HI IS IT POSSIBLE TO CONSIDER THE AIR FLOW INSIDE THE WIND TUNNEL WITH THE SPEED OF 20 m/s OVER A CYLINDER AS AN INCOMPRESSIBLE FLOW? WHAT KIND OF TURBULANCE MODEL WILL GIVE A BETTER RESULTS? THANKS REZA

IN VISCOUS SELECT K-EPSILON(2-EQUATION)

You can calculate the Mach number of the flow from the equation M=u/c, where u is the flow velocity and c is the sound speed and is calculated as sqrt(gamma*R*temp). Normally, at room temperature, air flow at this velocity (20 m/s) has a Mach number of 0.05 ...so it can be an incompressible flow. You can consider any flow at Mach numer less than 0.3 as incompressible. Since this is a wind tunnel flow, just take care of the temperature at the test section, which is normally not the room temparture.

Also calculate the Reynolds number of the flow to determine whether the flow is turbulent. Regrading the turbulent models, have a look at the fluent manual as a fisrt step and follow the instuctions..then compare the results with standard experiments. In your case, standard K-E is a good model to start with.

All the best Rajesh

Hello Rajesh,

I Have the same problem; do you know the values of turbulence intensity and length scale which i have to set at the velocity inlet?

Thanks

Are you doing the same problem with same domain and boundary conditions?

I have to simulate a bidimensional flow past a circular cylinder! The flow is air with temperature of about 15-20C at the velocity of 25 m/s! The domain is a rectangular domain of dimensions 25Dx32D where D is the diameter of the cylinder.

Never use a k-epsilon for flow over a cylinder whith no sharp edges since the k-epsilon "sticks" to much to the surface. Use a k-omega based model like the SST (Shear Stress Transport). Check this : http://www.pbase.com/michaelbo/cfd

- Michael Bo

the model laminar is good for this problem? if no, why?

What do you mean? If the flow is turbulent use a turbulence model, otherwise don't!

therefore, I have to calculate the reynolds number with the formula Re=density*velocity*D/viscosity, before making the simulation, to see if it is laminar or turbulence?

yes. Read this http://www.centennialofflight.gov/es...d_Flow/TH9.htm

Thanks for the link.

then is mistaken to carry out a laminar simulation with a value of Reynolds number of about 2000000?

Thanks alot. Reza

It is highly turbulent flow at that Reynolds no.

HI..everybody... I also have some problem about k-e in wind tunnel. In the k-e,it has three choice at NEAR-WALL Treatment,my Re about 25000;it has a heat sink in the wind tunnel.and which one is better for my case to use.(standard wall functions、non-equilibrium wall functions、enhance wall functions?)thanks.

Read the Fluent help section 11.9.1 and 11.9.2 - Michael Bo

I can't find the section 11.9.1 and 11.9.2?My user's Guide only have section11.5.6.ths