Gasket Installation & FRP Flanges 1

[mkkessel]

I am researching gasket installation on FRP piping and vessels and my search results have typically lead back to the vessel/piping or flange manufacturer. I'm curious if there's any empirical data on FG or FRP flanges that details maximum allowable stress? I've also been cautioned in vessel applications that the ability of the flange to accept stresses applied by sealing a joint are often secondary due to the relative strength of the neck coming from the vessel. If anyone has experience with written procedures on gasket installation in FG or FRP applications, it would be most helpful.

 

[StevenHPerry]

I'm not understanding the full problem here I suppose, but I'll venture an answer since no one else has.

With FG and FRP you aren't exactly dealing with high pressure situations. Throw in a full face gasket, make sure you got washers under the nuts, tighten by the usual pattern (i.e. skip around, don't go in a circle), and give it a go.

I would err on the under-tightened side and tighten joints as necessary during a hydrotest to achieve a seal, assuming that cleaning up the water would be less of a pain than replacing a fractured flange.

Is it really absolutely critical that you have a written procedure and achieve a seal on the first try?

- Steve Perry

http://www.linkedin.com/in/stevenhperry

This post is designed to provide accurate and authoritative information in regard to the subject matter covered. It is offered with the understanding that the author is not engaged in rendering engineering or other professional service. If you need help, get help, and PAY FOR IT.

 

[MikeHalloran]

I'm a little familiar with FRP marine mufflers.

My first choice for termination is a simple nipple, reinforced internally against crushing by a bonded stainless liner, joined to the remainder of the system by a silicone 'hump hose', with triple stainless hose clamps.

My second choice for termination, where flanged joints are required by a class society, are FRP/steel plate swivel flanges, and flanged rubber compensators. The flange bolts are tightened just enough to stop leakage, and silicone or polyurethane sealant is often used on the flange faces to further minimize the torque required.

In both cases, the goal is to minimize the possibility of moment loads from pipe movement being transferred to the (relatively thin) FRP shell through the termination.



Mike Halloran
Pembroke Pines, FL, USA

 

[jdarco]

The piping/flange manufacturer should have installation procedures for their flanges, but Mr. Perry pretty much has it right. Full face gaskets, typicslly a "softer" material (EPDM, Viton, Red Rubber), 1/8" or 1/4" thick, washers under the bolt/nuts (that is important).

Most piping/flange manufacturers will have a maximum bolt torque to acheive a seal.

 

[mkkessel]

Thanks for all. further research shows that ASME RTP-1 provides guidelines for material selection and installation. However manufacturers recommendations vary from this. None of the documentation I've found suggests the use of expanded PTFE, which is widely used in the field, although having much higher sealing stresses than the 60-70 durometer elastomers on the RTP-1 spec.

The purpose of this inquiry is/was to develop a series of guidelines for use of expanded PTFE in FRP and fiberglass flange applications. The physical properties and sealing characteristics of the material are well-known. Without some guidance on stress tolerances for FRP or FG flanges, the potential for catastrophic damage to vessels or systems is very real. So is the possibility of litigation resting with the gasket material manufacturer if this kind of failure occurs.

 

[StevenHPerry]

So is the possibility of litigation resting with the gasket material manufacturer if this kind of failure occurs.

Not likely. You'll get the cost of the gasket back, but that's going to be close to the limit of liability from any reputable gasket manufacturer.

Just start easy and tighten only as necessary.

A major FGD OEM uses a combination of split steel backing rings and steel spacers as field remedies whenever they get stuck with with a lap joint or raised face flange mating to an FRP flange. Keep the full face FRP flange from being sprung like a raised face.

Downgrading your bolts might be an option too. Pick one that will yield before you crack a flange. Perhaps A307-B instead of your normal A193s.

- Steve Perry

http://www.linkedin.com/in/stevenhperry

This post is designed to provide accurate and authoritative information in regard to the subject matter covered. It is offered with the understanding that the author is not engaged in rendering engineering or other professional service. If you need help, get help, and PAY FOR IT.

 

[Duwe6]

Sorry, but the flange stress introduced using an A307 Gr.B bolt and a standard washer will exceed the breaking stress of a typical FRP flange long before the bolt fails.

Please don't ask how I know this ugly fact.

 

[donf]

mkkessel,

While I can't answer your specific question, here's something you should consider:

http://vsptechnologies.com/wp-content/uploads/2011/04/vsp_opra_brochure_rev-6-010611.pdf

donf

 

[rconner]

Further to the astute revelation of Duwe6, in partial explanation flanges have been designed and applied most commonly in the last several decades in general with metallic materials that are much stiffer, less vicoelastic and stronger than the polymers/plastics (that have understandably exhibited relatively more problems).

The goal or guidance “tightened just enough to stop leakage”, though certainly well-meaning, is however in itself neither specific nor measurable, and furthermore may be of questionable practical value in the real world or field environment.

 

[Duwe6]

Most [all?] FRP manufacturers have recommended torque values. They may be a bit low, but these values are guaranteed to not ruin your flange. Read the torque carefully, as a lot of the values are inch-lbs, not ft-lbs.

 

[mkkessel]

Again thanks to all. Further research shows that ASME RTP-1 is most likely the prevailing document in this area. Two classes of FRP construction are considered. One has a maximum stress value of 1800PSI and the other 3000PSI. Torque recommendations do not exceed 40 ft/lbs. To a gasket material manufacturer these are extremely low figures.

Not only is the low available gasket load an issue with FRP/FG, but bolt stress, both initial and residual are called into question.

Given the limitations set up in RTP-1, gaskets must be highly conformable, and seat under extremely light bolt loads. It should also have excellent recovery characteristics, as residual tension in the fasteners is minimal. For materials other than low durometer elastomers, such as expanded PTFE, this has meant utilizing surface area reduction of the gasket itself to comply with the low load requirements.

90% of all gasket failure is related to things other than material selection. For purposes of making material recommendations, it is imperative to work backward from failure modes to do so. In this case design limitations and field installation practices are the driving forces to develop this recommendation.

 

[StevenHPerry]

I suppose I need to recalibrate my way of thinking...

Assembling low pressure flange joints doesn't even approach a scientific process in my mind. The pipe is already fit, the welds (or equivalent) are already made. I just send a bolting crew (foreman and a few helpers) out to drop in a gasket and tighten the bolts using a standard wrench. Once they claim they're done, I may or may not ping a few bolts before turning the hydro crew loose. If a joint leaks, the bolt crew gets to try again.

High pressure and/or critical joints are another story.

By some of the descriptions indicated here, I wonder if engineers are expecting skilled labor using torque wrenches, tensioners, etc to assemble every flanged joint to precisely the torque specified by an engineer. Is this really the norm?

- Steve Perry

http://www.linkedin.com/in/stevenhperry

This post is designed to provide accurate and authoritative information in regard to the subject matter covered. It is offered with the understanding that the author is not engaged in rendering engineering or other professional service. If you need help, get help, and PAY FOR IT.

 

[Duwe6]

Torquing is the norm on FRP flanges, at least until the pipefitters develop a "feel" for the correct, very low torque. Any pipefitter with a 12-inch crescent wrench can overtorque and destroy FRP flanges on lines up to 6"NPS. I've seen the aftermath.

I agree that this is not the Design Engineer's problem, and that it belongs to the Field Engineer, the Pipe Supt., and Quality Control. The problem is that our "Leaner, meaner, faster" management eliminated the QC guy, the Pipe Supt. is also filling in for at least one foreman, ahd Mr. Design Engineer is also expected to act as the Field Engineer.

Thus I highly recommend that the required torque values be prominently placed on the drawing(s) the pipefitters will be working to. Mr. Engineer has the resources to easily research the recommended torques. Mr. Pipefitter doesn't, and when he guesses, he guesses too high on FRP flanges.

 

[Hello000]

Recently i have done project with FRP/RTR lines upto 36 inches with pressure ranges from 10 bar to 27 bar. You shall follow only FRP/RTR flange manufacturer procedure for bolt tightening. For each flange size, manufacturer provides Torque increment & final torque value. You can get this table from manufacturer. Gasket material & type shall also be as per flange manufacturer recommendation. You can refer to guidelines available on manufacturer sites. Some of the good manufacturer are Bondstrand / Ameron / Aminatit, Future Pipe etc.