[MarkelRF]

Hi Everyone,

I am analyzing the turbulent flow over a cylinder in 2D (Re=10^4) with STD & realizable k-epsilon models. For each model, I am running two sub-cases:

1- Both kinetic energy production limiter and curvature correction switched off.

2-Both of the previous ones switched on.

I have obtained drag coefficients of around 0.55 for the sub-case 1 after having run with both STD & realizable models.

I have obtained drag coefficients of around 0.7-0.8 for the sub-case 2 after having run with both STD & realizable models.

Nevertheless, I have found in the literature that the drag coefficient for this Reynolds number should be around 1.2. I understand there might be some differences between both submodels due to artificial kinetic energy being generated in the domain area of the front stagnation point, where the flow is irrotational in the 1st submodel. But I do not understand why there is such a discrepancy between the calculated drag coefficients and the one found in literature.

PS: The reference values are all well entered, including the transversal area (radius of the cylinder x depth (1m)).

Sent from my Nexus 5 using CFD Online Forum mobile app

[karthikkumar]

What is k-epsilon method

Sent from my T02 using CFD Online Forum mobile app

[MarkelRF]

It is a two equation RANS turbulence model. It studies turbulent flow in an averaged way.

Sent from my Nexus 5 using CFD Online Forum mobile app