UW-12(d) & (e) for pipe use - Joint Efficiency E value? 1

[nachocivicoseng]

Hi,

I usually use pipe (both seamless or longitudinally welded) in the design of pressure vessels, and I do not fully understand UW-12(d) & (e) statements for E value: for the more conservative case of no RT, should I always use E = 0,85 for circumferential stress? Does not it collide with Table UW-12? Since circumferential stress is the controlling one over longitudinal stress, this decission on E value for shell calculations is crucial for me.

Best regards and thanks for your confirmation/comments!

 

[trottiey]

Yes, when using pipe with no RT, you should always use E=0.85 for circumferential stress. Yes, this part of the code is very confusing and is a common source of mistakes. It was best explained in an NBBI article from 5 years ago. See page 3 of the following link:

http://www.nationalboard.org/SiteDocuments/Bulletins/SU06.pdf

You would use 0.70 when making a shell from plates and welding your own category A welds. Note that when using welded pipe, UW-12(e) multiplies in another 0.85 factor, so the joint efficiency on welded pipe with no RT winds up being 0.85 X 0.85 = 0.72.

 

[nachocivicoseng]

Hi, trottiey, thanks for your fast reply & link!

Yet, I do not fully understand your second paragraph: when talking about welded pipe, I always mean "bought as such" to my raw material supplier, not fabricated by me from plates. I reviewed my 2004 Ed. of ASME VIII Div.1 UW-12(e) and it simply states "Welded pipe or tubing shall be trated in the same manner as seamless, but with allowable tensile stress taken from the welded product values of the stress tables, and the requirements of UW-12(d) applied.". Why do you apply 0.85 factor twice? I figure out that first 0.85 factor comes from UW-12(d), but what about the second 0.85?

Best regards and thanks again,

 

[trottiey]

The second 0.85 is included in the material allowable that you get from the Section II Part D stress tables. You'll notice that two sets of values are given for many pipe specifications, one for seamless and another that is 0.85 of that for welded pipe.

 

[TJOrlowski]

See appendix L, Fig. L-1.4-1. From the top:

Is full RT mandatory? No, proceed to next step. Is this a seamless section? YES per UW-12(e) (even if using SA-53 ERW Gr.B, or SA-312 Tp316L welded pipe). As trottiey pointed out, E=0.85 for your circ stress. For long stress, E will correspond to your RT on butt joints in Cat A, B, & C joints.

If you have corner joints on both ends of the pipe, E=0.85 for circ stress, and E=1.0 for long stress.

-TJ Orlowski

 

[nachocivicoseng]

Thank you very much, trottiey and TJOrlowski, the situation is clear for me now. Only one last doubt remains: how to deal with specifications such as "Smls. & wld. pipe SA-312 TP316L S31603" on ASME II Ed. 2004, Table 1A, p. 66-68? This case shows the highest stress values among all SA-312 materials, but being valid both for seamless and welded? And Note W14 did not help very much.

Thanks again,

 

[nachocivicoseng]

Hi, only to precise my previous post: I am focused on lines 1 to 4 of the referred pages, for my selection of 316L materials. Best regards,

 

[trottiey]

Under Section VIII, note W14 allows you to treat welded pipe as a shell with a Category A longitudinal weld, provided it was welded with filler metal. So you could do full radiography and root inspection to get a joint efficiency of 1.00. But if you're doing anything less than full RT, it doesn't make much difference.

Note that this line is the only one of the four which is permitted under Section III, which requires thorough examination of pipe anyway.

 

[nachocivicoseng]

Ok, trottiey, thanks, the situation is much clearer for me now. Best regards,

J.I.G.C.