CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Steady state flow over cylinder (http://www.cfd-online.com/Forums/cfx/85040-steady-state-flow-over-cylinder.html)

vmlxb6 February 15, 2011 17:53

Steady state flow over cylinder
 
I am trying to show von karman vortex street for a reynolds # of 200 (I know it is a simplest CFD problem, pls excuse me). I tried to run a steady state simulation in CFX and I dont see any vortex shedding. Its looks all symmetric. I tried changing everything. Convergence, boundary layer, Advection scheme number of iterations etc etc etc.
I ran the same case in FLUENT using Green Gauss Node based scheme. Same BC and Same initial conditions same mesh, and it shows vortex shedding. What could be the problem ????

I am attaching the velocity contours. Thank you guys.

http://img69.imageshack.us/i/cfx.png/

http://img13.imageshack.us/i/fluent.jpg/

Thank you.

Josh February 15, 2011 18:29

Quote:

Originally Posted by vmlxb6 (Post 295398)
I tried to run a steady state simulation in CFX and I dont see any vortex shedding. Its looks all symmetric.

Did you consider that it's not steady state? After all, vortex shedding is hardly a steady process. Run the simulation as transient.

vmlxb6 February 15, 2011 20:34

I realize that vortex shedding is not steady. But I need to know why is the contour plots different in FLUENT and different in CFX. At least the contours in CFX should give some kind of unsymmetrical vortex ???
It probably is the type of solver I am using or even the residuals at the end of convergence. CFX does a better job converging the solution (up to 10^-8). Am sure there may be other ways to converge the solution in FLUENT too, but who cares !!!!!!!

Thanks anyways. Appreciate your help !!!!

ghorrocks February 16, 2011 05:59

If CFX has given a symmetric solution then that is correct - when the flow is averaged out it will be symmetric. But the physics will be totally wrong so forget the steady state solution. Comparing a steady state CFX and a steady state Fluent simulation of a model which is transient is totally meaningless.

As Josh says, if the flow is transient you need to run a transient simulation.

vmlxb6 February 16, 2011 16:04

@ Josh and Glen : You guys kick Ass man !!!


All times are GMT -4. The time now is 12:18.