[SBE] Ultrasonic Leak Detection

[Edwin Bukont]

You too eh.  A TV stn a few hours from here has the underground route as well, but into a hillside, adjacent to another property, to get the bend from the building and out to the tower. soon after they built the transition,,,the grass of the hillside was paved over...god help them if they ever have to service that.Edwin Bukont CSRE, DRB, CBNT Comm-Struction and Services LLC P.O. Box 629;  Bel Air, MD 21014 USA V- 410.879.5567 F- 240.368.1265 C- 240.417.2475 ebukont at msn.com Member: IEEE, SBE, AES, PMI Digital Media and Power Systems Integrators. A Harris Broadcast Channel Partner > From: cspacone at socal.rr.com> To: sbe at sbe.org> Date: Sat, 17 May 2008 10:25:17 -0700> Subject: Re: [SBE] Ultrasonic Leak Detection> > Edwin,> > The short answers are no, I haven't disassembled the system and no I haven't> listened to the lines. Listening to the lines with a stethoscope sounds like> an excellent idea and one that I had not even considered (ahhh, the wisdom> of the mailing list prevails!). I will get a stethoscope and do so first> thing on Monday morning. Hopefully it will be noisy enough to be heard over> the background noise.> > A recap may be in order:> > I was initially dealing with the pressurization manifold. It consisted of a> compressed air supply from the house compressor that fed a regulator check> valve (limited to about 10 psi), a desiccant canister (which still needs to> be serviced), a low pressure regulator set to about 3 psi, then into a 3 leg> manifold. Each leg has a ball valve and pressure gauge that feeds the> associated waveguide via polyethylene tubing and compression fittings.> Because of construction in the building I had to remove the house air and> the DirEng said get N2 and install it ASAP. I complied.> > I found 3 major leaks on the manifold assembly described above. One was at> the house air to manifold interface. The second was a poorly assembled poly> line / compression nut (it was actually folded over!) and finally a 1/4"> brass 'Tee'. The brass tee was the hardest to find because it had the most> obvious leak! After much hunting I was exasperated enough to start wrapping> all of the fittings in a soaking, soapy paper towel. Voila! Found the tee> was actually cracked down the back. It was a cheap fitting, cast then> machined and over tightened by the original installer (probably in a fit of> rage while trying to get rid of the leaks).> > All those problems are gone now but the system still doesn't hold air (or in> this case N2) so clearly the problem lies downstream of the manifold.> Initially I started this thread because of the location of the elliptical> waveguide flanges and pressure windows: directly over the top of the HPA's.> Clearly, I am not putting soapy water anywhere near the HPA's (yes guys, I> realize I can do this safely, but I'd rather not tempt the Fates). So I> looked into leak detection. In a previous life I worked at a medical> instrumentation startup and used vacuum leak detection utilizing helium as a> tracer gas. I'm not about to invest thousands on a bad idea: pulling a> vacuum on waveguide is the antitheses of what I am trying to accomplish. I> have no intention of pulling beach air (I'm in LA about 2 miles from the> shore), laden with moisture and NaCl, into the waveguide.> > So, keeping the pressure applied (and admittedly delivering an unknown> amount of N2 back into the atmosphere) I decided to look for alternatives.> Being a HAM (KD6OUB) I had read a fairly recent article in QST regarding> using ultrasonic's to detect arcing on power poles in an attempt to find and> mitigate HF interference. I figured this method could be used to find gas> leaks. I did a quick search on the internet (Google is your friend!) turned> up a bunch of hits. That's when I decided to drop a note to the list and see> if anybody had experience with this leak detection method. Clearly, the> thread has wandered a bit but I'm taking away the conclusion that the method> is probably viable but may or may not be effective in my installation. In> any case I'll be asking the boss to buy one of the smaller, less expensive> (and probably low capability) units before I start shotgunning the problem.> This is our backup uplink and supports our ad hoc HD feeds so is considered> 'revenue generating'.> > I'm not about to do anything that would jeopardize that backup facility or> revenue stream. If it becomes an emergency then it becomes another story.> I'm not suggesting that you (or anybody else on the list) advocates> shotgunning, just stating my reluctance to do so.> > One point you make about ease of access is the unfortunate fact that the> three sections of elliptical exit the building and go underground thru> conduit to the dishes. God, please don't let it be in there!> > 73,> Chris> > -----Original Message-----> From: sbe-bounces at sbe.org [mailto:sbe-bounces at sbe.org] On Behalf Of Edwin> Bukont> Sent: Saturday, May 17, 2008 9:39 AM> To: sbe member discussion mail list> Subject: Re: [SBE] Ultrasonic Leak Detection> > > J. Fred Reily always said 'its not the tube' and the leak is not always on> the tower.> > Several engineers and tower folks had spent the better part of 15-20 years> dealing with a leak on 600' of 3-1/8 rigid line. By the time I got there,> the place was being pressurized with a dewar...the equivalent of I think 8> typical N2 tanks. I did replace that with a dehydrator, which had to be> serviced after about a year. New station owners said to have the tower crew> inspect the line again (mind you, it was always the same tower crew) when> doing the due diligence.> > In the process of their inspection, for whatever reason, I stepped up on the> base pier of the tower and wrapped my arm around the transmission line,> which was now about ear level....and thought I heard something rushing in> the line. A return visit with stethoscope indicated there was indeed some> sort of noise in the line. Working back along the line, it got louder as I> went towards the building. Since the level got louder, it was unlikely to> be air rushing in the line, but maybe something vibrating the line.> > It turned out to be where the line entered the building, and some knuckle> head had mounted the coax switch almost flat to the outside wall (straight> thru, not with elbows and into the switch, right above the dummy load, with> just barely enough room for the downward transition to the dummy load. They> used a short peice of solder flange line to get from the switch to outside,> such that the flanges were inside the block wall. The wall was sealed, of> course. The rigid line seal was damaged, probably from wrestling with the> too tight fit. The noise of the gas was not noticed inside the building> because the building is noisy and the gas was venting into the wall. All> that wasted time and cost to inspect a leak that could have been avoided by> proper, rather than cute, engineering. The building was big enough and had> enough ceiling clearance to properly mount the switch and bring a proper run> of line inside the room. > > The point being,,,has the system been fully disassembled at the easiest to> access end and fully inspected rather than assuming the leak is further> along. Have you tried listening for any noise in the line?> Edwin Bukont CSRE, DRB, CBNT Comm-Struction and Services LLC P.O. Box 629;> Bel Air, MD 21014 USA V- 410.879.5567 F- 240.368.1265 C- 240.417.2475> ebukont at msn.com Member: IEEE, SBE, AES, PMI Digital Media and Power Systems> Integrators. A Harris Broadcast Channel Partner > _______________________________________________> SBE mailing list> To unsubscribe, go to http://seven.pairlist.net/mailman/options/sbe> _______________________________________________> SBE at sbe.org> http://seven.pairlist.net/mailman/listinfo/sbe> > __________ Information from ESET NOD32 Antivirus, version of virus signature> database 3106 (20080516) __________> > The message was checked by ESET NOD32 Antivirus.> > http://www.eset.com> > > > __________ Information from ESET NOD32 Antivirus, version of virus signature> database 3106 (20080516) __________> > The message was checked by ESET NOD32 Antivirus.> > http://www.eset.com> > > _______________________________________________> SBE mailing list> To unsubscribe, go to http://seven.pairlist.net/mailman/options/sbe> _______________________________________________> SBE at sbe.org> http://seven.pairlist.net/mailman/listinfo/sbe