Branch Question: "Stub-in" Vs. "Stub-on" 1

[me1abpsu]

Hi Everyone,

My question is in regard to B31.3 FIG. 328.5.4D Acceptable Details For Branch Attachment Welds.

We have a fabricator that used sketch 1 where we had specified sketch 2. Will this adversely affect our design? I can't find anything in the code to suggest this. It seems both are acceptable as long as full penetration welds were done (and done correctly).

Background info if it matters:

Water Glycol Service

200 Deg F
150 PSI

36" C.S. Header 0.5" Thick
12" C.S. Branch 0.375 Thick

Modeled as Unreinforced Tee

SIFi=7.541, SIFo=9.721

 

[me1abpsu]

Anybody?

 

[ApC2Kp]

me1apsu,

There is not much difference for pressure design or bending stress. Most fabricators prefer the sub-on figure 1 with slightly less weld prep and weld volume. The final finish of the stub-in figure 2 might have some effects on flow, depending on how much the branch protrudes out into flow.

The biggest difference is how to inspect the weld for full penetration. The figure 2 weld requires a mirror to see the back of weld root. The figure 1 weld root is more of a direct view. The 36" header alows viewing from inside, so it's not a factor in this case. Visual verification of the weld prep and root gap is a good step to avoid questions of incomplete weld penetration, since radiography is difficult for the stub-on / stub-in weld geometry.

 

[me1abpsu]

That makes sense. Thanks for looking into this ApC2Kp!

 

[doct9960]

me1abpsu,

If your branch connection is subjected to severe cyclic conditions, sketch 2 is recommended. Refer to paragraph 306.5.2

 

[me1abpsu]

Thanks Doc.

After reading that para it sounds more like it's required. We're not seeing a cyclic condition though.

GREAT STUFF!