[achat]

Hello:

I am using ANSYS FLUENT v14.0.

i am solving a steady-state, fiber coating problem in a fairly complex geometry with surface reactions. The equations solved are momentum, energy and a few species concentrations. For the low diffusivity cases (low operating tempeartures) I seem to get converged solutions (residuals are very low and plateau out). But when I restart, the converged solution changes: the residuals spike and then they plateau out! This does not happen when diffusivities are high (high T).

Any ideas/suggestions as to why the solution gets 'trapped' in an artificial converged state for the low diffusivity cases? I am using 'coupled' pressure-velocity coupling with second-order upwind.

Thanks
athan