Isn't the joint design shown in Fig UW-13.1 of the code?
Most fabricators go with UW-13.1(j) and match the centerline of the head with that of shell. The head will have cutback. Part of shell will be tapered and become part of head.
jamesl,
That's how we normally make the joint, but, UW13.1 doesn't seem to specifically address hemi heads.
I'm just looking for a little enlightenment on the subject. Is that the only acceptable method of attachment?
Could you for example, not cut the head back and make the transition between shell and head thicknesses with weld metal?
But by "not cutting back" the hemispherical head you would not be placing the transition in thickness on the head's side of the tangent line as is required by Figure UW-13.1 (assuming the head is thinner than the cylindrical shell).
If I understand this correctly, the question relates to adding a hemi head to a shell with the weld at the tangent line. The proposal is to use weld metal buildup on the head to meet the transition requirements. In other words, the “triangular parts” in Fig. UW-13.1 (j) or (k) are formed with weld buildup.
Isn’t this covered in UW-13(b)(3)?
…When the transition is formed by adding additional weld metal beyond what would otherwise be the edge of the weld, such additional weld metal buildup…
I see nothing wrong with it provided all the requirements are met and the weld buildup isn’t excessive… A good welding engineer should glance at the proposal.
I think adding weld metal to head thickness to make the transition is an acceptable alternative. See UW-13(b)(3). People usually don't do that unless they did not buy long enough shell.
John, let's refer to Figure UW-13.1, sketch (j) or (k). These show, say, a top hemi head that is thinner than the adjacent cylindrical shell. The Figure shows the "tangent line", which is the 'line of demarcation' between the geometry of the head (hemispherical) and the shell (cylindrical).
What I was saying is that this figure shows the transition in thickness performed within the hemispherical geometry; ie: the cylindrical shell can't be tapered itself; at least, not to less than the required cylindrical thickness, within the "cylindrical geometry" that exists below the top head tangent line.
I have seen numerous hemi heads where the butt weld joint is an inch or two above the tangent line. This results in a hemi head that is not the theoretical 180° "hemi", but maybe 176°.
I may not have made that crystal clear. The point is that the cylindrical shell may extend *beyond* the theoretical tangent lines, at which point that shell material is no longer a "cylinder" but is "hemispherical".
