Learning Forum

[Shravan Bharadwaj]

Hello ! I am simulating an air bearing model in Fluent. I have a pressure inlet of 5 bar and atmospheric pressure outlet. I am using Laminar flow under steady conditions (since this is the assumption made in the books). I am getting divergence in the continuity equation and reversed flow. What to do ?

 

Ultimately I am computing the Surface integral of pressure to get load capacity.

[NickFL]

Here are some things to think about: 

1. The pressure-pressure boundary condition is mathematically challenging. Generally a more stable set of boundary conditions is a pressure and mass flow condition. One approach would be apply a given mass flow condition for the inlet. You can then run the simulation, see what the pressure drop is, then readjust the mass flow, and then repeat until you get the desired pressure drop. If you wanted to go further, you could then switch back to the pressure-pressure boundary conditions using the solved solution as your initial condition. Starting a pressure-pressure simulation with a poor initial condition will nearly always diverge as so.
2. The thin 20 µm region will likely need a large number of cells to accurately compute. I don't know what your dimension in radial is, but the cells here could be very stretched. I also don't know what the element count limit on Fluent Student is, but I would recommend that you start with a 2D axisymmetric problem to see what the required mesh is before moving to the three-dimensional problem.

[CFD_Friend]

Hi Shravan,

It all comes down to the boundary condition used and the mesh quality. Make sure you have a good-quality mesh in the first place. Please refine the mesh near the zones of interest.