CRN ferrules are 7 times more expensive! would like to avoid that 2

[wydim]

thread794-423427

designer here, trying to obtain a CRN for a vessel with standard Tri-Clamp ferrules (gasket and clamp not included). I have provided calculations for the ferrules as standard nozzles (weld, thickness and reinforcement check). The reviewer requests appendix 24 calcs. I say it's not applicable as I'm not including the gasket and clamp in the CRN.

one solution (in the reference thread) calls for ferrules with CRN. OK, but they're 7 times more expensive than regular ones that I want to buy (I still provide mill tests). do you have experience like that ?

regards,

 

[david339933]

Standard Designation: BPV Section VIII Div 1
Edition/Addenda:
Para./Fig./Table No:
Subject Description: Section VIII, Division 1 (1986 Edition, 1988 Addenda), U-1(e)
Date Issued: 07/07/1989
Record Number: BC89-186
Interpretation Number : VIII-1-89-112
Question(s) and Reply(ies):
Question: Is a clamp which fastens a piping component to a pressure vessel nozzle included in the Scope of Section VIII, Division 1?

Reply: No; see U-1(e)(1)(d).

 

[wydim]

david, I understand the clamp is not in the scope. But what about the ferrule that is welded to my vessel ? that's the issue.
my calcs treat the welded ferrules as standard nozzle (machined from bar stock SA-479 material)

this interpretation does not solve my issue.

regards,

 

[david339933]

I don't agree... App. 24 applies to the design of clamp connections....and the design of a clamp connection involves the selection of the "gasket, bolting, hub and clamp geometry". If you don't know these you can only really do one equation...hub hoop stress, equation 9 of 24-6..which is not per UG-27.

 

[cbPVme]

Tri clamp ferrules don't comply with Appendix 24. The scope of Appendix 24 states that the hub faces shall have metal to metal contact outside the gasket seal diameter. Sanitary type ferrules have a gasket across the whole hub face. Typically these need to be validated with a burst test.

Of course the CRN registered ones are more expensive. Someone has to pay for that certification.

 

[wydim]

the reviewer also asks for either CRN or burst test. help me understand the logic behind this requirement.
On the same vessel, I have a custom machined flange for a future agitator. I have provided calculations for this flange and they are accepted. So why is it acceptable for a machined flange but not accepted for a machined ferrule. I have mill tests, approved material and all that. why a burst test ?

 

[cbPVme]

Flange calculations are well established in ASME VIII-1 Appendix 2 for flanges. I am not aware of any calculations that cover these ferrules. Mill test reports only give material properties, they don't prove that the design of the ferrule is sufficient. How do you know the dimensions are acceptable for the pressure and temperature rating? What analysis have you provided the reviewer to show that these are acceptable for ASME VIII-1 use?

When no rules are given for a particular component, the component shall be verified with a burst test. See UG-19(c). U-2(g) would also permit alternative analysis procedures that are outside the code (FEA, textbook formulas, calculations from another code, etc.)

 

CRN ferrules are 7 times more expensive!

What is your time worth?
Express your answer in units of CRN ferrules.

"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[wydim]

cbPVme, rules are given for a standard nozzle wall thickness, welds and opening reinforcement. That's what I request to use. To my understanding, burst test are usefull for a complicated and complex geometrical shape like valve bodies, and very specific equipment.

ironic metallurgist: it literraly takes 5 min. with Compress software to add a custom nozzle size to a vessel. The $ is real.

I have for (2100$ of CRN ferrules) vs (300$ normal ferrules + 15 min. of calculations in Compress). that's for only 1 project. We have between 4 and 12 ASME rated project per year.

 

[wydim]

Another thing that poped into my mind while thinking about this : how can I do a burst test with an open component. If I put a tri-clamp cap and clamp the ferrule, then i'm not evaluating the ferrule alone, but the whole assembly. Nothing proves the ferrule by itself is strong enough! my CRN design request stops at the ferrule sealing face, not a mm further.

another thought: an existing CRN for ferrules covers the exact geometrical shape and the specific material (what else is there?!). If that's the case, then every ferrule that has this geometrical shape and material is covered ! by default, every ferrule has a CRN, only they are not marked ! I can prove the geometrical shape by drawing and the mill test takes care of the material.

 

[SnTMan]

wydim, making your case in this forum is all well and good, but gets you nowhere with the jurisdiction. Going to need to take it up there...

Regards,

Mike

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[marty007]

You are saying that your calculations only take into account the nozzle/shell junction (can be done in Compress in 15mins).

What calculations have you provided to justify the design of the sealing surface of the ferrule?

 

[wydim]

SnTMan : I guess I was trying to see if anyone had previous experience with the non-CRN ferrule on a code-vessel.

Marty007: none. And the discussion in this forum has showed that the sealing surface of ferrule/gasket is excluded from the scope of the code (and from my CRN request)

 

[TGS4]

It's not just the design of a component, but also the manufacturer and their quality program, that is covered with a CRN.

Bottom line is that you need a CRN for every pressure component in a Canadian system. Full stop. Either you get it yourself or you buy one from a supplier that has them. I recommend the latter.

 

[wydim]

"Bottom line is that you need a CRN for every pressure component in a Canadian system"

TGS4, no. I don't need a CRN for my machined flange that I provided calcs for.
UG-14 permit the use of rod and bars (provided with accepted material).

I'll let you know how it ends.

 

[TGS4]

For the machined flange that you provided calcs, you get a CRN for either that component or the piping system. But it does have a CRN.

 

[wydim]

TGS4, I'm sorry, I don't understand what you mean.
my machined flange is welded to my pressure vessel. I will eventually get a CRN for the vessel (as a whole), but the machined flange doesn't have a CRN on it's own. I'm looking to apply the same principle to the ferrules.