I tried to model sample of triaxial test in Abaqus program. I used hypoplastic subroutine. The parameters used in hypoplastic model was Hochstetten sand the given in website “https://web.natur.cuni.cz/uhigug/masin/plaxumat/node5.html”.
The used steps of analysis were as;
– First step is for applying confining pressure in the sample.
– Second step is for consolidation process (dissipation of excess pore pressure).
– Third step is applying deviator stress (axial stress).
The analysis was completed in first and second steps, but it was aborted in third step (note: the deviator stress was small).
Could anyone to solve this problem?
The boundary conditions are shown in attachments figures.
Dear Neif,
– Please first check the results of the first two steps to see whether they are sensible, even better you may compare them with results of some element test drivers from SoilModels: this will consfirm that you have properly set the parameters.
– Then make sure you subdivide the shear stage into many small steps. It is not clear from your post whether you use intergranular strain or not. In particular the istrain version may require small step size as there is significant stiffness change during loading.
– If the simulation crashes, it should produce error file (something like fort.1) where some crash information should be given. You may also see some messages in log files of Abaqus project, which could help to find the source of the problem.
Regards David
Dear Dr. David
– The intergranular strain concept was used with value δ=0.00001.
– The magnitude of axial stress that used at third step is 1.0 kPa.
– The type of steps is soils (transient)
– The time period and increment size of third step were 0.1, and 0.001 respectively.
In term of the results of analysis of first and second step, there was no change in void ratio.
the analysis is incomplete in third step.
I need for helping for this problem please?
Dear Neif,
It seems that the second step was failed if there was no void change.
Have you ever try a triaxial test with only one element?
I prefer to start with elementary tests without intergranular strain
Regards
Dear Wang
Thank you very much for your helping.
I considered your noting for modeling sample without intergranular strain, but the analysis was incomplete in step 3 (applying axial stress).
In term of void ratio, i mean that the void ratio in first step (applying confining pressure) is equal to the void ratio in second step (consolidation step). on other word, the value of void ratio is constant in first and second steps which is equal as the void ratio that set in initial condition.
Hi,
just a couple of observations:
Why are the bc in the lower side fixed in vertical and horizontal direction? You only need to fix them in the vertical direction (they must be free to move in the horizontal direction specially in the consolidation phase).
I would advise you to model this first with a 2D model in axial symmetrical conditions and without intergranular strain. Start with 1 element and then you can slowly increase the number of elements or even dimensions if you need it and afterwards add with the intergranular strain parameters.
If the void ratio did not change it means there is no volumetric deformation. This might be the case if the confinig phase is not modelled correctly (check if the stress state changes correctly) or if the sample is able to deform (drained conditions)
Good luck.
Dear Oscar
Thank for help. I considered your noting, but i have the same problem.
When Hypoplastic model were replaced by Mohr-Coulomb and elasticity, the analysis was completed and there were volume changes throughout steps.
When Hypoplastic model were used, and there were no drainage conditions at top and bottom of sample, the analysis was completed and here were no volume changes throughout steps.
Is there parameter in hypoplastic umat subroutine or in hypoplastic model (12 parameters) that control or prevent the volume change of the sample of triaxial?
I modeled initial void ratio as uniform value. If are there other method to model initial void ratio (predefined field), provide me this method please.
Hi Naif,
not to my knowledge. I have worked with the von Wolffersdorf hypoplastic model without a problem. I am thinking the problem lies in assigning the initial void ratio. You can check this value in Abaqus output by looking at the system defined variables (sdv).
Good luck solving your problem