Thread Content
The wall thickness of the cylinder is 28 mm, and it is formed by pressing using Q345R material with a thickness of 28 mm. The welded parts of the end caps will definitely become thinner; so what should be done regarding the circumferential welds that connect the cylinder to the end caps?
Could some expert give me some advice? It’s an emergency
Meeting the minimum thickness requirement is sufficient!
This post was last edited by zjq1962 on 2021-11-14 08:05. How should the circumferential seam mentioned by the original poster be dealt with? I don’t quite understand; handle it according to Appendix D of GB150.
This post was last edited by Paris on 2021-11-15 09:03. The drawing you’ve made is incorrect; since the head and the shell do not have the same thickness, a hemispherical head should be used. In theory, this ensures that the mean diameter of the shell aligns with that of the spherical (hemispherical) head. The inner diameter of the head should be larger than that of the shell. If the minimum thickness of the head is set to be the same as that of the shell, then a hemispherical head can be used, with both having the same inner diameter. The outer side of the head needs to be trimmed so that it has the same thickness as the shell, after which a groove can be created for welding. When a spherical head is used, the area to the left of the tangent line belongs to the shell, while the area of the shell to the right of the tangent line is part of the head. The addition of 10–12 mm of straight edge to the shell (as shown in the lower half of the diagram) is intended to prevent thinning of the shell due to machining errors during processing, which could result in insufficient thickness of the shell. But then again, this isn’t that thick either, so the pressure should be low as well. Why not use an elliptical head? {:1_90:}
Spherical heads are generally formed by drum forming, and the thickness reduction of heads produced by this method is very small. For spherical heads formed by hot stamping, the port area thickens under the stamping force and does not thin.
Thank you for the teacher’s patient explanations. What I’m considering is the HHA type hemispherical head; if it is processed and formed, according to the drawings, the thickness of the cylinder is DN800*28, while that of the hemispherical head is HHA800*28(22). If such a hemispherical head is formed, how should our manufacturing plant proceed? If we follow Appendix D, clause f, of GB/150.3, then the trimming line should be located on the cylinder itself. 2. According to the teacher’s explanation, may I understand that a spherical cap is treated in this way by trimming its edges? The length of the cylinder was set at 2000 mm; I then added another 10 mm to it, so the cylinder’s length became 2100 mm
It’s not exactly a teacher’s role, but we often work with high-pressure equipment, and such situations come up quite frequently. When creating drawings, we definitely take the minimum thickness into consideration. Since you require a minimum thickness of 22 for the head, then use the method I suggested: employ a hemispherical head. The total length of the shell should be the original length L plus 49 mm, with the head being shaped as a hemisphere with 49 mm removed from it.
I see, teacher. You’re the only one who has explained it clearly. In this case, what needs to be done is to cut 49 mm off the hemisphere and then transfer it to the cylinder.
This post was last edited by Paris on 2021-11-15 at 15:30. But the minimum thickness of the end cap on your device now is 22; why is the thickness of the material used for cutting so large? Around 24 should be sufficient normally. Is it to buy a board together? If the material thickness is 28, there should be chamfering shown in the drawing; follow what was mentioned above. But the actual cylinder body should not have any trimming; if trimming is done, there will be a groove after welding. So that’s basically how it should be for the actual product. However, it wouldn’t make much sense if the drawings don’t specify no trimming.