Learning Forum

[fz.erraghroughi]

Hello!

I have a question concerning the Dynamic Smagorinsky-Lilly LES Model. How can I enable it in Fluent turbulence panel?

I read the theory guide and I think I should enable the option "Dynamic stress" but I am not sure because the theory guide does not give information about those models.

Could anyone give me some guidelines?

Thanks in advance.

Regards.

[ai0013]

Yes, by enabling the "dynamic stress" checkbox Fluent sets the dynamic Smagorinsky model. Note that you are no longer asked to provide a value for Cs. Now Fluent calculates it automatically by applying a second filtering (at a test level) and using Germano's Identity.

You can run the case and display the 'subgrid dynamic viscosity constant'. Note that the contour will vary locally

[fz.erraghroughi]

Thank you sir for the reply!
Regards.

[fz.erraghroughi]

Sir, I have another question!
I am about to model particle deposition using the LES and the DPM model. I want to include the effect of particle dispersion. the DPM includes the Discrete Random Walk (DRW) model to account for the dispersion of particles due to turbulence. My question is when using the LES to simulate the flow field, should I enable the stochastic model (DRW) to account for the turbulence dispersion or it will be automatically included?
Thank you!
Regards.

[ai0013]

DRW model should be used only in RANS. Note that in RANS you solve only for the mean flow, i.e. velocity fluctuations are not solved, so particles need some source of "artificial" turbulence of the fluid phase to appear in the particle equation of motion. This is provided by the DRW model. For reference, see the work of Ioannides and Gossman from Imperial College.
Contrarily in LES, the instantaneous velocities of the fluid are now available. So DRW should not be used. Just be careful to perform a truly LES, check mesh requirements, discretization schemes, etc...

[fz.erraghroughi]

Okay sir, thank you very much for your kind help!
Regards.

[Rob]

Thanks much appreciated.
As an aside LES is run transient, so particles pick up the flow velocity at a given point at a given time so there are no stochastic tries. DRW is used in steady cases where (in theory) the flow doesn't change once the model is converged.

[fz.erraghroughi]

Thank you, sir!
If I understand, when enabling the DPM I should not disable the flow equations despite that the flow is converged? The flow equations must be run simultaneously with the lagrangian tracking?
Regards.

[Rob]

That depends on what you want to do with the results. I may run DPM coupled, or uncoupled; steady or transient it's all down to why I'm doing the model and how much time/cpu I can afford to spend.

[fz.erraghroughi]

ok, sir! thank you for the guidelines!
Regards.