Ball valve body-adapter flange thickness 1

[al1979]

All valve experts need your opinion :-
Why people insist to design ball valve body-adapter flange (thickness) to design as per ASME Sec VIII Div 1 Appendix 2?
Actually the adapter is guided in body and does not allow to create moment due to bolt load which is a typical case of end flange / pipe flange.
Any alternative methodology / design standard for this flange thickness calculation? How much FEA is reliable?

Thanks in advance!!!

 

[gerhardl]

Since nobody else has answered, let me try to help you starting a discussion:

Valve body adaptor flanges has obviously too be constructed, tested and produced to one or another standard. See typical example:

http://www.swagelok.com/downloads/webcatalogs/EN/MS-02-200.PDF

If there is a given standard, standards are just this: one exact way to do a thing, until somebody finds it unreasonable and changes the standard.

One example of trying to give an updated view and necessary, but 'trimmed down' economical sizes (and sterength) of valves and piping is the Norwegian based 'NORSOK' standard.

http://www.standard.no/en/Sectors/Petroleum/NORSOK-Standard-Categories/L-Piping--Layout/

The purpose of the standards is to ensure economical solutions, safety and shortest possible building time for offshore and subsea constructions. The standard refers to a large number of different international standards, but also point to a number of new exceptions, revisons and solutions.

 

[al1979]

gerhardl,

Thanks for the reply.
Norsok standard reference is indeed good but unfortunately related to end flange / pipe flange only.

We have been manufacturing ball valves from many years and our body-adapter flange calculations are based on some of Roark's formulas as ASME sec VIII div 1 appendix 2 can't be directly applied to this joint / flange design. We ensure bolting is as per ASME sec VIII div 1 appendix 2 which most of the piping specifications asks for and most of other valve manufaturer's comply to. However no one talks about thickness. These valves are hydro shell tested and are in service all over the world from so many years without any problem.

Now one of our customer is specifically asking for this thickness calculation as per ASME and hence I'm asking this question.

I think FEA is the only option left for us.

Any further input will be highly helpful.

Regards...

 

[gerhardl]

I am afraid I do not have the relevant background to give you an answer, and you might well be correct.

It could perhaps be interesting to bear in mind both the difference and similarity between valve flanges and pipe-flanges:
- For large valves normal mounting istructions specify the valve to be supported in such a way as not to transfer extra forces to the pipeline, eg. correctly mounted the pipeflanges and hence the valveflanges should not be overloaded past a normal valveflange's capacity.
- Longitudal forces, expansion and contraction of pipeline, including load from the valve, have to be under the pipelines load capacity, else compensated for.
- How large is the real on valveflanges?: No internal valve pressure could give an extra impact above the same pressure's impact on a pipeline flange????
- Eventual torque forces from opening or closing the valves have to be less than allowed forces for the pipeflanges?
- If the valveflanges in geometry is comparable to valvflanges in load distribution, or proven more solid than valvflanges (estimate or calculation ??? Including evaluation for material differences (in strength and tensile properties, temperature impact etc???). Then the strength of flanges should be OK?

Others have perhaps some comments on this?

 

[bcd]

This problem is really between you and the customer. You already know, as you have already stated, that the ASME formulas cannot be directly applied to a ball valve body joint. Loading such as the compression force of the ball and seat cannot be easily calculated and are normally determined by FEA or testing. The body joint flange thickness is determined by the manufacturer and is validated by FEA or testing. Most common method is to pressure test the valve per ASME code section UG-101. One option in UG-101 is to pressurize the valve to 2.5 times its rated pressure and no permanent distortion or leakage is allowed. If you pass the test, the body flange meets the ASME code.

 

[al1979]

Thanks gerhardl & bcd.

bcd :- The twist is valve is to be installed in EU so PED is applicable. EN 12516 part 1 & 2 doesn't clearly talks about body-adapter joint design. Regarding UG-101 i.e. burst test the same is covered under EN 12516-3 but product of DN x MAWP < 3000 for PED qualification which means very small range of valves will qualify as per this criteria.

I realy don't understand why people mix european and american standards?

Well good news is FEA has been accepted in accordance with ASME sec VIII div 2 part 5.

Thanks all for your responses.

Regards,

 

[gerhardl]

I agree with bcd here, if anything is unclear the enduser/customer must give a more detaild explanation of what exactly is expected and sufficiant regarding tests and calculations. This is especially necessary if different standards apply and might contribute to a'foggy' picture.

Regarding UN and EN norms together: if the customer has found this the best way to describe the wished for qualifications for his high-pressure valve, the customer is (as always) right. Your job is to see how close to his expectations you can come, and it is very often necessary to ask for clarification on technical points.