Eng-Tips is the largest engineering community on the Internet
Intelligent Work Forums for Engineering Professionals
-
Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
Non-circular pressure vessels
- Thread starter kstaylor
- Start date
kstaylor,
Have you reviewed the listings of bulletins for the "Welding Research Institute" ? These papers frequently become incorporated into various ASME Codes....
Try
Let us know if they are useful.....
MJC
Have you reviewed the listings of bulletins for the "Welding Research Institute" ? These papers frequently become incorporated into various ASME Codes....
Try
Let us know if they are useful.....
MJC
Hi kstaylor,
Have you found a solution for your reinforcement problem in flat heads? Im facing the same problem and the ASME code Division 1 is too confusing. If you have solved the problem or have any referral material please let me know
Thank you
Vijay
I have taken two approaches. First, use the largest dimension as the diameter. Second, use a algebraically equivalent dimension as the diameter (SQRT of (square of larger diameter plus square of smaller diameter)).
Advise if someone has a better method.
Where there is a will, there is a way.
Advise if someone has a better method.
Where there is a will, there is a way.
johnnymist2003
Mechanical
Hi all,
please refer to my post reply to Sam35 wrt "Large openings in rectangular flat heads".
John
please refer to my post reply to Sam35 wrt "Large openings in rectangular flat heads".
John
- Thread starter
- #7
I guess I should have been more specific. I am calculating minimum renewal thicknesses on the nozzles and the flat plates of air-cooled eat exchanger (ACHE) header boxes. I am looking for a published method that accounts for the effect of the opening with respect to the 3 plates involved (i.e. the plate that contains the opening and the 2 adjacent plates).
I am aware of the nozzle reinforcement rules in UG-36, UG-37, and UG-39. I have looked through the WRC bulletins and, unfortunately, nothing applies to rectangular cross-section vessels.
I am struggling with the requirement in 13-4(j)(2) that states:
4(j) Openings may be provided in vessels of noncircular cross section as follows:
(1) Openings in noncircular vessels do not require reinforcement other than that inherent in the construction,
provided they meet the conditions given in UG-36(c)(3).
(2) The reinforcement of other openings in noncircular vessels shall comply with UG-39 as a minimum. Compensation for openings in noncircular vessels must account for the bending strength as well as the membrane strength of the side with the opening. In addition, openings may significantly affect the stresses in adjacent sides. Because
many acceptable configurations are possible, rules for specific designs are not provided [see U-2(g)].
The ACHE header boxes have obround-shaped openings in the top and/or bottom plates with shaped transition type nozzles attached. These shaped nozzles are circular in cross-section where a normal pipe flange is attached and transition to the obround shape where the nozzle attaches to the flat plate. The cross sectional area is the same at both ends. The narrow dimension of the obround shape is 4", the same as the inside dimension of the header box between the two side plates. There is no additional reinforcing pad attached, and no room for it anyway. The shaped nozzles are either cast or forged and are somewhat thick relative to pipe. I believe that the combination of the relatively thick shaped nozzle and the thick plate it is welded to provide the reinforcement required for the opening.
Section 13-4(j)(2) requires that the bending and membrane strength of the flat plate where the nozzle is attached be considered as well as the effects on the strengths of the adjacent plates.
The thread794-78644 by John seems to be asking essentially the same reinforcement question with regard to adding a nozzle and how to account for the reinforcement requirements.
I am looking for something that can give me direction in assessing the effect of the obround hole on the 3 plates involved.
KST
I am aware of the nozzle reinforcement rules in UG-36, UG-37, and UG-39. I have looked through the WRC bulletins and, unfortunately, nothing applies to rectangular cross-section vessels.
I am struggling with the requirement in 13-4(j)(2) that states:
4(j) Openings may be provided in vessels of noncircular cross section as follows:
(1) Openings in noncircular vessels do not require reinforcement other than that inherent in the construction,
provided they meet the conditions given in UG-36(c)(3).
(2) The reinforcement of other openings in noncircular vessels shall comply with UG-39 as a minimum. Compensation for openings in noncircular vessels must account for the bending strength as well as the membrane strength of the side with the opening. In addition, openings may significantly affect the stresses in adjacent sides. Because
many acceptable configurations are possible, rules for specific designs are not provided [see U-2(g)].
The ACHE header boxes have obround-shaped openings in the top and/or bottom plates with shaped transition type nozzles attached. These shaped nozzles are circular in cross-section where a normal pipe flange is attached and transition to the obround shape where the nozzle attaches to the flat plate. The cross sectional area is the same at both ends. The narrow dimension of the obround shape is 4", the same as the inside dimension of the header box between the two side plates. There is no additional reinforcing pad attached, and no room for it anyway. The shaped nozzles are either cast or forged and are somewhat thick relative to pipe. I believe that the combination of the relatively thick shaped nozzle and the thick plate it is welded to provide the reinforcement required for the opening.
Section 13-4(j)(2) requires that the bending and membrane strength of the flat plate where the nozzle is attached be considered as well as the effects on the strengths of the adjacent plates.
The thread794-78644 by John seems to be asking essentially the same reinforcement question with regard to adding a nozzle and how to account for the reinforcement requirements.
I am looking for something that can give me direction in assessing the effect of the obround hole on the 3 plates involved.
KST
prex
Structural
kstaylor,
in trying to help you understand the meaning of 13-4(j)(2) I would comment as follows the two main points of this paragraph:
1) the requirement of accounting for both bending and membrane strengths relates to the fact that UG-37 requires full compensation for membrane stresses and UG-39 only 50% compensation for pure bending stresses; so in my opinion an acceptable procedure for checking the reinforcement would be as follows:
-take the membrane component only and check it per UG-37, using as the minimum wall required thickness the thickness t[sub]r[/sub]=t&[ignore]sigma[/ignore];[sub]m[/sub]/S (this holds because normally the membrane component is independent of the stiffness of the adjoining walls)
-in the case of an obround hole you should conduct two separate checks in the two main directions (if relevant, normally only one is of concern) using for each direction the opening diameter in the plane normal to it
-for the membrane+bending component it's more difficult, but I guess that you could check the reinforcement per UG-39 by using as the minimum required thickness t[sub]r[/sub]=t&[ignore]radic[/ignore];(&[ignore]sigma[/ignore];[sub]m+b[/sub]/1.5S) (not sure this is correct, take it as a proposal for further reasoning, however note that it works in case of pure membrane stress and of pure bending stress).
2) the requirement on the adjoining walls relates in my opinion to the fact that the (bending) stresses in a rectangular vessel depend on the relative stiffness of the walls, and of course the stiffness of a wall is impaired by the presence of the opening. My proposal is that you would be OK if you could check that the thickness of the adjoining walls is still OK when there is no opening, but the thickness of the wall with the opening is the minimum that satisfies the limit on both membrane stress and membrane+bending stress for that wall only. My way of reasoning is that, after compensation, a wall with an opening should be at least equivalent to a wall with the minimum required tickness and no opening.
Quite a complex matter to explain!
prex
Online tools for structural design
in trying to help you understand the meaning of 13-4(j)(2) I would comment as follows the two main points of this paragraph:
1) the requirement of accounting for both bending and membrane strengths relates to the fact that UG-37 requires full compensation for membrane stresses and UG-39 only 50% compensation for pure bending stresses; so in my opinion an acceptable procedure for checking the reinforcement would be as follows:
-take the membrane component only and check it per UG-37, using as the minimum wall required thickness the thickness t[sub]r[/sub]=t&[ignore]sigma[/ignore];[sub]m[/sub]/S (this holds because normally the membrane component is independent of the stiffness of the adjoining walls)
-in the case of an obround hole you should conduct two separate checks in the two main directions (if relevant, normally only one is of concern) using for each direction the opening diameter in the plane normal to it
-for the membrane+bending component it's more difficult, but I guess that you could check the reinforcement per UG-39 by using as the minimum required thickness t[sub]r[/sub]=t&[ignore]radic[/ignore];(&[ignore]sigma[/ignore];[sub]m+b[/sub]/1.5S) (not sure this is correct, take it as a proposal for further reasoning, however note that it works in case of pure membrane stress and of pure bending stress).
2) the requirement on the adjoining walls relates in my opinion to the fact that the (bending) stresses in a rectangular vessel depend on the relative stiffness of the walls, and of course the stiffness of a wall is impaired by the presence of the opening. My proposal is that you would be OK if you could check that the thickness of the adjoining walls is still OK when there is no opening, but the thickness of the wall with the opening is the minimum that satisfies the limit on both membrane stress and membrane+bending stress for that wall only. My way of reasoning is that, after compensation, a wall with an opening should be at least equivalent to a wall with the minimum required tickness and no opening.
Quite a complex matter to explain!
prex
Online tools for structural design
- Status
Similar threads
- Locked
- Question
- Locked
- Question
- Locked
- Question