How to implement wall function in my code?
 

Hello,everyone.I'm new to wall functions,since I don't have a thorough understanding of turbulent theory or boundary layer,I want to someone to help me about how to use the wall functions. The problem is I don't know how to use the wall shear stress and the wall heat loss.
Let me introduce my own understanding of the process of wall function.
First,using velocity at the first point P near the wall to compute the friction velocity u_star(either the log law in the log region or the linear law in the viscous sublayer, depend on the y_plus).
Second,after finding the friction velocity,we can calculate the wall shear stress as well as wall heat loss.This is my question,after getting these two parameters,how should I do with them?For the k-epsilon equations,we can get k using its original equation, and epsilon can be calculated by a simple algebraic formulation.
and for other flow variables, such as rho,u,v,w,p, Are they computed from the N-s equations?
Can someone give me some ideas,or give me a flow chart of implement the wall function,thank you very much.

It should be implemented in the turbulence model subroutine. Read some paper and it is not difficult.
Good luck!

yzeyue

Dear yzeyue,can you give me some papers? I couldn't find a useful article(i suppose) for topic,thank you~

For what concerns velocity and temperature, the wall shear stress and heat flux computed by the wall function just need to supersede the viscous term toward the wall.

For turbulence variables, instead, wall functions are usually used to fix them in the first cell near the wall

Thanks for your reply.
As for adiabatic wall,the wall heat flux was zero,for the isothermal wall ,we usually specify the Tw, and the heat flux can be derived from wall function,right?
Another question for the solution of turbulence variable k ,I read some papers and fluent's manual,one is solving k directly - from a simple algebraic formulation,another is specifying the production term of K from wall function,and solve this parameter as in outer layer. Are they different approaches to get the turbulence parameters?I'm confused by the papers.

You are right for the temprature: if you know the heat flux you don't need the wf at all; if you know Tw then you solve the temerature wall function after the velocity one (so that you know u_tau, which is probably needed in your temerature wall function).

I'm not very expert on the turbulence variables (i usually work in LES ;)), but i guess that following the Fluent references would not be bad. Another option is to check OpenFOAM or Code_Saturne implementation.