Dear CFDEM developers,
I am new to the CFD-DEM coupling, and doing a simulation of fluid flush on big particles. The big particles are modeled with overlapping spheres method in LIGGGHTS. The problem for the CFD-DEM coupling is: due to the overlapping volume of spheres, the sum volume of all spheres is about 200% - 300% of original particles, therefore in CFD model the void fraction of fluid is extremely low - e.g. 0.1, and the simulation blows up in few seconds.
So I would like to ask, is it possible to exclude some certain part of spheres so that the solid particles' volume is scaled correctly for the void fraction calculation in CFD?
what I did is to make a group of spheres whose sum volume is more or less equal to the sum volume of real particles, and therefore fluid drag not acting on all of spheres.
----------------------------------------------------------------------------
group cfdem_spheres id 3 8 19 20 24 28 35 .................
fix cfd1 cfdem_spheres couple/cfd couple_every 100 mpi
fix cfd2 cfdem_spheres couple/cfd/force
fix integr all nve/sphere
---------------------------------------------------------------------------
However, it failed because the void fraction of fluid remains the same. I wonder the cfdem scheme still takes all spheres into account for the void fraction calculation.
So how to figure this out? modify the source code, or still possible to use the existing features to solve this issue?
Any tips and recommendation are highly appreciated.
Best,
Joran