I am trying to run a study in which a liquid flows through a very narrow annulus. 

I tried to set the mesh so that i would have a minimum of 3 cells along the thickness of the annulus, but that happens only at certain locations, while the number of cells increases needlessly. I maxed out the capacity of the computer and run the analysis for days but the solution never converged and the results were extremely different than the ones collected during the actual tests.

As for the Boundary Conditions, I tried Pressure at the inlet and the outlet, or Pressure at the inlet and "wished" flow rate at the outlet. None of these trials yielded any close results to the test data.

I also tried to analyze it as an equivalent flat configuration by projecting the circular surfaces onto plane, but i ran into the problem of setting the boundary conditions where the circle was split.

Does anyone know what will be the best approach in this type of problem?