Finding leaks in double wall pipe 1

[h0ckeypuck]

I have miles of double wall (pipe in a pipe) piping at my semiconductor facility for chemical and wastewater distribution. All is well until a leak detector is activated and I have to find the source of the leak. I've tried thermal imaging and super sensitive audio detection to try find the leak without success. Many of these pipelines are heat traced and insulated to keep them from freezing which complicates matters. Does anyone have a suggestion other than cutting windows or holes in the containment pipe to find the source of the leak?

 

[Montemayor]

h0ckeypuck:

Could you describe your jacketed pipe (pipe in a pipe) with specific details? The reason I request this is that there are different configurations, depending on your scope of work. I've used a similar installation in a process plant and my scope was to maintain the process fluid in the liquid state by using a continuous steam jacket on the pipe. The process fluid was a solid at ambient conditions. My pipe ran was extensive and I recall we used 1-1/4" pipe inside 2"; both were sched.40. We used 20-ft lengths of the 1-1/4" - all butt-welded and 100% radiographed. The buttwelds were exposed and not jacketed. All 2" jackets were approximately 19 ft long and we used jumpers from jacket to jacket. This was done to allow flanged connections every 2 or 3 pipe lengths and gave the ability to test these sections individually for leaks by selectively inserting blinds at the flange connections, with a pressure gauge on the jacket. I realize this type of configuration starts to add up in costs due to the jumpers, but the maintenance flexibility and relative ease of locating any leak in a minimum of time gave the justification. Maintenance time was important due to the need to minimize unit downtime.

Your configuration might be different due to a different scope, but staging your jacketing also has other advantages such as good distribution of expansion stresses on the pipe. If you are jacketing for the principle of "double containment (containment of leaks) Then it would seem more justifiable to expend more money towards staging the jacketing - although you might not want to have any "bare" pipe spots at all. That's why I request more details of your needs and scope.



Art Montemayor
Spring, TX

 

[weeeds]

h0ckeypuck,
What activated the leak detecter? What are your sensing?

 

[h0ckeypuck]

Art/Widla,

The outside (containment) pipe is installed to contain any leaks of wastewater from the primary pipe so that they are not released to the environonment. This particular pipe is put together with solvent welded joints (both primary and containment pipe). The containment pipe is then heat traced (to prevent freezing of the primary), insulated and then jacketed to keep the insulation from getting wet. Electronic leak sensors were installed on "drip legs" at 100 foot intervals and are suppposed to indicate approximately where a leak in the primary pipe occurred. Unfortunately the drip legs penetrate the bottom of the containement pipe too far and liquid is traveling several hundred feet before activating the leak detection. Bottom line is I don't want to make swiss cheese out of the secondary containment pipe (my usual method)if there is another way to locate the pinhole leak.

 

[sailoday28]

HOw about an acoustic device?
If located on the bottom outside steel od, the time for sound wave to travel across annulus will differ greatly between a voided space and one containing a liquid.

 

[JEB66]

To add onto sailoday's advice, try these guys:

http://www.acousticmonitoring.com/

I have used these sensoros to "hear" leaks inside the steam generators with pretty good sucess.

 

[MikeHalloran]

One thing to consider is using a portable drill and a back- spotfacing tool to reduce the inside projection of the drip legs, so the leak detectors can work as intended.



Mike Halloran
NOT speaking for
DeAngelo Marine Exhaust Inc.
Ft. Lauderdale, FL, USA

 

[weeeds]

Another option to consider is to charge the containment space with a few PSI of Helium then sniff for higher levels of helium at various locations from the outside. There is a process to do this but you may have to bring in an outside contractor. Do a google search for Helium Leak Detection and see if there is anyone in your area. If not then let me know where you are located and I can give you some contacts. You should be able to locate the leak quickly and accurately.

 

[bvi]

If the system has a pitch, the leak will probably be higher than the highest leak detection point. The only reason it wouldn't be is if you flooded the containment pipe. If you did have sufficient leakage to flood it, you can drain the leaked fluid, and then see where you start detecting leakage first.

 

[hydrodude]

It sounds to me like your pipe is above ground or you would not need heat tracing. Why not use an infared thermometer to check the temp in the area you suspect and then push heated or chilled fluid through the line at the same time you are using the infared, you should have an almost instant temp change where the leak is.