Hello Yoav,
I am struggling on modeling natural convection in a air box and I saw that you gave the most detailed helpful answer to this type of problem on COMSOL discussion Forum. I will be so grateful if you could help me with this problem. I do appreciate your kind help in advance.
here is the problem:
Hello All,
I want to model a simple natural condition with buoyancy force but I constantly face convergence issue.
The model is based on having a hot plate in room temperature and I want to see what happens in the air field surrounding it. So what you can see in the model is a square domain for air. and the boundary conditions are:
Laminar flow:
Volume force: Fy=g_const*nitf.rho*(T-T0)/T0
This is to set the buoyancy force with Boussinesq Approximation.
Boundary conditions:
1) wall at the bottom boundary which is heated (hot plate)
2) symmetry on the side walls
3)outlet on the upper boundary
* these were the ideal cases for boundary conditions to be realistic but I changed all these to many different conditions but still the model could not converge.
Heat Transfer:
1)Temperature: T=393.15 K for the lower boundary which is heated (hot plate)
2) symmetry for the side boundaries.
3)outflow or temperature=293.15K at the upper boundary
I also tried to create extremely fine mesh at the edges and also tried to fully couple and use highly non linear solver!
I don't really know what else I can do and I truly appreciate if you can help me.
The model is attached,
Thanks,
Mohammad