Cohesive element and geostatic stress

[Faranak Sah]

Hello,

I am trying to model frozen soil-pile interaction using cohesive elements. when I apply the geostatic stress as a pre-defined field to soil, I expect the lateral pressure of soil element adjacent to cohesive element to be the same as the pressure applied to the cohesive element in its thickness direction. However, this does not happen. for example if the lateral pressure of soil is 60kPa in a soil element, the pressure on cohesive element (in its thickness direction,S33) is less than 1kPa, which means that it can not read the geostatic stress from the adjacent soil correctly.
I am not sure that if I am doing every thing correctly. I would greatly appreciate your comments and ideas about that.

Faranak

 

[FEA way]

Are you using a geostatic analysis procedure ? Do you have an initial step defined to establish equilibrium ?

 

[Faranak Sah]

Yes, I define a Geostatic step and apply geostatic stress predefined field at 2 vertical elevations .Then I apply gravity in geostatic step. When I use the ordinary elements at soil-pile interface it works well. But when I use cohesive element at soil-pile interface, there is a discontinuity in stress distribution because of interface elements.

 

[FEA way]

Well, according to the documentation, geostatic initial stresses can be used only with continuum elements. Special-purpose elements (like cohesive ones) may not work properly with this type of predefined field. Are there no warnings in output files regarding this limitation ?

 

[Faranak Sah]

There are no warnings regarding the invalidity of cohesive elements. This is What I read from the documentation:

"Any of the stress/displacement elements in Abaqus/Standard can be used in a geostatic procedure. Continuum pore pressure elements can also be used for modeling fluid in a deforming porous medium. These elements have pore pressure degree of freedom 8 in addition to displacement degrees of freedom 1–3. However, the enhanced procedure can be used only with continuum and cohesive elements with pore pressure degrees of freedom and the corresponding stress/displacements elements. "

So it seems that it should work for cohesive elements.

 

[FEA way]

Yes, but this just means that cohesive elements can be used in a geostatic step (otherwise there would be an error). But they may not support initial geostatic stress.