Hi,

I'm seeking some publication about the stability of the collocated grid arrangement for turbulent flows. I'm especially looking for some comparisons between the staggered and collocated investigation of a LES or DNS turbulent channel flow. I don't know if collocated turbulent channel flow has been done before!! If you know, please share your sources with me. Sincerely,

Frederic Felten

Hi,

It is quite feasible to perform LES computations on a collocated mesh. For an example, please visit

http://www2.inha.ac.kr/~fnc/

The calculations were done with ANSWER -- a commercial code developed by ACRi, http://www.acri-us.com/

Feel free to contact me if you need more details.

Best wishes.

Madhu

Hi,

I have no doubt that ones can perform LES investigations with collocated grid, but my real expectations are some references of some publications.

I appreciate your concern, and just for the tought, the ANSWER code is not exactly collocated arrangement. If you check the ANSWER manual page 62 (http://www.acri-us.com/pdf/answer.pdf), it's explicitly written that: " All the state variales except for velocity are defined at the grid nodes".

Anyway, thanks for your help.

Sincerely,

Frederic Felten

You can have a look at

A. G. Kravchenko and P. Moin, "On the Effect of Numerical Errors in Large Eddy Simulations of Turbulent Flows", Journal of Computational Physics 131, pp 310-322 (1997)

Regards

Dear Frederic,

Thank you for bringing the manual to my attention. The line should read, "All state variables except fluxes are defined at the grid nodes."

I will try to hunt up some references for you.

Sincerely, Madhukar Rao

Thank you very much Madhukar and sergey for your informations. I do apprecaite your concern. All the best,

Frederic Felten

Hi Frederic,

A colleague of mine (Dr. Wen-Ping Wang) at Iowa State University extensively worked on this issue for his PhD with Prof. Richard Pletcher. You can probably contact Prof. Pletcher and request a copy of his thesis.

It dealt with LES of a channel with high heat transfer. After a lot of attempts with colocated grids he finally did most of his calculations with a staggered grid. The colocated grids with central difference schemes are the most unstable. Some ideas from Kravchenko's paper can be used to make it more stable.

chidu...

[mazdak]

Hi all freinds

i am writing a code for turbulent 2-d channel flow using collocated grid.

i have some qustion.

first is about grid arrangment.i wanna know that do i put some nodes on wall or like staggered grid i must put some nodes out of wall.

i mean the first and the last row of grid must be puted on wall or else?

i know that we calculate all variables at same nodes and i must use Rhi-Choi inter polation.

another question is about treatment of V-velosity near the wall when we PUT SOME NODES out of wall.

as you know for finite volume method we capture the wall boundary conition by adding one terms for coefficient of u equation.

i wanna know how mus i treat for V in this situation?

thanks and pleas do me a favor and reply these questions.

i am waiting for your replys.