I think the best way to go about this is to define the air-oil interface as a heat transfer coefficient, modelling the air convection cooling.
Let me know how it goes 
2 Likes
I did not even know this was possible! Thank you very much again, Darren. I think the simulation is going to work this way. I will let you know if it really does.
Hi @gcleite96, how are you getting on did it work? would be more than happy to take a look
Best,
Darren
Hi @1318980! I think it worked quite well.
The thermal plume developed was not 100% perfect, but the results were reasonably good. I am now trying to make both simulations (for air and oil) better.
Best,
Gabriel
@gcleite96 great! If you do asymmetric simulation, i.e. a half model along the symmetry, convergence will be better, this is because we know that the results will converge to a symmetrical state eventually, if it doesn’t it is because it has not yet converged (for steady state). Therefore, this will make your plume much better because it will be symmetric as expected.
Good luck,
Darren
I see your point, @1318980. To be honest, I thought about using this strategy, but I was afraid of something going wrong.
I will try it later!