Thermal stress in pipe flanges 1

[bulkhandling]

A flange connection is designed on a condensate blowoff pipe (cycles can be 50 to 200). The design temperature is 1050F and pressure is 50psig. Flanges are Class 300 with one material 304ss, the other P91.
My question one is, during start-up time, High temperature steam can make the inside temperature to 1050F while outside of the flange can be ambient. How shall I consider the internal stress due to the temperature difference.
Question 2, thermal expansion of the two flanges are different, is there any problem?
Thank you.

 

[prex]

1)With that number of cycles you shouldn't worry about the thermal fatigue in the flanges.
2)Differential expansion of the flanges is easily accomodated in the joint: even if you had a RJ I wouldn't worry, except that after a number of cycles the RJ could start to leak (but it would be sufficient to retighten a bit, no problem with other common types of gaskets)

prex

http://www.xcalcs.com

Online tools for structural design

 

[bulkhandling]

Thank you, Prex,
But how can I do the calculation to support your conclusion? Or how about if the calculated radial or tangential stress at flange is bigger than the yield stress at the temperature?
Do you have practical experiece about it?
Thanks again.
Echo

 

[prex]

You must use ASMEVIII Div.2 App.4 to assess the calculated primary + secondary stresses in the flange.
The limit for primary + secondary stresses is 3S_m, an allowable that may be close or equal to 2Y.
Moreover secondary stresses need be obtained be means of a linearization through the wall of the element (the flange in your case). What you calculated is (partly) a skin effect that gives rise to peak stresses: peak stresses are limited only by a fatigue analysis (App.5 of Div.2), that, for your number of cycles, is unlikely to be of significance.

prex

http://www.xcalcs.com

Online tools for structural design

 

[bulkhandling]

Thank you Prex,
Your response is of great help!
Echo