CFD Online Discussion Forums

CFD Online Discussion Forums (
-   OpenFOAM Running, Solving & CFD (
-   -   Coupled solver (

weiss April 16, 2005 14:13

Hello everybody, I would li
Hello everybody,

I would like to couple an OpenFOAM-solver to a particle solver and so need to extract some informations from the field/mesh.

1. How can I get the coordinates of a field-value or vice versa? (field[x] <-> mesh[x]: do they correspond? Any more sophisticated way?)

2. How can I create a field that includes some specific mesh/field-points, referring to the original field ((maybe by adding them pointwise))?

I'd be very thankful for any suggestions.


mattijs April 18, 2005 12:26

1. fvMesh::C() is a volVectorF
1. fvMesh::C() is a volVectorField of the cell centres. So field[cellI] corresponds to the value at mesh.C()[cellI]

2. Don't know about this. You'll have to do it yourself (guess with labelList for selected cell labels + reference to original field)

eugene April 18, 2005 12:35

You realise foam has a built-i
You realise foam has a built-in particle tracking layer? IMO the best thing to do would be to track the particles on the foam mesh, which automatically keeps track of cell locations and boundary collisions. If you simply have to calculate the particle trajectories with an external app, then you can map this back onto the foam particles before updating your tracking. You can also use the "interpolate" functions to do second order mapping of field values onto the particles.

What exactly does your particle solver do?

weiss April 19, 2005 12:30

Thanks. I think the particl

I think the particle tracking layer wouldn't help, hence the coupled solvers should exchange their fluxes.

Selecting the cells by hand seems complicated for complex geometries. Maybe there is the possibility to define something like a region in the blockmeshdict-file, so that I can select the interesting area at once?

mattijs April 19, 2005 13:17

Unfortunately there is no such
Unfortunately there is no such possibility in blockMesh. Would be quite handy though.

All times are GMT -4. The time now is 11:28.