[nico.frisone]

Hi guys,
i'm studying with Ansys fluent (16.0 version) 4 different cases for driven cavity flow, the geometry is a square of 1.75m x 1.75m with an inlet and outlet of 0.5 m. i have to study:
Re=10
Re=500
Re=10000
As Laminar flow,
and the fourth one is with Re=10000 using the k-epsilon model for the turbulence.
In order to have a significant velocity inlet for the Re=10 case, i chose as a fluid the engine-oil already present in the fluent database.
I used a quadrilateral mesh first, but it gave me a reynolds cell number > 2 for the k-epsilon case, then i refined it along the walls using the bias factor and the problem disappeared.
All the solutions goes to convergence using residuals with a order of magnitude of 10^-6, except for the Re=10000 Laminar case.
Here the residuals don't diverge but they fluctuate around 10^-1 for continuity and 10^-3 for the velocity. I tried to decrease the urfs for pressure and momentum but it doesn't work,i tried to use a discretization at the first order for momentum, i also tried to refine the mesh again but it doesn't work, i don't understand how to solve this problem so i hope somebody can help me with this.