[ecto]

Hi,

I need to run a steady simulation of an aluminum heatsink in a volume of air (ideal gas) @fixed T at rest in order to calculate its T field and understand the flow field that originates externally thanks to natural convection. I set up the heatsink with regions considering the two PCB, the led (volume heat power source), some internal plastic components and the internal and external air. I set up Interfaces Contact In place to allow heat transfer. The external air is a big cylinder, and here comes the trouble. Which BC to be applied at the inlet and outlet circular surface of it?

With Stagnation Inlet @gauge p = 0 and Pressure Outlet @gauge p=0 I obtain a FLOW FROM TOP TO BOTTOM. Why? That's the effect of gravity? But I expect a flow in the opposite direction, because air is going to be heated (so density is going to decrease).
With the same inlet and free flow at outlet I obtain a flow correctly going from bottom to top. But in theory everything should go correctly also with the first set of BC.

Any ideas?

Thanks

[ecto]

I forgot to say that's it's laminar.

[ecto]

No idea at all?