Customer wants 1500 flange but code requires 2500# 2

[Eng822]

Lets say there was a vessel that was designed back before I was born and the code calculation specified a 2500# flange but referenced dimensions for a 1500# flange (basically, there appears to be an error in the design) and the vessel and piping system were built with 1500# flanges. Fast forward several years and now the client wants to replace the vessel but won't accept that they may need to replace their existing piping to match the 2500# flange that should have been there in the first place. The vessel is not installed in the States and their justification is they have never had a problem with the first vessel.

Before it gets mentioned, an app. 2 flange could work but it requires 1-3/4" bolts where the #1500 flange on the existing piping uses 1-5/8" (B7 High strength).

The vessel is to be U stamped. I considered a Div.2 Part 5 approach, but I don't see how FEA could be used to justify the flange since this situation does not fall under U-2(g) and really any form of justification could not be used to say the design is at least as safe as the rules of Div.1.

I have never tried to justify a design based on previous working history so I am asking if anyone knows if there is even an option for that. Destructive testing perhaps but again how is that at least as safe? I think most PV Engineers I talk to have the same understanding that Div.1 is highly conservative less accurate standard so it is not a surprising that the 1500# flanges have worked without any issue. Any experience or opinions on the matter would be appreciated, thank you.

Note: The client is requesting the shell material be upgraded but the flange material matches the existing vessel so it is not an exact copy.

 

[XL83NL]

Why go in to so much trouble and not tell yor client where it's really at?

 

[DSB123]

If the "design" is not safe then if anything happens then it will be you (and your CEO) who goes to jail not the Customer!!!

 

[oknow]

Without knowing all the particulars perhaps replacing with a 1500# WNF, end the code boundary at the weld joint. Do not include the 1500# WNF on the ASME MDR. Run by your AI/AIA.

 

[LittleInch]

What pressure, temperature and materials are we talking about here?

If the connection piping is class 1500 this implies that the service conditions coming out of the vessel are within class 1500 flange limits?

ASME B16.5 flanges are good for 1.5 times rated CWP for hydrotest purposes, so not surprising they work, but this sounds like you need to go back to design basis and first principles here.

currently something doesn't add up here.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Eng822]

Oknow,

That is in interesting approach.

By the way, Design P = 4375psig, Design T= 500degF, chromoly material.

So let me see if I am understanding your logic. The customer wants a coded stamp vessel but the flange wont meet code, therefore we stamp the vessel but consider the pressure boundary to end at the circ. seam before the flange. I have done something similar with heavy bar nozzles that had welded pipe stub ends so the vessel could be installed with an orbital pipe welder but I never considered doing it with a flange end because I thought that was not allowed. I will discuss this option with our AI as he will be at our facility today.

----

I can not really comment on the safety of the existing piping because I did not design it and I have not reviewed the calculations. What I do understand is that this plant uses the same 1500# flange all over on other equipment that operates at the same pressure and temperature with out issue. Keep in mind that this plant is overseas and although ASME BP&V is widely used to govern pressure vessel design all over the world it is not not always a requirement like it is here in the states, and anyone who has ever had to redesign a vessel that was built to some other European standard knows, it is very conservative.

Thank you everyone for your response.

 

[marty007]

Could you make your 1500# App.2 calc work with stronger bolts (ex. SA-354-BD)?

 

[Eng822]

Marty,

That material did work, I was able to reduce the bolts down to 1-5/8", I will discuss this option with my client.

Thank you,

Brian