Checking for weld defects using low pressure pneumatic testing

[tmengineer]

Hello,

For a 100 m^3 tank would a test pressure of 0.6 barg be sufficient to detect if there was a defect/ leak in any of the welds?

A small overpressure is required for safety concerns as the tank would have to be located in close proximity (>10m) to the workshop as that is where the compressor is.

How long would the tank have to be left for before a leak was detectable?

The MAWP for the tank is 1.96 bar. A lot of tests talk about test pressures of 1.1 x MAWP (2.156 bar), but I think that is to test the integrity of the tank rather than checking for defective welds.

Any feedback would be greatly appreciated!

Tom

 

[racookpe1978]

If you did a fluid (soap film) "bubble test" on the outside you would see a nasty leak in minutes.

But a slow leak? Probably would nt be detectable at that low a pressure. The tank heating up/cooling off unevenly due to sunshine would overwhelm any such test.

Do a NDE dye penetrant test of the joint welds from BOTH sides. If no surface defects were found on either side of any weld joint or bolted joint, you probably have no through cracks that would yield gas leaking out.

 

[metengr]

Was the tank in-service and weld repaired?

 

[itdepends]

Avoid pneumatic testing if at all possible due to the large release of energy if it fails. Pneumatic testing to check weld integrity begs the question- what happens if a weld fails? Once a failure point starts it may propogate and then it'll get real interesting.

Second the NDE dye pen tests suggested above.

Can you fill the tank up and do a standard hydrotest?

As a chem eng/metallurgist the first part of any answer I give starts with "It Depends"

 

[Gregorydas]

I thing a DP and hydrotest will do the job for you.