Modeling Stress Relaxation in ABAQUS

[rm447]

Hi,

I am trying to model a stress relaxation experiment in ABAQUS. I am using cartilage material in the test and for modeling that I am using the soil analysis. The displacement of the indenter for each step is a ramp rising at constant velocity for t0 seconds and then constant over some prescribed time period. In order to do that, I defined two steps. One where the cartilage plug is being loaded, for which I set a velocity type boundary condition at the top surface, and a the second step where I remove this BC to fix the position of the top surface and let the plug relax. When I plot the results it I don't see any relaxation. The second step shows constant stress (S22) - it's just looks like a ramp. I tried increasing the permeability just to see if it had any effect. However, it seems that the problem has nothing to do with the material properties, but with the way I am modeling the condition. This may be very trivial but I am just beginning to learn ABAQUS and appreciate any thoughts on this.
I can post my input file if needed.

Many thanks!

[Jorgen]

First, I would check that the applied strain history that Abaqus is using is indeed what you have in mind. If that looks correct the the problem is likely caused by the material model that you are using. Are you using a soil model to represent the cartilage? That sounds like a strange choice to me. Can you explain why you selected that model.

- Jorgen

[rm447]

Dear Dr. Bergstrom,

Thanks for your reply. Yes, the strain history is correct. I am using the soil model because it can be used for analysis of partially or fully saturated fluid-filled porous media. As for the material model, I am using the properties that I obtained fitting Van Mow's biphasic model to my experimental data. I defined the elastic properties of the solid matrix, the permeability and initial pore pressure and void ratio. The reason for choosing this method was to use ABAQUS functionalities rather than defining a user material, since I am new to ABAQUS and thought that would not be an easy way to start. Is there a better way for doing this than using the soil model?
In order to check if my input file made sense I also tried using the material properties from an example in ABAQUS documentation on the soil analysis (Terzaghicpe8p.inp for example) using my own loading history. In the ABAQUS example a load is suddenly applied and then the soil is allowed to reach equilibrium. Using my strain history I still don't see any relaxation.
Is the way I'm loading the tissue correct?

[Jorgen]

OK, I see. I have used the similarity between the porous media soil models and the biphasic model in the past. If I remember right, there are numerous input parameters that are needed: the material parameters, pore pressure, permeability, boundary conditions, and the Abaqus solution parameters that control how the transient behavior is captured.

I don't see anything wrong with the boundary conditions that you applied
Are you sure the different parameters are consistent?

- Jorgen

[rm447]

Right, I hadn't used the right permeability. The permeability in ABAQUS is in m/s and is equal to the specific weight of the liquid*the actual permeability (m^4/N/s). It works now Though I am not sure this is a good way to do it since you need to have the specific weight of the tissue matrix as well (in order to find the pore ratio, but I only have the dry weight percentage).
Thanks a lot for your reply!