PJP Groove Welds from 1 side in cyclically loaded structures

[trainguy]

Has anyone noticed that in AWS D1.1 (steel) and 1.2 (Alum), that these types of welds are dis-allowed?

I can completely understand why, but in the Eurocode standards, these welds are actually permitted and are part of the available fatigue detail categories!

Why would these be allowed in Europe but not allowed in NA?

tg

 

[Ron]

Possibly there's a difference in the level of quality assurance/quality control that is mandated. In NA, QA/QC is not really mandated except mostly on government projects. For private projects, QA/QC ranges from little to significant, but there is no real driver for control.

Another reason might be differences in allowable stress levels, which are significant in the control of fatigue.

 

[MintJulep]

AWS is extremely conservative regarding fatigue.

 

[dhengr]

Trainguy:
Since the Europeans are much older than we are as designing nations, they have undoubtedly already been through many more cycles of stress and fatigue than we have. They have become kinda immune or less sensitive to fatigue. You certainly must know that many of them have stress intensity factors, that are in the ‘laid back’ range, much lower than ours. More wine and beer I think, sauteing and softening their weld roots.

I suppose if you can eliminate any tensile stresses or bending and prying type stresses across the unprotected root, that is, running perpendicular to the line of the weld, you would be o.k. with a PJP weld from one side. That would be akin to the shear stress on the throat of a fillet weld which is preventing two plates from being sheared past each other, parallel to the planes of the plates, along the line of the weld, a shear flow.

How are our disallowed PJP welds loaded and stressed vs. the Eurocode’s welds. I never gave a second thought to a PJP weld, from one side only, same as a fillet, on a web to a flange joint on a built-up box member; as long as we had good fit-up and good penetration of the root pass, and no possibility of prying causing tension across that root. The prying would be caused by rolling (bending) the web over toward the flange and putting the root in tension.

I think we spend too much time trying to codify every possible condition and detail. And, then not enough using general weld design rules and common sense; and not enough time thinking how is this joint or member going to act w.r.t. this weld? How is this weld loaded and stressed, maybe I shouldn’t use that detail, in this particular case; but that doesn’t make it wrong in all cases. Then, as Ron said, QC is all important, since one weld defect can defeat the whole thing. Poor root detail, fit-up, and access; or lack of penetration and fusion, undercuts, cracking due to weld bead proportions, etc., etc. will all likely undo the thing quicker than the fact that you welded from one side. The operative situation is an inferior and unprotected root, and there are all kinds of ways you can get those. Some of this concern probably came about after the Northridge EQ, or there abouts, and our greater concern about what we thought were moment connections which failed from the inferior root.