In this case two RANS-based turbulence models were tested, i.e. the two-equation linear eddy viscosity k-ε-model, where k is the turbulent kinetic energy and ε the turbulent dissipation rate, and the four-equation, non-linear eddy viscosity k-ζ-f-model, where ζ is the velocity scale ratio the MathML source. The k-ε-model is the most widely used turbulence model, and has been implemented into most CFD codes despite its shortcomings. It assumes homogeneous isotropic turbulence close to the wall and a damping function is applied in that region. The k-ζ-f-model replaces the damping function by introducing an additional term, i.e. the velocity scale ratio. Therefore, a so called Hybrid Wall Treatment is used, which ensures a gradual change between viscous sublayer formulations and the wall functions. It has been shown that the k-ζ-f-model allows for the application of a coarser grid close to walls which may save computation time [15].