Maybe you are looking at this the wrong way. In every threaded connection is there not always a spiral shaped hole leading from inside to outside that follows the thread valleys? A threaded steel connection is not sealed by filling the thread valleys with the thread peaks by forcing steel to deform such that there is no gap. The sides of the valleys have some chance to be very close but that is not true for the valley bottoms. The pipe dope’s job is to plug that spiral hole. I would wager a guess that a combination of a few factors are involved including high system pressure and something wrong with the dope product or procedure. I’ve heard accounts of a contractor keeping a fitter full time for one year on site to fix leaks just like you mention until the occurrence rate dropped below some level. The fix was to replace the fittings and the drop at each discovered leak. They looked at everything; threads, tools, fittings, fitter crew arm strength and so on without finding anything definitive.
It may also not be possible to torque check any existing connections with any meaning if that is what is planned. The break torque for any connection increases with time. Since it is the dope that seals the spiral hole, and since dope products vary in material and application, which would influence the turning force, then it may not be possible to develop a torque metric for a no-leak installation. Allan Seidel St. Louis, MO > On Jan 13, 2016, at 5:54 PM, Steve Leyton <[email protected]> wrote: > > I have an interesting and challenging situation: > > My firm is overseeing remediation of rampant leaks on all floors of the > sprinkler systems in a 6-story building. Our first suspect was a really bad > MIC infection, but that's not it. What we think we have is situation where > the threads were not cut correctly or where arm-over and drop piping wasn't > correctly made into the T.I. ready shell system branch outlets. The owner > has hired a metallurgist to do all sorts of analysis on the piping and one of > the things we want to do is torque test some of the samples to determine > whether they're made in correctly. But we need a benchmark: I know that > there are different training methods for piping installation and that making > pipe in 4 or 5 turns by hand and then a certain number of turns by wrench is > one of the metrics that has been widely used over the last few years. But > we have not been able to ascertain whether or not there's ever been a torque > load value established for making in threads. Can anyone help me find > publishe > d standards of care and practice for threaded piping make-ups in terms of > torque or turns? > > Thanks in advance, > Steve Leyton > Protection Design & Consulting > San Diego, CA > > > > _______________________________________________ > Sprinklerforum mailing list > [email protected] > http://lists.firesprinkler.org/listinfo.cgi/sprinklerforum-firesprinkler.org _______________________________________________ Sprinklerforum mailing list [email protected] http://lists.firesprinkler.org/listinfo.cgi/sprinklerforum-firesprinkler.org
