CFDEM®coupling - User Forum

hi, when install the cfdem(openfoam2.1.0,git-liggghts), the follow error occured:

subModels/locateModel/engineSearch/engineSearch.C:64: error: no matching function for call to ‘Foam::meshSearch::meshSearch(const Foam::fvMesh&, Foam::Switch&)’
/home/zp/OpenFOAM/OpenFOAM-2.1.0/src/meshTools/lnInclude/meshSearch.H:180: note: candidates are: Foam::meshSearch::meshSearch(const Foam::polyMesh&, const Foam::treeBoundBox&, Foam::polyMesh::cellRepresentation)

Hi!

I would like to try cfdem in order to model geotechnical problems, such as triaxial shear tests or oedometer tests. I already tried to compile the software twice, but I always get error messages. Particularly, when I type "which liggghts" after install liggghts, nothing happens, when I try to compile src I get a message:

Make/linuxGccDPOpt/twoWayMPI.o
In file included from subModels/dataExchangeModel/twoWayMPI/twoWayMPI.C:30:0:
subModels/dataExchangeModel/twoWayMPI/twoWayMPI.H:47:62: fatal error: lammps.h: No such file or directory
compilation terminated.

Hi,

Is there a step-by-step tutorial for a simple CFD-DEM test case?

Best regards,
Jan

Dear all,

I currently do NOT recommend to run CFDEM with OF 2.1.x as there is an incompatibility with the engineSearch model.
This is gonna be fixed for future CFDEM versions.

see forum post for quick fix if you really want to use 2.1.x:
node/680#comment-2153

Cheers,
Chris

hi Alice,

from the installation instruction of CFDEM, OpenFoam 1.7.x (Source Pack) and Ligghts 1.5 (Source Pack) need to be compiled at first, then CFDEM can be compiled successfully.

I have installed Ubuntu 10.04 LTS, and I was wondering:

1. Can we just simply install the lastest Version of OpenFoam and LIGGGHTS from Deb or RPM PACK, and then compile the CFDEM?

2. If not possible, can we compile the source pack of OpenFoam 2.1 and LIGGGHTS 1.5, and then CFDEM 2.3.0; or CFDEM is only based on OpenFoam 1.7.x?

Hi,

I am trying to track submicron particles, where I recieve error of small particles sizes, useing fix pour . however, in creation of single atoms there is no problem as I define diameter rather than radius there.

my units are all cgs. is there a way to solve this problem?

Thanks,
Vajiheh

Hello everybody,

I tried to compile CFDEM from the instructionfile

system/files/setup_LIGGGHTS__OpenFoamR_CFDEM_2p0_on_Ubuntu1004_24052011.txt

At this point ...

#####################################
# compile src
#####################################

#- change the user-1.7.1/src/lagrangian/cfdemParticle path:
cd $WM_PROJECT_INST_DIR/$USER-$WM_PROJECT_VERSION/src/lagrangian/cfdemParticle

#- to compile the src type:
rmdepall
wclean
wmake libso

... I get a compilation error:

Hi All,

I am running a coupled CFDEM case, with time steps of around 10e-10 , and coupling of every 10 steps. the simultion stops half way through with below error, I appreciate any help.

Thanks
ime = 6.554e-05

Courant Number mean: 2.82015e-05 max: 0.0036778
- evolve()
Starting up LIGGGHTS
Executing command: 'run 10 '
run 10 Setting up run ...
Memory usage per processor = 15.5843 Mbytes
Step Atoms KinEng 1 Volume
655391 1 0 1.3533016e-11 5.4e-09
655401 1 0 1.3533016e-11 5.4e-09

Thanks to all of the developers of CFDEM. I have been obtaining some encouraging results- http://youtu.be/mQN9WmHIW2c . One thing that was bothering me was dealing with pizza.py. It works well, but if it generates output at each coupling step, the dump file becomes enormous. I have attached two header files that output particle positions, radii, and velocities directly to Paraview readable file directories. Each folder has the same time stamp as the OpenFoam files. Thus, both pizza.py and foamToVTK become unnecessary.