Also I use a simple broom handle waved in front of me before I stick my face or other body parts up close to weld seam before inspecting a pressurized vessel. The general rule I follow is to hold the test pressure for 10 minutes after closing the valve. If I see a drop in the pressure reading on the gauge then I reduce the pressure a good bit and start looking for the leak/leaks. The broom handle technique may seem a little simple but I have seen a high pressure pinhole do allot of damage to a broom handle when it could have been me. I do at least 1 hydrostatic pressure test a month and some months several more. So far I have not had a safety incident by taking these simple safety precautions. Another thing I would be careful of is testing boilers with feed water. I was inspecting welds inside a boiler at about 500 psi when a bonnet on a feed water valve broke and suddenly the pressure went up to over 1200 psi. Needless to say I shot out of there pretty fast and I was very lucky that all the welds and tubes held. The trick to a safe hydro test to me is to use a pressure source that allows you to bring up the pressure on the vessel very slow and controllable and never get in a hurry. A small 5 HP pressure washer and a couple of manual ball valves along with a bleed off and safety relief valve works very well for safely testing most vessels once the air has been bleed off IMO