Dear Yvan,
Thank you very muck for providing this clarifications!
As I told, thanks to Code_Saturne I want to simulate the atmospheric dispersion of a pollutant emitted from a point source, first in case of a flat terrain. For this,
- I'm using the version 5.0,
- I used the Graphical Interface (.xml file, please see attachment),
- I created a structured mesh (please also see attachment).
So, I would like to insert a point source of a passive scalar with an ejection speed, viscosity and Schmidt number that I must specify.
Thank you in advance!
Best regards,
Y. Oukali
point source of a passive scalar
Forum rules
Please read the forum usage recommendations before posting.
Please read the forum usage recommendations before posting.
-
Yvan Fournier
- Posts: 4251
- Joined: Mon Feb 20, 2012 3:25 pm
Re: point source of a passive scalar
Hello,
There is no "true" point source in Code_Saturne, only volume sources, but you simply need to select a small volume containing you point.
Be careful, selection criteria such as "box[]" select only cells whose center of gravity is in the indicated volume, so a better selection approach might be to use a user subroutine, and select (using "findpt") the cell whose center is closest to your given point. Otherwise, with the GUI, you can also create additional postprocessing meshes using a given cell selection, so you can use trial and error to select a single cell and visualize it, if you do not want to use user functions.
If the point does not move between computations, an alternative option might be to insert a small boundary portion in the mesh to handle this as an inlet (easier to control velocity and direction, as you otherwise probably need both a scalar source term and a velocity source term).
Best regards,
Yvan
There is no "true" point source in Code_Saturne, only volume sources, but you simply need to select a small volume containing you point.
Be careful, selection criteria such as "box[]" select only cells whose center of gravity is in the indicated volume, so a better selection approach might be to use a user subroutine, and select (using "findpt") the cell whose center is closest to your given point. Otherwise, with the GUI, you can also create additional postprocessing meshes using a given cell selection, so you can use trial and error to select a single cell and visualize it, if you do not want to use user functions.
If the point does not move between computations, an alternative option might be to insert a small boundary portion in the mesh to handle this as an inlet (easier to control velocity and direction, as you otherwise probably need both a scalar source term and a velocity source term).
Best regards,
Yvan