CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Main CFD Forum (https://www.cfd-online.com/Forums/main/)
-   -   Implementation of the Low-Reynolds-number Chien model (https://www.cfd-online.com/Forums/main/74118-implementation-low-reynolds-number-chien-model.html)

 om3ro March 24, 2010 19:16

Implementation of the Low-Reynolds-number Chien model

I'm trying to implement the Chien model with OpenFoam, but I have some general doubts on the description of y+.

The Chien model uses y+ for the definition of the damping factors.

Suppose to describe a 2D duct flow in with the following configuration:

HTML Code:

```<pre>     /\  y       |       |       |       ------------>  x ____________________________________________ ->                                                                    -> ->                                                                    -> ->                                                    *            -> ->_________            _ ,    °  .________,*________->                   |        |  |        |                                  |____ _|  |___ __| </pre>```
y+ is defined as: y+ = (U_t * y) / nu

and U_t(x) = sqrt(tau_w(x) / rho)

y+, U_t, tau_w are functions of x!!!

It is necessary to define for each point (x,y) of the numerical domain the x coordinate in which define y+, U_t and tau_w.

An example:

In both points identified with " * ", tau_w must be evaluated in " , ".
But in the point " ° ", shall tau_w be associated with " , " or with " . " ?

Does anybody have experience on this aspect? What is the principle to adopt in this task?

Marco

 All times are GMT -4. The time now is 10:05.