CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   Question on moving mesh, mesh velocity is really small! (http://www.cfd-online.com/Forums/main/110579-question-moving-mesh-mesh-velocity-really-small.html)

ripperjack December 14, 2012 12:15

Question on moving mesh, mesh velocity is really small!
 
Hi guys,

I am coding a finite volume code to calculate the unsteady pressure for a moving airfoil. I used Roe scheme, and I am really confused with the moving mesh method.

I calculate the mesh moving velocity and it is really small (V_g=1m/s), and the flow velocity V is larger than 200m/s. And it seems that the mesh moving velocity has no influence on the solution.

As I understand, for Roe scheme, the cell surface flux for moving mesh is nearly the same as for stationary mesh, the only different is that the velocity V in eigenvalue for moving mesh is replaced by V_r. (V_r=V-V_g) . However, since V_g is very small, it seems that it has limited influence on the solution.

Can you guys who had the experience in moving mesh help me? Thanks!

orxan.shibli April 27, 2014 12:03

Not only eigenvalues but flux vector changes, too. Considering 3D flux in normal direction:

F_0 = \rho (u_n - u_b) \\

F_1 = \rho (u_n - u_b) u + p n_x \\

F_2 = \rho (u_n - u_b) v + p n_y \\

F_3 = \rho (u_n - u_b) w + p n_z \\

F_4 = (u_n - u_b) E + p u_n \\

levivad April 28, 2014 13:37

from a physics stand point, it makes sense that if the boundary moves slowly relative to the flow velocity it will be insignificant ( unless there's some special case). This is true until the geometry changes enough to impact the flow.
It might be considered quasi-static.


All times are GMT -4. The time now is 14:38.