Modeling rigid basement

[Jamie112]

I have this building that has two levels of basement and I want to assume the base is on the ground floor.
What I have done in my model is I include these basement walls in my model and at the same time I assign point restraint UX UY to four corner joints of my ground floor slab (shell element) and all nodes of the ground floor share one diaphragm.

Is this approach correct and if not, what do you do?

 

[KootK]

This is a complex issue and, as far as I know, there really is not a "right" answer. Some common design approaches:

1) What you've suggested is as good as anything.

2) In the old days, and often still, once just stopped modelling at the ground floor, thereby enforcing the kind of behavior that you are trying to create.

3) This fine article describes some of the nuances associated with modelling the flexibility of basement and ground floor slab: Link

 

[slickdeals]

I would prefer to model a very stiff spring in lieu of a unyielding support. The reason is that you will end up with some artificially high shear reversals with Ux, Uy restraint. Also, be careful with rigid diaphragm assignment on a level that has rigid supports.

 

[Jamie112]

Thank you guys.
Can you please explain this a bit more?

I would like to model spring but I find it hard to assume the spring stiffness. What is the stiffness you would use? (ground floor to basement 1 is silty clay while basement 1 to basement 2 is basalt rock). The geo report doesn't mention this. Is there anyway to calculate based on the soil parameters? I know I better ask geo engineer but what is the stiffness (kN/m) you would recommend for this soil type? Also, do you apply point spring to the ground floor nodes? or line spring, or area spring to the wall?
I found using spring is making it complicated

 

[bones206]

I agree that using springs in a model can sometimes help avoid an over constrained model. It’s a bit of an art form and can take some trial and error to get a feel for the global behavior under a range of spring values. I would try to work out the expected behavior using simple hand calcs first, then adjust the boundary conditions until you achieve the expected results.