Flanged Steam Piping - Proper installation method 9

[Geby]

Leaks have occurred in flanged piping joints at a 125psi working steam pressure. Piping is schedule 40 steel. Leaks have occurred in pipe sizes 4", 6", 8", & 10". I have been tasked with identifying what needs to be specified for future installations to ensure that no leaks occur. I speculate that the leaks have occurred for one of the following reasons:

*Wrong bolt type used for flanged joint connections. Grade #2 bolts have been used thus far.
*Improperly torqued bolts.
*Wrong gasket type used.
*Poor pipe alignment prior to bolting flanged fittings together.

I need to identify the "right way" to make flanged steam piping connections. Specifically, what type of bolts should be used? Is a higher grade of bolt needed? What should the bolts be torqued to, and what kind of care should be taken to ensure that the bolts have been properly torqued? Should a torque wrench be used? Should the bolt threads be lubed prior to assembly?

Apparently, various companies sell "flange bolt kits" that include all of the bolts, nuts, and gaskets needed to make a single flanged connection. Pennstan Supply sells one that comes with Grade #2 bolts, but Fastenall's kits only come with Grade #8 bolts.

I have got a retired engineer telling me that only Grade #8 plain finish bolts should be used for steam service, because grade #5 will stretch over time due to the heat, even though it is only 300-400 degrees.

I will appreciate any advice. Thanks. Geby

 

[codeeng]

Our gasket standard for this service is 304SS spiral wound, with graphite filler. Per B16.5 either intermediate strength or high strength bolts should be used. Our standard is SA193-B7/194-2H (high strength). Bolts > 1 1/2" we normally like to see hydraulically tensioned. For smaller bolts we use the recommended torque noted by Garlock in the attached link:

http://www.sealsolutions.com/techinfo/

 

[codeeng]

Here's a better catalogue for torque data on spiral wound gaskets:

http://www.garlock.com/products_cat...tal+Gasket+Catalog.pdf&submit.x=33&submit.y=5

Go to GMG-1-1

 

[TBP]

Lots of cut "paper" gaskets will work in this pressure/temperature range as well - you just need to make sure that the material is rated for the temp, and is no thicker than necessary.

Ensure that the maintenance guys are "hot bolting" as well - going back after a day or so, and retorquing. Lots of times, you won't get anything on the nuts, other times, it's amazing how loose they are.

Make sure that the system is designed properly. If there a flange that can become "cocked" when the steam line heats up and expands, that will often be the source of a leak.

 

[svanels]

1) Check bolt thighting sequence
2) Check faces for dents, burrs etc.
3) Check tools the fitters are using (you have canibals which preferred tool is a sledgehammer when they need a screwdriver)
4) I wouldn't use grade 2 bolts, normally grade 5 and up
5) Are there bolts, nuts and gaskets in the stores? If not they (mechanics) are capable of anything to fix the job (even tie- raps!) The temporary solution that becomes permanent.

 

[JohnBreen]

Hello,

May I suggest the companion document to the B16.5 Standard - it is ASME PCC-1-2000, Guidelines for Pressure Boundary Bolted Flange Joint Assembly (An American National Standard). This document provides good guidance for solving the leaking flange issue. Also, from ASME Code for Pressure Piping, B31.3 - Process Piping, paragraph 335.1.1(c):

(C) Flanged Joints. Before bolting up, flange faces shall be aligned to the design plane within 1 mm in 200 mm (1/16 inch/foot) measured across any diameter; flange bolt holes shall be aligned within 3 mm (1/8 inch) maximum offset.

There are those who believe this tolerance is excessive and who recommend that half the B31.3 tolerance (1/2 mm in 200 mm) be used.

Then of course there is the issue of excessive bending moments being present at the flange pair due to thermal expansion and/or inadequate support. The design of the piping system (and especially the flexibility and stress analysis) phase, must address this issue. It may be worth the effort to have a stress analysis performed of the "as-found" piping system if there is a persistent flange leakage problem. If the bending stresses are too high, no amount of optimizing the bolt-up procedure will solve the problem.

Regards, John.

 

[StressGuy]

I'll second John's call that you may want to have the stress rechecked (for a 300F to 400F system, the piping system should have had a stress analysis done to begin with, though not all stress engineers think to check flange bending stress). Also, even if the stress analysis was done properly, if a flange set is added later to facilitate field assembly, the stress engineer may not have been told.

With regard to bolts, you need to check material compatibility with regard to thermal expansion. I've seen systems leak because the bolts were incompatible and had a higher thermal expansion factor than the flanges. This will inevitably lead to the joint loosening.

Edward L. Klein
Pipe Stress Engineer
Houston, Texas

"All the world is a Spring"

All opinions expressed here are my own and not my company's.

 

[sjrfc2]

I've seen specs that require the flanges in steam service to be re-torqued following the first thermal cycle. This is to address gasket creep. The gasket creep will occur in spiral wound as well as sheet gasket due to the temperature. There are several WRC bulletins on gaskets and some of them show this.

You don't call out what gasket your using but we've recently had problems with some elastomer/aramid sheet gaskets in steam service. They became brittle and broke apart. One thing that we noted was test data showing degraded performance of the elastomer aramid gasket especially if gasket seating load was low.

If you can use them I think the spiral wound/graphite gaskets are going to work the best but B16.5 has recommendations on gaskets/bolts/weld necks if you're using 150 class flanges.

 

[unclesyd]

A lot of good points have been made on possible problems and solutions thereof.
One thing touched on by sjrfc2 is the use of spiral wound gaskets with Class 150 flanges. We have found this to be especially true with the 3" and 8" Class 150 flanges.
If you do have Class 150 flanges the makeup of the flange is very critical in that the flanges have to kept parallel, pulled up evenly, and tightened in the proper sequence. It has been my experience that overall you can better accomplish this using studs.
Making up a flange is actually a state of mind based on mechanical realities. If you monitor the maintenance mechanics you will find some will never have leaks while a few will always have trouble with leakers.

Talking Points
The mention of Grade 2 bolting being used on any process equipment scares the hell out of me. Grade 2 bolting shouldn't ever be allowed on a plant site. Grade 5 bolting isn't for process equipment, including utilities. The proper fastener material, readily obtainable, should be ASTM A193 Gr B7 or B16 with 2H nuts, as stated by codeeng, depending on temperature.

Watch these gasket kits as 99.44% of the folks making them up don’t have a clue. A point of history, the idea of flange kits was started by the utility industry due to the character of the work force. This is especially true for some of the big named companies.

I’ll have to get on TBP for using “hot bolting” where he should be using re-torquing or hot torquing as referenced by sjrfc2. Hot bolting is what it says, changing out bolts one at a time

For some older systems where the flanges are adequate to properly seat a spiral wound gasket we use metal inserted Grafoil ring gaskets.

If possible on spiral wound gaskets we like to have the inner ring. This is must on 12" and above.

Never reuse a spiral wound gasket no matter how good it looks. You can check the effectiveness of the flange makeup procedure by checking the thickness after use. It is a good practice to observe a flange and gasket that is going to be reworked due to leaks. You cannot depend on the observations of others. Check every thing out with an open mind.

Now matter what tightening procedure used train the mechanics to use their eyes to ensure that the flanges are parallel prior to tightening the bolts.

If you are to reuse a bolt make sure it is clean and guages out.

Use a thread lubricant that everyone is happy with. Working with steam lines watch out if using aerosol cans. Someone always sits a can on a operating steam trap.

Make sure you have the proper tools and they are in good condition. We don’t allow open end wrenches, crescent wrenches, or 12 point box ends or sockets to be used for tightening bolts. In some critical areas no slip-joint pliers.


Never tape across a gasket, either ring or spiral wound.

Lefty Lousy, Righty Tighty

 

[TBP]

I dunno, unclesyd, this is what "Bolt Science" gives as a definition:

"HOT BOLTING

This term is used for the completion of maintenance work on a bolted joint when the joint is under loading. This can involve the replacement of individual bolts. There are risks both to the joint itself and to health and safety associated with this technique."

One more question for geby - is your system full of 125# cast iron flanged components, and are you mating those flat faced flanges to 150's with a raised face? There's a bunch of issues if that's the case - including what bolt grade is used. If I remember correctly with 125 CI, you have to use something like A307 Gr B, that has a MAXIMUM listed tensile strength. You've also got to remove the raised faces off the 150 flanges, use full face gaskets, etc. It's quite a production.

 

[Geby]

I would like to thank each of you for taking the time to respond to my post. This is the first time that I have posted a question to this forum and I can't believe how much good advice you all have offered. What a resource this forum is.

I now have my work cut out for me reading through the appropriate ASME standards and investigating the existing leaky flanged connections.

 

[Gadge]

*Wrong bolt type used for flanged joint connections. Grade #2 bolts have been used thus far.

Use correct studs, washers & nuts.

*Poor pipe alignment prior to bolting flanged fittings together.

Check flange faces for damage. Use taper drift to align bores. Check facing gap with a feeler gauge, not by eye. Start tightening up from the wide side.

*Improperly torqued bolts.

Use correct bolting procedure. Always use correct torque indicating tool. Repeat procedure after system is at working teperature/pressure.

*Wrong gasket type used.

Never re-use old gaskets. Check on the side of the flange for spex, this should be specified in vendor drawings.

Is the flange facing flat face, raised face or ring joint?

You can find most of the manufacturer spex on Google.

 

[tombmech]

TBP and Gadge mentioned what I was thinking - the flange faces. Raised faces are a good practice. Visual inspection and identifying leaks are much easier, and it's harder to hide a shoddy installation.

If this is a systemic problem, JohnBreen and StressGuy are right. If the bending stresses at the flanges are too high, gaskets and bolts don't matter.

Finally, I agree with Geby - lots of good info!

 

[TBP]

A huge (and I mean HUGE) percentage of industrial steam systems that operate at 125 PSIG are simply filled with 125 CI valves, control valves, strainers, trap bodies, etc. The safety valves, however, typically lift at 150 PSIG. If the safeties lift at any steam pressure over 125, all of these bits don't meet Code, right off the bat. Even if the safety valves lift at 125 or less, the mating 150# flanges have to be flat-faced, along with rest of the requirements.

I have a deep, deep hatred for 125 CI in steam & condensate service.

 

[doye]

I will want to know if venting under the sea is acceptable practice in Oil and Gas Design.