Hi All,
I have used "twoWayMPI" as a data exchange model, and now I try to use "oneWayVTK" model. So I changed the 'couplingProperties' file as follows:
//for twoWayMPI//
dataExchangeModel twoWayMPI;
twoWayMPIProps
{
maxNumberOfParticles 10100;
liggghtsPath "../DEM/in.liggghts_bonded";
}
//change these for oneWayVTK//
dataExchangeModel oneWayVTK;
oneWayVTKProps
{
DEMts 0.001;
relativePath "../DEM/post";
couplingFilename "pre%4.4d.vtk";
maxNumberOfParticles 30000;
}
It works until data saving time step; it means if I set the time interval for CFD calculation as 0.001 s and for CFD data saving as 0.01 s, the simulation ran well until 0.009 s but it had an error at t = 0.01s. Do I have to change other things besides changing the codes in couplingproperties? The error on terminal was as below:
Time = 0.01
Courant Number mean: 0.00448055 max: 0.180375
timeStepFraction() = 0.01
createTimeDir(path_), path="./lagrangian"
lagPath_=createTimeDir(fileName(lagPath_/lagrangian)), lagPath="./lagrangian"
[0] #0[1] #0 Foam::error::printStack(Foam::Ostream&)Foam::error::printStack(Foam::Ostream&) at ??:?
[1] #1 Foam::sigSegv::sigHandler(int) at ??:?
[0] #1 Foam::sigSegv::sigHandler(int) at ??:?
[0] #2 ? at ??:?
[1] #2 ? in "/lib/x86_64-linux-gnu/li in "/lib/x86_64-linux-bpthread.so.0"
gnu/libpthread.so.0"
[0] #3 Foam::basicIO::dumpDEMdata() const[1] #3 Foam::basicIO::dumpDEMdata() const at ??:?
[0] #4 Foam::cfdemCloud::evolve(Foam::GeometricField&, Foam::GeometricField, Foam::fvPatchField, Foam::volMesh>&, Foam::GeometricField, Foam::fvPatchField, Foam::volMesh>&) at ??:?
[1] #4 Foam::cfdemCloud::evolve(Foam::GeometricField&, Foam::GeometricField, Foam::fvPatchField, Foam::volMesh>&, Foam::GeometricField, Foam::fvPatchField, Foam::volMesh>&) at ??:?
[0] #5 at ??:?
[1] #5 ?? at ??:?
[0] #6 __libc_start_main at ??:?
[1] #6 __libc_start_main in "/lib/x86_64-linux-gnu/libc.so.6"
[0] #7 in "/lib/x86_64-linux-gnu/libc.so.6"
[1] #7 ?? at ??:?
at ??:?
paul | Wed, 07/11/2018 - 09:35
set IOModel to off.
set IOModel to off.
Soo-Min Ham | Fri, 07/13/2018 - 03:12
Thanks, the code runs!
Thanks, the code runs!
medvedeg | Thu, 07/12/2018 - 11:38
Hi,
Hi,
I am afraid oneWayVTK may not work and it is no more supported by us. Please use twoWayMPI.
Soo-Min Ham | Fri, 07/13/2018 - 03:16
Hi,
Hi,
Can I ask why oneWayVTK does not works? Then, how can I set the coupling condition to run a case with a fixed DEM?
Thanks,
Soo-Min
alice | Tue, 07/17/2018 - 08:19
Hi Soo-Min,
Hi Soo-Min,
oneWayVTK stems from the first days of the coupling and is very slow and inefficient in comparison to mpi coupling. You can fix the DEM results be removing the integrator (fix nve/sphere), this will cause the freezing of the particles on the DEM side (using a fix freeze command is another option, but if you want to freeze the entire DEM side, the removal of the integrator will be fine as well).
Cheers,
Alice