My Project
programmer's documentation
Data Fields
cs_wall_functions_t Struct Reference

wall functions descriptor. More...

#include <cs_wall_functions.h>

Collaboration diagram for cs_wall_functions_t:
Collaboration graph

Data Fields

cs_wall_f_type_t iwallf
 
cs_wall_f_s_type_t iwalfs
 
int iwallt
 
double ypluli
 

Detailed Description

wall functions descriptor.

Members of this wall functions descriptor are publicly accessible, to allow for concise syntax, as it is expected to be used in many places.

Field Documentation

◆ iwalfs

iwalfs

wall functions for scalar

  • 0: three layers (Arpaci and Larsen) or two layers (Prandtl-Taylor) for Prandtl number smaller than 0.1
  • 1: consistant with the 2 scales wall function for velocity using Van Driest mixing length

◆ iwallf

iwallf

Indicates the type of wall function used for the velocity boundary conditions on a frictional wall.

  • 0: no wall functions
  • 1: one scale of friction velocities (power law)
  • 2: one scale of friction velocities (log law)
  • 3: two scales of friction velocities (log law)
  • 4: two scales of friction velocities (log law) (scalable wall functions)
  • 5: two scales of friction velocities (mixing length based on V. Driest analysis) iwallf is initialised to 2 for iturb = 10, 40, 41 or 70 (mixing length, LES and Spalart Allmaras).
    iwallf is initialised to 0 for iturb = 0, 32, 50 or 51
    iwallf is initialised to 3 for iturb = 20, 21, 30, 31 or 60 ( $k-\epsilon$, $R_{ij}-\epsilon$ LRR, $R_{ij}-\epsilon$ SSG and $k-\omega$ SST models).
    The v2f model (iturb=50) is not designed to use wall functions (the mesh must be low Reynolds).
    The value iwallf = 3 is not compatible with iturb=0, 10, 40 or 41 (laminar, mixing length and LES).
    Concerning the $k-\epsilon$ and $R_{ij}-\epsilon$ models, the two-scales model is usually at least as satisfactory as the one-scale model.
    The scalable wall function allows to virtually shift the wall when necessary in order to be always in a logarithmic layer. It is used to make up for the problems related to the use of High-Reynolds models on very refined meshes.
    Useful if iturb is different from 50.

◆ iwallt

iwallt

exchange coefficient correlation

  • 0: not use by default
  • 1: exchange coefficient computed with a correlation

◆ ypluli

ypluli

limit value of $y^+$ for the viscous sublayer

ypluli depends on the chosen wall function: it is initialized to 10.88 for the scalable wall function (iwallf=4), otherwise it is initialized to $1/\kappa\approx 2,38$. In LES, ypluli is taken by default to be 10.88. Always useful.


The documentation for this struct was generated from the following files: