Hydrotest Criteria 2

[bob2005]

I am installing a concrete lined, ductile iron water pipeline. The pipe is 1000mm (40”) diameter and each section is 6meters long and the connections are push-on bell & spigot type. The pipeline is 104km long. (60 miles). I have valves at a spacing of: 1.5 km minimum & 5km maximum. Test Pressure is 1847kPa (18.47Bar, 268psig)and will be held for 24 hours. The pipe will be buried in the ground 1meter below grade. Surrounding the pipe will be brackish ground water. I will be using Potable Water to do the hydrostatic test. Due to expediting the construction schedule, I am forced to perform the test between the valves. Obviously the maximum length of pipe to be tested will be 5km with approx. 850 connections that can leak.
My problems is Leak Detection.
1) How can I detect the exact location of any leaks, over this length.
2) What method and equipment do I use.
3) Name of equipment supplier and web address.
I am thinking of an acoustic system, like ultrasonic.
I have no experience of leak detection over this distance.
Any advice would be appreciated.
Regards
Bob2005.

 

[vzeos]

Bob,

If the pipe will be surrounded by brackish ground water, you might think about doing a pre-test with high-pressure air and look for the air bubbles in the surrounding water. Then if you need to do a hydrotest for regulatory reasons, you can follow-up with an official hydrotest. In either case, testing against valves is not a good idea. If your valves are passing, you may not be able to sustain your test pressure for 24 hours. I would suggest temporarily removing the valves and installing end caps or frying pans for the test.

 

[zdas04]

If you put in skillet blinds, make sure that they are thick enough. For 1,000 mm pipe tested to 150% of 1.8 MPa, I get a required thickness of 34.1 mm. If you skimp on this then you create a metal structure that looks very much like a wok and is really difficult to get out of the pipe. If the wok deforms around a ball valve ball then you may be taking it out with a cutting torch (and then starting the test over again).

Don't test against block valves unless they are trunnion-ball valves and you can put a gauge on the body cavity (i.e., leakage past one seat gives you pressure in the cavity, leakage past the second seat will tell you why your test failed).

David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

The Plural of "anecdote" is not "data"