Guidelines for “reasonable” 2d simulations

Hi,
Though downscaling from 3d to 2d cannot be performed properly in CFDEM simulations, as some of the feature still consider the atom type to be sphere rather than a disk, there are times which one just wants to check whether a concept would work or not, and an initial 2d simulation would be sufficient. I’ve checked the following link on steps to take for performing 2d simulations
http://www.cfdem.com/media/DEM/docu/Section_howto.html

In short:
- Use the dimension command to specify a 2d simulation.
- Make the simulation box periodic in z via the boundary command
- In using the create box command to define a simulation box, set the z dimensions narrow.
- Use the fix enforce2d command as the last defined fix to insure that the z-components of velocities and forces are zeroed out every timestep.

I’ve applied these modifications to the settlingTestMPI:
In the ligggths input script:
//-------------------------------------------------------//
…
atom_style granular
dimension 2
boundary f f p
region reg block 0 0.1 0 0.1 -0.001 0.001 units box
…
fix zwalls1 all wall/gran model hertz tangential history primitive type 1 zplane -0.001
fix zwalls2 all wall/gran model hertz tangential history primitive type 1 zplane 0.001
create_atoms 1 single 0.05 0.025 0 units box
…
fix integr all nve/sphere
fix 4 all enforce2d

//------------------------------------------------------//

Also adjusted the blockMesh by narrowing the z-dir length, 1 cell in z-dir, and empty-type BC for front and back faces. Also adjusted the 0 folder for the fields. Attached figure is what I get which is completely off.

On the following webpages of CFDEM:
http://www.cfdem.com/featured-work,
the developers show an example of a 2D simulation “Coupling ParScale®, LIGGGHTS® and CFDEM®coupling: 2D fluidized Bed with detailed temperature modelling inside the particles using ParScale". So this shows that it is possible to carryout reasonable 2d simulations.

So I would appreciate it some suggestions as what is it that I'm missing here?

Thank you.
Ali