masonry building problems

[schvetz]

Hey everyone,

I am currently working on my master's degree thesis model of a masonry building with concrete slabs (portuguese "placa" building) so the slabs could be modeled as diaphragms, even though I decided to use thick shells to model it because of the different dead load cases. My problem is that when running the earthquake combination the deformed is completely wrong, like the building is split in two and with high shear deformation in-plane, even with diaphragms on the slabs this problem still happens (as seen in the pic).

Anyone have any idea why this could be happening? I applied area edge constraints where walls meet with slabs, I re-designed some of those shells, among other things.

Sorry for my english, it's not my native language.

Thank you in advance

 

[Shz713]

Did you try thin shell elements? it follows the Kirchhoff plate theory with no shear deformation

Also, try not to use area edge constrain, instead restrain the boundary condition by using joint restraint depending on boundary condition. In case you need to consider soil interaction, youse joint spring with appropriate stiffness in each direction. Lastly, make sure the automatic mesh option is not selected. It is safer to tell this guy (SAP2K) to divide the mesh by max size, in your case as it is huge building, I gues 1-0.5 should be fine.

Cheers

 

[schvetz]

Thank you for your reply Mohandes2018. I thought the Kirchoff vs Reissner-Mindlin difference was in shear deformation out of plane and not in-plane, so that's why I didn't try it. In the meantime, I found out that this deformation was only happening in the load combination 1DL + 1E + 0.3LL which presents the max values of each, so I think that's what was causing the problem. Can you tell me the downside on using area edge constraints? Since I pretty much put them in every wall intersection and where meshes don't match.

Thanks!

 

[Shz713]

As you said, mesh don't match due to area edge restraint. Keep it as rule of thumb, try to use restraint in joints rather than other elements as this may cause error in output. For better meshing, why don't use manual meshing instead of auto meshing. For details of edge constrain, you better refer to CSI reference manual.

 

[Shz713]

Out of sudden, I was working on a bridge and modeled the kerb as solid shell and wanted to connect to deck, I divided the areas to match the division of deck shell elements. Then I assigned Body Constraint with 6DOF fully fixed, it works perfectly. Here's the quick guide for you:

 Rigid Body: fully rigid for all displacements
 Rigid Diaphragm: rigid for membrane behaviour in a plane
 Rigid Plate: rigid for plate bending in a plane
 Rigid Rod: rigid for extension along an axis
 Rigid Beam: rigid for beam bending on an axis

 

[schvetz]

Thank you for the advice Mohandes2018, maybe I will have better luck assigning body constraints to the joints.

Cheers