Hi,
I'm using the Launder-Sharma low Re model, and I had a query regarding the exact expression for turbulence dissipation that is being used in the model and its BC at solid wall. There is the classical definition of turbulence dissipation: epsilon and a modified expression varepsilon (defined using gradient of sqrt(k)).
Which variable is being solved and output for postprocessing? I am also looking for the subroutine where the BC at wall on this quantity is also set.
Many thanks for the clarification
Michael
Epsilon output in Launder Sharma model
Forum rules
Please read the forum usage recommendations before posting.
Please read the forum usage recommendations before posting.
-
Yvan Fournier
- Posts: 4204
- Joined: Mon Feb 20, 2012 3:25 pm
Re: Epsilon output in Launder Sharma model
Hello,
Not sure about this. In any case, the variable output for preprocessing should directly be the one solved (except possibly for some clipping).
The Launder Sharma model is only used in a few specific cases as far as I know, so even the main turbulent experts in the code might not have very "fresh" memories of the details of this implementation. Unless someone with this knowledge pops in on this forum, I would recommend checking the code itself to see which differences in implemetation are present relative to the other k-e models.
Checking in src/turb/cs_turbulence_ke.c/cpp, I would guess the changes and damping term are reflected mostly in the boundary conditions, but I had only a quick look. This seems consistent with your observations If I understand correctly.
Best regards,
Yvan
Not sure about this. In any case, the variable output for preprocessing should directly be the one solved (except possibly for some clipping).
The Launder Sharma model is only used in a few specific cases as far as I know, so even the main turbulent experts in the code might not have very "fresh" memories of the details of this implementation. Unless someone with this knowledge pops in on this forum, I would recommend checking the code itself to see which differences in implemetation are present relative to the other k-e models.
Checking in src/turb/cs_turbulence_ke.c/cpp, I would guess the changes and damping term are reflected mostly in the boundary conditions, but I had only a quick look. This seems consistent with your observations If I understand correctly.
Best regards,
Yvan