Relative Stiffness of Steel Frames and CMU Shear Walls

[LPPE]

I've got a steel framed multi-story building. I'm using ordinary concentrically braced frames (or at least trying to). I've put frames everywhere I possibly can in the structure, but the foundation forces are still unacceptable. A face of the building is CMU, as are two stair towers and the elevator shaft. In addition to my OCBF's, I'd like to engage these CMU walls.

What would be the relative stiffness between the OCBF's and CMU shear walls? What procedure would be used to determine both? E of steel and E of CMU are quite different (roughly 29:1), and I don't think my OCBF's will take 29:1 times more load than the CMU shear walls.

 

[IceNine]

Well,

Assuming you have rigid floor diaphragms, you should distribute the load based on the stiffness of the elements, such that they deflect the same amount. The stiffnesses depend on E as well as I.

To determine the relative stiffnesses, model the braced frame with an arbitrary load (say 10K) and model the shear wall with the same load. Calculate the deflections, and the ratio of the deflections will give you the relative stiffnesses of the walls/frames.

If the center of mass and center of stiffness aren't the same, don't forget to take torsion into account.

 

[haynewp]

It depends on the inertia also. You could put a unit load on the cmu stair and elevator shafts and get the deflection as compared to the deflection of the braced frames under a unit load at the same height, figure the stiffness=1/deflection. Then compare the relative stiffness of the cmu versus the braced frames. The cmu will have shear and bending deflections and the braced frames have tension and compression contributions.

You have to get the inertia and modulus of elasticity of the cmu shafts and if you can input this into a computer model or figure the deflections by hand as compared to that of the braced frames.

 

[whyun]

Distribution between the brace and CMU walls are based on the relative stiffnesses of the two systems per the two responses above.

If the CMU ends up veing very rigid compared to the OCBF, consider using a two stage static analysis and deliver all of the load into the CMU walls. I would still design the OCBF for the force distributed per relative stiffness.