Hi,
I use Code_saturne v1.3.3. And I am trying to launch a case : NACA_0012 in a flow with an angle of incidence equal to 10°. I have a problem with convergence of the model. And I have some doubts concerning the right definition of the k-epsilon model.
I joined to this post my fortran subroutines : usini1; usclim; inini.
I think that I have forgotten to define some variables.
I can join the mesh as it's too big...
Thanks for your help.
Gauthier
What are the inputs to define a k-epsilon model ? without GU
Forum rules
Please read the forum usage recommendations before posting.
Please read the forum usage recommendations before posting.
-
David Monfort
Re: What are the inputs to define a k-epsilon model ? without GUI with Saturne v1.3
Hi Gauthier,
I've found some small errors in your setup, and have a few remarks/questions herebelow.
First, at line 701, you've set IPHAS to 2 whereas the maximum number of phase is 1. The compiler won't tell you anything but it could lead to strange behaviour while running the simulation. Likewise, IPHAS shouldn't be set to 0 at line 973.
You should also let the variable ILEAUX set to 1 when restarting a simulation, otherwise the mass flux won't be read (actually all values at faces) which could lead to discrepancies in the first iterations of the restarting simulation.
Setting gravity to something non-zero is useless if you don't have some buyoancy flow (or similar).
Concerning the boundary conditions setup, this seems fine. I'm just wondering why you check against the color 7... Isn't it enough to search faces accordingto their group names?
Note also that the example we give for turbulent setup at the inlet in usclim.F is more intended to deal with pipe flows (or similar). You might want to use some kinetic energy level to initialize your turbulence at the inlet.
Finally, concerning your use of iniini.F, you can change the same parameters in usini1.F (in the usipsu subroutine). Just copy and paste the piece of code you want to change.
Regards,
David
I've found some small errors in your setup, and have a few remarks/questions herebelow.
First, at line 701, you've set IPHAS to 2 whereas the maximum number of phase is 1. The compiler won't tell you anything but it could lead to strange behaviour while running the simulation. Likewise, IPHAS shouldn't be set to 0 at line 973.
You should also let the variable ILEAUX set to 1 when restarting a simulation, otherwise the mass flux won't be read (actually all values at faces) which could lead to discrepancies in the first iterations of the restarting simulation.
Setting gravity to something non-zero is useless if you don't have some buyoancy flow (or similar).
Concerning the boundary conditions setup, this seems fine. I'm just wondering why you check against the color 7... Isn't it enough to search faces accordingto their group names?
Note also that the example we give for turbulent setup at the inlet in usclim.F is more intended to deal with pipe flows (or similar). You might want to use some kinetic energy level to initialize your turbulence at the inlet.
Finally, concerning your use of iniini.F, you can change the same parameters in usini1.F (in the usipsu subroutine). Just copy and paste the piece of code you want to change.
Regards,
David
-
gdambrin
Re: What are the inputs to define a k-epsilon model ? without GUI with Saturne v1.3
Hi,
Thanks David, I will check this things. But I have an other question concerning the definition of the number of processors. I can use 4 processors and I'm actually using only one. But when I changed the Nb_Processeurs equal to 4 in the file lance in the folder SCRIPTS, I obtained an error line 1248 if I remenbered. I think I have to change other things but I don't know where !!!
I didn't write something in list_processeurs, as I really don't know what to write ......
thanks for your help.
Gauthier
Thanks David, I will check this things. But I have an other question concerning the definition of the number of processors. I can use 4 processors and I'm actually using only one. But when I changed the Nb_Processeurs equal to 4 in the file lance in the folder SCRIPTS, I obtained an error line 1248 if I remenbered. I think I have to change other things but I don't know where !!!
I didn't write something in list_processeurs, as I really don't know what to write ......
thanks for your help.
Gauthier