[pzoum974]

Hello,
I am modelling transient cooling in a small fuel tank (20mmx20mm) with adiabatic vertical walls and the horizontal plates are at the same temperature (290K). I am in unsteady state condition and the initial temperature is 319K. I am using a fuel (JetA) and for the properties of the fluid are polynomial profiles in the materials menu.

I set the residuals for energy at 10^-10, x-velocity 10^-4, y-velocity 10^-4. The simulation converges well.

The mesh contains 490,000 cells and is very fine next to the horizontal edges. I first used a mesh with 135,000 cells and the mesh with 490,000 gives me "better" results.

I have to run the simulation until 800s to see the natural convection in it. I have experimental results but the results from my simulation do not match them.

If you have any suggestions, could you help me?

Thanks a lot[IMG]file:///C:/Users/Julien/Documents/Research%20project/Rapport/Finest%20mesh.jpg[/IMG]