in-service torque procedure

[yamoffathoo]

In order to reduce the risk of over compressing a leaking gasket or over stressing the fastener, would it be worth the risk to apply breakloose torque (CCW) first, before applying CW torque? The added risk is local relaxation of gasket area affected by fastener, which might temporarily increase existing leak. Is permanent gasket degradation possible with this approach? Line is in depressurized state (class 900 flange with 20 psig static head of water).

 

[HeviiGuy]

Whew... this is one of those tricky ones. Chasing a leak at this point can be like chasing one's tail. :-(

I don't suppose that the flange is set-up for tensioning is it - ie, studs extend beyond the nut by at least one diameter. If so, and if you absolutely can't take the flange apart to re-seal it, simultaneous tensioning would be one avenue to explore. By pressurising each tensioner, holding the pressure and turning all of the nuts down, you may find one or two which didn't have sufficient load. Since you'd be getting consistent gasket compression in this manner, there shouldn't be an issue of gasket damage. Or, rather, perhaps I should say further gasket damage.

Whatever you end up doing, it'll be a shot in the dark. If you really want to make sure that the flange won't leak, break it out and start over. Or, next best thing: Box it up and squirt some epoxy goop into it. This might buy some time until the next scheduled turnaround/outage.

Ciao,

HevïGuy

http://www.heviitech.com

 

[rconner]

Wow, it's kind of amazing a somehow assembled Class 900 flanged joint apparently won't hold 20 psi static water pressure! While I guess there could be lots of reasons, (including the obvious many folks would think about in wondering about alignment, cleaning and tightening etc. procedure), I guess I am first curious however what specific types of gasket and type/material of bolting is involved? Was any kind of gasket compound used in the initial making of the joint, and if so what specifically was used?

 

[HeviiGuy]

I was under the impression that the process is down and that the 20 psig within the pipe/flange is simply due to the head above the joint.

Ciao,

HevïGuy

http://www.heviitech.com

 

[yamoffathoo]

I should have mentioned that yes, it is nuclear and no, flange removal is not possible while at low pressure (normal operating is 1200 psig) and no, studs are 2-3 threads proud of nuts. Engineering was criticized for recommending CCW break away torque because of an OSHA standard that forbids pressure boundary fastener removal while pressurized...we argued that loosening is not removing. Loosening then tightening seemed to be less risky than hoping to break-away while tightening without overtightening. We were also counting on loosening to require 20% less torque than tightening. The application is a vertical 6"nps vent pipe with blind flange, spiral wound flexitallic gasket, no goop and normal operating temperature of 500 degF heavy water.

 

[HeviiGuy]

Yamoffathoo, I'd have to agree with the authorities: While loosening isn't technically fastener-removal. The effect could be the same. Whereas backing off and then trying to reseat may result in additional compression, it may also provide an opportunity for Pandora and her friends to come out and say "hello".

I'd still suggest that if there's no way whatsoever of opening the joint and re-doing it now. Put an epoxy-filled box around the thing and let it run until the next shutdown. Meanwhile, review the procedures and quality control process to ferret out what caused this problem in the first place. Although there are many factors at play, one of the most common in situations like this is inconsistent residual bolt load caused by inconsistent applied load -further exacerbated by load transfer during tightening. This is, granted, assuming that somebody hadn't used the wrong gasket in an already-proven installation ;-)



Ciao,

HevïGuy

http://www.heviitech.com

 

[yamoffathoo]

Your reply is a tad cryptic, so I'll try to summarize. Which method (CW or CCW) of checking torque and applying proper preload will minimize the risk of degrading the joint? Studs will be ground flat and ultrasonic inspection used to look for cracks. Gasket is original (25 years of service).
1) adjacent stud integrity: very low risk but higher with CCW.
2) torqued stud integrity: higher risk than 1) and higher with CW
3) increased leakage rate due to gasket over-compression, loss of resiliency at temperature: higher with CW
4) increased leakage rate due to unloading and "disturbing" interfaces: lower risk than 3) and higher with CCW
In summary CCW is lower risk than CW...agreed?