CFDEM®coupling - User Forum

hi, all-

Sorry if this is a stupid question. I've run tutorial cases like ErgunTestMPI in cfdemSolverPiso and twoSpheresGlowinskiMPI in cfdemSolverIB. Both of them can be run in parallel (like 4 cores or 2 cores) with no problem.

However, when I tried to run them in only one core (in serial), I got errors as below:

--> FOAM FATAL ERROR:
bool IPstream::init(int& argc, char**& argv) : attempt to run parallel on 1 processor

From function UPstream::init(int& argc, char**& argv)
in file UPstream.C at line 79.

I am trying to simulate a pneumatic conveying problem but the particles are not moving with the air, the simulation got crashed very quickly, I am getting the following error:
I have tried with changing the density or particles, the velocity of air in OpenFOAM, if anyone can help me, I'll be really thankful, I have attached the LIGGGHT file and couplingproperties and controlDict files.

Hi CFDEM members,

The log_file created after running a case is sometimes really large in size (e.g., 19Gb in my case), this might cause a problem to the memory capacity of computer. My question is how to manage this log_file effectively. Is there any possibility to change this file, for example, control the interval time to write log_file ? (e.g., write in every 50 time steps).

Thank you,

Regards,

Hi, i want to simulate the cylinder-shape particle settlement, but the use of multi-sphere method is not included in Public version.
Is there any method to adjust the shape effect of perfect sphere particle in order to simulate non-spherical (like cylinder shape) particle settlement? or any useful document related to shape-effect?

(The total number of particle is the order of million, so the multi-sphere method may not proper to my case...)

Thanks!

bc

Im trying to insert particle into a cylindrical container and apply axial load on it over the top plate. But it is showing error
ERROR: Fix wall/gran (id geometry): More than one wall of type 'mesh' is not supported (/build/buildd/liggghts-3.0.3+repack/src/fix_wall_gran.cpp:505)

# Shear Cell example
### Initialization
# Preliminaries
units si
atom_style sphere
boundary f f f
newton off
communicate single vel yes
#processors 2 2 2

Hello,
I started to experiment wit the cfdemSolverPisoSTM.
At the DEM-part, I can find the lines
# species transfer
fix specCap all property/global speciesCCapacity peratomtype 1.0
fix cfd4 all couple/cfd/speciesConvection speciesName speciesC species0 1.0
But I don't understand what they do and how to adapt them, when I use more then one atom-typ.

Hi,

I got this compilation error when executing: cfdemCompCFDEM: I used additionalLibs_3.0.x for locating turbulence models
and my OpenFOAM version is OF 3.0.1.
Please help.

Hello everyone,

I wish to make a probability distribution function of particle Reynolds number for all the particles in the simulation.

What is the best possible way to achieve this on the go? It appears there is no 'dump'-like LIGGGHTS feature in CFDEMCoupling.
Sadly the fluid velocity is not accessible on the LIGGGHTS side.
More general question would be, how to dump any similar data from CFD side to a file?

Thanks and kind regards,
Vinay

Hi, All:

I am running a CFD-DEM coupled simulation using CFDEM, the number of particles is large while the fluid domain is quite simple and small, so I want to use many cores on the DEM side (e.g. processors 5 5 20) and few process on the CFD OpenFOAM side (total domains 4, using 2 2 1), is that possible to be done using the current CFDEM coupling? Does the DEM has to be the same domain decomposition as OpenFOAM?

Thanks for any suggestions!

Roy

Hi,

I'm using cfdemSolverPisoScalar for solving the convection equation between particles and air. I'd like to know what equation is being implemented by this solver. In the documentation it was mentioned that the equation was implemented according to Nield & Bejan, Convection in Porous Media.

Can anyone direct me of the exact way it was implemented? I want to know how the heat transfer is being calculated from the air to particles. Is it based on the Li and Mason model or something else?

Can anyone help me understanding it.