Does this connection work?

[CURVEB]

I'm reviewing a project and it has a framing detail similar to the attached sketch. This condition isn't really covered under any AISC design guide or typical detail that I've ever seen, and I was wondering if anyone has any insight. The concern is that the brace being connected to the face of the beam puts the beam in torsion. In this case, they are showing a double-angle bolted shear connection at the end of the beam, but they are still relying on the web to transfer the torsion into this connection.

The constraint is that the brace and column cannot lie in the same plane as the beam due to the layout of the interior space.

The loads are on the light end, and this is in a low-seismic region (SDC B or less).

What's your opinion on the feasibility for this connection to meet AISC provisions? If the beam had a torsion-resisting connection at the end would it work?

 

[DaveAtkins]

I don't like it.

If the lateral load is small, maybe you can justify it. Still, it looks awfully messy.

DaveAtkins

 

[Berniedog]

Very unconventional. I doubt you could calc. it

 

[Hokie93]

As drawn (designed?), the connection looks very suspect and highly atypical. It may be possible to adequately reinforce the beam given the "light" loads but I would try like heck to secure a change in geometry (or change in braced frame location) so as to ameliorate the situation. Has anyone tried talking sense into the architect or interior designer?? Let me guess: the carpet pattern is dictating the column locations?

 

[TLHS]

Sure you could calc it, it's just really annoying. Getting an upper bound stress of some sort probably isn't too hard. Your critical spot is likely the coped bit of beam. You'll have a moment and a torsion through there along with the shear. Flange bending where you have the plate welded on is also a reasonable place to look for problems, and it's a spot where you add a lot of flexibility. There's also a worrying implication when you load the beam in torsion every time you have a lateral event. Your wind or seismic event could end up inducing a torsional buckling failure in the gravity load path if you aren't careful.

If I had to do this, I'd likely take the brace to the side of the column instead of the side of the beam so I could eliminate the annoying flange plate. Plus it's HSS so it would easily carry the torsion and resolve it into the beams up top.

 

[CURVEB]

@TLHS - good thoughts. The brace into the side of the column might be the solution that works.

 

[KootK]

I'd get rid of the side plate and attach the clip angles / plates directly to the beam web. That way you get rid of the flange bending failure mode and you at least give the connection the option of attempting to deliver the load concentrically. The eccentricity in the connection will generate a moment that has two places to go:

1) Torsion in the beam.
2) Flexure in the brace.

It may well be that the brace is stiff enough to compete.

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.