ASME Flange Design

[gpape4]

Trying to qualify an existing vessel with no existing paperwork. Built 1939. Everything checks out except for inspection opening. Non-standard flange, flat plate, intergral/slip-on. The deficiency report keeps failing the gasket seating or fails at the g1 weld requirement. Used existing info gathered in the field and have latest thickness readings. It has held pressure for over 70 years, but the numbers say the flange is not good. Tried different gasket, larger bolts, different weld combinations; probably have done 20 iterations, still nothing works. Any ideas?

 

[fegenbush]

What type of gasket are you trying? Depending on the service, many of the old vessels I have looked at use a tackycloth type gasket that has some very low m and y values.

 

[gpape4]

Based on the information on the drawing from the old equipment file, it was an Armco soft iron. Based on an old Lamons catalog, m=3.25 and seating stress is 7,600.

 

[TJOrlowski]

Don't you just love it when "new and improved" design formulae condemn equipment that has been working beautifully for decades?!

Suggest you try a cammprofile gasket or reinforced grafoil gasket, something with a low m and y value as fegenbush alluded to. You may also want to try low-stress bolts, or accounting for the addition of a thread lubricant with a low coefficient of friction - something that would reduce the required bolt load.

Also, are you trying to qualify the flange to withstand the force during hydrostatic test (usually the controlling load on new vessels), or just doing a rating for MAWP?

-TJ Orlowski

 

[BigTank]

Is this a flat-faced flange? Will the service possibly allow you to spec. a self-energizing gasket? Can you make the gasket narrower and 'N' closer to the bolt circle? Those things might help.

Low-stress bolts: also a good call.

Stainless steel or carbon steel flange and bolting?

--------------------------------
Fitter, happier, more productive

 

[DavidCR]

Sometimes a FEA analysis is good to compare the real stress condition with the code prescriptive formulas. If you don´t have a FEA tool it is very difficult to understand how safe is the design compared to code rules.

 

[gpape4]

TJOrlowski (Mechanical), BigTank (Mechanical),
Thank you for the replies. I ended up finding a gasket, Grafoil GHR laminate, that solved the flange situation and then, Compress would not allow me to just PWHT the flange. Turned out that there was a bug in the program. Had to run the shell cylinder, as well as the nozzle in PWHT condition, then break the report apart to submit. Compress said they would look into it.
Thanks again.