CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   BC in k-eps solver (

nileshjrane April 6, 2010 11:06

BC in k-eps solver
Hi folks..
I am using a research code for my masters thesis. Its an k-eps(jones-launder high Re model) multiple species chemical reactive code code which was written for something else and i am trying to work in for my case. I want to apply viscous wall BC in the code. I know that the boundary conditions basically should be like this:
@ boundary u,v,w,k,eps all equal to zero.
now i can easily set this by making the values at ghost cells just negative of whatever the value at the cell next to wall ( this results in zero at wall by extrapolation) but what about the normal temp, vibrational temp, energy, density, pressure, vibrational energy,and the species concentrations variables??? how to specify the BC for them?? and what conditions should be specified on the flux values for these variables??
Can anyone help???

ztdep April 7, 2010 01:00

finite volume method or fdm. by the way, at the solid wall ,now all these quantitied equal to 0

nileshjrane April 7, 2010 04:48


Originally Posted by ztdep (Post 253470)
finite volume method or fdm. by the way, at the solid wall ,now all these quantitied equal to 0


Say the wall is j=2 plane. now there is one ghost cell j=1. So for u,v,w,k,eps things are easy.

u(k,1,i) = -u(k,3,i) so that resultant at (k,2,i) would be zero..and so on for v, w, k, eps.

Now temperature i can give isothermal conditions. But i have no idea about vibrational temperature. There must be some formula right?? and what about energy variables e and ev (vibrational energy)?? How to specify condition on that???

Also what about the fluxes du, dv etc?? can somebody tell me how to put conditions on this??

ztdep April 8, 2010 08:46

Dear friends:
My experience is that I assemble the matrix only with the inner grid points. We have known the boudnary grid points value for the first kind boundary condition. for the second and third kind bc, i implemented it with the "additional source term method". all the flux was trasformed into the source term of the first inner control volume.
Hope this help.

All times are GMT -4. The time now is 08:10.