Hi I found very interesting post about drag cylinder but I have a question about the setting of turb at inlet ; I am doing a 3D , steady, k-e simulation, Y+<3 ; I am setting turb Intensity to 0,01 (1%) and lenght scale (to 0,07Lh) (according to some previous post it is what should be used (?)). The problem is that when I am decreasing the Turb Int at inlet the drag is decreasing .

a)- Does anyone knows which value should be used (to compare with the experiments in the literature).

b)-what approx should the distance between the inlet and the cyl (in terms of diameter?) .

c)-is it nornal that there is such big link between the decrease of a Turb Int at Inlet and the Drag (it is physical)?. Does it means that if the experiment of drag is done in a tank with a moving cylinder through a steady water, the drag would be less that what is in the litterature ?

Thanks for any kind of help

Hi PK,

I'm just going to show my point of view but I'm not really sure about it,

a) The value of turbulence intensity should be based on experimental data or your own estimation. I mean, modern wind tunnels are reaching as low as 0,05 %. But, be careful, I think that in the boundary conditions panel you are asked directly for percentaje, so if you want 1 %, type 1, not 0,01.

b) I'd put at least 10 times the carachteristic length (diameter).

C) I'd say that it is perfectly physical as the boundary layer has got enough energy to stay attached to the cylinder. What is not very lovely is the fact that for lower TI you get less drag. But I don´t understand your last question exactly. In fact, the drag should depend only on the Reynolds number and the shape of your case. So, turbulence couldn´t affect the drag but it helps to keep the boundary layer attached. I don´t know if I'm wrong, so I don´t know if, assuming a link turbulence-drag, it is as important as you are getting in your case.

Hope this helps just a bit, good luck and, specially important, don´t forget to change the reference values any time you change the case.

Hi, Thank for your feedback. a)ok

b)ok : in fact I have more than that so my question is not related to some boundary effect.

c)In fact in my 3D simulation when I use 1% as the turb Int at Inlet I got a drag around my cylinder of about 1. for a Re of 30000 ; I run again the same (same Re) simulation with a 0% turb Int at Inlet and I get a Cd of 0.85. I have check with some "ready made" 2d fluent case for a cylinder and I got the same behaviour. There was a post sometimes ago mentioning the fact that FLUENT was using directly the value at the inlet (?).

d) About the drag on the cylinder, I use to calculate it directly from the total pressure given by FLUENT (So I don't have to change the reference values (?)

Thanks