[retech]

Hi everyone,

I trying to run an LES using oodles, with initial conditions mapped from a steady RANS solution gotten from simpleFoam. When I execute oodles, I get a curious error where oodles complains about "different dimensions". I've checked all of my object files to be sure I've got the dimensions OK, and couldn't find any mistakes there. (B/t/w, I'm trying to use the dynamic Smagorinsky SGS.) Part of the error message follows.

"Create mesh, no clear-out for time = 0

Reading field p

Reading field U

Reading/calculating face flux field phi

Selecting incompressible transport model Newtonian
Creating field Umean

Creating field R

Creating field Bmean

Creating field epsilonMean

Creating field pMean

Creating field pPrime2Mean


Starting time loop

Time = 5e-07

Courant Number mean: 0.00687179 max: 2.76796
[1]
[1]
[1] --> FOAM FATAL ERROR : Different dimensions for =
dimensions : [0 2 -1 0 0 0 0] = [0 1 0 0 0 0 0]
#0 Foam::error::printStack(Foam:stream&)
#1 Foam::error::abort()
#2 Foam::dimensionSet::operator=(Foam::dimensionSet const&) const
#3 Foam::GeometricField<double,>::operator=(Foam::tmp <foam::geometricfield<double,> > const&)
#4 Foam::LESmodels::dynSmagorinsky::correct(Foam::tmp <foam::geometricfield<foam::te nsor<double>, Foam::fvPatchField, Foam::volMesh> > const&)
#5 Foam::LESmodel::correct()
#6 main
#7 __libc_start_main
#8 __gxx_personality_v0 at ../sysdeps/x86_64/elf/start.S:116
[1]
[1]
[1] From function dimensionSet::operator=(const dimensionSet& ds) const
[1] in file dimensionSet/dimensionSet.C at line 156.
[1]
FOAM parallel run aborting..."

Any thoughts on what this is? I'm a VERY new Foam user, so not so much in Foam is obvious to me just yet ;)

Thanks,
Joe

Hi Joe,
this message means really that some value is defined differently in your boundary conditions files and in the programm.
Have you checked the boundary files, whose dimension corresponds to [0 2 -1 0 0 0 0] ?
It could help for finding the bug, or at least which equation is concerned.
There is maybe an equation adimensionned by the velocity.

[retech]

OK, that points me in the right direction. I suspect my problem arises because I've mapped an incompressible RANS field, based on the k-w SST closure, into what I hope to be an incompressible LES field, based on the dynamic Smagorinsky sgs model.

In the RANS field, I specified BC files (and schemes/solvers) for k, nuTilda, omega, p, R, and U. For the LES field, I'm specifying BC files (and schemes/solvers) for B, k, nuSgs, nuTilda, p, and U.

So it would appear that the LES solver is looking for two fields with dimensions (0 2 -1 0 0 0 0), i.e., nuSgs and nuTilda, whereas the field mapped from the RANS solution only had one variable with that dimension, namely nuTilda.

So I think my question now is, how do I map the RANS field to the LES case while ensuring that the LES solver is getting the field it's looking for?

Joe