Distribution of Force in Interacting Concrete Shear Walls

[joeycheng]

Hello Peers,

I am struggling to make sense of ETABs results for interacting concrete shear walls. I have the following ETABS Model:

Two cantilever shear walls with pinned bases connected with a rigid link (to simulate a concrete floor with no coupling force, could also be done by assigning diaphragms). I get the following shear outputs from ETABs.

Based on conventional shear theory (shear distributed by stiffness, i.e. 2nd moment of inertia) there should be a linear increase of shear in each of the walls (backed up in pg 381 of Paulay and Priestley). However, from ETABs, the shear in each wall varies. I suspect this is due to some coupling action which I have not accounted for, or due to the deflected shape. However, I'm abit worried regarding how much this differs from conventional theory.

Anyone care to shed some light?

Regards,

Joey

 

[Agent666]

Try any of the following:-

Infinitely small supports. Basically the shell deform unrealistically locally affecting results.

Shear deformations also come into play. Not based purely on flexural theory as you've noted.

Application of a point load to one side of the wall has the same effect as the supports, deforming the shell locally. Providing a distributed load across all walls lessens this effect.

Highly dependant on the mesh size adopted. Reduce mesh size until answer converges.

Might depend on the constraints you have on the links. If you are matching the X deflection at the two attachment points, see points above about locally deforming mesh. Also axial elongation in the walls will artificially constrain x deformations. One wall edge is going down, other up, affecting actual distance between attachment points.

 

[r13]

Did you check wind in both direction, do the results agree, make sense?