[emat]

Hello,

I am trying to model the heat dissipated from a hot body, in still air, by natural convection. I am using snappyHexMesh and chtMultiRegionFoam to create a 2-region conjugate heat transfer model (air and pipe), and appear to have successfully run the model in the case where the hot body is sitting in an enclosed air space.

But what I would really like to do is configure the boundary conditions of the air space so that it is not enclosed. I would like the air to be entrained by buoyancy forces, resulting in it entering from the bottom and exiting at the top.

In the user guide section "5.2.3.2 Entrainment boundary conditions", it is explained that such entrainment conditions are "extremely common". Unfortunately I can't seem to get it working in my model, using a configuration similar to that applied in the damBreak tutorial, and I can't find a tutorial that more closely ressembles this heat transfer situation.

The attached case file (entrainedSnappyMultiRegionHeater.tar.gz) actually runs, but the results look unlikely. Velocities pointing in horizontal directions, and impossibly high heat transfer to the surrounding air. Would anyone be willing to take a look at it and suggest where I've gone wrong?

Many thanks for any suggestions,

Eddie

EDIT: I double-checked that the problem is indeed coming from my entrained inlet/outlet boundary conditions, by enclosing the air space with walls and making the space taller. The results from this are a lot closer to what I would expect. So the problem definitely stems from my use of "totalPressure" and "pressureInletOutletVelocity". Case files for this enclosed version attached (enclosedSnappyMultiRegionHeater.tar.gz).

[emat]

My question doesn't seem to have attracted much interest! That's ok. Would anyone have any suggestions on how I might go about troubleshooting this myself? I don't have an error message as such, so it's a bit difficult to start researching where I've gone wrong.