Hello, I think the probability of torsion of shear wall under transverse load is small or very small. Considering torsion, the whole deformation of structure will cause torsion of shear wall under earthquake.

If center of mass and rigidity have eccentricity then torsion will result. However in comparison to columns shear wall are more stiff and therefore they will be also subjected to similar behaviour but less in magnitude.

Yes, It is possible.

Certainly it is possible. For an extreme example, consider the case of perimeter shear walls, two in the N/S direction and a single wall in the E/W direction. This system would be stable with a N/S eccentricity of 1/2 the building dimension for the case of uniform area mass distribution.

If shear walls are oriented along different directions in the floor-diaphragm (e.g. some walls along X-dir and other walls along Y-dir), then applying only the Dead Loads may induce torsional deformations. The diaphragm's rotation is due to the different deformation modes between the strong direction of a wall (cantilever bending type) and the weak direction of another wall (counter-flexure (frame) bending type). Because the constrain action of the diaphragm, which imposes common floor displacements along a bending plane, the rotation of the floor-diaphragm is inevitable. The induced torsion increases with the number of stories and so, it can be significant in multi story buildings.

Thank you Athanasios Bakalis. But I could not imagine the meaning of different deformation modes between the strong direction of a wall (cantilever bending type) and the weak direction of another wall (counter-flexure (frame) bending type).

Thank you again for your reply.

Thank you Timothy Huff and @Om Prakash Chhangani. But could the eccentricity between the mass and rigidity center induce torsional rotation due to only dead load?

Torsional rotation will be caused by earthquake force which is acting at centre of mass. If there is no earthquake load then DL alone will not result in torsional rotation. The DL has to be accompanied with EQ load.

To better understand, model a 4 wall, 3D multi story building assigning diaphragms at each story. Connect the walls with beams. Assign a different orientation only to one of the four walls (i.e. 3 walls having their strong direction along X axis and 1 wall with strong direction along Y axis). Apply the Dead Loads only. In the deformed shape, notice the different deformation shape of each wall along a bending plane (X or Y). The wall with its strong direction along the bending plane shows a cantilever type bending. On the contrary, the wall having its weak direction along the bending plane shows a frame type bending (counter-flexure). In order to understand the difference, model a second multi story, 4 wall building, similar to the previous one, but this time all walls have the same orientation (e.g. all walls have their strong directions along X axis or Y axis). This time, the deformation shapes of every pair of walls along each bending plane is the same (cantilever or frame type) and so, there is not diaphragms' rotation and therefore, torsion is not induced.

Athanasios Bakalis

The left building (have two walls in x direction and one wall in y direction) exhibit torsional rotation and the right one have major bending (three wall in x direction)