Leak testing of large volume fuel gas pipework

[MediaDaveT]

We are planning a test under AS5601 section E of a fuel gas pipework. The pipework volume is huge, with more than a 1000m of various radius pipework from 2" up to 8" for more than 16,000 litres. The test is at 700kPa and although the code is vague on this, we have agreed a 24 hour hold time. We will use N2 as the test medium. Technically we are not supposed to have any pressure drop but we will see some thermal movement overnight.
I am looking for advice from anyone who may have pneumatically tested pipe volumes of this size and issues we may have.

 

[brpillai]

Pneumatic testing, involves risk. It is reccomended only when hudrostatic test is not possible. Anyway 700 kPa is not very high pressure.
Are you doing testing of all lines together as one test pack or are they sepererate? if sperated, risk is less.

When you do the test for 24hrs, the atmospheric temperature should be back to the starting value. I feel better to start in the morning and finish next day morning to avoid more temp fluctuations than by doing from afternoon to next afternoon.

 

[LittleInch]

16 cubic metres might seem large to you, but in pipeline terms that is really quite small.

If this is a leak test, presumably you are testing joints, valves, screwed connections etc with water? then leaving it to see if there is a leak in a section you can't test.

If you can sectionalise it you will be much better off trying to find any sections with a leak. The smaller the volume the easier it will be to find any leak. Temperature effects with gas are much lower than liquid as the temperature is measured in degrees Kelvin so a few degrees centrigrade difference will make a very small change in pressure.

Beware the nitrogen might initially be quite cold coming form a bottle so give it some time - say an hour - to get up to temperature.

Is this buried? If so you're much better off than above ground.

There is a lot on this site about pneumatic testing which yu can read, but basically look at each place where pipework is above ground and say what happens if that thing (valve, end cap etc) flies off at some speed. what / who would it hit? How do I protect myself from this.



My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[MediaDaveT]

Thanks guys thats good feedback.
The pipework is all above ground and the test has to be pneumatic as this is part of a code requirement...my big concern is related to getting the system stable prior to starting the test.
Technically the test needs to show zero pressure drop after 24 hours.

 

[racookpe1978]

OK. So how are you going to assure that the start and end temperatures are "exactly" the same? I know what the "spec" says, but that won't help when your pipe measurements show an increase in fluid inside (temps rise) or a decrease (temps fall) in sunshine, rain, wind, overnight, different clouds, etc.