CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CD-adapco (http://www.cfd-online.com/Forums/cd-adapco/)
-   -   interpolation scheme? (http://www.cfd-online.com/Forums/cd-adapco/54846-interpolation-scheme.html)

HADD September 6, 2005 04:39

interpolation scheme?
 
Good day

I'm very interested in what manner STAR-CD interpolates the pressure values at the cell faces

Any suggestions or references will be highly appreciated

Daria


4xF September 6, 2005 07:46

Re: interpolation scheme?
 
As far as I remember, 2nd order central differencing is used.

HADD September 14, 2005 10:34

Re: interpolation scheme?
 
Thanks

And for interpolating face velocity in continuity equation - a face momentum equation is used? Or I`m not right?


4xF September 15, 2005 08:06

Re: interpolation scheme?
 
Actually, there is no such thing. As far as I reckon (please anybody corrects me), the face velocities are not used in the continuity (or in this case the "pressure" equation) but rather the mass fluxes. These are evaluated in the same way as in the momentum equation in order to have a consistent discretization practice.

HADD September 15, 2005 08:37

Re: interpolation scheme?
 
As I understand, if mass fluxes in "pressure equation" is interpolated in usual way (linear interpolation of cell-centered velocities to the face), then so called checker-boarding of pressure may occur (in other words - oscillations)

4xF September 15, 2005 13:20

Re: interpolation scheme?
 
Of course, but there is a special procedure implemented (also known as Rhie & Chow correction) to get rid of this problem :)

HADD September 15, 2005 14:03

Re: interpolation scheme?
 
Thanks

But I wonder, if for interpolating mass fluxes Rhie and Chow procedure is used, why for interpolating pressure in momentum equation the 2nd order central differencing scheme (as you wrote) is implemented? What is in this case with checker-boarding ?


4xF September 16, 2005 07:13

Re: interpolation scheme?
 
I think there is some kind of misunderstanding here. In the mass conservation equation which degenerates in the case of incompressible flow into a Laplace equation on pressure, the pressure stencil uses second-order differencing. For consistency, the mass fluxes are evaluated in the same way they are evaluated in the momentum equations. Then, for avoiding the checkerboard patterns on pressure, you have an additional term based on the Rhie&Chow procedure. This is mandatory when you use a colocated arrangement of variables.

I hope this is clearer now. For more details, I would suggest to read the book by Ferziger & Peric: "Computational Methods for Fluid Dynamics".

HADD September 16, 2005 07:42

Re: interpolation scheme?
 
Thanks a lot!

I understand, that for Laplace equation the second-order differencing is used. But what scheme is used for pressure in momentum equation in STAR-CD?

Thank you for the book recommended!


4xF September 16, 2005 13:08

Re: interpolation scheme?
 
Gauss' theorem with pressure at the face centers, hence second-order interpolation as in the mass conservation equation (for consistency).


All times are GMT -4. The time now is 04:47.