[Tobi]

Hi all,

I made a comparrison of some schemes:

https://holzmann-cfd.de/

[makaveli_lcf]

Great job Tobi! I did something similar for OSCIC'11 conference (see https://docs.google.com/file/d/0BzCp...it?usp=sharing). What I planned to try but still didn't implement it is a mesh dependency of the results:

1. Use unstructured grid
2. Vary an angle of your step profile on structured grid

Would be great if some one tests that))))

[Tobi]

Hi,

nice hints! If I have time I would do it

[s.m]

Originally Posted by alberto

Check your local (cell) Peclet number. The linear scheme requires it to be less than 2 (see Ferziger and Peric for a reference).

In your application you might want to use limitedLinear or linearUpwind anyway, which ensure the boundness of the solution.

Best,

Dear alberto,
my Pe number is too high, what should i do to reduce it?
what scheme should we use because you said, linear schemes need Pe number lower that 2,
Time = 2000
Reading phi
Calculating Pe
Selecting incompressible transport model Newtonian
Selecting RAS turbulence model SpalartAllmaras
SpalartAllmarasCoeffs
{
sigmaNut 0.66666;
kappa 0.41;
Cb1 0.1355;
Cb2 0.622;
Cw2 0.3;
Cw3 2;
Cv1 7.1;
Cv2 5;
}

Pe max : 309910

End

can you give me some guidance,
Thank you very much.

[Tobias Adam]

The new website of Tobias Holzmann:
http://www.holzmann-cfd.de/index.php...erical-schemes

I found it on my research about the different numerical schemes and what they do. Right now I still don´t know how the maths is done by the different numerical schemes.
Does anybody have some good papers or links where the interpolation schemes are explained?

Best regards Tobi

[makaveli_lcf]

Tobi, did you check the reference section of the cfd-online?)))) http://www.cfd-online.com/Wiki/Appro...grids_-_Common for example...

Also a PhD thesis of Prof. Jasak has a lot of maths concerning the descretisation. It is also on web.

Regards,
Alex

[AlmostSurelyRob]

Hello everyone,

sorry for necrobumping this thread. I was wondering if I could get you interested in my battle against a Delaunay mesh which seems to be very sensitive to fvScheme setup too. I am reporting it here:

http://www.cfd-online.com/Forums/ope...tml#post516017

Perhaps you will find it also interesting - getting OpenFOAM to work on this type of meshes would be a significant win for everyone. They can mesh pretty complex geometries although we're test driving it on laminar pipe flow at the moment. We have some successes and some setbacks (paralle runs).

[qutadah.r]

how do i check the local (cell) Peclet number ? Thanks!