Interaction model for separation of edges in 2D

[WayneKagawa]

Hi all
I am modeling the slip and separation of a structure resting on soil in 2D when affected by both transverse and vertical TH of an EQ.
I am using interaction properties as mentioned below:
Tangential: Penalty -> Coeff of Friction, Shear Stress limit, maximum elastic slip
Normal: Linear -> Stiffness value
I have define the contact between the soil (master) and the structure (slave) at the edge on top where they meet.

The slip values are coming fine. It's the separation that is bothering me.
The paper I am referring to uses interface elements as a joint between the soil and the structure. The Shear and Normal stiffness values of the interface elements are provided along with plots for both shear and normal behaviour of the elements. I've posted images of shear and normal respectively.

The normal behaviour is such that it takes no tension and is activated (transfers pressure) only during compression. Hence I used a soft contact with a linear variation. Acc to the manual, it transfers pressure only when the overclosure is above zero.

The vertical displacement values I am getting are way higher than what I am expecting. I expect a peak of 10 mm max for the structure and 6-7 mm for the soil. I am getting double the values almost.

I am extremely confused as to how to solve this. Any help would be appreciated.

 

[Mustaine3]

What exactly is the problem? Is the penetration higher than expected?

What normal behavior did you used? Some kind of penalty stiffness or a pressure-overclose definition with a linear stiffness?

 

[WayneKagawa]

The problem is that the penetration and separation of surfaces is higher than I expected.
I used a pressure-overclosure definition with linear stiffness.

Also the EQ time histories have been applied at both the sides and the base.