I have been looking for a good reference with computational results for the flow over a cylinder for Re = 200-20,000 [or any Re = 0(1,000)]. The reason is that my 2-D computations seem to estimate a higher CD than experiments even after various grid etc studies, and I have successfully computed airfoils with the code.

I know that the shedding is 3-D for some Re > 200, and thus would need to be modeled in 3-D. But has anyone quantified the effect of the 3-D shedding?

I've looked into the "Flow around circular cylinders" by Zdravkovich.

My computational results (both CD and ST) match those reported by W.Dettmer and D.Peric Comput.Methods Appl.Mech.Engrg.192 (2003)1177-1226

Thanks.

-- Jarmo

I have carried out some calculations with my code and, for 2D simulations, obtained Cd=1.7 and St=0.24 for Re=10000. The problem of the overprediction of Cd for high Re numbers is the reason why in general 2d simulations are not carried out for Re greater than 200. It is interesting that some researchers simulate the 2d flow around a cylinder using first order convective interpolations and "validate" their codes. They don´t notice that the high numerical diffusion compensate the absence of 3d effects.

Thanks for the response. Your results are consistent with my findings, so I am a bit more relieved now!

Hello !

I have tested some solvers that I've written with a flow over a cylinder in 2D and I have the same experience.

Have a look at the article:

'Effect of three-dimensionality on the lift and drag of nominally two-dimensional cylinders' by R.Mittal and S.Balachandar in Phys. Fluids 7 (8),1995

Petri Majander

Kiitoksia! I will go and get that asap.

Dear Petri,

Is there any way of getting that article online?

Pascale