[tkolba]

Dear users

I am trying to study the influence of dimples on a golf ball. The results i should obtain can be seen in following picture: http://www.aerospaceweb.org/question...phere-drag.gif

But even the smooth sphere gives me trouble. I made a solid sphere and added an inclosure and then uses boolean to subtract it etc. Also the boundary conditions setup with an inlet and outlet seems to work. I tried 3 models allready but can't get the right number for de drag coeff. Here some values for the speed of 39m/s (Re approx 1E5)
k - epsilon model: 0.238
k - omega model: 0.385
SST model: 0.295

Looking at the expected outcome this should lie around 0.5 for a smooth sphere. I also don't have the dip in drag coeff but a more linear reduction.

Could you guys help me out, i really don't know where my errors are. (I'm new at ansys fluent)

Kind regards
Tom

[tschultz]

modeling a sphere in FLUENT may not yield the perfect Cd vs. Re relations as you might hope for. Don't be surprised if your numbers don't correlate exactly as in experiments. However, some things that you can try:

-refine your mesh near the boundary of your sphere. you can use the inflation tool and the sizing tool when creating the mesh to do so. youtube videos will show this method.

You shouldn't be using turbulence models all willy nilly either. There are pros and cons to each of them. Try Realizable Ke. Reference http://www.engr.uconn.edu/~barbertj/...nt%20Flows.pdf
for more info on turbulence models.

-Check your timestep. I assume you are doing transient (which you should be if youre not). You can find a Strouhal vs. Reynold number plot, find your strouhal number and calculate the frequency of the vortices that would be shed. Take the period and divide it by 10 or so to find your time step.

Those are just some ideas. I am working a cylinder problem similar to what you are doing so I can relate to your issues. good luck

[ck3]

http://www.dtic.mil/get-tr-doc/pdf?AD=ADA494935