SORF to WNRF flange 1

[XL83NL]

I have a question, hopefully some expert(s) may enlighten me on with some remarks.

I have an SORF flange, which will bolted onto a WNRF flange, both cl. 300. Design temps around 350 C, 17.5 bar design pressure, medium is N2/H2/light hydrocarbons.
Spiral wound gasket with inner and outer ring to be used. Design code B31.3, normal fluid service.

I think the Code doesnt 'disallow' this bolted joint, but is there any 'potential for problems' here, say to so? Perhaps gasket stress seating, unequal stress in the SO flange, .. ?

 

[pennpiper]

There is no problem bolting a Raised Face Weld Neck (RFWN)flange to a properly welded Raised Face Slip On (RFSO) flange.

 

[XL83NL]

Thanks pennpiper. I've just found a similar discussion with similar replies in thread 297409.

Quickly did an VIII Div 1 app 2 calc. No real problems there, however as mentioned by moltenmetal thread 297409,
would you maybe worry about the gasket getting torn apart?

 

[moltenmetal]

Spec the correct gasket and the joint will not be a problem.

Spec the WRONG gasket and you're in trouble.

The ID of a spiral wound gasket needs to be no smaller than the OD of pipe (NOT the ID of the pipe!) any time a slip-on flange is used. Otherwise the potential for the unsupported and uncompressed portion of the gasket to fail inward by coming un-wound definitely exists.

 

[Duwe6]

Raised Face = RF. RF-to-RF is a standard connection. Just use a regular gasket.

 

[XL83NL]

Thanks all, Duwe6, I understand RF needs to go to RF. I was more concerned on the SO to WN connection.
I was 'hoping' for more mechanical answers in that direction.

As implied by moltenmetal,
the SO flange obviously doesnt have the same raised face area as the WN flange.
This allows for the potential of gasket seating overstressing, perhaps? Or what else?

 

[moltenmetal]

If I could find the )(#@$ cable to connect my camera, I'd post a picture of a gasket which failed in the way I'd described, then perhaps Duwe6 would believe me.

Regular spiral wound gaskets DON'T WORK PROPERLY with RFSO flanges, either between two RFSOs or between an RFSO and any other RF flange- take my word for it. Even the sales staff at the local gasket houses are unaware of this, but we had to have a couple hundred of them replaced on a job early in my career so I wasn't soon going to forget that lesson! That's why I keep the sample here, to show others who do not know or don't believe this will happen.

Gaskets for RFSO flanges need to have an ID no larger than the ID of the RFSO to work properly, which by the way is more than just slightly LARGER than the OD of the pipe when compared to the width of the winding layers in a spiral wound gasket. That means that you cannot use a piece of sch5S pipe for the inner ring and be safe. The example gasket on my desk has a sch5s inner ring and it STILL came unspooled after the joint was torqued.

As to the OP's original question: assuming that the B16.5 people have given us flanges that are adequately stiff, the reduced gasket seating area available when one or both flanges is a RFSO will give you more gasket seating stress at a given bolt tension. Accordingly the joint is actually less likely to leak, unless the flange faces are damaged.

 

[XL83NL]

Thanks again moltenmetal for your useful contribution.

Curious for that picture though.
I know it's always a hassle with those camera cables

 

[StevenHPerry]

Molten,

When you say the ID of the gasket needs to be no smaller than the OD of the pipe, do you mean the ID of the windings? Or the ID of the inner ring?

At first blush, a partially unsupported IR doesn't sound too bad.

- 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.

 

[moltenmetal]

Since the ID of the RFSO flange is larger than the OD of pipe, it is also larger than the OD of a slice of sch5 pipe when it is used as the inner ring, which is common practice for small gaskets.

ANY unsupported windings ARE a problem.

Obviously that problem is greatest when you're using the gasket between two RFSO flanges in a line where there's a flow with considerable velocity.

 

[moltenmetal]

Finally found the stupid cable...here's the picture of what happens when you match a STANDARD spiral wound gasket with inner ring with a RFSO flange.

 

[XL83NL]

Thnx moltenmetal!
Forgot to mention that our SPWND's also have an internal ring; would that reduce or prevent the failure as posted/described by you?

 

[Duwe6]

Internal rings [type "CGI"] would help some, but if you have unsupported windings like moltenmetal's photo shows, it won't help very much.

Damn shame it took me this many years to learn something so basic. In the various ASME Codes, SO = WN for design, thus like most folks, I've been using them interchangeably. Thanx, moltenmetal

 

[gasketguru]

Agree totally with moltenmetal here - the ASME 16.20 spirals are no good on the smaller sizes - typically 1.1/2" and below with slip-ons (the standard says this but most people don't actaully read the table in the standard). On small bore SOW I would use kammprofiles instead as they do not share this problem.

 

[moltenmetal]

Any of the major mfgs of spiral wounds will make you a double tacked inner, no inner ring gasket with an ID just greater than the OD of the pipe, ie. greater than the ID of the RFSO flange. They work, but even the gasket suppliers don't know to ask you whether or not you need these.

Yes it's a problem especially from 1/2" through 2", but you can experience the same problem with 6" and even larger, depending on the construction of the gasket.

The inner rings are useless in this case unless they take the inner diameter of the winding to a diameter greater than the inner diameter of the RFSO. Some designs of inner ring do this, as they're washers made from plate. Most inner rings do not, as they're made from pieces of sch5S pipe.

 

[11echo]

moltenmetal ...I have to ask, this problem was from a RFSO flange bolted to a RFWN flange? I could see this issue if it was RFSO to RFSO, but with a RFWN the gasket would be supported to the ID of the gasket and the commodity flow couldn't get behind the gasket to deform it.

 

[gasketguru]

It is not the media pressure pushing the gasket inwards - a spiral-wound gasket has a radial stress within the element under compression (imagine pulling a tape measure or a catherine wheel tight - the filler and steel are reacting to one another as the sealing element compresses).

The internal stress buckles the element inwards, and it is too close to the bore to start with. - Even the narrow inner rings of ASME 16.20 sometimes buckle inwards (more so than the BS3381 sizes typically), and indeed many sizes have been revised because of this since it was adopted from API 601. (We also use plate not pipe to make our inner rings - it is just the dimensions that give problems.)

If the element goes into an unsupported region on either side it collapses into that void and loses the energy that you are trying to "store" in it from the bolt forces.

Again kammprofiles do not suffer from this problem - see EN 12560-6 for dimensions. - I find these easier on the small SOW flanges.

 

[moltenmetal]

Gasketguru has it right: it's the unsupported edge which won't resist the compression that's the problem, not the drag caused by the media. The gasket photo I posted was merely inserted, torqued, then removed- no media present, ever.

Kammprofiles are easier, and they seal at a lower gasket seating stress. They're also quite a bit more expensive (at the small sizes, they're often as expensive as the flanges they're put between). Unlike spiral wounds, which in practice can be safely removed and re-installed multiple times before they must be discarded, kammprofiles must not be re-used unless they're re-faced (at the factory) first.