CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Main CFD Forum (https://www.cfd-online.com/Forums/main/)
-   -   Courant number in 2D and 3D (https://www.cfd-online.com/Forums/main/69526-courant-number-2d-3d.html)

harbinyg October 26, 2009 09:23

Courant number in 2D and 3D
 
Hi.

Given the velocity as u, v and w separately for the x , y ,z direction.

How to define the Courant number in 2D and 3D?

Thank you so much.

harishg October 26, 2009 18:39

You can use c0={sqrt[u^2+v^2+w^2]+c}dt/dx < 1 (assuming dx=dy=dz). If they are not equal you can try using dx=sqrt(dx1^2+dx2^2+dx3^2). For some schemes used for incompressible flows you might need c0 < 0.5.

NewtonKF October 27, 2009 03:48

Hi folks... It is known that CFL criterium is linked to convective dominant problems. Is there another criterium when considering diffusive dominant or balanced diffusive-convective problems???

thanks...

harishg October 27, 2009 12:39

For diffusion you can use \mu * dt / dx^2 =0.5.

NewtonKF October 27, 2009 12:45

Harishg, thanks for you reply... Can you indicate some papers that use this criterium???

thanks again...

harishg October 27, 2009 13:18

The criterions are usually derived by using Von Neumann stability analysis. The use of the above criterion covers most of the methods. You can find detailed information on the stability criterion in the book of Hirsch.


All times are GMT -4. The time now is 15:25.