- **Main CFD Forum**
(*http://www.cfd-online.com/Forums/main/*)

- - **interpolation on faces and wall boundaries
**
(*http://www.cfd-online.com/Forums/main/15661-interpolation-faces-wall-boundaries.html*)

interpolation on faces and wall boundaries
I'm afraid that this is a very basic question as I'm pretty new to CFD and the following is a first principles argument..
I wish to model an incompressible fluid on an arbitrary staggered 3D hexahedral mesh. 3 layers of the u-component cell are shown below. s is a scalar variable (T, turbulent KE & dissipation etc). s_B, s_N, s_S, S_T are the interpolated scalars on the faces. .................................. w(i,j,k+1).....s_T.....w(i+1,j,k+1) .................................. -------â€"------------ v(i,j,k)......s_N......v(i+1,j,k) s(i,j,k)....u(i,j,k)...s(i+1,j,k) v(i,j-1,k)....s_S......v(i+1,j-1,k) -------------------- .................................. w(i,j,k-1).....s_B.....w(i+1,j,k-1) .................................. -------â€"------------ Bearing in mind that the mesh spacing is not uniform, what is the best way of interpolating the scalars? 1. Bilinear interpolation? 2. Area weighted interpolation of the interpolated values at the edges? (since we have displaced the mesh) 3. Average the interpolated values at the edges? __________________________________________________ ______ The second question is how best to incorporate wall functions into the finite volume difference equations. For example there is a wall boundary on the bottom face. I wish to incorporate moving walls and so the wall velocity may not be zero. Set a von Neumann boundary on the bottom (i.e. a_B =0) and set momentum source terms? |

All times are GMT -4. The time now is 18:05. |