CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   Problem in Natual Convection (

Cyrus69 June 11, 2018 04:42

Problem in Natual Convection
1 Attachment(s)
Hi all
I'm simulating a natural convection problem in fluent.this case hase been simulated in comsol but my homework is to do it in fluent.the problem is simulation of convection & conduction in a vacuum flask.flask contains of 90c coffee that's insulated by foam and covered by steel.flask is in ambient temp(25c).purpose is to model natural convecion in 10hours.i calculated rayleigh number and that's about 10^7(laminar).
i created my geometry in GAMBIT.meshed it fineand created zones for fluids & solids.then I imporeted mesh in fluent with this settings:

Model Settings
Space 2D
Time Unsteady, 1st-Order Implicit
Viscous Laminar
Heat Transfer Enabled
Gravity Enabled
Solidification and Melting Disabled
Radiation None
Species Disabled
Coupled Dispersed Phase Disabled

now I have some problems.i wanted to set the initial coffee zone temp as 90c and air zone pressure as 1atm so i patched those faces,but when i ran calculation and see temp countor i see every where is 90c.i've tried to adapt regoins but my geomtry isn't simple(maybe I sould make smaller simple regions).I'm realy stucked!
Attachment 63921
my second problem is in boundary conditions.I've set all of them as "wall".when i import it in fluent i can just coupled thermal condition and I cant assign H.T.C and temp.

rsankar1069 June 11, 2018 06:36

Problem in Natural Convection
Hello Cyrus69,

The simulation seems interesting. I have good experience working on Conjugate Heat Transfer simulations. If interested, please drop me a mail at

Cyrus69 June 11, 2018 09:02


Originally Posted by rsankar1069 (Post 695454)
Hello Cyrus69,

The simulation seems interesting. I have good experience working on Conjugate Heat Transfer simulations. If interested, please drop me a mail at

i sent u geometry and case file.

Cyrus69 June 11, 2018 15:33

i encounter another problem.i've used boussinesq for density but after calculation i don't see any density diffrence in it!

rsankar1069 June 12, 2018 01:51

Problem in Natural Convection
Hello Cyrus :),

1) Looking at the case files, I see that you have not defined an interface connection between various domains. You will have to define interface connections between two different domains in contact in order for the heat dissipated from one body to transfer to the other body.

2) It would also be better to specify the operating density under the operating conditions tab close to the density value of air at ambient conditions. Buoyant forces are driven through the difference in densities calculated with operating density as reference.

Please let me know if the case runs fine with these changes made.

Cyrus69 June 13, 2018 04:46

3 Attachment(s)
thanks dear sankara for ur tips.
i've made some changes.i've a created a fluid zone with water properties for coffee zone and i created interface bc as wall,so fluent made coupled Bc automaticaly.
i made boussinesq density for air with 298k operating temp and 1.225 operating density.i set interface between air and coffee as wall!(i know its not correct but it gives me reasonable result.attached results).
ididnt find any answer for density problem.velocity countor is good but shows 0.5 m/s speed at the top corner of flask(i think that's high).
i'm still workin on it.i'll be glad to know ur ideas.
Attachment 63997

Attachment 63998

Attachment 63999

All times are GMT -4. The time now is 17:53.