CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   - to make them fall (

laura_c April 5, 2013 03:03 to make them fall
Hi guys, I am modelling natural convection on Fluent. It is a simple box with 2 inlet holes and one outlet. I am using an automatic mesh with inflation on the edges of the holes and on the heater which is providing heat to create flow...

the solution is not converging, but the surface integral is constant - facet temperature (outlet). the residuals are stabilized but high.

I am trying to change the under-relaxation values, but I think it will take ages to converge... does anyone know what other paths are available to converge the solution?

ztdep April 12, 2013 09:22

for natural convection problem in open system, have you add fictious domain for your open holes. the computational domain should be larger than the physical size of the problem.
could you please attach your grid system.

laura_c April 12, 2013 12:39

What do you mean by fictios domain? I am modelling both inlets and outlets as pressure inlets/ outlets as i have no info on the flow rates taking place..... i will be attachinv the mesh as soon as i transfer it....

All times are GMT -4. The time now is 13:49.