|June 5, 2010, 11:09||
Reynolds Stresses in Fluent
Join Date: Jun 2010
Posts: 9Rep Power: 8
I am working the Re=3900 2-D cylinder problem for a turbulence class in graduate school. We were to use the k-e and k-w turbulence models and plot uu Reynolds stress profiles to compare to Kravchenko's paper.
How do I back out the Reynolds stress components I need from the turbulent quantities given by these two models? I imagine I just use the Bouss Approx, and write a UDF to plot them, but I've never used the UDF's.
Can someone give me some idiot-proof step by step instructions on how to do this?
|February 1, 2012, 13:57||
Join Date: Mar 2011
Posts: 7Rep Power: 7
You have to use Bous. Approx.
Use the Define->Custom Field Functions to calculate
delu=dx-velocity-dx + dy-velocity-dy + dz-velocity-dz
then calculate strain rate tensor S_uu
S_uu=dx-velocity-dx - 1 / 3 * dukxk
-u'u'=2*viscosity-turb*S_uu/density - 2/3*turb-kinetic-energy
|September 15, 2013, 07:29||
Join Date: Mar 2009
Blog Entries: 6Rep Power: 43
|April 14, 2014, 16:26||
Join Date: May 2013
Posts: 32Rep Power: 5
I need to define Reynolds stresses in fluent with k-e model. I also could not understand the term dukxk and you define in the beginning something called delu but you do not use it anywhere. could you explain what is that one also?
Thank you very much, I really appreciate it.
|December 1, 2015, 11:16||
Boussinesq aproximation solves the problem
Join Date: Jan 2014
Location: Brno, The Czech Republic
Posts: 6Rep Power: 4
Yesterday, I found this thread and spent almost whole day by searching for answer with some, let's say, proof or explanation. I found that the problem raises from the transport equation of k in two equation models; and it's one of the problems in turbulence modeling. In general, we have more unknowns to solve than available equations; and Boussinesq aproximation is used to enclose transport equation of k in one- or two- equation models.
And now, back to problem:
let's have the Boussinesq aproximation
is Kronecker delta
and also have those two vectors:
and velocity vector
if the flow is incompressible, then
Now, it only depands what kind of Reynolds stress you want. For instance, if you want to know normal stress simply put and ; then, substitute first element in and to the Boussinesq aproximation and do the math. You will obtain:
And if you want to know shear stress , just put and and you will get:
Apply same principle for any other Reynolds stress you want to know and just simply put this to Fluent using Custom Field Functions
Wilcox, D. C., 2006, Turbulence Modeling for CFD, 3rd ed., Dcw Industries, Incorporated.
|boussinesq, reynolds, stress|
|Thread||Thread Starter||Forum||Replies||Last Post|
|Obtaining Reynolds Number in FLUENT||Emmanuel||FLUENT||2||April 7, 2016 01:46|
|Two questions on Fluent UDF||Steven||Fluent UDF and Scheme Programming||4||September 20, 2013 16:30|
|RMS velocity and reynolds stres with fluent||gina||FLUENT||1||May 20, 2013 04:05|
|Reynolds stresses without velocity correlations?||Phil||Main CFD Forum||1||October 2, 2006 04:29|
|Reynolds stresses and second law||JF||Main CFD Forum||6||June 2, 2005 22:57|