omegaWallFunction implementation

Horus · April 9, 2013, 13:37

Hello,

I'm studying the implementation of the incompressible omegaWallFunction.

http://foam.sourceforge.net/docs/cpp/a09165_source.html

Code:

00207         scalar omegaVis = 6.0*nuw[faceI]/(beta1_*sqr(y[faceI]));
00208 
00209         scalar omegaLog = sqrt(k[faceCellI])/(Cmu25*kappa_*y[faceI]);
00210 
00211         omega[faceCellI] = sqrt(sqr(omegaVis) + sqr(omegaLog));

$\omega = \sqrt{\omega_{vis}^2 + \omega_{log}^2}$
$\omega_{vis} = \frac{6 \nu}{\beta_1 \cdot y^2}$
$\omega_{log} = \frac{k}{\sqrt[4]{C_\mu} \cdot \kappa \cdot y}$

This clearly becomes oo when y approaches 0. Does this account for the turbulence damping near the wall?

Next question: What is G?

Code:

00173     volScalarField& G =
00174         const_cast<volScalarField&>
00175         (
00176             db().lookupObject<volScalarField>
00177             (
00178                 turbulence.type() + ".G"
00179             )
00180         );

And set here:

Code:

00213         G[faceCellI] =
00214             (nutw[faceI] + nuw[faceI])
00215            *magGradUw[faceI]
00216            *Cmu25*sqrt(k[faceCellI])
00217            /(kappa_*y[faceI]);
00218     }

The paper mentioned in the source code is available nowhere...

I'm especially clueless that the last part with the G is about?

Thanks!
Florian

fumiya · April 9, 2013, 18:08

Hi Florian,

For the second question, the followings will be of help:

The term G is the production of turbulent kinetic energy.

Hope this helps,
Fumiya

dancfd · April 28, 2014, 21:10

Hello all,

I noticed that the logarithmic component of omega1 in the code is not representative of the Menter 2001 paper.

The code is:

Code:

scalar omegaLog = sqrt(k[faceCellI])/(Cmu25*kappa_*y[faceI]);

Which gives:
$\omega_{log} = \frac{\sqrt{k}}{\sqrt[4]{C_\mu} \kappa y}$

Menter 2001 gives:
$\omega_{log} = \frac{u_\tau}{\sqrt[4]{C_\mu} \kappa y}$
Where C_mu = 0.09.

Why the difference? A message above makes reference to the following website: http://www.cfd-online.com/Wiki/Two_e...bulence_models

The last term of the last equation in the "Standard Wall Functions" section at that link matches the code, but it is still not clear why the code chose the equation that was selected (Wilcox 1994 eq 6.84) and not that which was defined in Menter 2001.

Does anyone know why the code is what it is, and not in line with Menter 2001?

Thanks,

Daniel

mooby · September 21, 2015, 05:54

Hi,

I am also wondering why is there difference between Menter Wall function definition and OF wall function definition. do you have an answer now ??

Blanco · September 23, 2015, 08:05

Hi,

do you have the Menter 2001 paper? I'm not able to find it anywhere...but all my references and docs are citing the

$\omega_{log} = \frac{\sqrt{k}}{\sqrt[4]{C_\mu} \cdot \kappa \cdot y}$

which is what is implemented in OF (correctly IMHO).

Best regards,

Andrea

mooby · September 23, 2015, 08:59

Hello Andrea,

Unfortunately I also dont hv Menter 2001 Paper. Nevertheless, I have requested someone on CFD online. He hopefully will share it with me. If I receive it, will surely share it with you

Blanco · September 23, 2015, 09:41

Ok thanks,

because I'm wondering if dancfd was correctly citing Menter formula in his post... I mean, if instead of dancfd cited formula which is

$\omega_{log} = \frac{u_\tau}{\sqrt[4]{C_\mu} \kappa y}$

we have

$\omega_{log} = \frac{u_\tau}{\sqrt{C_\mu} \kappa y}$

then everything is solved because we know that

$\omega_{log} = \frac{u_\tau}{\sqrt{C_\mu} \kappa y} = \frac{\sqrt{k}}{\sqrt[4]{C_\mu} \cdot \kappa \cdot y}$

Let's see if someone could provide Menter paper to check.

Best regards,

Andrea

dancfd · September 23, 2015, 21:08

Hello,

The COBEM paper is available in the public domain here:
http://www.abcm.org.br/anais/cobem/2001/home.htm -> just go to Lectures link

It was posted by someone else in this thread, #43:
http://www.cfd-online.com/Forums/ope...t-paper-3.html

It looks like you are correct, there must have been an error in my post above because sqrt(Cmu = 0.09) = 0.3 and then everything matches up (if it is Cmu^1/2 instead of ^1/4). Apologies for the confusion caused by this error.

Dan

mooby · September 24, 2015, 01:59

Hello Andrea,

Daniel "Dancfd" have sent me the Esh & Menter Paper. If you wish I can forward it to you or you can request Daniel for it

Blanco · September 24, 2015, 02:59

Thanks a lot Daniel and Ali,

so everything is ok now, we can say OF is correct!

Best regards,

Andrea

mooby · September 29, 2015, 08:52

Hi Adrea/Daniel,

I was wondering the Omega_Log term is defined by Menter as

Omega_Log = U_tau/(0.3*kappa*y)

where U_tau is calculated from U_Vis = U1/y & U_Log = U1/((ln(y)/k)+c)

U_Tau = ( (u_Vis)^4 + (U_Log)^4)^(1/4)

But in OF omega_Log is approximated by I think Spalding assumption

i.e. U_Tau = cmu^(1/4)*sqrt(k)

So, what if Menter U_Tau is used in omegaWallFunction .. will that improve the results i.e. mesh independence

April 9, 2013, 13:37	omegaWallFunction implementation	#1
Horus Member Florian Join Date: Nov 2009 Posts: 59 Rep Power: 16	Hello, I'm studying the implementation of the incompressible omegaWallFunction. http://foam.sourceforge.net/docs/cpp/a09165_source.html Code: 00207 scalar omegaVis = 6.0nuw[faceI]/(beta1_sqr(y[faceI])); 00208 00209 scalar omegaLog = sqrt(k[faceCellI])/(Cmu25kappa_y[faceI]); 00210 00211 omega[faceCellI] = sqrt(sqr(omegaVis) + sqr(omegaLog)); $\omega = \sqrt{\omega_{vis}^2 + \omega_{log}^2}$ $\omega_{vis} = \frac{6 \nu}{\beta_1 \cdot y^2}$ $\omega_{log} = \frac{k}{\sqrt[4]{C_\mu} \cdot \kappa \cdot y}$ This clearly becomes oo when y approaches 0. Does this account for the turbulence damping near the wall? Next question: What is G? Code: 00173 volScalarField& G = 00174 const_cast<volScalarField&> 00175 ( 00176 db().lookupObject<volScalarField> 00177 ( 00178 turbulence.type() + ".G" 00179 ) 00180 ); And set here: Code: 00213 G[faceCellI] = 00214 (nutw[faceI] + nuw[faceI]) 00215 magGradUw[faceI] 00216 Cmu25sqrt(k[faceCellI]) 00217 /(kappa_y[faceI]); 00218 } The paper mentioned in the source code is available nowhere... I'm especially clueless that the last part with the G is about? Thanks! Florian

April 9, 2013, 18:08		#2
fumiya Senior Member Fumiya Nozaki Join Date: Jun 2010 Location: Yokohama, Japan Posts: 266 Blog Entries: 1 Rep Power: 18	Hi Florian, For the second question, the followings will be of help: http://www.cfd-online.com/Wiki/Two_e...bulence_models (last eq.) http://www.tfd.chalmers.se/~lada/pos..._report_WF.pdf The term G is the production of turbulent kinetic energy. Hope this helps, Fumiya Horus, kiddmax and bmikuz like this. Last edited by fumiya; April 10, 2013 at 04:13.

April 28, 2014, 21:10		#3
dancfd Senior Member Daniel Join Date: Jul 2009 Location: Montreal, Canada Posts: 156 Rep Power: 16	Hello all, I noticed that the logarithmic component of omega1 in the code is not representative of the Menter 2001 paper. The code is: Code: scalar omegaLog = sqrt(k[faceCellI])/(Cmu25kappa_y[faceI]); Which gives: $\omega_{log} = \frac{\sqrt{k}}{\sqrt[4]{C_\mu} \kappa y}$ Menter 2001 gives: $\omega_{log} = \frac{u_\tau}{\sqrt[4]{C_\mu} \kappa y}$ Where C_mu = 0.09. Why the difference? A message above makes reference to the following website: http://www.cfd-online.com/Wiki/Two_e...bulence_models The last term of the last equation in the "Standard Wall Functions" section at that link matches the code, but it is still not clear why the code chose the equation that was selected (Wilcox 1994 eq 6.84) and not that which was defined in Menter 2001. Does anyone know why the code is what it is, and not in line with Menter 2001? Thanks, Daniel Last edited by dancfd; April 28, 2014 at 21:25. Reason: More info added.

September 23, 2015, 08:59	Menter 2001 Paper	#6
mooby New Member Mubashir Ali Join Date: Apr 2013 Location: India Posts: 10 Rep Power: 12	Hello Andrea, Unfortunately I also dont hv Menter 2001 Paper. Nevertheless, I have requested someone on CFD online. He hopefully will share it with me. If I receive it, will surely share it with you

September 23, 2015, 21:08		#8
dancfd Senior Member Daniel Join Date: Jul 2009 Location: Montreal, Canada Posts: 156 Rep Power: 16	Hello, The COBEM paper is available in the public domain here: http://www.abcm.org.br/anais/cobem/2001/home.htm -> just go to Lectures link It was posted by someone else in this thread, #43: http://www.cfd-online.com/Forums/ope...t-paper-3.html It looks like you are correct, there must have been an error in my post above because sqrt(Cmu = 0.09) = 0.3 and then everything matches up (if it is Cmu^1/2 instead of ^1/4). Apologies for the confusion caused by this error. Dan Last edited by dancfd; September 23, 2015 at 22:45.

September 21, 2015, 05:54	Implementation of SST Wall Function (Menter)	#4
mooby New Member Mubashir Ali Join Date: Apr 2013 Location: India Posts: 10 Rep Power: 12	Hi, I am also wondering why is there difference between Menter Wall function definition and OF wall function definition. do you have an answer now ??

September 23, 2015, 08:05		#5
Blanco Senior Member Blanco Join Date: Mar 2009 Location: Torino, Italy Posts: 193 Rep Power: 17	Hi, do you have the Menter 2001 paper? I'm not able to find it anywhere...but all my references and docs are citing the $\omega_{log} = \frac{\sqrt{k}}{\sqrt[4]{C_\mu} \cdot \kappa \cdot y}$ which is what is implemented in OF (correctly IMHO). Best regards, Andrea

September 23, 2015, 09:41		#7
Blanco Senior Member Blanco Join Date: Mar 2009 Location: Torino, Italy Posts: 193 Rep Power: 17	Ok thanks, because I'm wondering if dancfd was correctly citing Menter formula in his post... I mean, if instead of dancfd cited formula which is $\omega_{log} = \frac{u_\tau}{\sqrt[4]{C_\mu} \kappa y}$ we have $\omega_{log} = \frac{u_\tau}{\sqrt{C_\mu} \kappa y}$ then everything is solved because we know that $\omega_{log} = \frac{u_\tau}{\sqrt{C_\mu} \kappa y} = \frac{\sqrt{k}}{\sqrt[4]{C_\mu} \cdot \kappa \cdot y}$ Let's see if someone could provide Menter paper to check. Best regards, Andrea

September 24, 2015, 01:59	Menter 2001 Paper	#9
mooby New Member Mubashir Ali Join Date: Apr 2013 Location: India Posts: 10 Rep Power: 12	Hello Andrea, Daniel "Dancfd" have sent me the Esh & Menter Paper. If you wish I can forward it to you or you can request Daniel for it

September 24, 2015, 02:59		#10
Blanco Senior Member Blanco Join Date: Mar 2009 Location: Torino, Italy Posts: 193 Rep Power: 17	Thanks a lot Daniel and Ali, so everything is ok now, we can say OF is correct! Best regards, Andrea

September 29, 2015, 08:52	Smoothing Function for Friction Velocity	#11
mooby New Member Mubashir Ali Join Date: Apr 2013 Location: India Posts: 10 Rep Power: 12	Hi Adrea/Daniel, I was wondering the Omega_Log term is defined by Menter as Omega_Log = U_tau/(0.3kappay) where U_tau is calculated from U_Vis = U1/y & U_Log = U1/((ln(y)/k)+c) U_Tau = ( (u_Vis)^4 + (U_Log)^4)^(1/4) But in OF omega_Log is approximated by I think Spalding assumption i.e. U_Tau = cmu^(1/4)*sqrt(k) So, what if Menter U_Tau is used in omegaWallFunction .. will that improve the results i.e. mesh independence

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Free CFD code with implementation of Menter's k-omega SST model	nikola_m	Main CFD Forum	2	May 8, 2020 12:13
a little confused with the implementation of compressibleInterFoam	su_junwei	OpenFOAM Running, Solving & CFD	2	May 7, 2014 17:33
Implementation issues of fvSchemes / laplacianScheme, in particular gaussLaplacianSch	thomek	OpenFOAM Programming & Development	0	October 18, 2010 05:10
Request for help so as to debug MAC implementation	HectorRedal	Main CFD Forum	0	September 13, 2010 16:52
Please explain the implementation of species transport Eqn in reactingFoam	kallipygian	OpenFOAM Running, Solving & CFD	0	October 13, 2008 07:29