wall heat flux 2D to 3D
Dear all,
I'm doing unsteady 2D computations of a heat transfer from a wall to my fluid, with a imposed heat flux. It is natural convection, with fluid density varying with temperature. I got relevant wall temperature (in comparison to theory). In order to pursue my study, I tried a 3D case with the same numerical settings (cells size, controls, schemes). However, the temperature is here meaningless (10 times higher and osciallting a lot). The residuals are not as good as the 2D case, but it still seems to converge. Of course the heat flux is the same in both cases. I computed the total power on the wall surface in Fluent, which is 120W. Do you have any idea for the reason of the bad result with the 3D case ? If the mesh is enough fine in 2D, it should be ok for 3D ! Best, Mat 
Is your 2D case axisymmetric?

No, its not 2D axisymmetric.
There is a plane of symmetry in the real geometry, and the 2D case is computed in this plane. 
I'm really stuck with this problem.
Why I get different temperatures for the 2D and 3D cases, with the same power ? (160W) Knowing that for the 3D case, the temperatures are meaningless (really too high). Thanks in advance for your help. 
All times are GMT 4. The time now is 18:08. 