ASME VIII combined Loading Calc.

[mtvessel1]

I have a calculation on a conical section under combined loading, including earthquake. The calc is to PD5500 Appendix B, and PD5500 Annex A (which is a compressive/Buckling check), and PD5500 Section3, covering reinforcement at the small and large ends. The Internal Pressure is less than 1bar and is less significant than the external loading.

I am required to change the calculation to conform to ASME VIII.

I have produced a calc using Div1 Appendix L Example 2 for combined loads, and also Appendix 1-5 for reinforcement. However Appendix 1-5 states that when FL is compressive and greater than PR/2,(this is true in my case) then follow U2(g), which states that the designer must provide safe details of design.

From here I have used DIV2 AD-340 for cylinders under axial compression.(which seems to be the same as UG23(b) in hindsight).

My questions are:-
1)Are the methods above acceptable and more importantly, sufficient for design of the cone to ASME?
2)Is AD-340/UG23(b) acceptable to use for Cones?(It seems to quote for cylinders and not cones) and if so, is the Radius replaced by Radius/cos(alpha)?
3)Does U2(g) mean that I can use PD5500 Appendix A for compressive/buckling limit, as this specifically covers conical sections.

Thanks in advance to you ASME VIII experts.

 

[prex]

My answers:
1)Yes, except for what is left to U-2(g)
2)AD-340 is for cylinders and you can't extend it automatically to cones. However you should simply use R/cos[α] (this is by the way what PD5500 does), but of course the final decision will be up to the design reviewer, whoever will be.
3)The method of PD5500 (if I recall it correctly) is not adequate in my opinion. This method essentially consists in the determination of the allowable longitudinal compressive membrane stress, but you have this also in ASME and you should follow the latter.
What 1-5(d) addresses by sending you to U-2(g) are the discontinuity stresses at the junction, that must be checked against the limits in 1-5(g). The calculation of those stresses is not a simple matter, unless you turn to St.FEM: the Roark has some formulae for conical sections under end loads, but you'll need my best wishes of good luck before you engage in that direction.

prex

http://www.xcalcs.com

Online tools for structural design

 

[mtvessel1]

Thanks for your help Prex.
Does this mean then that one cannot realistically design a conical section under compressive loads to ASME without using FEA?

 

[prex]

After reconsidering the matter a bit more I think that the main concern of 1-5(d)&(e), when the net load at the junction is compressive, is more the circumferential compressive hoop stress than the longitudinal compressive stress or the discontinuity bending stress. Under those conditions the reinforcement ring (or the increased wall thickness) is at risk of buckling.
I think that you could demonstrate the stability of the junction by simple conservative assumptions, and then use the reinforcement area per code formulae even if it is said in the code that they cannot be used for compression loads. Otherwise can't see how to proceed without the use of FEA, with still the need of proving the stability with some conservative assumption.

prex

http://www.xcalcs.com

Online tools for structural design