|
[Sponsors] |
November 7, 2003, 02:43 |
anisotripic results
|
#1 |
Guest
Posts: n/a
|
Hi all,
i´m trying to model flow instabilities in an internal flow. I was suggested to choose a turbulence that is known to produce ansiotropic results (e.g. Launder eddy viscosity model). What does it means "anisotropic results" ? I don´t have the Launder eddy viscosity model. The turbulence models that i have are the follows: Standard k-eps Model Zero Equation Model RNG - (Re-normalized Group Model) NKE - (New k-eps Model due to Shih) GIR - (Model due to Girimaji) SZL - (Shi, Zhu, Lumley Model) which one should i choose ? |
|
November 7, 2003, 11:01 |
Re: anisotripic results
|
#2 |
Guest
Posts: n/a
|
Anisotropic turbulence is turbulence that is dependent on direction, i.e. (u'u')x != (u'u') y != (u'u') z. The standard k-e model for instance assumes isotropic turbulence, ie. turbulence that is independent of direction.
What u need is a model that is not based on the conventional eddy-viscosity assumption by Boussinesq (i.e. Reynolds stress tensor = (eddy viscosity x mean strain rate)-2/3 x ro x k x g^(beta x alpha)) but one which extends the stress-strain relationship by introducing non-linear elements. These models are called anisotropic or non-linear eddy-viscosity models (NLEVM's). Since u seem to be using ANSYS/FLOTRAN I am pretty sure (except of SZL) that none of the mentioned models incorporate the anisotropic eddy-visocity approach. To make sure, though, I suggest u check out the code's manual and look out for the relationship between stress and strain, e.g. Reynolds stress tensor = (eddy viscosity x mean strain rate)- (2/3 x ro x k x g^(beta x alpha))+ (2 x eddy viscosity) - ((4 x eddy viscosity/(ro x cp x k)*(A x ...)) + (B x ...) + (C x ...) where A, B and C are the closure coefficients. These coefficients were determined to get a good match for shear flows. Hope this helps. |
|
November 20, 2003, 08:17 |
Re: anisotripic results
|
#3 |
Guest
Posts: n/a
|
Just stumpled across a few pieces of info and thought of your inquiry. It's late but better late than never:
You can definitely use the following models to model anisotropic problems: a) GIR - (Model due to Girimaji) - non-linear (probably differential) Reynolds stress model b) SZL - (Shi, Zhu, Lumley Model) - algebraic Reynolds stress model Still not sure about the new K-e model due to Shih so better don't ... Hope it's not too late, let me know how u r getting on ... |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
OpenFOAM - Validation of Results | Ahmed | OpenFOAM Running, Solving & CFD | 10 | May 13, 2018 18:28 |
Velocity spots in openFoam results | Valle | OpenFOAM Running, Solving & CFD | 4 | August 19, 2009 05:53 |
How to plot a function over a time period? | Cirion0000 | CFX | 4 | July 18, 2009 12:48 |
how to view the complete results | shokry | FLOW-3D | 5 | February 3, 2009 13:56 |
CFX cylinder or sphere benchmark results | Mel | CFX | 1 | August 8, 2005 18:47 |