CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Natural convection (http://www.cfd-online.com/Forums/fluent/30147-natural-convection.html)

JF August 8, 2002 09:11

Natural convection
 
Hi,

I need to simulate the flow due to the presence of a cold wall in a room. I have tried to simulate it but it doesn't converge. I have improved my mesh but it doesn't work.

Has anybody an advice on the question ?

Thanks a lot.

JF


david August 8, 2002 19:12

Re: Natural convection
 
Hello there, What kind of relation (density vs temp) did you use? It should converge (even for bigger time step), however, the motion of the airflow would be very slow if density diff is the main driving force in the room.

DC

Jin-Wook LEE August 8, 2002 20:27

Re: Natural convection
 
For natural convection, you should be careful that,

1) enough grid should be embeded in the boundary layer

2) in general, unsteady calculation is needed for high Rayleigh number

Sincerely, Jinwook


JF August 10, 2002 04:34

Re: Natural convection
 
Hello,

Thank you for your answer. For the density, I use the Boussinesq relation(Define in material). I'm also not very sure of my viscous model, I use the K-eps Standard with an Enhancend Wall Treatement so I can simulate the Thermal Effect end the Buoyancy Effects.

JF

sri September 5, 2002 12:48

Re: Natural convection
 
Hi JF,

Body-force weighted scheme should help you.

sri


All times are GMT -4. The time now is 05:44.