[stay31]

Hey all,

I have read many threads about the drag coefficient for a cylinder and have tried all the suggestions to no avail. I am at a loss and really hoping to achieve a relevant answer. I am using gambit 2.4 to mesh the model and FLUENT 12 to do the analysis
Here is what i have done:
I am doing a steady flow, 3-D case with the standard k-epsilon model with standard wall functions. I have tried two flow speeds, one at 4.5 m/s and another at 50 m/s and have tried changing the turbulent kinetic energy ratio and dissipation rate as well as the other values found under Turbulence quantities. I have changed my yplus and ystar from min=50 and max=500 to min=30 and max=300. I have also ran the simulation using realizable k-epsilon model with enhanced wall treatment. I scale the mesh when I import the mesh into FLUENT from gambit because gambit writes everything in mm. With all the above tried, and more, I still obtain a Cd~0.18-0.2 for both of the above stated flows.
Any help and/or solution is greatly appreciated, thank you!

stay

[esozer]

I suggest you try:

1. Refine near wall grid to about y+=1
2. Use Menter's SST turbulence model
3. Don't use wall functions for this case

Wall functions don't really work well for separated/high curvature flow regions. SST model scales the eddy viscosity near the wall to account for low-Re region. So you can directly integrate to the wall.

[stay31]

esozer,

thank you very much for your input. I will try your suggestions and keep my fingers crossed that it works.

regards,
stay