Calculix has *soil?

1

As in the title, I wonder if Calculix could simulate *SOIL.

2

Here are the material models available in CalculiX:

image
image1283×627 186 KB

Unfortunately, there are no built-in material models for soils. You would have to create one using UMAT subroutine.

1 Like
3

Hi ,

Synt commented recently on the Prepomax forum about an old ccx distro with some (Mohr Coulomb & Drucker Prager) UMAT’s included. Check at the end of the post.

1 Like
4

This model is also mentioned in CalculiX’s documentation. It’s used as an example of UMAT usage.

1 Like
5

there’s many models for soil material, classified as clay, sandy, and rock or between. soil is highly nonlinear and inhomogeneity, for brittle type material many available trough linking with library MFront.

right, based on request of member Yahoo Group in 2015 i posted a source web links of Umat for Drucker Prager, Mohr Coulomb and Hoek Brown material models the work of Prof. J. Clausen from Aalborg University.

the code has been proposed to have a permissions to redistributed in CalculiX and it was done by Vittorio were in contact the authors directly. however some benchmark, verification and validation not yet finished. might it be the reasons to not officially supported and documented.

i’m only tested for Mohr Coulomb (Modified) to modeling concrete materials and it seems fast in convergences, also numerically stable inside CalculiX. cohesion values (Rankine, tension cut-off) can be lowest at half and more of Abaqus input which lead to begin a problems.

thanks for the hints, i’m not yet tested since it seems a new. also Mfront library given availability of Drucker Prager material including caps.

1 Like
6

Thank you all for the great discussions. I will check out the Umat for Drucker Prager, Mohr Coulomb.
I wanted to use *SOIL as there is a need to model the pore pressure (through *SOIL consolidation) during groundwater pumping. That requires C3D4P poroelastic elements or similar. However, I am not sure if the CCX UMAT could support this as well. The above worked well in Abaqus.

7

thanks to be more specific, no couple for hydro-mechanical problems. as i know CalculiX only supported thermo-mechanical coupling problems.

if you still look an opensource ones, you may goes to OpenGeoSys and it seems Mfront material library is supported also. GMSH or CGX can be use to generate model mesh .

OGS is more advanced and complex than Abaqus since it developed to capable to perform simulation of thermo-hydro-mechanical-chemical (THMC) processes in porous and fractured media

8

Hi all, I am responsible for porting Prof. Clausen UMAT for MC model in Calculix. Not a big merit, but the UMAT works for some classical problems. I tested it in bearing capacity problems but had not enough time to make a well organized report with the validation. The subroutine was compiled along with a 2.8 CCX release both under Ubuntu 16 and Win/Cygwin. As any UMAT it requires a recompilation of the program, so using MFRONT should be easier.

2 Likes
9

hi @vttrlc you’re welcome

due to lack of brittle material in CalculiX it could be large benefit when someone in their expertise maintenance this UMAT (Modified MC, Hoek Brown, Drucker Prager) by providing documentation and verification/validation.

best,

10

xyont, sorry for my late reply.
Just for point out that Modified MC, Hoek Brown, Drucker Prager, are only failure criteria, but the post-failure behaviour of most material models using these criteria is not brittle but perfectly plastic, without softening (some harde. Brittle material models usually (or even softening models) also require a solution procedure able to follow the decay of structural strength (e.g. displacement control).
BRG

11

i mean brittle material as categories nonlinear plasticity with limited tension capacity still can predict shear brittle failures. When tension softening being considered it’s another advanced ones and categories as damage/fracture.