Timo, I saw you are still in Stuttgart, so I hope you still post in this forum. What you mentioned about the order of the regular sst and the sas-sst model in the code: Did you get any information about that? We would need to have a look at Fluent or CFX code to find out what Menter really does here, but as I guess OpenFoam programmers didn't have a chance to do that, so this differences are possibly just not done on purpose?

Hi Philipp,

I have not checked the code anymore about the order e.g..
And I am using 1.6-ext with hard coded wall functions in the SAS-model.
I would be careful with the SAS in OF in general, cause I have never seen a good validation with that model.

Best,
Timo

Hi, I implemented the kOmegaSST-SAS as a RANS model. Feel free to test it, it's also the more recent formulation of the model.

http://www.cfd-online.com/Forums/ope...nce-model.html

Dear Foamers,


I'm facing the same error like podallaire at the very beginning of this thread, with the difference that I want to do a laminar simulation, which is in my mind completely weird.


I've set the simulationType to laminar in the turbulencePropertiesDict, as well as I've chosen laminar for the RASModel in the RASPropertiesDict (and of course turbulence off...).


When my solver tries to compute the discretization matrix tmp<fvVectorMatrix>, which is needed during the PIMPLE process, I get confronted with the following error message:



--> FOAM FATAL ERROR:

request for RASModel RASProperties from objectRegistry region0 failed
available objects of type RASModel are
0()

From function objectRegistry::lookupObject<Type>(const word&) const
in file /usr/lib64/OpenFOAM/OpenFOAM-2.3.0/src/OpenFOAM/lnInclude/objectRegistryTemplates.C at line 198.



I'm wondering myself, why there is a request for a RASModel at all (since I'm forcing a laminar computation by setting the simulationType to laminar) and how to solve the problem.

Can anybody please give me a small hint or any additional informations about how to handle this problem?

Kind Regards
Tom

Hi,


I am also trying to use the kOmegaSSTSAS model. Is it still a low-Re Model or also viable for high-Re-flows?




Kind regards,
shock77