[raza22]

I am new to CFD. A cube sitting at the bottom of the wind tunnel. What boundary conditions do I apply and what are the general steps I need to take to run an analysis?

[Time4Tea]

First question is what CFD software package are you using? To be honest, it sounds like you could really do with going through a basic tutorial or training course for that package. Perhaps it has online tutorials that you can do?

[raza22]

I'm trying to understand CFD for academic purpose.

[Time4Tea]

@raza22: your question is very broad, so I will only be able to give a very general answer:


For a cube in a wind tunnel, you would need to apply non-slip wall boundary conditions to the walls of the tunnel and cube, as well as flow inlet/outlet b/cs at the inlet and exit of the tunnel.


In terms of the general steps to run the analysis, it would just follow the same basic steps as any CFD analysis:


1. Construct the model geometry
2. Mesh the flow passage
3. Apply material and physical models (e.g. turbulence, energy)

4. Apply boundary conditions
5. Select and set up the solver (plus any monitors)

6. Initialize the flow field
7. Run the solver, until a desired level of convergence is achieved
8. Post-process and interpret the results

[raza22]

What advantages do CFD simulations have over the wind tunnel
experiments?

[FMDenaro]

Originally Posted by raza22

What advantages do CFD simulations have over the wind tunnel
experiments?

This question can be totally reverted in: what is the advantage of a wind tunnel experiment over CFD?
I strongly suggest to understand first that CFD is a field that starts from physics and mathematics. You have to understand your physical model, what mathematical PDE and BC.s can be used to model correctly the problem, what discretization can be suitable and then what is the solver to use.
If you have to compare CFD and experimental measures you have to understand what kind of measure are available.


In your case the flow over a bluff body is a very generic problem. Is the flow assumed compressible or incompressible? Are you considering a Reynolds number that ensure laminar regime or you are facing transitional and turbulent conditions? Are you correctly modelling the wind tunnel geometry in your grid?

[raza22]

This answer helped me more than anything. Thank you