Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

Re: Stresses at Shell Attachments

[Subject Prev][Subject Next][Thread Prev][Thread Next]
>Analysis of stresses at attachment to a spherical or cylindrical tank is
>generally based on formulas, charts and graphs found in Welding Research
>Council bulletin WRC 107,  WRC-297 and Pressure Vessel Design by
Actually these days it's generally based on finite element analysis, if 
only because it's simpler and faster. For nozzles, the ASME Code 
requirement for internal pressure is to be met irrespective of external 
loading, then beefed up to account for external loading. Just the manual 
arithmetic to analyze external loading always cost me a day or two and a 
splitting headache. I can do a better job today in an afternoon, and it's 
a helluva lot easier to explain to an AI. BTW, by 'clip' do you mean a 
lifting lug or a specific nozzle detail like a doubler plate or perhaps a 
weld saddle?

Bijlaard's paper (Welding Journal Research Supplement December 1954)  
specifically discourages rectangular attachments to spheres under 
torsional loading, but WRC-107 explains how to do it for other loadings. 
If you're worried about torsion, you'll probably do better to put a round 
pad on the vessel (use a drop from a manway) and attach the lug to that.

> Designs of this kind have existed even before the
>advent of computers.  How was the design performed then?
The procedure for internal pressure still used is the area replacement 
method in the ASME Code. The vessel is considered adequate if the 
material added for reinforcement equals that removed in making the 
opening. It's a little more complicated than that, but easy with a 
spreadsheet. Section VIII Div 1 requires that external loading be 
addressed, but conveniently leaves the methodology to the responsible 
engineer and provides no stress criteria. Div 2 and the nuclear codes 
aren't any better about providing methodology, but at least they provide 
stress criteria. Most people who do nuclear vessels use finite element 

The classical approach is the Bijlaard analysis (comprising the guts of 
WRC-107) which came out in the '50's, and I have references that go back 
further than that, since the problem an old one. Snoop around in any 
document repository and you'll get as much as you can handle.

If only because I've been in the pressure vessel biz for a while, your 
question seems a little like a request to modify an existing vessel. If 
you haven't done this before (and please forgive me if I sound 
patronizing--nothing of the sort is meant) be aware that every state but 
one has very strict rules about such things. It's not a trivial matter. 
People die when botched pressure vessel repairs tear loose. Authorized 
shops do this according to National Board Standards but only on ASME Code 
vessels. If someone has asked you to modify a non-code vessel, I 
recommend you run away. And if 'clip' means a lifting lug make damn sure 
your welding is high quality--nothing like local loading to initiate 
fatigue failures

Christopher Wright P.E.    |"They couldn't hit an elephant from
chrisw(--nospam--at)        | this distance"   (last words of Gen.
___________________________| John Sedgwick, Spotsylvania 1864)