How to simulate different strain rates in undrained triaxial test with Plaxis3D

Question

I would like to determine the effect of different strain rates in the undrained triaxial test with Plaxis3D. For example, how to input strain rate as 10e-5 per second with an undrained triaxial test in Plaxis3D, is it possible to do this case?

Dennis Waterman · Answer

Dear Kanthima, 
 Yes, that can be done. But you would first have to ask yourself what kind of strain rate effect you would expect to have. Because apart from consolidation, most strain rate effects would require a constitutive model that actually has those effects included. For instance, if you would expect to see a creep related effect, you would have to use a constitutive model that includes creep - you won't see it with Hardening Soil as that model doesn't have creep. In SoilTest the standard option for a triaxial test doesn't have time, so you should use the General option to prescribe the boundary conditions as well as the time interval. 
 If you're interested in consolidation effects, be aware that SoilTest doesn't have consolidation included, so you will have to make a small finite element model representing the triaxial test (an axisymetric model, horizontlally fixed on the axis of symmetry and vertically fixed on the bottom - load on the outside (cell pressure) and a prescribed displacement on top to apply the strain) and then you can just specify the prescribed displacement at the top and a time interval to create the strain rate. You would have to calculate the application of the strain rate as consolidation phase of course. With kind regards, 
 Dennis Waterman

Dennis Waterman · Answer

Dear Kanthima, 
 First of all, if you have the possibility to do it in 2D I would recommend that. It's far easier to use a 2D axisymmetric model for a triaxial test than using a 3D model. 
 If you do have to use a 3D model, you would have to model it as a cylinder. Apply the cell pressure radial on the cylinder, and the prescribed displacement on the top. 
 1) I presume you're referring to the top surface where you apply the prescribed displacement. Yes, there you should set the x and y directions as free. 
 2) Yes, that's possible. Though you may want to do your calculation in seconds rather than in days. 
 3) I wouldn't use a flow boundary at all, what would you need it for? Just prescribe a degree of saturation to the soil sample itself by setting it to "unsaturated". 
 With kind regards, 
 Dennis Waterman