Torsionally Pinned Connection

[RFreund]

I know this subject came up once before (see here: thread507-204599 ) However there seemed to be a couple of different opinions and it didn't seem like there was a clear answer (which may still be the case).
I would like to know if there is a design procedure or a detail for a torsionally pinned and flexurally pinned connection. Specifically I would like to know how to analyse a single shear plate and double angle connection to be used as a pinned torsion connection.

I thought for a single shear plate you could analyse the weld for shear and torsion, but not exactly sure how to address the plate.

Thanks

EIT

 

[a2mfk]

I would do as many suggested on that other thread- design the connection for shear using a standard AISC shear connection. Then place angles or plates at the top and bottom flange for the torsion, design them for simple shear.

My one concern is I would want to ensure the shear plate bolts are in full bearing before welding/bolting the torsion bracing. Maybe top/bottom angles at the flanges could have vertical slots so they were only taking horizontal torsional forces, and not accidentally taking vertical load.

Or maybe if your structure will allow it, you could weld an end plate to an HSS member and then bolt or weld this to your supporting member. Could get tricky from an analysis standpoint, since your welds would take shear and torsional shear. Also this would most likely be a rigid connection.

 

[JoshPlumSE]

To me a torsionally pinned and flexurally pinned connection is a double clip angle. It can restrain torsion, but it cannot restrain warping. It cannot restrain flexure. It can be handled using the AISC design procedures or the simplified hand calc methods (i.e. the equivalent tee analogy). It should be relatively easy to resolve the torque into shear forces on the bolts. So, clip angles should be straight forward.

IMHO, a shear tab should not be used any time there is signficant torsion in a connection. For existing structures I'm not sure what you would do, but I would not make changes to my structure that would introduce torsion into these types of connections.

 

[RFreund]

Thanks,

I agree. I don't think I was grasping the angles on top and bottom concept as I was thinking that they would start transferring moment (horizontal forces) into the supporting member however they angles if bolted would be relatively flexible. I imagine plates would start to transfer moment though, no?

EIT

 

[a2mfk]

RF- if you used plates top and bottom you could weld them to the support, then provide slotted holes and bolts to the top/bottom flanges in the axis of the beam to keep the connection pinned for moment and restrained for torsion.