CFD Online Discussion Forums

CFD Online Discussion Forums (
-   FLUENT (
-   -   supersonic solver settings (

Scott March 25, 2004 16:55

supersonic solver settings
Hello, I am a beginner with fluent and am trying to simulate flows at high mach numbers. I am having a bit of trouble finding the correct solver settings for this type of flow. My main question involves whether to use the steady solver or the unsteady solver. I have read that in the early days of cfd, high mach number flows with detached normal shocks were easiest to solve when modeled as unsteady flows due to the nature of the PDE's involved (hyperbolic vs. elliptic). The fluent manual makes no mention of this, so I am curious as to whether it is necessary. Also, how does one go about determining the Courant number? Thanks in advance


Jane March 25, 2004 17:55

Re: supersonic solver settings
As for Courant number, my experience is you have to try to see whether it converges or not. I know very limited about this, but I think for supersonic flow, you should use implicit solver (although I am not exactly sure what does this mean, my guess is momentum equations are coupled with energy equation since it's compressible flow.) That's all I know.

Nandu March 26, 2004 03:05

Re: supersonic solver settings
Detached shocks tend to be problematic when you use a steady state approach to solving the problem. strong shocks are better solved using the coupled explicit, unsteady solver in Fluent. the courant number is dependent on the speed of the flow and on the smallest grid size in the domain. you'd be better off simply setting the courant number to a value of say 0.8, running the coupled, explicit, undteady solver until the residuals converge or they flatten out.

if you adapt the mesh to improve the resolution of the flowfield, use the coupled steady solver to accelerate the solution, as the initial guess i good.

"Importantly use the coupled solver". the continuity and momentum equations are solved simultaneously and the shock resolution is singnificantly better than when using the degregated solver. the resons are pretty self explanatory as in the coupled solver there is no need have an interpolatory scheme (such as SIMPLE) to provide the pressure-velocity coupling.

cheers nandu

All times are GMT -4. The time now is 21:26.