Simulation of Rayleigh-Benard convection
I have decided to simulate the Rayleigh-Benard convection.I have never simulated this phenomena with STAR-CCM before. But its the programm, I know most about. After two days, working on the simulation of the natural convection to get the Rayleigh-Benard convection (I tried really everything), I realised that I'm not able to cope with this problem in CCM+. (The CFD should be done in 3D.)
I asked a friend, if I could use his ANSYS fluent to essay the simulation of the Rayleigh-Benard convection phenomena. After 30 minutes (I have never worked with fluent before), I got the phenomena. Clear visible!
Now I'm thinking about: What was wrong with the settings in CCM+? To cope with this I'm trying to do the connection between the different namings in the programms. And that's why I post it in the Main-Forum.
In Fluent I used these "General" options:
Type: Pressure-Based [?]
Time: Transient [in CCM+ I used "implict unsteady" for this.]
In Fluent I used these "Solution Methods":
-Pressure-Velocity-Coupling [probably the "Segregated Flow Model" in CCM+]
Scheme: SIMPLE [is standard in the seg. algo.]
Gradient: LEAST SQUARES CELL BASED [?]
Pressure: BODY FORCE WEIGHTED ["gravity"?]
Momentum: FIRST ORDER UPWIND [?]
Energy: FIRST ORDER UPWIND [?]
I know that there is an option of Buayancy Driven (Xu) in CCM+. This option is able to be used if the k-epsilon model is invoked. (By the way the k-epsilon and k-w are recommended in the userguide to simulate the natural convection). Spalart-Almeras is not allowed to be used for this kind of phenomena, according the userguide.
Well, could anyone help me with the "translation" to the STAR-CCM+ settings?
Or, what are generally the proper settings for the simulation of the Rayleigh-Benard convection, natural convection and stack effect?
Thanks a lot for your help!!
PS: Does anyone have some papers about the Rayleigh-Benard convection and how to simulate it to get all the important characteristics of the phenomena?
I never tried with star-CCM+ but I did it with STAR-CD and I remember that it was not straightforward despite this type of flow is easy to catch with any simple CFD code.
|All times are GMT -4. The time now is 08:40.|