Flat plate machined concave 1

[zakk]

If I have a flat plate blanking off the end of a vessel and machined it so the inside surface is concave to pressure and the outside remains flat, do I still need to use the Sec VIII rules governing flat plates to determine minimum wall thickness or can I now use other rules, such as those for sphereically dished covers for example? In other words, would the stresses change from bending to tension by changing the shape of one surface? I do not have access to FEA so I cannot determine easily what the result of this action would be.

 

[deanc]

I would suggest that you do calcs. per UG-34 first. If it makes it,fine,if not check with your AIA to see if they would be willing to go with U-2g and UG-32 justifcation.
If this is a flanged attachment do not forget the requirements of Appendix 2.

 

[arto]

Watch out for laminations!! Like in Sec. VIII Div1 Appendix 2-2-d-2

You're basically creating a "hub" on the skirt of the head, and both menbrane & bending stress are tending to peel across any laminations.

 

[LSThill]

I agree with arto (Mechanical)

zakk (Mechanical)you may want to deep water U.T., This can be a bad problem.

Also, for FEA use FE/PIPE - Finite Element Analysis for Piping / Vessels

Paulin Research Group
11211 Richmond Avenue
Suite 109,
Houston, Texas 77082.
Phone: 281-920-9775, FAX:281-920-9739

http://www.paulin.com

E-Mail: tony@paulin.com



Leonard

http://www.thill.biz

 

[deanc]

Excellent point "arto".There could be grain structure problems,laminations or not.

 

[zakk]

I should have clarified that the 'flat plate' would be manufactured as a forging, essentially a forged flat plate.
What I am inquiring is by concaving the pressure side only of a forged flat plate do the stresses go from bending towards membrane, hence I could use thinner material to cap the vessel.

 

[prex]

A design principle that I personally adopt is that, when checking primary stresses (the only ones considered in Div.1), when you (virtually) cut away some material from a vessel part, you get a part of strictly less strength, and if you satisfy code rules with this carved part you are OK. I don't know exceptions to this rule, but of course this is only valid when no secondary or peak stresses (fatigue) are involved.
In your case: imagine cutting away from your head on the flat side some material in order to get a constant thickness head of, say, spherical shape. By doing this you also must fulfill the requirements at the attachment (knuckle radius, skirt length, butt weld and examinations of this weld, and so on).
If the included head satisfies code requirements then you are OK.
In all other cases you would need a detailed approach as figured out by other posters, but I guess this wouldn't be worth the cost of a flat head. prex

http://www.xcalcs.com

Online tools for structural design

 

[SHG]

The contour of the internal and external surface of the head and the shell shall be consider. if you have not a transition region for stress flow then it seems that it shall consider as flat head then you shall have the min. thickness (at bottom of concave) equal to calculated flat head thickness otherwise (have not transition region) best way for thickness is stress analysis and following the rule in ASME Sec.VIII Div. 2.
Of course the material checking such as chemical, mechanical, soundness and so on shall be carefully be cosidered.