Calculation of yPlus in MRF simulations
 

Dear All,

I'm simulating airflow passing a propeller using MRF. I'm using incompressible solver simpleFoam and kOmegaSST turbulence model with nutkWallFunction as a wall function at nut. I don't have problems with simulations, it converges beautifully to nice results, but I would like to ask how do I calculate yPlus in my situation?

Since it is High-Re type of simulation I ran yPlusRAS utility, but resulting yPlus is not ok even though I recalculated U by adding \omega x R (rotating term).

After a little bit of research in doxygen I found that since I'm using nutkWallFunction the utility yPlusRAS consequently uses equations: u_{\tau}=C_{\mu}^{1/4}\sqrt{k} together with y+ definition y+=y u_{tau}/\nu. Since turbulent kinetic energy k is being calculated in rotating frame of reference it apparently doesn't result in correct y+ calculation?!?

Then I changed kOmegaSST in constant/RASProperties to SpalartAllmaras and I changed all nutkWallFunction in nut file to nutUWallFunction in the last simulation time folder. Then I ran yPlusRAS again and this time it calculated "logical" y+ distribution.

nutUWallFunction uses an iterative calculation in order to calculate y+ described in: http://www.cfd-online.com/Wiki/Two_equation_models (standard wall function).

Since I ran simulation with nutkWallFunction and calculated yPlus with nutUWallFunction I don't know what to think about these results?!?

How do you calculate yPlus in MRF?

Thank you very much for your help.

Best
Matej

Dear MatejAndrej,

Excuse for using this same topic but I think you can help me the little I need.
I am trying to use the same approach that yours to simulate a wind turbine using MRF approach. All goes ok until I activate turbulence ( I am trying to detect where the problem is).
For the walls inside the rotating region I have put nutkWallFunction, KqRWallFunction and epsilonWallFunction.
I have also tried nutUSpaldingWallFunction with no luck

Is it needed to change in some way the wall function for MRF? The k and epsilon values at wall increase a lot and then the simulation crash.

I am using OF2.3.x and kEpsilon model ( the mesh is ready for y+ around 100)