CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)

 Mehmet Sonmez December 29, 2007 11:44

Unsteady Laminar Flow Around a circular cylinder

Hi,

I am investigating the behaviour of liquid water around a circular cylinder. Reynolds number is 178, and the velocity is 0.0008942m/s. But I couldn't get the velocity contour as it should be.

Here is the image of how it should be. http://img186.imagevenue.com/img.php..._122_425lo.JPG

And this is what I got(It is definitely wrong) http://img11.imagevenue.com/img.php?..._122_930lo.JPG

Can anybody help me? I chose unsteady solver and laminar for viscous. Time Step Size, Number of Step Size, Max Iter. per Time Step, have they any impact on the result?

 Ramesh.K December 29, 2007 12:55

Re: Unsteady Laminar Flow Around a circular cylind

Hi, when u are solving unsteady flow time step plays a major role in predicting the physics of the flow. Timestep is calculated based on v=del(x)/del(t) where del t should be ur time step. u can see it in fluent manual how to calculate the time step. if accurate time step is not specified the solution will not be correct.

bye

 Mehmet Sonmez December 29, 2007 15:03

Re: Unsteady Laminar Flow Around a circular cylind

Thank you very much, Ramesh.K

 Sham December 31, 2007 01:13

Re: Unsteady Laminar Flow Around a circular cylind

Ramesh is right that time step is very important. In choosing a time step, try to keep the cell CFL (Courant) number to be below 1 near the cylinder. Also, you have to run it long enough to be able to get some vortex shedding and may be need to introduce some perturbation to stimulate shedding.

Sham.

 vinse January 3, 2008 06:28

Re: Unsteady Laminar Flow Around a circular cylind

hi, How to calculate the number of time step per iterations in an unsteady problem. thanks in advance

 Sham January 3, 2008 06:33

Re: Unsteady Laminar Flow Around a circular cylind

In my opinion thats does not really matter as long as your time step size is small enough. In m simulation, it takes about 5 iterations per time step.

 All times are GMT -4. The time now is 02:41.