8 replies to this topic

[Zos]

I am offshore and a section of the plant needs to be leak tested. The procedure calls for high pressure nitrogen to be used, but that is not available (within the timeframe needed.) The commissioning manager is now planning to use process gas (i.e. mostly methane) from another section of the plant to do the leak testing. The section to be tested has previously had both hydrostatic and pneumatic pressure testing done, but a large number of flanges have been removed and replaced since then along with valves pulled for work and then reinstalled - so in my opinion it is likely there will be some leaking.

I would appreciate any experienced members viewpoint on (1) the safety of doing this and (2) whether it is common practice in the industry to do this (i.e. has it been done in your plant?)

Thanks.

[breizh]

Hi Zos,
As long you removed pieces of equipment you need to do pressure test ?
It's on purpose.
Breizh

[comp_bg]

It is common in on shore gas transmission (not sure about off shore work practices) but only after a full hydro test has been conducted to establish mechanical strength of system components. Initial pressure up = 30% of final pressure, leak check flanges and fittings. Subsequent pressure increases should not exceed 10% * MAWP (maximum allowable working pressure verified by the hydrotest).

Always blow down to atmosphere if a leak requires fixing. Do not work on flanges under pressure regardless of the testing medium.

first time poster, just my two cents.

[Elvin]

Comp_bg,

It is a normal practice when nitrogen is not available. is it named as service test and is equivalent of leak testing with nitrogen or air. Even if is a high pressure system, is it planned to be hydro tested and fully reinstated, and a derogation to do it with gas is required. As well it is a MUST to have flare system available and in operation (preferable).
A detailed M/S or procedure with a TRA attached is required as well. Pressurisation MUSt be done on steps, 10 to 20 bar increments up to normal operating pressure.
I think now is already done.

Take care

[kkala]

My only contact with commissioning was the first semester of this year (2011), when I had office work within a commissioning team for a project in a refinery (onshore). Conversations with commissioning team, resulted in following practice:
1. Hydrostatic (or pneumatic) test occurs with instruments removed from the pipes.
2. After reinstalling the instruments, leakage test occurs with air or nitrogen at a pressure lower than design pressure. If air has been used and the pipe is to transfer a flammable liquid, air is displaced by nitrogen.
3. Using flammable gas in leakage test instead of air or nitrogen was never specifically discussed. But the care of expediting nitrogen unit start up indicated that this was not thought of, apparently due to safety (risk of leakage).
It is pointed out that above has not been applied yet, so they may change a bit. But (mainly for my own knowledge) I would like to ask members to advise / comment on them. I see that leak test is implemented at higher than design pressure (http://www.cheresour...ng-of-a-plant/'> http://www.cheresources.com/invision/topic/12859-software-list-for-commissioning-pre-commissioning-of-a-plant/ and wonder whether a code permits the use of flammable gas as test medium. It seems this is permitted for offsore facilities on sea (see above posts), but what about a refinery build on land? Is above practice too conservative, or represents another practice also used. Hope of some advice.

Edited by kkala, 30 August 2011 - 12:09 PM.

[Himanshu Sharma]

'Leak Test' if this is the term and motive then the above procedure seems correct even for a Petroleum Refinery.IMHO, leak test and Pneumatic testing are not synonymous terms !

Pneumatic testing is carried out for the lines where the supports or structures are not designed for line full of liquid ,normally flare headers.Pneumatic testing is for mechanical integrity check while leak testing is for simple detection of leaks and may not necessarily be carried out at or above design pressure.

As per my experience Being part of a few commissioning teams for high pressure Hydro processing Units ,i have witnessed the following sequence for Leak testing of Reaction Section/HP section.

• Hydro testing About 1.3 times design pressure Multiplied by stress correction factor depending upon design temperature
• Leak testing with Nitrogen in Various Pressure Levels

• Low pressure about 7 kg/cm2 g
• Then about 50% of Design pressure ,with the help of RGC.

3.Then comes the Leak testing with Hydrogen in various pressure levels

• 7 Kg/Cm2 g
• 50% of Design pressure ,with the help of RGC
• Operating pressure ,with the help of RGC

RGC-Regeneration Gas Compressor

We often detected leaks while testing with Hydrogen that went undetected while testing with nitrogen,because of size of smaller Hydrogen molecule.Whole of the reaction section has to be depressurized to flare and inertized again for attending leaks ,majorly shell side exchanger flanges.

[kkala]

Former posts indicate that leakage test is not performed only with nitrogen but also with the actual fluid to be tranported (it may be called service test) in a later stage, at least in case where this fluid is a gas lighter than nitrogen. Onshore facilities sometimes proceed directly to leakage test with actual fluid, when nitrogen is not available. For the refinery where I worked in 2011, bypassing leakage test with nitrogen would be too risky.

[Babu Prasad]

It is depends on the stage. If your unit ready for commissioning then you can use Methane gas for leak but you must take care of H2S content in the Methane which will be higher from offshore gas. I suggest better follow the procedure given by your plant designer.

General Method followed in our Gas processing plant

• After hydro test finished, line has to be air blown to clear any foreign material and it will be leak tested with air initially as most of leak will known by low pressure itself. probably air pressure will be available at 9 bar
• Better prepare small loops which contain same operating pressure of piping/vessel/columns and lineup thro control valve and bypass valve. Do pressure test with air & eliminate the leaks. Try to hold for 12 hrs see the rate of leak.
• Once satisfied, inert the same loop with N2 and eliminate O2 content. Lean sweet Methane gas can be used in the steps of 5 bar until operating pressure of the loop. In my experience you can eliminate most of the leak at minimum pressure itself.
• Rotating equipments can be isolated and & it can be done separately with its design pressure once it connected with process system after flushing the related pipelines.
• If your gas is sour better use higher pressure N2 to avoid hazard conditions as most persons will be new to pant & not familiar with hazard conditions which may lead to human loss.

[lucaa]

Hello I am a chemical engineer and I have to write a thesis on the leak test oil refinery. So I'm looking for any material that can help me on this topic. Thanks in advance for your contribution