valve body thread leak 1

[someguy79]

I have a few brass bodied valves with threaded inlet connections. The inlet connections are leaking.

The application where they're used requires them to hold pressures of 0-300 psig at temperatures ranging from ambient to -200F.

I have not been able to confirm with the plant operators whether or not any sealant was used on the threads. I expect that none was used. Because of the outlet connection needed, only certain positions (every 90 degrees) can be acheived for tightening down the valve body.

The knee-jerk solution is to tape the threads with teflon tape or similar material, but I've come across a few cautionary tales about tapes and puttys. Apparently, typical white teflon tape will creep and squish out over time. eventually it will not provide a good seal. (or so the story goes)

I'm also concerned that because the coefficient of thermal expansion is so much greater for the teflon than either the brass valve body and 304L pipe. It might shrink more than the metals and cause leakage to be worse.

Can anyone provide examples of similar situations, and how it was dealt with? What (if any) sealing materials might be able to alleviate this problem?

 

[unclesyd]

I would give these people a call. The make the X-Pando thread sealant.

http://www.xpando.com/

 

[iainuts]

Hi Someguy, Some folks don't like pipe threads, mostly aerospace. I worked on cryogenic equipment in the rocket launching industry (Cape Canaveral & Vandenburg) for 8 years when I got out of college, and for threaded valves, we'd use a straight thread with some type of crush metal and/or pressure energized metal gasket. They weren't perfect, but they worked. Our standard was never to use pipe threads, and in fact, the range safety document which dictates such things, tells you pipe thread is not acceptable for cryogenic service.

I got out of aerospace and have now worked on the cryogenic side of the industrial gas business for 12 years. Here, we use pipe threaded joints almost exclusively. I've had better success with pipe thread than the straight threads. I'd never go back. In fact, some of the equipment I design has to go onto launch facilities at Cape Canaveral, Kennedy and Vandenburg, but we've always taken exception to the rule that says no pipe thread.

Even at 6000 psi and -452 F, pipe threads work much better than anything else I've used. The threads must be clean and well formed, 3 wraps of Teflon tape (keep it off the first male thread so it doesn't get into your flow stream), and generally some halocarbon grease on top of that, and you shouldn't have any problem. When the thread is made up, it sqeezes out the Teflon tape to the point it's all but invisible, it's so thin. Because it's such a thin film, there really isn't any concern about differential thermal contraction of the Teflon, nor of it cold flowing. There's just not enough left to move around. You might be just a tiny bit concerned about the stainless pipe and brass body, but your installation sounds a lot less demanding than some of the ones I've seen and I've had lots of good experience with this simple sealing technique.

 

[EdStainless]

iainuts hit on the key points. Remember that pipe threads cannot seal leak tight by design. Well formed threads and a very small amount of tape will give you good results.

= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection

http://www.trent-tube.com/contact/Tech_Assist.cfm

 

[ballvalveguy]

Hi Iainuts,

You seem to have a lot of experience with cryogenic applications and I have a question (for you or anyone else in this forum) about the pressure relief hole in a ball valve. My understanding was that the pressure relief hole should be in the stem slot (to relieve pressure between the seats to the ball cavity) and on the upstream side of the ball to relieve pressure to the process. However, I was talking to someone in the "fracking" industry where trucks put liquid nitrogen in to the wells to extract oil. His understanding was that the hole should be on the downstream side so that when the trucks are traveling, any liquid that was in the ball cavity could be released to the atmosphere.

 

[iainuts]

Hi ballvalveguy. Floating ball valves will have the hole in the upstream side of the ball so that the ball seals against the downstream seal. The seats may have springs on them, but the primary load on the seat is from pressure which forces the ball against the downstream seat. The vast majority of the valves we use are of this type - at least for the smaller line diameters of a few inches or less.

Trunnion ball valves on the other hand, have a seat seal on the upstream side, so I would expect them to have a hole in the downstream side, though I believe valves can also be designed without the hole, and they rely on one or the other un-loaded seats being the leak path. I can't remember ever using a trunion style ball valve in cryogenic service though.

 

[someguy79]

I was looking through some other posts here and realized I never posted the solution that was found (months ago). The thread sealing was not the root cause of the leakage as many here suspected.

The relief valves I discussed were installed incorrectly. The valve has a bushing at its inlet (part of the valve design, not mine) which is in turn threaded into the main valve body. Some brilliant person likely screwed the valve body down too tight and bungled up the seal between the body and the bushing. It is also possible that after doing so, the threads were backed off to align the valve at the correct 90 degree increment.

 

[unclesyd]

someguy79,

Thanks for the update.