CFD Online Discussion Forums

CFD Online Discussion Forums (
-   FLUENT (
-   -   Cd for 2-d cylinder (

Mike Lamar October 8, 2002 11:35

Cd for 2-d cylinder
hi, everyone

I am trying to ues FLUENT to show the relationship between Re and Coeffiecient of Drag for a 2-D cylinder. I tried the range from 10-1E5. Half of the cylinder is applied since I only need mean performance. However, I could not get the correct relationship. The Cd predicted by FLUENT is quite away from classic fluid dynamics book. here are some details of my approach. the half cylinder is in the center of the bottom of the domain, with the other bottoms are set as symetric condition. The diameter of the cylinder is 2. and upstream is x=-5, set as velocity inflow bc. Outflow bc is set at x=+5. At y=5, also set as velocity inflow,use the same value as the inflow boundary. For all the cases, steady and laminar model are usd. And the values in the /reference values/ are correctly adjusted. But the results are still far away. Can anybody give some help on this. Thanks very much.

Chetan Kadakia October 9, 2002 18:51

Re: Cd for 2-d cylinder
How fine is the mesh in your boundary layer?

CFDfan October 11, 2002 11:12

Re: Cd for 2-d cylinder
Dear Mike,

The main reason you get such bad results is your problems set-up. If your cylinder has a diameter 2(units) then you should set your your outer boundaries at 15-50 (units) or even more.

My advice:

- use far field at 50 units

- use velocity inlet for all boundaries except the outflow boundary where you should use pressure-outlet or outflow BC.

- use a grid with uniform stretching which should not exceed 1.1 - 1.5, so that the grid should have enough resolution close to the cylinder

- for turbulent flow regime use a turbulence model (SA, ke, komega, RST, etc.) make sure that if you use a high Re nr model (e.g. using law of the wall) y+ = 30-100, etc.

Good luck my friend!

All times are GMT -4. The time now is 22:20.