CFL-Number
I know CFX is an implicit solver nevertheless i wanted to know which best practice range for CFL number does exist for multiphase flows.
thanks Julienne |
There's no recommended range. For example, as you refine your mesh your CFL number would change, but in general you wouldn't need to adjust your timestep.
|
From theory we know, that an implicit solution is stable independent from the time step. I guess does not include source terms from reaction or multiphase. For that reason in that applications you may require smaller time steps.
|
To expand on the previous comments: Explicit solvers have a stability limit at a Courant number of 1. You simply can't go faster than that. But CFX is an implicit solver so this restriction does not apply (but a very common mistake is to forget this and run CFX with a Courant number <1).
So the timestep required for CFX should be determined with a sensitivity study on a case by case basis. The allowable timestep is often much bigger than 1.0, but exactly how big is OK depends on the simulation. |
All times are GMT -4. The time now is 01:13. |