CFD Online Discussion Forums

CFD Online Discussion Forums (
-   ANSYS (
-   -   How to print out coordinates of cells in Icem (

mrMunson July 13, 2009 12:06

How to print out coordinates of cells in Icem

I would like to print out the coordinates of the cell centers at the inlet selection on a mesh to a separate data file. Does anyone know a way to do this in icem cfd? The mesh is created in icem, but I got other software such as fluent available if it's necessary to export the mesh into another format.


Tushar@cfd July 17, 2009 12:09

No you will need to change format.
You cannot just import the *.uns, or *.blk of Icem in Fluent. You need to change the format, there's a small procedure for that, I guess you know these stuff..If not, mail me back.

Tushar Chourushi

Tushar@cfd July 17, 2009 12:12

Talking about the first question that you asked.
Well Icem don't have this facility, but it can be made possible manually, you need to do some more stuffs for that.... anyways I am afraid you will not get the results as such. If you do get some, just reply me back would like to hear new things.

Tushar Chourushi

obraun July 28, 2009 10:49

Well, in my experience ICEM is not really strong in exporting simple ASCII of whatever he might be meshing.
For structured mesh, I managed to get out some pure 2D mesh using the .geo export format as the most readable one that I quickly identified.
Once you have gone unstructured, it is a relative pain to reconstruct the mesh structure. But if you are unstruct, there is no other way anyway.
Anyway .. face centers?? What does ICEM know about what your solver considers to be a cell center (center of gravity or whatever). But if you know how, your triangles might help.
One way, transfer the mesh to CFX to make the dumbest of setups possible, just what you need to write a definition file. Open this in the post-processor and export the region as csv. You get a list of node coordinates and a connectivity table telling you which nodes make up the triangles.
Another way is to parse the cfx export format written by ICEM. The node coordinates are a first long list of coord triples. In the rest, the volume elements and faces are defined. Search for the name of your region and you get the indices of the nodes in the triplet lists that define your tris / quads. Up to you to know how the face center is defined from that. With MATLAB at your hand or other tools of your choice, you can get everything, but it will always be a good moment to spend with coding this, depends if you need it once or very often...

sundark28 May 3, 2012 08:53

groovyBC problem: unread condition

I read posts, which are related to groovyBC but i can still not understand why my boundary condition is not accepted by OpenFOAM. Here is my boundary condition on the velocity profile:

type groovyBC;
variables "rvc = 0.2*0.020; vmax = 5;
r = sqrt(pos().x,2)+pow(pos().y,2));teta = arctan(pos().y/(pos().x));
ux = (pow(r,2) < pow(rvc,2) && 0 < pow(r,2)) ? -vmax*r*sin(teta)/rvc:-vmax*rvc*sin(teta)/r;
uy = (pow(r,2) < pow(rvc,2) && 0 < pow(r,2)) ? vmax*r*cos(teta)/rvc:vmax*rvc*cos(teta)/r;";
valueExpression "vector(ux uy 5)";

OpenFoam returns this message:
"> FOAM Warning :
From function groovyBCFvPatchField<Type>::groovyBCFvPatchField(c onst fvPatch& p,const DimensionedField<Type, volMesh>& iF,const dictionary& dict)
in file groovyBCFvPatchField.C at line 119
No value defined for U on EINLASS therefore using 260{(0 0 0)}
Reading/calculating face flux field phi"

It seems that OpenFoam does not accept my boundary condition:confused: or does not read it. Can anyone give me some hints to solve this ?

Thanks in advance! :)


Darwin R.

All times are GMT -4. The time now is 21:22.