CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Fidelity CFD (https://www.cfd-online.com/Forums/fidelity-cfd/)
-   -   an error in fine (https://www.cfd-online.com/Forums/fidelity-cfd/98071-error-fine.html)

bluelotus March 2, 2012 07:50

an error in fine
 
I created grids of a rotor, but when I started euranus, an error occurred as flows:
[INDEXERR] Illegal index(0) for collection with 0 element
this error occurs only when I use "I" type grids.If I change "HOH" Grids ,the euranus is ok. And it is strange that if I import the CGNS file of "I" grids to CFX , it works well,too.
So I wonder why this error occurs?
Can you help me?? Thanks ~ :)

mohsenrahmani March 23, 2014 07:13

Quote:

Originally Posted by bluelotus (Post 347278)
I created grids of a rotor, but when I started euranus, an error occurred as flows:
[INDEXERR] Illegal index(0) for collection with 0 element
this error occurs only when I use "I" type grids.If I change "HOH" Grids ,the euranus is ok. And it is strange that if I import the CGNS file of "I" grids to CFX , it works well,too.
So I wonder why this error occurs?
Can you help me?? Thanks ~ :)

i have this problem also
and the other problem that i have is
when i solve a centrifugal compressor in Fine Turbo of Numeca with spart-alarms i have output but when i use another one, i have negetive domain density,pressure
can any one help me with this problem?

Hamidzoka March 25, 2014 05:09

Dear All;
regarding the grid type, note that "I" type is not flexible and can only be applied to limited types of blades. the best configurations are HOH and default topologies proposed by Numeca. This error generally arises when an improperly defined FNM (full non matching) boundary exist in the domain. Sometimes changing the number of streamwise grids may be a solution to this problem.

regarding the turbulence model, I think "mohsenrahmani" is right.
unlike the Spalart-Allmaras model, Euranus solver is not compatible with low-Re two equation turbulnece models such as SST. So if you wish to have a high resolution run with a better turbulence model, easily try other solver!

Regards

colinda1 March 26, 2014 04:41

Quote:

Originally Posted by Hamidzoka (Post 481950)
regarding the turbulence model, I think "mohsenrahmani" is right.
unlike the Spalart-Allmaras model, Euranus solver is not compatible with low-Re two equation turbulnece models such as SST. So if you wish to have a high resolution run with a better turbulence model, easily try other solver!
Regards

EURANUS has been validated and used extensively with models like SST by many FINE/Turbo users. If you have a problem on a particular case running SST while Spalart-Allmaras does not, it may be good to look at your project settings and in particular the initial solution and boundary conditions for k and epsilon.

Hamidzoka March 26, 2014 07:54

Dear Colinda;
it is not matter of initial or boundary conditions. it has difficulties in convergence even in simple multi-stage simulations.
if you have successfully used SST in such simulations please share your experiences with us.

Regards

colinda1 March 27, 2014 03:08

The centrifugal compressor tutorial in the FINE/Turbo documentation is one of the many examples of a case that was run successfully using SST model. Just give it a try. If you don't succeed to get it converged, your local support office will surely be able to help you correct the project to get it to run.

Hamidzoka March 27, 2014 04:59

My concern is mainly axial machines. are there validated cases with SST as well?

colinda1 March 27, 2014 09:34

Sorry, as "mohsenrahmani" was writing about centrifugal compressors I assumed incorrectly that you were working on similar configuration. The tutorial cases delivered with the latest release also include a multistage axial compressor case and a multistage axial turbine case. Both run well using SST, aiming for y+ values of 1 at the walls and setting the correct boundary conditions and initial values for the turbulence as recommended in the manual.

mohsenrahmani March 28, 2014 02:57

thank you very much
my problem have been solved by increasing my mesh number in autogrid 4 about negetive cell

Hamidzoka April 7, 2014 10:10

Dear All;
My problem with SST in axial flow machines was solved and I would like to share it with you!
neither changing the number of cells nor solver settings resolved the problem. I got a converegd solution only through changing the topolgy of the mesh. It may seem strange but worked in my case!!!

Regards


All times are GMT -4. The time now is 02:57.