can you share cylinder forces results ?
 

hello!

I'm trying to learn how calculate forces with OpenFOAM 1.6 and I'm doing simulation for a 2D cylinder of 1m diameter in an incompressible viscous flow with a velocity inlet of 1m/s (Re=1.10^6). I'm using simpleFoam and k-epsilon turbulence model. I'd like to know if someone already runned this simulation with thoses values (or even other) and would like to share the results for the forces on the cylinder.

Thanks a lot in advance.

Marine

Kind of
 

I've done similar, but with k-omega SST and on a geometry I can't provide you, but I did match drag results within 15%. I'd be happy to discuss, provide setup, discuss getting forces, etc.

i3enhamin@gmail.com

I remember that k-epsilon is not good for such simulation.

Hello,
The results are based on Std-Wallfunctions with High-Re Models for smooth cylinder. I havent tested nutSpallartAllamarsWall function.

The experimental values are also varying very much. I recommend you to have a look at this paper for more Refernece about the experimental data.

http://ctr.stanford.edu/ResBriefs01/wang.pdf


Have a look at the Plots and Tables as attached.

Good luck !

Thanks for your answers and for the paper !

Lin which model do you think is better for such simulation?

Just try some other two-equation turbulence closure other than standard k-epsilon. I think those would be better.

Hi!

I tried differents simulations and I have very stranges results, they are so differents and the convergence of the calcul vary a lot between the simulations.
My BCs and results are in cylinder2d.zip, if you are interested in it and can't open the file I can send you by mail.
so I'd like to know if anyone knows why the results are so differents, and if I can improve my boundary conditions or my turbulent model?

thanks a lot !

Marine
[IMG]file:///C:/DOCUME%7E1/ADMINI%7E1/LOCALS%7E1/Temp/moz-screenshot-1.png[/IMG][IMG]file:///C:/DOCUME%7E1/ADMINI%7E1/LOCALS%7E1/Temp/moz-screenshot.png[/IMG]

Hello,
I hope you fixed the errors. If not just check these variables.

1. Type yPlusRAS and see the values and plot them in paraView. May be you are using wall function with inappropiate near wall spacing. Check it out.

2.check the discretization schemes as well. I guess you are use 1 st orders scheme for U.

3 Regarding BC's just review turbulence modelling for wilcox. Hope its clarifying you..

Good luck