Torsional load on clip/lug weld

[jamesl]

When we design the lifting lugs, we usually have to calculate the shear stress in the fillet weld caused by twisting load. Pressure vessel design handbook by Bednar recommends using Sshear = TJ/c, where J is the polar moment of inertia. I believe this is the method that everybody is using. However, Design of weldments by Blodgett pointed out that formula is good only for circular closed sections.Both his book and Roark's Formulas recommends using R(or K) in other types of cross sections. If we follow Blodgett or Roark's, the fillet weld size will be huge, and I never saw anyone using this method to design the lifting lugs. I would like to hear some expert's comments. Thanks.

 

[prex]

Bednar's position seems to me more realistic, though not necessarily always safe.
The reason for this position is that the formula T/Jc corresponds to Coulomb's theory for torsion, that assumes the sections remain plane. This is of course true only for circular sections, and is not true especially for long beams where warping of sections (in other words: their departure from remaining flat) is not prevented.
Now it is clear that a beam formed by a fillet weldment around a lug (or also a lug, if it is full penetration welded) cannot be considered as a long beam, and it is quite unlikely that it could display significant warping (though this will depend on the vessel wall thickness).

prex

http://www.xcalcs.com

Online tools for structural design

 

[hiliter]

You are correct in that the Blodgett method is used in determining weld size. What you have to keep in mind is that your maximum fillet weld size that you should consider in your calculations is limited to the thinest section you are welding less 1/16". What will and does happen, is that the weld will be strong enough, but the plate will tear. So what do we do in the design of lifting lugs? Typically, for a vertical vessel, a bar is added between the top of the lug and the head. Other options include: making the lug longer and cutting a recess on the inside of the lug. If there is still a problem with torsion on the shell, consider increasing number of lugs or going to a lug design which will rotate and will not transmit torsionsl load.

 

[jamesl]

Thanks for comments. Does anybody use Blodgett's method to check the weld size in his rountine design? It would leads to a weld size several times bigger than the Bednar's method. Over the years, I have used Bednar method without doubts. I just used Blodgett method to check some lifting lugs we designed years ago. The shear stress is above the tensile strength. But we never had any lifting failure.Is Blodgett method overconservative or is there any other reason?

I am not doing any lug design right now. The question is just for my personal interest.

 

[LSThill]

Hi jamesl (Mechanical)

look at the two different lifting lug calculation on the photo of

http://www.thill.biz

Leonard Thill
Leonard@thill.biz

 

[jamesl]

I just found that Blodgett also used Tc/J (section 6.3)to calculate weld size. He did use R (section 3.6), which is usually a fraction of J for beam-type cross sections, to design structure members. There is no explanation in his book.

 

[chicopee]

If I remember right, Bednar formula is for an eccentric load in the same plane of the lug and is not for a load orthogonal to the plane of the lug.

 

[LSThill]

Hi jamesl (Mechanical) and all Members

WRC 198 has the computer program and the update and revision to WRC 198 IS WRC 448 with update computer program.

http://www.forengineers.org/pvrc/WRC Publications Catalog 479.htm

198 (1) Secondary Stress Indices for Integral Structural Attachments to Straight Pipe, by W.G. Dodge; and (2) Stress Indices at Lug Supports on Piping Systems, by E.C. Rodabaugh, W.G. Dodge and S.E. Moore. September 1974 (45 pp) (ISBN #1-58145-197-0) ©¤ $63.00

448 Evaluation of Welded Attachments on Pipe and Elbows by E. C. Rodabaugh, E. A. Wais , G. B. Rawls. ¨C January 2000 (40 pp) (ISBN: 1-58145-455-4) - $96.00