Natural convection model
 

I want to be able to compare the effect of forced vs. natural convection on a heat generating element. Right now I am not sure how to model a natural convection simulation. I have a cube with a constant heat flux on each side sitting in a big room. The top of the room is a pressure outlet with all sides being walls. The initial condition of the room is 25 C.

I have selected:

-Boussine model
-Constant density
-Gas
-Gravity
-Implicit unsteady
-Segregated flow
-Segregated fluid temperature
-Laminar
-3D

The problem is when I run it the residuals do not converge and I keep getting reversed flow on all the faces which does not go away. Can someone give me some tips on this? Thanks.

Switch off the Bussinesq model and change constant density to ideal gas. Although it should be good for that, Boussinesq seems not to work correctly in some versions.

Reversed flow can be neglected, it should just tell you that some inflow occurred at an (pressure) outlet. This is nothing to worry about, since something needs to flow into your domain when something flows out of it.

Well, how high is temperature difference in your flow problem?

Set the thermal expansion coefficient to >0 or 1/K where K is your operating temperature.

For some reason they set the default value to be 0 which makes Boussinesq not working

Would you please tell me how did you set the ambient temperature for the block?by setting some constant temperature to the block's wall or...?
Also,didn't you use any velocity inlet?
Thank you in advance