Simple Double-Angle Connection

[spats]

I have a project that is under construction where I have specified mezzanine beam connections to metal building columns using simple double-angle shear connections welded to the columns. In accordance with the AISC Steel Construction Manual and the 360-05 Specification, I specified that they weld the outstanding vertical legs, with a 1/2" return at the top. The specific references are Part 10, pgs. 10-7 & 10-8, and Spec Sections J1.2 & J2.2b. The way I read it, the requirement is that you must provide flexibility in the angles, allowing end rotation, in order to design the connection for reaction shears only. Limiting the return across the top allows the beam to rotate because the angles can flex.

As you may have guessed, they fully welded the angles, all three sides. Many of these are fairly light connections (edge girders), typically with a 3 row bolted connection at the beam web, with a maximum reaction of about 18 kips. But there are interior girder connections, 4 row, with a maximum reaction of 32 kips. All are composite beams. Question is, do I need to ask them to remove the additional weld at the top? They're likely to do more harm than good, and there's not enough room for them to be able to get in and remove the welds with a grinder.

 

[delagina]

Following this one...

I had cases where I had a completely welded shear tab connection that is highly not recommended but we have no choice because it's a modular structure and everything will be shipped welded.

 

[KootK]

The only advice that I can provide is anecdotal:

1) In Canada, we're a little behind the times and we still weld all three sides of our connection angles all the time. Nothing bad seems to happen.

2) I've seen the welded shear tab business used as a fix often and without incident. For future designs, perhaps you could use a welded version of the extended end plate connection.

In steel ductility we trust. I guess.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[delagina]

we assumed welded shear tab as shear in analysis

 

[KootK]

Where I mentioned extended end plate connection above, I meant extended shear tab connection. I can never seem to keep those straight.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[SAIL3]

I would also be inclined to let it go.....however, if the flanges on the bm are coped I would, at least, check the web of the bm as that would likely be the weak point. The problem is trying to approximate the rotational fixity provided by the clip angles welded as such.
For a conservative upper bound I would assume the outstanding legs of the angles as cantilvered off the col and calculate the moment of inertia of these legs....then I would run one of these bms on the computer with a corresponding rotational spring @ the supports and get the moment reaction. With this moment and shear I would check the bolts and bm web. If all is good , you are done. If not, then, I would have to refine the analysis which may require doing some research into the actual theory and behavior of these angles.
These type of connections are a matter of relative fixity and there is no such thing as a pinned connection without a dedicated effort.There is also the question of how much restraint is bled-off by the deck in this case.At the end of the day the OP will have to make an engineering decision based on a rational analysis and , if required, a reasonable check of the underlying theory and existing research.
Normally, I would not envision a problem from adding a few inches of weld, but with both flanges coped I am not so sure....anyway, I have never liked the case where both flanges of the bm are coped.