Learning Forum

[samstudent]

Hello Ansys Fluent Users,

I am a beginner with anys fluent aiming to simulate a natural convection heat sink.

the outer top and side surfaces are pressure outlets

the mesh size is 4mm of the main air volume, 2mm for the inner air section, and 1mm for the heat sink itself.

the control volume is 22 inches radius with a height of 12 inches

I have the flow set as incompressible to save some time with running the simulation

I am not sure why the streamlines have 2 vortices, I think the flow should be smoother as it moves inwards and up.

If anyone has any ideas about how these results can be improved, that would be appreciated!

Thanks, Sam

Eigth T.PNG

[Rob]

If the flow is incompressible how is it buoyant? Note, staff are not permitted to open attachments so please repost the images.

[samstudent]

[samstudent]

Ignore the different colour of the heat sink, it has a separate local temperature contour. As far as the buoyancy, I believe the density is changed as a function of temperature but the flow is modelled as incompressible, and a different gravitational force is applied based on that different density. This I am not fully certain about to be honest.

[Rob]

You need to find out what's driving the flow. Otherwise the result could be noise on a "nonsense" flow result. The bulk of the domain looks to have near enough zero velocity so the flow may just be chaotic. Vectors will be more useful to show what's going on.
What did you set the operating density as? That will govern whether flow "falls" through the domain. Is back flow "from neighbouring cell" or "normal"? The latter can do all sorts of strange things in buoyant flow.