Correct Boundary Condition for an Open Atmosphere?

[waaizhaq]

Check these boxes if you have followed the posting rules.

• Q&A Navier Stokes is the most appropriate category for my question
• I have consulted the posting Guidelines on the Discussions front page
• I have searched the Discussions forum and my question has not been asked before
• I have searched the MOOSE website and the documentation does not answer my question
• I have formatted my post following the posting guidelines (avoid screenshots if possible, triple back quotes before/after pasted text, etc)

Question

Hello,

I revisited one of my early MOOSE practice problems about natural circulation flow from a heated wall (constant temperature), and open to the atmosphere. This problem was taken from Bergman and Lavine Heat and Mass Transfer Textbook:

I defined my mesh as a simple 2D rectangle, where the left boundary represents the heated wall, the bottom boundary represents the ground, and the right and top boundaries represent ambient air. I had to compute the heat rate coming off the wall to compare it to the one from the textbook. However, the value I computed was less than the textbook's example: ~ 232.42 Watts < 1108.24 Watts. I'm not sure if I computed the heat rate wrong or if I set my boundary conditions incorrectly. Currently I have a dirichlet boundary condition on the right and top boundaries of 296 Kelvin, but I don't think that represents ambient air accurately. I tried to use an outflow condition with INSFVMomentumAdvectionOutflowBC for velocity and FVNeumann for the temperature both set to 0, but the solve was unstable and didn't converge.

Additional information

Mesh size and type: 30x30
Reynolds number:
Discretization (finite element CG/DG, finite volume, etc): finite volume
Models (turbulence, porous media, etc): natural circulation
Solver method (fully coupled, segregated, multiapps, etc):
Base input you started from:

[GiudGiud]

Currently I have a dirichlet boundary condition on the right and top boundaries of 296 Kelvin

Setting the temperature would not influence the result if it were very far from the heat source. Right now it seems close, and will impose a strong decrease in the temperature to meet this boundary condition.

FVNeumann for the temperature both set to 0

This sets the heat flux to 0. This is in effect imposing a symmetry of the system, which is also not realistic.

I think the right solution for temperature is no boundary condition on the right and top, letting the INSFVEnergyAdvection kernel advect energy through those boundaries.
Momentum and MassOutflow should be correct too imo.

Can you get the flow equations to converge with a fixed temperature profile (using an exodus restart from the current T profile for example)

[GiudGiud]

I think if we dont have a set temperature anywhere but on the heated area, at the final time the temperature would just be the heated area

[waaizhaq]

So even if I have momentum and mass outflow boundary conditions? Now I'm curious as to why my solution didn't increase to the heated area at the final time when I first ran it without a set temperature on the right and top.

[GiudGiud]

So even if I have momentum and mass outflow boundary conditions?

if you have strictly mass outflow then it wont matter!
but there has got to be an inlet somewhere, and the temperature set on that boundary will matter for the steady state of the system

[waaizhaq]

So even with imposing a temperature DirichletBC on the other walls, I still get the same behavior where the fluid is only heated directly adjacent to the heated wall. It kind of looks like the energy isn't really advecting out. Do you think maybe my energy advection could be wrong?

[GiudGiud]

We dont have an EnergyAdvectionOutflowBC but feel free to code one