Case XII - Typical Column Splice 1

[MollyK]

I am working on connecting a W10X88 Column to an HSS8x8x5/8 for a combination of shear, axial, and an applied moment. It is an architecturally exposed connection, and with that the architect would like to have a 1" butt plate. I am having a hard time proving that a 1" plate is adequate; I initially treated a portion of the plate that was not "braced" by the welds like a beam. Unfortunately based on that check, a plate greater than 4 inches would be required (which is a bit ridiculous). This leads me to believe that I am doing something completely wrong. The connection I have does not 100% conform to the type XII typical column splice (Table 14-3 13th ed.), although it is similar. I was hoping someone could point me to the AISC Specification that is references to select the butt-plate thickness. I can easily check the shear and tension, but it’s the moment that is causing my grief.
Any help would be appreciated; I have also attached a snap-shot of what is going on for this connection.

 

[bootlegend]

First off, I would ask why? I've never seen anything like that. Why is the tube needed?

If there is a satisfactory reason, then there is a lot of stuff to check, but you can quickly check the cantilever of the plate and see where you are at.

I'm scratching my head still though.

 

[MollyK]

The architect has put these things all over this job, he/she finds them aesthetically pleasing (no really a satisfactory reason, I know). So far the EOR has provided us with a plate size per the architects requests, unfortunately not for these connections.

Thank you for your response, I'll see if that comes up as adequate.

 

[IsaacStructural]

I don't like it, seems really strange to put a splice so close to the column base like that, very unconventional. You'd probably have to do quite a Frankenstein calculation to adapt the provisions of AISC to feel confident that you've designed it correctly.

M.S. Structural Engineering
Licensed Structural Engineer and Licensed Professional Engineer (Illinois)

 

[JLNJ]

Can you split a tube and weld it into the column section above the splice? You'll need to shim between the inside of the flange and the outside of the split tube.

 

[JAE]

Whatever you do, the key is to have the column sections "mimic" each other up through the splice so that the plate has to do minimal work to transfer the load.

This would maybe mean trying to match the WF flanges with the flange walls of the tubes.

 

[KootK]

With a load path so convoluted, you'd have a hard time convincing me that the connection truly represents "fixity" from a stiffness perspective.

I vote for running the W10 right down to the bolted flange plate. If a box & plate looking thing is required aesthetically, accomplish that look by welding in side plates etc.

Is the top of the floor slab located at the top of the cap plate by chance?

 

[MollyK]

Thanks for the responses!

JLNJ - I cannot visualize "splitting the tube". Unfortunately we dont have a lot of room to change section sizes at this point. I am confident that this connection will be more than adequate with a 1" plate, but it is just presenting it to the EOR with a viable check. This connection only supports a small roof with minimal loading, aka just minimum snow/wind/dead loads for a roof (more like an awning).

JAE & KootK - Unfortunately not to much room to change the section sizes.

I am going to assume that all of the loading is going to transfer through the flanges and treat it like a simply supported beam, find the maximum moment and move forward. I believe that this is an extremely conservative approach,but at least this way I will also check for "punching shear" as well.

Thanks you all for your input

 

[bootlegend]

I think everyone is questioning why the tube is needed at all when you could simply run the column all the way down to where the bottom of the tube is shown in your picture. The connection shown is going to be many times more costly on top of having inefficient load transfer.

I think JLNJ meant that you could run the column down and then use a split tube as a facade to give the appearance of what you have sketched.

 

[MollyK]

It is not that the tube is needed for any structural purpose, but rather an aesthetic purpose. I am not an architect, but I can tell this concept of box to wide flange has been repeated throughout this job. My issue, is that at every other location, I have not had to size the plate, it has been provided in the structural details. (It is provided in the architecturals as a 1 inch plate, but I cant refer to the Arch. on my connection design). So if I can come up with some reasonable way to check the adequacy of a 1" plate, that will keep the EOR, the Architect and me happy.

 

[structSU10]

i dont undertand how this is aesthetically pleasing. If anything, use a tube column all the way up to the roof if you want that look. And do you have access to FEA? That could be a way to see how the load travels through the assembly.

 

[Archie264]

And I don't understand why this isn't the responsiblity of the Engineer of Record. Having a fabricator size standard AISC connections is one thing but the design something this unusual should be the responsiblity of the EOR, in my opinion.

 

[MollyK]

Haha I agree, unfortunately its the architect's vision. It's a 12 story building, and much of the structural steel is exposed, so we've been stifled before by what the architect wants.

And unfortunately I do not have access to FEA, but it would be awesome. I have it "worked" out, I went back to the basics, and very conservatively checked the plate, and the architect is just going to have to live with a 2.5" plate.

 

[MollyK]

Archie264 the connection design for this project was sub-ed out the the company I work for by the fabricator, and the EOR reviews all of the connection designs.