CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   SAS-SST and transitional model have disappeared (https://www.cfd-online.com/Forums/cfx/79754-sas-sst-transitional-model-have-disappeared.html)

Josh September 2, 2010 15:11

SAS-SST and transitional model have disappeared
 
Hi all -

We recently updated our ANSYS Mechanical and CFD licenses. I was running several simulations with the SST Transitional model. Before we updated the license, the transitional model worked fine, but now it seems to have disappeared from Pre along with the SAS model. I am using version 12.1.

Has this happened to anyone before? Tech support doesn't know what the problem is. I have a feeling they accidentally sent us version 11, or something.

ghorrocks September 2, 2010 18:46

V11 has the SAS model. Just do help/about to confirm you are on the correct version of CFX.

Almost certainly what has happened is you do not now have the advanced turbulence license. You need this (or a full capabilities license) to access models like transitional turbulence, SAS and the other snazzy turbulence models.

Josh September 2, 2010 20:27

Hi Glenn -

I checked the About file first thing to ensure I was on 12.1. Strange about that license - I called the distributor and they said we have the full capability license.

Man, am I sick of this licensing trouble. All year I've put up with time consuming license problems.

Josh September 2, 2010 20:37

Quote:

Originally Posted by ghorrocks (Post 273808)
You need this (or a full capabilities license) to access models like transitional turbulence, SAS and the other snazzy turbulence models.

The LES and DES models were still available, strangely.

stumpy September 3, 2010 09:03

LES is available with the basic license, but DES and SAS need the full license. So the fact you can see DES but not SAS is very strange. If CFX-Pre is filtering out these models based on license checks, then set the following environment variable to disable that:
CFX_DISABLE_PHYSICS_LICENSING=1

Josh September 3, 2010 09:27

My mistake. DES is not available. It must be a licensing issue. It's strange, though, as they claim they sent us the full capabilities release.

I tried entering the aforementioned environment variable in the Pre command editor, but received the following error:

CCL validation failed with message:
Error: Parameter 'CFX_DISABLE_PHYSICS_LICENSING' is not allowed in /

stumpy September 3, 2010 11:08

Environment variables are entered at the system level. Before starting CFX, enter it under "Start > Control Panel > System > Advanced > Environment Variables" (that's for XP)

Josh September 3, 2010 12:29

Thanks for the help, stumpy, but I'm afraid it didn't work.

stumpy September 3, 2010 13:40

The case is transient right? CFX (including the launcher) was restarted? The variable name is shown as "CFX_DISABLE_PHYSICS_LICENSING" and it's value is shown as "1" (without the quotes)? One other thing - if you update the version without updating the client AND server license managers then funny things can happen.

Josh September 3, 2010 14:21

The case is transient (I've seen others on this forum looking for steady state DES, etc., so I can understand why you're asking).

It worked this time. I'm never altered the environment variables before and I mistyped it (I didn't separate the variable and value).

Thanks, stumpy. Would ANSYS be bothered by this resolution, or could I tell them the problem is fixed?

stumpy September 3, 2010 15:00

I got that variable from somebody in ANSYS, so somebody must know about it :-)

Josh September 3, 2010 15:03

Thanks again, stumpy. The amount of technical problems I've had in the past three months is beyond frustrating.

ghorrocks September 4, 2010 06:45

Does the variable simply make the options visible in CFX-Pre (which would not help much) or make it available in the solver (which would appear to bypass licensing)?

Josh September 5, 2010 18:05

The wall scale variable seems to have disappeared in the Solver, so I'm confident it worked.

I'll be sure to check Monday, though.

Josh September 7, 2010 14:59

The wall scale variable is still being calculated in the Solver. It's residual RMS value is 4.4E-16. However, the momentum thickness Reynolds number and intermittency residuals are also being calculated, indicating the transitional model is being used.

Is it possible that the wall scale is only there because it was used for the standard SST model in a previous run?

ghorrocks September 7, 2010 18:45

The turbulence transition model needs the distance to the wall, so will calculate the wall scale equation.

Josh September 7, 2010 19:17

Sorry, I was confusing the wall scale for the wall function. Long day.

Thanks for all the help.


All times are GMT -4. The time now is 10:06.