I'm a Mechanical Engineer and I have been asked to reserch methods of leak detection for a massive piping network. The basic idea is that we need a non-invasive method of detecting leaks in a dry environment whereby the transported fluid is potable water. So invasive pigs etc would be too suceptable to polluting the potable supply and this leads me to my posting;

Can 'ground penetrating radar' penetrate as deep as 20m to give accurate depictions of leaks. I have researched this type of system, but it seems to be quite experimental at any real depth (i.e. most of it still seems academic.) Can anyone help or suggest any other methods?

Views: 85

Replies to This Discussion

My field is Electronics Engineering, so I can only offer rough guidelines, but first I think the problem needs to be defined a little tighter.

* Is the concern about leaks from fear of the water 'contaminating' the environment or the environment outside the pipe contaminating the water?

* Are you looking for a periodic inspection system or constant monitoring?

GPR sounds like a periodic inspection method, or perhaps an attempt to detect the location of the leak after its existence has been confirmed y other means. It also sounds like the symptom being looked for is water entering the environment. I'm not sure if GPR can detect the 'wet soil/dry soil' boundary, the most I've heard it used for is detecting 'ground/air' boundaries to seek out hidden tombs or 'hard rock/porous rock' boundaries to spot oil pockets, both od which have a clearly defined transition point to create the reflections (and in the case of the oil pocket, usually large enough that even if you drill a few hundred meters off center of the target, you'll still hit the oil.)


Other methods would depend on one crucial factor: is this an existing network already 20 meters down and buried, or is it a planned network waiting to be installed? I can think of a few monitoring-style systems, that would measure the capacitance or inductance at regular intervals to spot the leak by the change in what should be static values, but that would be difficult to install if the pipe is already underground.
Just to let you know.
GPR is ideal for finding wet/dry soil boundaries for shallow leaks.
But 20m is a real challenge. I've penetrated 60m in clean sands
on Cape Cod, but typical soils will only let you go about 5-10m,
and clay soils won't even allow 2m.

You could do resistivity to map a long profile, or even EM measurements
depending on budget and time constraints.

Andrew's comment seems good, if nothing is yet installed.
What type of soil are you dealing with?
the concern is two fold - leak to the environment will count as financial losses to the client (as well as the structural implications), and similarly impurity ingress would lead to loss of 'clean' potable water supply for downstream use.

The network has already been installed, the inductive/conductive methods you're refering to I have looked at (permeable wiring looms etc) but the invasive nature of retrofitting several hundred km of pipeline is not feasable.

The soil content is predominantly sillica (sand) with some very small concentrations of other materials, with only a very small portion (in comparison to the overall size of the network) being laid on groundrock.
My company manufactures equipment specifically designed for detecting leaks in underground piping and storage tanks using helium (totally inert and non-toxic) as a trace gas. If helium is introduced to the system our instrument can detect the helium as it reaches the surface. Detecting leaks as small as 0.05 gallons per hour is not a problem. We have an office in Oxford and local personnel in Manchester. Let me know if I can be of assistance.
John said:
My company manufactures equipment specifically designed for detecting leaks in underground piping and storage tanks using helium (totally inert and non-toxic) as a trace gas. If helium is introduced to the system our instrument can detect the helium as it reaches the surface. Detecting leaks as small as 0.05 gallons per hour is not a problem. We have an office in Oxford and local personnel in Manchester. Let me know if I can be of assistance.
Attachments:
HI David, I've been looking into this what type of setup did you use for the GPR, did you use a mounted setup on a portable unit (to pass over pipes), or was it antenna locatted along the length of pipes emitted to a reciever?

David C said:
Just to let you know.
GPR is ideal for finding wet/dry soil boundaries for shallow leaks.
But 20m is a real challenge. I've penetrated 60m in clean sands
on Cape Cod, but typical soils will only let you go about 5-10m,
and clay soils won't even allow 2m.

You could do resistivity to map a long profile, or even EM measurements
depending on budget and time constraints.

Andrew's comment seems good, if nothing is yet installed.
What type of soil are you dealing with?

RSS

Members

© 2020   Created by Marshall Matheson.   Powered by

Badges  |  Report an Issue  |  Terms of Service