Learning Forum

[psh1988]

Hello 

I am doing a transient structural analysis on a slope(Mohr-Coulomb soil material under a time history of acceleration). As the third dimension of the slope is greater than the other two, I am inclined to model it in 2 dimension to save time as I am going to use my FE model for validation(hundreds of analyses).

Accordingly, I should validate my 2d results with those of 3d to be sure that my 2d model is as accurate as 3d one. In this regard, I modelled a 3d slope( with 1 m thickness) and applied the desirable boundary conditions (fixed base, constant displacement in side boundaries in x direction, constant out of plane displacement for both front and back side of my model) and acceleration but the results differ from my 2d plane strain model. I realised that the non-zero shear stress results (s(yz) and s(xz)) in 3d modelling(which is zero in 2d-plane strain case)might be the reason of difference.

P.S. I did the same modelling for elastic material and i could get the same result for both 2d-plane strain and 3d under the same loading and boundary condition as for Mohr-Coulomb material.

Could you please tell me what should i do in order to get the same results as 2d from my 3d Mohr-Coulomb material model??

so , I will be able to simplify my FE model to 2d model and save so much time!

Thanks

[psh1988]

and here is my plane-strain model.

[psh1988]

Hello everyone..

As i have not reach a conclusion, I am still waiting for your kind response ,especially, engineer Peteroznewman....

[Aniket Chavan]

Ansys Employees can not download attachments.

Are you sure that your inplane dimensions are sufficiently small as compared to the thickness so it can be safely assumed as a 2D plane strain model?

-Aniket

Guidelines on the Student Community

How to access ANSYS help links

[psh1988]

Hello

My model is different...I am going to model 1m thickness of my lengthy model and apply plane-strain like BC. However, the results are not  the same(comparison between 2d plane strain and 3d 1m-thickness model).

How can I upload my model, so ANSYS employees be able to view it??

 

Bests

[Aniket Chavan]

Sorry, there is no way in which you can upload the model at which we can take a look. Other members can check your model though.

How are you applying plain-strain like BC in the 3D model?

And again, what are your in-plane model dimensions?

-Aniket

Guidelines on the Student Community

How to access ANSYS help links

[peteroznewman]

I looked at your 3D model and see these boundary conditions to confirm that B is Z=0 on both sides leaving X,Y Free, and C is X=0 on both ends leaving Y,Z Free.

Here is some detail on the mesh.I suggest you use Sweep and Inflation to get a better mesh.

        

Here is your Plane Strain model.

Here is the 2D model Static Structural result in 1 G gravity.

Here is the 3D model Static Structural result in 1 G gravity

It looks to me like the 3D model gives the same exact result as the 2D Plane Strain model.

Aniket may have some comments to add.

[psh1988]

Hello

Many thanks for your helpful answer... In fact, I do not know how to use Sweep+ inflation in my case, though i have watched all the videos and read Ansys tutorial. 

Could you please help me with this matter?

Bests

[psh1988]

Or Could you post your 3d meshed file here?

 

[Aniket Chavan]

check sweep mesh:

https://ansyshelp.ansys.com/account/Secured?returnurl=/Views/Secured/corp/v195/wb_msh/ds_sweep_method_option.html

Multizone method:

https://ansyshelp.ansys.com/account/Secured?returnurl=/Views/Secured/corp/v195/wb_msh/ds_multizone_method_option.html

-Aniket

Guidelines on the Student Community

How to access ANSYS help links

[peteroznewman]

Attached an ANSYS 2019 R3 archive of the 3D swept mesh.

[psh1988]

Thank you Peter,

I could finally achieve results from the3d model similar to 2d with your helpful ansys file...However, the crest displacement in 3d model (using different range of elements) is about 0.1 of that in 2d model while displacement in middle and toe of the (2d and 3d) slopes are more or less the same....What is your idea?Do you think that that through some changes in the meshing pattern around slope crest similar results can be achieved??  this file contains the results for 2d vs 3d cases. 

[psh1988]

I wonder  why i can not upload any files!!!  Could you please tell me what meshing pattern should i use to derive the same displacement at the slope crest for 2d and 3d  model? 

One sample answer:

2d:  number of elements: 2675     crest displacement(x):-0.39    middle displacement(x): -0.85          toe displacement(x): -0.51

3d:   number of elements: 3406    crest displacement(x):  -0.068      middle displacement(x):  -0.5   toe displacement(x):-0.5

[peteroznewman]

Here are instructions for uploading files.  Note that ANSYS staff are not permitted to open attachments.  You can get their help by putting up a detailed report of your model your results and your question.