[reza_k]

Dear experts,
I want to modify turbulent prandtl number (Prt) as a function of "Yplus". I defined an expression in CFX-pre and then the new expression was added to turbulence setting... and there was no error message. However, during solution CFX gives the error below:

"Error in subroutine get_YPLUS_TURB
Error calculating Yplus
GETVAR originally called by Subroutine DEF_DIFTRBC"

I will be greatful if someone expert help me handle this problem.
Thanks.

[Lance]

It is hard to help by guessing...
What was the CEL expression? Show your ccl?

[reza_k]

I did this:

"************************************************* ***
Insert/Epressions,.../Expression

then, I defined a new expression named "Prt2"

.../Definition

if(Yplus<=11.6, 8.8e-4*Yplus^3-0.0239*Yplus^2+0.238*Yplus+0.0836, 0.75)

then

Domain/fluid model/Turbulence/Turb. flux closure for heat...

Turb. Prandtl Num. : Prt2
!************************************************* ***

Thanks.

[Lance]

I dont get it.
The Y+ variable is a surface quantity while the turbulent Prandtl number is present in the entire domain, so I cant see why your expression should work. What is it based on? What do you want it to do? What about the freestream values?

[reza_k]

as you know, Prt is just a relation between "vt" and "Kt". some turbulence models like standard k-e keeps the Prt fixed(0.85 or 0.75), however it isn't correct in near wall region and in cases in which nearwall heat transfer is important may lead to large discrepancy. the relation I use is based on DNS simulation of Lu and Hetsroni "Direct numerical simulation of turbulent open-channel flow with passive heat transfer"... the relation just corrects the Prt for near wall cells based on their non dimensional distant to the nearest wall. anyway I think the error isn't physical!! may be its about the solver?!

[reza_k]

Dear Lance
this relation is used in TLVA, " a DNS Based turbulence model".

Also see

" Near-wall Modelling of turbulent convective heat transfer in film cooling of turbine blades with the aid of Direct Numerical Simulation Data",Djamel Lakehal

[lycastrog]

Hello, (Reza_K) I have the same problem, but with another expression. Could you solve your problem?
I have dealt with several expressions depending on Yplus and Temperature, on both occasions it gives me errors, similar to yours.

[FWP]

Dear Reza_K, I have the same problem, could you please help me to solve this problem? I just want to modify the Prandtl number which is a function of Y plus. Thanks a lot.

[Opaque]

Originally Posted by FWP

Dear Reza_K, I have the same problem, could you please help me to solve this problem? I just want to modify the Prandtl number which is a function of Y plus. Thanks a lot.

May I ask why you want to do that? What is the theoretical basis, and compatibility with the implemented models?

The fact a parameter exists, being a constant or a function, does not mean it is free for all to change. Walk that line carefully.