mojijoon |
March 12, 2014 07:47 |
convergent channel
3 Attachment(s)
I am trying to model unsteady flow of gas at low speed in a convergent channel with finite volume method, using unstructured and co-allocated grid, with SIMPLE algorithm. But the answers from my code are diverging and I don't know what causes the problem.
I only consider the flow in 1 direction. Is this a wrong assumption?
velocity does not change respect to so the grids become 2D, but when I want to calculate the area of faces I use area of 3d shape and for the normal vector I consider a 2D shape. Is it wrong? should I create a 3D grid regardless of velocity component?
|