[levka]

Hello everyone.
Here are everything you need to compute DNS in channel vs OF 2.1.
Attached pdf is comparison this solver with DNS data of Kawamura and "Kim and Moin, 1987". Max divergence by mean "u" vs Moin is 2%, vs Kawamura 0%. Also in pdf there is view of full turbulent velocity profile that was taken during average procedure(10,000-30,000).
Following reference is to download zip file:
http://www.sendspace.com/file/xw4myd

Zip contains:
- solver (modified ico_Foam);
- utility to create tangential mesh spacing;
- test case with perturbed initial data;
- computed averaged data+log file;
- manipulation with data Excel file;
- paper Kim and Moin 1987;
- DNS data from paper above;
- DNS data Kawamura;
- master thesis of Steven van Haren"Testing DNS capability of OpenFOAM
and STAR-CCM+";

In case zip is not available i will e-mail it .


Have fun guys!

[Goutam]

Thanks for your contribution. What about LES for a pipe flow?

[alberto]

You might want to consider to submit this validation study to the Foundation, once they will start accepting code and contributions. It is surely valuable.

[levka]

Nice idea! But how to do it ?

[alberto]

There is no defined procedure yet. When I tried to submit some code I used OpenCFD mantis, and they said they are defining the rules for contributors, and will contact me when ready. Maybe you could ask them if they are interested also in validation studies.

[ehsan]

Dear all

Could someone help me in taking a test of DNS over airfoil?

Thanks

[levka]

Hello ehsan,

I suggest you to download solver solver from the first post and utilize that for your purpose. Just change mesh and BC according to your case, not forget to use initial conditions with perturbations( you can map them from my case to yours using mapField utility). And enjoy. Be in touch. I am interesting in the results.

[ehsan]

Dear Levka

Thanks a lot, we will work on this topic and will be in touch. However, does 2D DNS solution is very time consuming?

Regards
Ehsan

[levka]

Originally Posted by ehsan

Dear Levka

Thanks a lot, we will work on this topic and will be in touch. However, does 2D DNS solution is very time consuming?

Regards
Ehsan

That depends on your mesh and mesh it self should be defined according to comparison some integral parameter (for example U+ profile or Cf, Cd ) with some closest benchmark problem. Got it?
Cheers

[ehsan]

Dear Levka

In fact, I recently saw this paper:
http://www.sciencedirect.com/science...45793004001203

where the flow simulation over two dimensional and three dimensional airfoil had been reported, in Fig. 7 the U velocity had been reported and we will try to compare our results with this paper.

In fact, I am not familiar with DNS at all while I had worked on LES before, so it took me some time to understand DNS. I downloaded the file you prepared and work on it now.

Regards and thanks
Ehsan

[Farshad585]

Hi everybody
I want to simulate an airfoil and analyze it by dnsFoam. I've used boundary condition as it is in incompressible/simplefoam/2Dairfoil
I've started to run by dnsFoam, but there are the below errors:
--> FOAM FATAL ERROR:
calculated number of cells is incorrect

From function Kmesh::Kmesh(const fvMesh& mesh)
in file Kmesh/Kmesh.C at line 84.

FOAM aborting

#0 Foam::error:rintStack(Foam::Ostream&) in "/opt/openfoam210/platforms/linuxGccDPOpt/lib/libOpenFOAM.so"
#1 Foam::error::abort() in "/opt/openfoam210/platforms/linuxGccDPOpt/lib/libOpenFOAM.so"
#2 Foam::Kmesh::Kmesh(Foam::fvMesh const&) in "/opt/openfoam210/platforms/linuxGccDPOpt/lib/librandomProcesses.so"
#3
in "/opt/openfoam210/platforms/linuxGccDPOpt/bin/dnsFoam"
#4 __libc_start_main in "/lib/i386-linux-gnu/libc.so.6"
#5
in "/opt/openfoam210/platforms/linuxGccDPOpt/bin/dnsFoam"
Aborted


Could anybody help me what should I do?

[levka]

Hello,Farshad585
that error maybe coursed by either you did not run blockmesh, either your initial field is generated not for your mesh.
Best

Originally Posted by Farshad585

Hi everybody
I want to simulate an airfoil and analyze it by dnsFoam. I've used boundary condition as it is in incompressible/simplefoam/2Dairfoil
I've started to run by dnsFoam, but there are the below errors:
--> FOAM FATAL ERROR:
calculated number of cells is incorrect

From function Kmesh::Kmesh(const fvMesh& mesh)
in file Kmesh/Kmesh.C at line 84.

FOAM aborting

#0 Foam::error:rintStack(Foam::Ostream&) in "/opt/openfoam210/platforms/linuxGccDPOpt/lib/libOpenFOAM.so"
#1 Foam::error::abort() in "/opt/openfoam210/platforms/linuxGccDPOpt/lib/libOpenFOAM.so"
#2 Foam::Kmesh::Kmesh(Foam::fvMesh const&) in "/opt/openfoam210/platforms/linuxGccDPOpt/lib/librandomProcesses.so"
#3
in "/opt/openfoam210/platforms/linuxGccDPOpt/bin/dnsFoam"
#4 __libc_start_main in "/lib/i386-linux-gnu/libc.so.6"
#5
in "/opt/openfoam210/platforms/linuxGccDPOpt/bin/dnsFoam"
Aborted


Could anybody help me what should I do?

[Farshad585]

Originally Posted by levka

Hello,Farshad585
that error maybe coursed by either you did not run blockmesh, either your initial field is generated not for your mesh.
Best

Thanks a lot

[aerospain]

Hi levka,

Could you explain how to proceed with the zip file you made available in order to build the ico_DNS and run it, please?

thanks!

[Farshad585]

Hi everybody
I've used least squares and fourth order for discretization for pressure gradient (grad(p)), but I don't know the performance of least sqaures and fourth order. Could somebody give me some information about it?
regards

[student_07]

Hallo Levka,



I have one question to your solver. I'm still a beginner to OpenFoam, so don't wonder, if it is obvious to you.



I ran the test file and it works fine. Now I want to modify it and want to use a constant inlet velocity as boundry condition. In "constant/transportProperties" you set a pressure gradient "gradP". But there is also a "transportProperties_channel" file in which U_bar (probably an average velocity) can be set instead of the pressure gradient.


How do I have to change the solver to use a constant inlet velocity? When I just use Ubar, OpenFoam still requires a pressure gradient?


Regards,


Klaus

[Farshad585]

Originally Posted by student_07

Hallo Levka,



I have one question to your solver. I'm still a beginner to OpenFoam, so don't wonder, if it is obvious to you.



I ran the test file and it works fine. Now I want to modify it and want to use a constant inlet velocity as boundry condition. In "constant/transportProperties" you set a pressure gradient "gradP". But there is also a "transportProperties_channel" file in which U_bar (probably an average velocity) can be set instead of the pressure gradient.


How do I have to change the solver to use a constant inlet velocity? When I just use Ubar, OpenFoam still requires a pressure gradient?


Regards,


Klaus

hi
you shoul go to system/fvscheme and set pressure gradient as it show gradP

[Farshad585]

hi
you shoul go to system/fvscheme and set pressure gradient as it show gradP

[levka]

I suggest you to look at regular channel solver and it's tutorial. "Ubar" is used over there without any gradP.

[lixx]

Nice work!

In the test folder, I can not find the 0/ folder. So you just started from a precursor run or else?

Originally Posted by levka

Hello everyone.
Here are everything you need to compute DNS in channel vs OF 2.1.
Attached pdf is comparison this solver with DNS data of Kawamura and "Kim and Moin, 1987". Max divergence by mean "u" vs Moin is 2%, vs Kawamura 0%. Also in pdf there is view of full turbulent velocity profile that was taken during average procedure(10,000-30,000).
Following reference is to download zip file:
http://www.sendspace.com/file/xw4myd

Zip contains:
- solver (modified ico_Foam);
- utility to create tangential mesh spacing;
- test case with perturbed initial data;
- computed averaged data+log file;
- manipulation with data Excel file;
- paper Kim and Moin 1987;
- DNS data from paper above;
- DNS data Kawamura;
- master thesis of Steven van Haren"Testing DNS capability of OpenFOAM
and STAR-CCM+";

In case zip is not available i will e-mail it .


Have fun guys!