Thin Plates for Moment and Torsion connection

[MiguelS]

Hi guys,

I have a bit of an unusual situation. So, in reality I'm designing some stairs in which the flights/landings cantilever off of columns in the central well of the structure.

All sections of the structural members will be hollow section (beams and columns) and the senior engineer doesn't want to use hollo-bolts or anything similar.

His suggestion for the connection is illustrated in the picture below:

My concern is, this thin plate connection would need to resist bending plus torsion, would it be any effective in doing so? Or am I overthinking it?

If you have any suggestions on how to design the connection and make sure it works as intended that would also be great!

Thanks.

 

[briancpotter]

I think instead of torsion, your P4 load is more likely to induce a couple on the two plates, with positive shear on one and negative shear on the other (superimposed on top of the vertical loads and the bending from the uniform loading).

Brian C Potter, PE
Simple Supports - Back at it again with the engineering blog.

 

[rb1957]

yes, 99% or the reaction will be shear, but there will be a small torsion on these plates due to curvature of the beam.

I'd be more worried about the welds in bending (due to off-set loads).

another day in paradise, or is paradise one day closer ?

 

[MiguelS]

Thanks for that. Yeah, bending was the other concern. The codes tend to not recommend this types of connections for bending, but I couldn't really understand why. In theory a pair of forces would form on the bolts to carry the moment to the column. And the connection shouldn't be particularly flexible I imagine, if it failed would be catastrophic failure.

Any ideas why this connections are usually considered pinned and not recommended for moment connections?

 

[briancpotter]

Not sure if this is the reason, but I'd be a little worried about the weld unzipping due to load concentrations at the top/bottom of the weld Typically when moment is placed on a weld, the weld is detailed to have an extra line of weld at the top and the bottom to give it a C-shape, but your geometry doesn't allow that here.

Brian C Potter, PE
Simple Supports - Back at it again with the engineering blog.

 

[BAretired]

I don't like the suggested detail because the three bolts must resist both shear and moment. They are not sufficiently spread out to offer much moment resistance. Any play in the bolt holes results in slope of the landing which is difficult to adjust.

I would prefer an HSS bracket shop welded to the column and extending to the outer beam. The inner and outer beams could bear on the bracket, preferably but not necessarily, on top of it.

BA

 

[MiguelS]

Yeah, after doing some quick calcs it will be very, very difficult to carry the moments in the small welds. The main factor for this arrangement is that this is highly architecturally driven, and they want to try to avoid any visible connections.

Sorry BAretired, what do you mean by HSS bracket shop?

 

[jayrod12]

I may be wrong but I think he's indicating welding an HSS outrigger/stub to the column in the shop and then site welding the beam to it. This would allow for a "hidden" connection in that it would just be a fully welded connection. Which when done by a good welder, looks beautiful.

 

[rb1957]

it would be "better" if the plates cut-thru the beams ... break the beams at the plates and weld everything back together. Now at least there's a good loadpath into the plates.

another day in paradise, or is paradise one day closer ?

 

[BAretired]

Jayrod12 has it right. I should have used the term outrigger instead of bracket. It would be shop welded to the column.

BA

 

[rb1957]

does it look like a flimsy loadpath to you guys ?

another day in paradise, or is paradise one day closer ?

 

[wadavis]

I've avoided these connections out of concern for bolt slip creating a sloped landing (unless they are fayed surfaces). Do you think this is an unwarranted concern?

Structural, Alberta

 

[DETstru]

Is this cantilevered in two directions off the column or is the far left beam a support?

 

[Toby43]

"The codes tend to not recommend this types of connections for bending, but I couldn't really understand why.."
It would not have enough stiffness to attract significant moment in a frame system. As a cantilever as you have it, I would be worried about slope of landing as others have mentioned. As BA mentioned could an outrigger be used? Even take it full width of landing.