CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   wall boundary condition (

woytex August 1, 2004 04:59

wall boundary condition
I am coding a 3D RANS/k-eps cfd code and testing 2D now. In the flow solver at the wall I set the normal stress tauxx=tauyy=0 and tauxy=tauyx=tauwall (tauwall is based on the wall low). Is it correct? Who knows, who has experience? Thanks. WX.

Jonas Bredberg August 19, 2004 07:11

Re: wall boundary condition
Hi there,

You may either implement the wall friction as a source term in the code as: tau_w*A, where tau_w=rho*u_tau*U*kappa/ln(9*y+) from the law-of-the-wall and A is the wall-area using this routine you should override the normal routines in the momentum equation

or through a modified eddy-viscosity as: mu_eff=rho*u_tau*y*kappa/ln(9*y+) and include that instead of the normal eddy-viscosity.

The normal stresses are very small at the wall as: tau_xx=mu*dU/dx tau_yy=mu*dV/dy for which both the velocity gradients and the (effective) viscosity are small.

Cheers Jonas

All times are GMT -4. The time now is 05:13.