Bounding Cam Clay Model

[g8star]

Hi STKO,

I have a simple slope stability problem and I would like to use the BoundingCamClay material to represent critical state soil behavior. During the gravity stage, the model exits with a -3 error flag. I'm using the material property values from the example problem. https://opensees.berkeley.edu/wiki/inde ... g_Cam_Clay

The model works just fine with the PIMY NDMaterial. Is there a specific element or preprocessing step needed for the BoundingCamClay model? There is very little documentation of the model online.

I've attached the .scd file for your reference.

cut_inwater.zip

(96.08 KiB) Downloaded 317 times

[STKO Team]

It seems it's giving a singular matrix.
We don't know if there is an error in the code, there is not a sufficient documentation to test it

[g8star]

Yes, I agree with the limited documentation. I would like to use a soil model that captures volumetric hardening associated with lightly overconsolidated clays. Is there another soil model available? Or should I reach out to the authors of the BoundingCamClay model for advice on the material model?

[STKO Team]

Yes you can contact the authors, but first try to search online some example, or try to run a simple 1-element example and export the stiffness matrix to understand why it is singular and in what DOF

[g8star]

I've developed a simple 1-element test, but it doesn't provide much insight. The stiffness matrix appears to develop values of -nan between all translational dof's within the first time step. I've attached the .scd file and the stiffness matrix output.

I've searched online for additional documentation, but there are no complete examples of the BoundingCamClay model. Are there other soil models that exhibit volumetric hardening? Most other FE and FD package have a model for clay with a volumetric cap. For example, Plaxis has the Hardening Soil Model and FLAC has a soft-soil model.

ccel.zip

(74.38 KiB) Downloaded 295 times

[STKO Team]

Check this:

ccel.zip

(95.17 KiB) Downloaded 324 times

We looked into the source code.
It looks like you first need to do a gravity pseudo-static analysis to obtain vertical stress due to self-weight.
During this vertical analysis, the material response must be enforced to be elastic (using the parameter materialState,... don't ask me why they decided to use another parameter and not the usual updateMaterialStage...)

Then you can switch to plastic and go on with the transient analysis

[g8star]

You are all amazing. I knew their had to be some initial step during gravity loading but I never would have found the materialState parameter. The single element seems to converge now, so I'll try the material in a more complex model.

Thank you so much! Your excellent responses and customer support are key reasons that I'm using STKO instead of other FE packages.

[STKO Team]

You're welcome.

Unfortunately, when the documentation is not sufficiently complete, the only way to understand some results/parameters of Element or Material in OpenSees is to go into the source code and search the appropriate methods.
For parameters, the method is setParameter:

BCC.png (90.87 KiB) Viewed 5567 times

Just remember that I created a selection set for the setParameter command in STKO with that surface. When you have a larger model, just make sure to create a selection set that contains all geometries (volumes in 3D or faces in 2D) whose Physical Property is the one that needs the update

[Lei Lei]

Dear STKOteam,

I have replaced the soil elements from 'quadUP' to 'quad' and conducted a static gravity analysis on your updated model.

However, I encountered the same issue previously observed. Could you please offer further suggestions on how to proceed?

Looking forward to your response.

Best regards.

[STKO Team]

Check your selfweight, as you put it is 15 MPa, are you sure?
Then use adaptive time step