W Beam Welded to side of HSS Column

[Structural Matt]

How would you go about analyzing the connection pictured below? For more context, the beam is 15'-0" long and has a typical shear connection at the other end. W12X14 beam and HSS10X10X3/4 column. This connection was supposed to be a shear connection at the centerline of the HSS column, but for reasons currently unknown, it was installed like this.

The total load in the beam is about 2 KIPs, so I believe this connection should be fine for what it is, but I also don't exactly know how to analyze and prove it works. Next time something like this happens and the loads are higher I'll want to be better prepared.

 

[JAE]

I don't think it is fine at all.

The fillet welds on the flange tips will be placed in bending about the weld longitudinal axis.
That is a poor situation and possibly dangerous if the welds are only on one side of the flange vs. top AND bottom.

It looks like there's a bottom weld on the top flange but even so you are creating a bending moment at the flange tips where the welds would have to work in tandem with a very small moment arm to resist.

2,000 lbs. isn't much I agree but I'd want to analyze the actual load on the welds to check it with the bending from the flanges.

 

[STpipe]

One thing for sure to consider would be the effect of torsion. You're going to get a tension-compression couple and would likely check that force against the requirements for Chapter K (off the top of my head) for the side wall of the column yielding and punching failure. Also would also include the effect of the eccentric moment on the design of the column.

Edit: Also good call on JAE for pointing out flange bending as a potential failure mode.

 

[JAE]

I forgot to mention that if you have only welds on one side of each flange then the weld will be flexed across its root and possibly fracture.

 

[driftLimiter]

I pray that you the OP never has to encounter this again.

A simple analogy for this would be two plates cantilevered off the column (hopefully by double sided fillet welds).

Ideally the whole thing is controlled by bending of the flange plates. The bending moment into each weld is particularly problematic.

Easy enough to fix with an end plate, or maybe a plate through the beam web.

 

[JoelTXCive]

Could you add a little corbel or seat below the beam?

 

[phamENG]

This should be fixed with an angle seat to the column and probably an end plate since web stiffeners won't be an option.