Head to shell weld - missmatch on ID or OD

[dbday]

I have a supplier who wants to weld the head (10mm thick) to the cylindrical shell (8mm thick) of a vessel he is making for us by matching the outside diameters of the head and shell.

This will give a miss-match at the internal diameter / root of the weld.

Is there any rule about this ?

I would think that you would not want a miss-match at the root and that at the outside diameter such a miss-match doesn't really matter as it will be taken out by the weld seam.

 

[fegenbush]

I am going to assume you are working to ASME Section VIII Div 1. The process is similar for other codes as well. UW-13 specifies the details of attachment of heads to shells. Additionally, Figure UW-13.1 has a graphical representation.

The short answer is to grind a taper onto the thicker member with a minimum of a 3:1 transition between the thicker and thinner components.

 

[austsa]

Dished end suppliers have their own standard sizes. Some are based on ID whilst others are based on OD. Generally, client specifies vessel shell ID, and the fabricator selects a standard head size of best fit. Thus, sometimes 1:3 taper is on the inside, whilst at other times, it's on the outside.

 

[SNORGY]

In a horizontal vessel, I would view it as a good thing to have the head thicker than the shell and match the OD. Then the ID mismatch, tapered or otherwise, does not give rise to ponding of whatever liquids are present right against the weld root.

 

[SNORGY]

Of course if liquids are there in any detectable quantity, they will be on the weld root anyway...

 

[chicopee]

I agree with Fegenbush on the 3:1 transition.

 

[bernoullies123]

I generally use a 1:4 taper (14 deg). This is the maximum slope permitted by PD 5500 Figure 3.10-2. A bit more grinding, but smoother transitional has greater benefits.

 

[chaulklate]

In addition to all above, depending on vessel diameter and size of manway, welding can be performed from inside so your outside root matches.