Lines of Support - Vessels under vacuum 3

[michigander]

Can anyone comment on the L dimension for the line of support in figure (b) of Fig. UG-28.1? This is the vacuum configuration that has a shell, as well as a top and bottom cone. Are the cone to shell junctures always a line of support? The referenced note 3 says " When the junction is a line of support", then do this which leads me to believe that it is not always a line of support. If I had a cylindrical vessel with a top head, bottom cone with a drain nozzle directly out the bottom of the cone, would this configuration be more attributable to figure (a-2) of the same section?

Thanks in advance...

 

[TomBarsh]

As you've seen, the Code permits you to optionally consider a cone-cylinder junction as a line of support for external pressure. When the junction is treated as a line of support it must meet the stiffness (moment of inertia) requirements of Appendix 1-8. The benefit of considering the junction a line of support is that it reduces the unsupported length "L" of the cylidrical sections and will allow a lesser shell thickness. But in some cases this may require the addition of junction stiffening rings.

If the junction is not treated as a line of support you no longer have to meet the stiffness requirements of Appendix 1-8 (but you still have to meet the requirement for reinforcing area; see UG-33(f)(1)(a) Step 9 and (b) Step 5). You must also meet the requirments of Fig UG-28.1 Note 1 that the cone thickness is not less than required thickness of adjacent cylinder. The benefit is that you do not need to have a stiffener at the cone-cylinder junction.

It may be worthwhile checking the design both ways. I have seen some vessels that required a stiffener ring if the junction was treated as a line of support. But when the junction was not a line of support the vessel was okay based on existing thicknesses. Generally, the vessel MAEP was less in this second case but as long as it is above the design external pressure (full vacuum, etc), there is no problem.

 

[michigander]

Tom;

Thanks for the quick reply. I assume, then, that in the case of figure UG-28.1 figure (b), that L would be from top tip to bottom tip of the cones, if I do not consider the juncture a line of support. Would you agree?

Thanks again...

 

[TomBarsh]

Whoa, I should have read your original post more fully with specific reference to sketch (b). In the case of sketch (b) the cone-cylinder junction is treated as a line of support and L is the length of the cylinder.

I agree that if the cone-cylinder junctions are not treated as lines of support for sketch (b) configuration then L would be taken to the bottom and top tips of the cones.

my pleasure!

 

[TomBarsh]

Whoa, I have to give that star back!

This is a ticklish issue, not allowing the cone-cylinder junction in sketch (b) to be a line of support.

The Code does describe "conical heads". So presumably conical heads may have the line of support applied at 1/3 the depth of the head just as for formed heads. So in this case "L" is not from tip to tip, but is from 1/3 depth of head to 1/3 depth of head, just as though the heads were ellipsoidal, etc.

But say we snipped off the end of the cone and added a tiny cylinder of very short length with a tiny formed head. Then per a strict application of the Code rules the line of support would be at the 1/3 point of the tiny head. But say this is so tiny we may as well say that the line of support is at the small end of the cone.

In the second case, one could say that this is no different than a nozzle on a formed head. But per Code rules the cylinder at end of cone is not a nozzle but is a regular cylinder; ie: there are no area replacement rules for the opening as per UG-37.

So there are two conflicting interpretations for the case of cone heads (remember them from old Saturday Night Live?).

I think now I will change my vote to allow the line of support at the 1/3 height of the cone. Anyone else?

 

[prex]

TomBarsh,
see point (1) in the definition of L:conical heads are excluded from the one third rule.
The reason is that a cone with a small opening angle would be substantially equivalent to a cylinder, but counted for only 1/3 of its length.
My answer to michigander is:
-if the cone cylinder junction is not a line of support (per 1-8) then L includes the lenght of the cone down to the flange of the drain nozzle (that should easily satisfy the requirement for being a line of support)

prex

http://www.xcalcs.com

Online tools for structural design

 

[TomBarsh]

prex...A Star for You!

 

[michigander]

Thanks to prax and Tom for the exchange. These are all the points I have pondered (such as toms point on the potential conflict). Good, salient answers.

Two stars to both of you for answering on a holiday weekend!