Steel column made of three parallel pipes - checks

[mikesg]

Hi,

I have to design a column made of three steel pipes (5" diameter each), that will be loaded by biaxial moments and compressive force. As a cross-section the pipes will be placed in the tops of a equilateral triangle and have clear space between them about 4-5". I intend to provide connection between them with steel plates welded to the pipes with a certain spacing between them.

Can someone please provide a pointer to code soecifications or a book regarding the checks that need to be made and the requirements towards the plates and the connections in order to design the column as a single element (i.e. a frame element with section having the geometric properties of the three circular sections).

What I have found in our code is that the plates must be spaced at maximum 40*i (radius of gyration of single pipe) and be designed to carry a fictitious shear force.

I will make additional checks by modelling the column as a vierendeel space truss and design the welds and the plates using the forces obtained, but I want to be sure I am not missing some checks.


Thanks a lot and I wish you a productive week!

Mike

 

[Ron]

Mike,
Your approach is reasonable. Have done similar with 4-piece columns. Mine was a bit easier because of the symmetry.

Be sure that the top and bottom plates of the column are of sufficient stiffness to transfer the load equally to all the pipes, otherwise that fictitious shear becomes real. Check for local crushing of the plates as well.

Ron

 

[ishvaaag]

In any case it would be as easy to model the thing, tying plates and segments in-between as line elements, then getting direct check from a program like RISA-3D. Any gemetrical or construction imperfections, P-Delta and delta can be dealt in a variety of ways, leading to safe results for K=1 and lesser.

 

[mikesg]

Thanks for both answers and for your suggestions for approaching the problem!

Mike