[robinfisichella]

Hi, I am currently conducting an investigation into the aerodynamics of a human powered vehicle such as seen here http://www.speed101.com/photos.htm

My aim is too closely match a published figure for a certain HPV of Cd=0.045

I am having some issues currently, my console is giving me limited tubulent viscoscity problems.

Could someone please suggest if any of my input conditions seem off?

Star CCM+

Sparat Allmaras Model
Velocity=32.25 (0 yaw)
Turb viscosity ratio=5
or
Turb intensity=0.01 (value as seen, this means 1% right?)
Length scale 20m (half the domain height)
velocity scale = 0.4m/s

Im using a trimmer mesh, the object is close to the ground, no rotating wheels. I am using a wind tunnel type domain with slip walls as the far field boundaries.

Cheers

[gocarts]

If you are going to use the turbulent intensity and turbulent length scale option then I'd guess (I'm not a Star CMM+ user) that:

• Turbulent length scale should be a fraction of the vehicle length not the overall domain: http://www.cfd-online.com/Wiki/Turbulence_length_scale
• Velocity scale (reference?) should be your freestream velocity so that your turbulent intensity can then determine your turbulent velocity fluctuation: http://www.cfd-online.com/Wiki/Turbulence_intensity
• 0.01 Turbulent intensity = 1%, correct. That's the level you'd expect in a good wind tunnel

Matching absolute drag values from experiments is a tough call, just ask the drag workshop: http://aaac.larc.nasa.gov/tsab/cfdlarc/aiaa-dpw/

In my opinion CFD is better suited to computing drag for different designs and using the relative difference (from a baseline) in value to make design decisions.

Hope this helps

[robinfisichella]

Originally Posted by gocarts

If you are going to use the turbulent intensity and turbulent length scale option then I'd guess (I'm not a Star CMM+ user) that:

• Turbulent length scale should be a fraction of the vehicle length not the overall domain: http://www.cfd-online.com/Wiki/Turbulence_length_scale
• Velocity scale (reference?) should be your freestream velocity so that your turbulent intensity can then determine your turbulent velocity fluctuation: http://www.cfd-online.com/Wiki/Turbulence_intensity
• 0.01 Turbulent intensity = 1%, correct. That's the level you'd expect in a good wind tunnel

Matching absolute drag values from experiments is a tough call, just ask the drag workshop: http://aaac.larc.nasa.gov/tsab/cfdlarc/aiaa-dpw/

In my opinion CFD is better suited to computing drag for different designs and using the relative difference (from a baseline) in value to make design decisions.

Hope this helps

Thanks that re-assures me on a few things.