Reinforcement neer a Flat Head

[neilbonner]

I have a 6” nozzle located very close to (4”) a flat head Fig UG-34(b–1), I can’t move the nozzle any further away from the flat head.

I’m assuming A1 will have to be calculated from the excess thickness in shell and the head.

Also what do I do about the knuckle,Fig UG-34(b-1)?

Thanks

 

[johnnymist2003]

Hi,

I assume that the flat head and cylinder are of the same material, and that the circumferential weld is near the flat head knuckle...Caluculate the maximum distance from the nozzle centreline to the start of the flat head knuckle (ie. the cylindrical portion). For area A1, take the diameter available for reinforcement as twice (2x) the distance calculated. Thus, A1 = [(2xDistance)-OD nozzle] x [(t actual-t required)].This method will reduce the available area in the cylindrical section.You may then need to increase nozzle thckness to accomodate the lesser area available in A1.Alternatively, ignore completely the area iavailable in the cylindrical section, and reinforce the opening with the excess area available in the nozzle neck only.You cannot take any consideration of the material in the flat head.

 

[neilbonner]

Yes the head & shell are the same material.

I don’t know were your statement “You cannot take any consideration of the material in the flat head” comes from.

Why can’t you take the excess thickness in a head into account in the reinforcement calculations?

 

[prex]

Can't locate a rule where the code explicitly forbids to intermix reinforcement available in different structures of the vessel (though I'm convinced it exists), but I fully agree with johnnymist2003: you are not allowed by code formulae (unless you want to revert to U-2(g)) to use reinforcement in a different structure. The same would apply to an opening in a formed or flat head, where you can't use the excess material in the shell.
A formal way to justify this with the code is that the limit of reinforcement is measured along vessel wall, and no rule is given on how to extend this, when the direction along which to measure changes, because you are in another structure.
A more technical reason is that only the material that is in or along shell wall will effectively participate in place of the metal cut in resisting pressure stresses. After all, as the limits are calculated differently for the shell and for the nozzle wall, you would require a new limit definition for the flat head, but this would require a separate analysis not covered by code.

prex

http://www.xcalcs.com

Online tools for structural design

 

[johnnymist2003]

Hi Neil,
sorry for the delay in the response to your question. Bear in mind that the shell is seeing circumferential stresses due to pressure (see UG-37, definition of tr), whereas the flat head is seeing bending stresses. I believe that it is the intent of the code that one should not "mix" stresses. If you refer to UG-37 & UG-39, there are completely different rules for the reinforcement of openings in "shells and formed heads" and for openings in "flat heads". I agree with prex, you could use U-2(g) to try to justify your method. The solution I gave is by far the easiest to justify in terms of my understanding of code rules, however that is not to say it is the only solution. One could revert to the use of FEA, for instance, however one would need to weigh up the costs of the additional calculations vs the cost of extra material. Unless the thicknesses you are using are very close to the required thicknesses for the components, then the additional material costs (if any) may not justify the use of a more complex design tool. Why not have a chat to your AIA, I'm sure they would give you some guidance also.
John

 

[chicopee]

whatever you do, dont weld on knuckles.

 

[jte]

Yeah, gotta watch out for those knuckles on flat heads!

 

[GenB]

HOW CLOSE TO THE KNUCKLE is the nozzle located?
If it is too close, I think you will endup with a very funny reinforcing pad outside the vessel.
I need to know how close to assess the problem
genb