LES for 2D case?
 

Dear Foamers,

When I use LES model for a 2D case, it suggests that 'LES is not strictly applicable.....'
So, it means it is not ok to use LES for 2D case? A little bit confused.

it can be used numerically, but when i asked an expert in turbulence it says using LES for 2D cases is non-sense as it LES using spatial filter

Two-dimensional turbulence has different physical behavior than three-dimensional turbulence.
That is why LES on a 2-dim. domain should not be used to obtain reliable results for 3-dim. flows.

Emmm, I am using a 2D case to test my own developed solver, coz 3D case is bloody time-waste, if put too many grid inside the layer Y+<30.
I mean, is it a possible way to confirm the LES model's working well?

If you are comparing your results to the results from another 2D LES solver, you should be fine.

Hi Xiangyin,


Both Alex and Nima are correct. You need to make a distinction between validating your implementation and validating your model.
* If you purely want to validate your solver (so the numerical implementation), you can compare your results with other solvers.
* However, if you want to validate it, doing a 2D LES is basically pure non-sense. Turbulence is by definition a 3D effect. As soons as you neglect this, you will probably end up with pure rubbish.

Cheers,

Lieven

Thank you guys! That's really what I want to know.
My plan is to make sure every step is OK before using real 3D running, otherwise it is too long to try something in setup.
Thanks again!

Why 2D pitzDaily with "LES" in tutorials?
 

Hi,
Why OpenFOAM contains 2D pitzDaily with LES in ?
Obviously it does not need to validate implementation, on the other hand, based on your saying, it is completely rubbish!
Then why?

OpenFOAM tutorials are neither validation nor verification, but only a set of illustrations of the functionalities of OpenFOAM, so that users can fetch parts useful to them.

By the way, I don't think they put a wrong case just for illustration purposes; Existing a 2D LES case somehow depicts the allowability of using LES with 2D geometries, at least in some situations.

turbulence is by definition a three-dimensional phenomenon, except for some simplified cases of atmospheric flows; therefore, 2D LES is almost always nonsense since any vortex stretching mechanism is theoretically nonexistent in the governing equations (note: I do not refer to statistically independency or homogeneity). accordingly, they have definitely put the tutorial as an illustration, in my opinion.

Hi
Your guids and helps are so good, I use them
Thank you very much.