CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   LES fully developed channel flow (http://www.cfd-online.com/Forums/fluent/36607-les-fully-developed-channel-flow.html)

Gem May 9, 2005 11:16

LES fully developed channel flow
 
I am testing Fluents 6.1 Large Eddy Simulation capability using fully developed channel flow as the test case. I have carried out the simulation, however i am having difficulties with the post-processing stage. I have enabled the data sampling for data statistics option in the iterate panel and ran the simulation for 8 flow through times. However when I display the velocity contours for the instanteous and mean x, y and z velocities they are exactly the same, the RMS values are very small. How do I enable Fluent to correctly output the instanteous, mean and RMS x, y and z velocities to display the velocity contours/data? does anyone know if there is a bug in fluent when using the data sampling option, I think it might be collecting data over one time step only and not over the 8 flow through times.

anindya May 10, 2005 03:23

Re: LES fully developed channel flow
 
I faced the same problem as yours but could not find any solution even after posting it here several times.

I have ran different cases with different settings, but everytime I saw that the mean and the instantaneous results were the same even after running them for several thousand time steps (hundred thousand or more iterations).

I strongly feel that there is a problem with LES model in Fluent 6.1.

Anindya

Gem May 10, 2005 05:32

Re: LES fully developed channel flow
 
Anindya,

Thanks for your post. Did you manage to solve the problem by using a different version of Fluent?, if so what version did you use? Did you find any alternative way of outputting the mean and instanteous velocities?

Thanks

Gem

Anindya May 10, 2005 18:26

Re: LES fully developed channel flow
 
I just had fluent 6.2 installed. I am going to run it and see if the new les models/settings helps me to get better results.

I will let you know if I manage to get better results. Btw what was your problem size (# 0f cells)and time step size and no of time steps?

Gem May 12, 2005 05:30

Re: LES fully developed channel flow
 
the size of the computational box is 7.85 (x), 2(y), 4.71 (z). the grid size is 64x82x82. no of time steps = 3,500 timesteps (3 flow through times) prior to setting up the data sampling for data statistics.


All times are GMT -4. The time now is 16:11.