Knuckle Radius & Table 1-4.4

[takiyasamsama]

Hi,

I'm in a situation where there's an existing plant to be installed with a reactor (a vessel actually). The problem is that the vessel itself was fabricated in another plant and running well for now I presume but when I calculate the required knuckle radius it is less than minimum required based on ASME Section VIII.

In ASME Figure UG-36, it is mentioned there that the radius at the large end of the cylinder to cone shall be greater than either of 0.12(R+t) or 3 times the thickness (if not mistaken). Unfortunately our client specify the knuckle radius as 80mm where in the calculation it should be at least 138.6 mm.

This value, 80 mm, was based on the existing plant which the plant already operating. However, besides the GA drawing they didn't have any calculation sheet or provide with any other relevant documents besides the drawing itself.

Referring to Table 1-4.4 also I couldn't get the ratio right (definitely), from r/D ratio we could easily get the height of the cone itself and as predicted, the height also is way off than what is required.

Does anyone have any advice regarding this matter and kind enough to share their valuable opinion.

Thank you

 

[racookpe1978]

Have you measured the actual knuckle (outside, under the insulation)?

If all you have is a general arrangement (assy dwg) it too might be a simple case of "draw a curve, nobody will ever care what actual knuckle is installed" approximation by an earlier draftsman.

 

[SnTMan]

takiyasamsama, Table 1-4.4 applies only to elliptical heads of certain proportions, see 1-4(f)(2).

If your (presumed) toriconical head fails to meet the knuckle radius requirements of UG-36, well given that the vessel(s) are apparently in service, I'm not sure where that leaves you. Vessels are no longer under ASME Code, but more likely NBIC. Maybe a matter for the jurisdiction.

Regards,

Mike