Beams on Column Cap Plates

[JT-1995]

I am designing a series of shop fabricated horizontal weldments made from HSS12x8's that will be flown in and supported by an existing grid of WF columns. The idea was to minimize the plant outage by replacing the equipment floor but reuse the existing column structure. Note, the floor is a machine floor that is only about 7.5ft above the concrete foundations below.

All seems great in the computer models and connection calculations I have done. Everything is "pin" connected at the bottom of the HSS sitting on cap plates we will put on the existing columns. We will weld the HSS to the plate and the plate to the columns. I have two thoughts growing in my head.
- be careful not induce moment at the top of the column by inadvertently making the connection between the HSS/plate/column top too stiff. But I really don't think it is possible to make that connection too stiff. Could someone agree or disagree? As the design matures I will do some analysis with moments entering the column (exist columns are W8x31's) just to confirm that they are not overstressed if there is moment transfer.
- All of the new weldments will sit on top of existing columns except for three places where we will have three new HSS8x8 posts. Originally I was going to hang the new weldment off the face of the new columns but the connection (HSS beam to HSS column) can be cumbersome with the open ended beam, so I thought to just bear on a top plate at the new column too. All that lead to feeling like I had a house of cards sitting on pegs. The columns are all braced amongst each other and all analysis results in the system being stable, but I am starting to get concerned with the machine load being 13.5" above the top of the column since the machine sits on a 3/4" bearing plate that is welded to the top of the HSS12 beams, that in turn has a 3/4" bottom plate that is the column cap. There is a some horizontal load at the machine base, again the analysis says all is stable. But I am concerned that there are inadvertent stability issues with EVERYTHING bearing on the top of columns and nothing having a common horizontal plane with the load.

I would appreciate any thoughts on the stability of beams sitting on a system of column cap plates. By the way, the column grid is 22 existing columns and 3 new columns and the machine floor covers about 20ftx30ft area.

 

[human909]

be careful not induce moment at the top of the column by inadvertently making the connection between the HSS/plate/column top too stiff. But I really don't think it is possible to make that connection too stiff. Could someone agree or disagree?

This has been discussed here previously, several times. Here is one time:

https://www.eng-tips.com/viewthread.cfm?qid=475312

-A beam across the top of a HSS with a cap plate will usually behave as a moment connection in most respects and so should be treated as one unless you specifically design it so it isn't.
-In many respects what you lose in column strength from induced moment you gain in column strength from the increased effective length.
-Visual deflection of the column CAN be a problem especially if you choose slender columns. The column might be strong enough but it could look like a banana.

If you want a pinned connection then design it as one. I'd use welded cleats and a single bolt to act as a true pin.

 

[JT-1995]

Thanks for the thoughts and links. There is good discussion there.