Full-Pent. Cantilever Beam Extension

[krus1972]

We are adding on to an existing beam that now stops at the centerline and bearing on top of a column. After adding on to the beam we will be creating a cantilever and the beam will be continuous over the the top of the existing column when the work is finished.

The connection is a complete full pentetration weld on the top & bottom flange and the web.

The weld inspector instructed the contractor to cut "rat" holes in the web at the top & bottom at the new joint. The reason for this is so the welder can provide a continuous welding across the top & bottom flanges using a continuous back plate. The inspector ultrasounded the top & bottom flange and the web and the Full Pent welding is good.

My question is:

This is now a continuous cantilever beam with the new joint over top of a column. When the load is placed at the end of the cantilever there will be a negative moment AND a high amount of shear at the new joint. The loading will be cyclic so fatigue is an issue. Will the rat holes in the web cause web cracking over time?

Can anyone point me to a reference / paper describing the behavior of the shear with rat holes in the webs?

Thanks,

Jeff

 

[UcfSE]

Have you already exhausted the information in the AISC Manual, or does the building code in your area put you under a different code?

 

[Ron]

I would not put rat holes in a cyclicly loaded element. They create stress risers and unless they were drilled, they are probably ragged.

The inspector overstepped his bounds. He should not be directing the contractor to change a design element without the structural engineer's approval. He seems to have made the erroneous assumption that the flange takes all the tensile stress. The complete penetration weld could be provided without the rat holes and you would not have the issue of the stress risers.

Go hand the inspector his "rear end" and let him know that when you decided you wanted the full element with complete penetration welds, that didn't mean one with holes in it.

As for the issue of the rat holes and the shear at that location....the shear is affected by the reduced cross section from the rat holes.

 

[HgTX]

It's not that rat holes *can't* be used in cyclically loaded applications. Steel rolled beam bridges use them routinely. It's a matter of properly sizing, porportioning, and finishing the rat hole and the end of the weld, and accounting for the loss of section in the web.

None of which appear to have been done properly in this case but it doesn't mean it's a lost cause.

What's the standard method for soundly splicing two beams without rat holes?

Hg

Eng-Tips policies: faq731-376

 

[Ron]

HgTX...you're exactly right...it's not that they can't be used, it's just that fab shops that are not used to doing fracture critical work don't understand the nuances of such. In this case, I do not believe the rat holes were necessary to accomplish a proper weld and probably do more harm than good.

 

[kslee2006]

Field welding is quite difficult not to mention full pent around W-Shape under a tight space. The welding inspector may have do you and the welder both a big favor to provide a practical solution and to ensure a sound weld joint at a difficult situation (the beams may be distorted due to the high heat and subsequent cooling, and the weld lines can be full of defects due to working under odd position..just mention a few concerns over field weldind). But, the inspector is err in not to consult the Design Engineer prior to alter the connection details.
As of the shear concerns, I suggest to check using finite element model with reduced web section, and place the column reaction near the column faces (per AISC) to determine whether the connection is acceptable or not.

 

[HgTX]

And grind those rat-holes smooth.

Hg

Eng-Tips policies: faq731-376