[Pepsi]

Hello, I'm new to ANSYS Fluent and I'm trying to model natural convection. My setup involves a heated square box with a circular opening at the top. The box is heated to 50°C, while the surrounding environment is at 23°C. I've filled the box with air, but I'm unsure about the appropriate boundary condition to ensure that the heated air inside naturally exits through the circular opening at the top.

I'm considering using a pressure outlet at the box opening, but I'm unsure if this is the correct approach given that there is no inlet. Any suggestions or insights would be greatly appreciated. Thank you!

[Svetlana]

There are two options

1. Use a cylindrical domain syrrounding the heated box. The pressure-outlet BC on its top and veryical boundaries allows backflow. Cylinder needs to be sufficiently large. (this can be done as 2D test if setup with cyclic symmetry)

2. Use a rectangular domain similar to a wind tunnel and set a low inlet speed. This may make the solver more stable. Experiment with 0.1 m/s, 0.01 m/s etc as the speed and try to reach a point when your result is not that affected by the further reductions in the speed anymore. (this test can be done in 2D, though results will not be realistic, they will allow you to assess the speed required. then proceed to 3D simulation)

[Pepsi]

Hi! Thank you for your reply! I'm curious, would it be possible to add a pressure outlet at the opening instead? I'm wondering if that approach would still simulate natural convection.

[Svetlana]

just modeling the box and no adjacent air may be insufficient as this does not accurately represent the pressure distribution. i might suggest trying two simulations with only the box in one and the box with additional included air in the other, and comparing the results: my expectation is that the results will be different.

[Pepsi]

Alright! I'll give it a try. Thank you!!

[Martin_Sz]

Remember to set boundary conditions for natural convection with energy balance