Skirt to bottom head weld

[jtseng123]

All,

For multiple pieces of formed head, do you weld the skirt crossing the head meridional seams, or skip the weld at the head seam ? This is the first time I saw a vendor skip the weld at the head seam, and I do not think that is a common practice. Although spec may say attachment shall not be welded on top of long or girth seams, but I think skirt is an exception that no need to follow it. What is your common practice ?

 

[dhengr]

Jtseng123:
I don’t know exactly what ASME has to say on the matter, but you cite some general verbiage “Although spec may say attachment shall not be welded on top of long or girth seams...,” and that seems right to me. Good welding practice and engineering judgement suggests you should not weld across another weld seam, or from two or three directions (multiple directions) into an inside corner. When you do this you set up some nasty multi-axial, tri-axial, residual welding stresses and/or some complex/nasty multi-axial loading stresses, essential at a point. This is never a good condition if it can be avoided. Then, these may be added to your basic PV pressure/temp. stresses, and you don’t know how they might combine either. It is good welding practice to stop a weld an inch or so short of a weld it would cross. You shouldn’t have any trouble finding enough weld length for the skirt to shell loads, so those missed four or six 3" lengths of skip welding is not significant in the scheme of things, as long ast the starts and stops are nice and clean, without creators, etc. Also, we stop most welds an inch or so from an inside corner. You often see gusset pls., stiffener pls, etc. clipped where they meet in an inside corner, for fit-up and other reasons, one of the reasons being that it also stops people from welding into the corner. Then, I’ve seen welders trying to do a really really good job of welding trying to fill those clipped corners with weld. They’ve put enough weld and heat into that one spot to weld the whole rest of the structure.

 

[SnTMan]

Not at all my class of work, but I think I'd want the skirt coped where it crosses as per dhengr.

Regards,

Mike

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[BJI]

Without knowing details of the head size and thickness, quite often heads are normalized after forming, thereby mitigating the issues raised about residual stresses. However, typically the welds are primarily skipped for inspection purposes (or lack thereof) and in a lot of cases the stress concentrations cased by skipping the weld is far worse than just welding across. I would heat treat and NDT the welds then treat it like any other head and continuously weld all round the skirt. Also, if there are high thermal loads on this vessel I would try to avoid introducing another discontinuity in this region.

 

[jamesl]

If the skirt is not continuous welded, and bottom head is insulated, how do you prevent rain water getting into insulation? All the specs I have seen requires continuous weld at skirt attachment if the specs cover the subject

 

[marty007]

In our shop, we would normally have a continuous weld. I have never seen a customer spec that requires skipping this area.

 

[jtseng123]

All,

I have never seen this issue until recently we use a new vendor who copes the skirt and skips the weld at the head meridional seams. I believe my conclusion is, just like jamesl and marty007 stated, skirt shall be continuous weld across head seams as a common industry practice. Thanks.

 

[r6155]

Skipping the welds facilitate in-service inspection (better interpretation of UT of meridional welds from inside).

Regards
r6155