[sasanghomi]

Dear Sirs,

I am focusing on turbulent fluid flow, these days. My question is pertinent to SST-SAS model. According to what I understood from articles, SAS does not resolve any region of the fluid flow directly and it just model all regions by using URANS. However, this model adds a source term to W equation in order to capture some specific length scales.

What are the resemblances between LES and SST-SAS ?

Please let me know if I am making any mistake in the above-mentioned explanation.

Best Regards
Sasan Ghomi

[ghorrocks]

Isn't this all explained in the documentation? Have a look in the theory manual. It also gives references to some literature and you could chase those papers up too.

[sasanghomi]

Thank you dear Horrocks for your continuous support.
However, when I ask a question which is mentioned in documentation briefly. It means that I am not 100% sure about my perception. So, I believe that it is not worthless to discuss deeply the topics which are even explained in documentations. As you know, commercial software programs are such black boxes and we need to go through details so as to get the best out of them.

Best Regards

[ghorrocks]

The modelling guide explains this quite well:

SAS is an unsteady RANS based model where a von Karman length scale is introduced into the turbulence scale equation. This allows it to produce results similar to LES in some flows. But note some flows SAS will not work properly and will just produce a steady state flow field.

LES is a fundamentally different approach. The intention of LES is to allow the mesh directly model the large turbulence scales while leaving the small turbulence scales (which are simpler and more isotropic) to a simple sub-grid turbulence model. LES does not use RANS equations, it uses filtered Navier Stokes equations which are a little different to the RANS equations. It also uses a sub-grid model to model only the turbulence length scales smaller than the mesh resolution.