Understanding Seismic base shear distribution at podium level 2

[Usman3301]

Hi there,
I have this habit of going through ETABS models of buildings being designed in our office to understand the behavior of structures. So I was just going through this multi-tower model with initial framing and parameters as per instructions and submitted it to my senior who'll carry out the design.

I have been trying to understand the behavior of shear walls, loss of stiffness due to cracking, and how cumulative story shear (Base shear) is distributed at podium level and is further attracted by basement walls. Previously, @Kootk helped me out by introducing the idea of coupling beams and that piqued my interest to improve my understanding.

Multi-towers with common podium

What I am currently struggling with is very high seismic base shear demands in basement walls. What my understanding in this regard is that total base shear V = 2200(tower 1) + 3650(tower 2) = 5850kips is to be resisted at the podium level and basement periphery walls will act as absorbers and attract most of this shear demand. But when I consider the pier design for the left basement wall as shown in the figure attached, applied shear is way too high. One thing which comes to mind is the 5% eccentricity considered in the design to account for accidental torsional eccentricity. But still, 17500kips shear on a basement wall seems way too much.

It would be great if someone can help me understand why it is happening this way.



BSc. Graduate

 

[driftLimiter]

First thing with almost any modeling issue is check the applied load matches the base reaction.

You have indicated the shear from the two above regions adds up to 5850k, what about the force from the podium itself. It should all add up to the total base reaction in that direction.
If the computer is generating lateral loading, Spot check the loads per floor to start. Work your way down until you convince yourself that the loading is correct as you would like it.
If your using modal analysis, use ELF to confirm base shear.

Beyond this check, consider what rigid diaphragm analysis is and how it is impacting your results. If you have a big difference between the center of the applied loading and the center of stiffness then you should expect rotation, and with rotation comes coupling effects and this could contribute to 'odd' looking results.

Don't let the model drive you, you drive the model

 

[KootK]

The issues that you're having may well be centered around the backstay effect which can create very large shears in subgrade level walls. Here's an excellent article on the issue to get you started: Link. Quite often, one needs to "soften" parts of the their main floor in the modelling in order to keep the results manageable.