CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (http://www.cfd-online.com/Forums/openfoam-solving/)
-   -   LESdelta cubeRootVol dimension problem (http://www.cfd-online.com/Forums/openfoam-solving/60374-lesdelta-cuberootvol-dimension-problem.html)

vvqf December 14, 2005 11:12

dear all, i am using oodles
 
dear all,

i am using oodles solver to solve a 2D case.
with delta cubeRootVol
But only ran into error :
´╗┐ FatalErrorIn("cubeRootVolDelta::calcDelta()")
<< "Case is not 3D or 2D, LES is not applicable"

later i tried to change delta to smooth(just let the case run at first), by changing the entries in turbulenceProperties for delta, then I got another error:
------
Selecting incompressible transport model Newtonian
--> FOAM Warning :
From function dictionary::read(Istream&, const word&)
in file db/dictionary/dictionaryIO.C at line 92
Reading "/auto/people/evebe514/ewenqui/OpenFOAM/ewenqui-1.2/run/project/cases/cavity3/co nstant/turbulenceProperties" at line 175
could not add entry
smoothCoeffs
{
delta smooth;
smoothCoeffs
{
deltaCoeff 1;
}
maxDeltaRatio 1.1;
}
on line 175 of dictionary "/auto/people/evebe514/ewenqui/OpenFOAM/ewenqui-1.2/run/project/cases/cavity3/co nstant/turbulenceProperties"


--> FOAM FATAL IO ERROR : Attempt to return dictionary entry as a primitive

file: /auto/people/evebe514/ewenqui/OpenFOAM/ewenqui-1.2/run/project/cases/cavity3/con stant/turbulenceProperties::smoothCoeffs from line 124 to line 124.

From function ITstream& primitiveEntry::stream() const
in file db/dictionary/dictionaryEntry/dictionaryEntry.C at line 79.

FOAM aborting


Questions:
1. is the problem with geometry mesh?
I read ´╗┐cubeRootVolDelta.C
´╗┐ const Vector<label>& directions = mesh().directions();
label nD = (directions.nComponents + cmptSum(directions))/2;

if nD is neither 2 nor 3, then the error will popup. And I think it might be the problem of the mesh.

2. how could i set the proper entry for delta smooth,
if i switch delta to smooth can help solve the problem?

Thank you in advance.

eugene December 16, 2005 14:45

For delta smooth; you wan
 
For
delta smooth;

you want

smoothCoeffs
{
delta cubeRootVol;
cubeRootVolCoeffs
{
deltaCoeff 1;
}
maxDeltaRatio 1.1;
}

This is a nested structure.
You cant derive smooth delta from smooth delta. You have to derive it from a more basic type, like cubeRootVol.

vvqf December 19, 2005 04:40

Thank you very much for the re
 
Thank you very much for the reply.

I took the turbulenceProperties file from the tutorial case "oodles/pitzDaily". At the very beginning to avoid the "Case is not 3D or 2D" error, I tried to apply delta smooth and I just changed one line at the top of the turbulenceProperties file:
//delta cubeRootVol;
delta smooth;
But I still got the same error "Case is not 3D or 2D".

So later I guessed there might be more to change. Then I just roughly substituted all "cubeRootVol" with "smooth". -- That's the mistake you pointed out above. But after reading your reply, my understanding is you meant I just need to change delta cubeRootVol;
to
delta smooth;
as i did before.
However, the mistake remains.

Do you think it's the geometry problem of my mesh ? Or I did any mistake in using delta ?

Sorry to make the post long, but I want to state my problem clear. And as you can see i do not have this experience.

Expecting responding..
Thanks.

eugene December 19, 2005 07:02

Yes. Have a look at /users/s
 
Yes. Have a look at
/users/software/OpenFOAM/OpenFOAM-1.2.1/src/LESmodels/LESdeltas/cubeRootVolDelta /cubeRootVolDelta.C
from line 45 to 77
Also
/users/software/OpenFOAM/OpenFOAM-1.2.1/src/OpenFOAM/lnInclude/fvMesh.H
line 87

You probably have to many empty patches, that make the mesh think it is neither 2 or 3D.

Check your boundaries and make sure the only empty patches you have are the front and back of the the 2D geometry. If that still doesnt work, make the existing empty patches symmetry planes. This is bad practice and much more expensive, but it might help you find your problem.

vvqf December 19, 2005 09:00

Thank you so much for the time
 
Thank you so much for the timely responding.

I did as you suggested, and also with reference to the tutorial case "oodles/pitzDaily". -- I made a lot changes comparing with the initial case I got.
Now the case runs!

There are now convergent problem: extremely huge Mean and max Courant Numbers. :s
I now have to have a look into it.

Thanks again.

vvqf December 19, 2005 09:48

one more thing: the only empt
 
one more thing:
the only empty patches of my case are indeed the front and back of the 2D geometry.
Now I changed them into symmetryPlane.

Earlier I also made 'meanlingless' change of the boundary file.-- only to combine side1(front) and side2(back) into defaultFaces:
defaultFaces
{
type empty;
physicalType empty;
startFace 39176;
nFaces 39300;
}
still, FOAM FATAL ERROR : Case is not 3D or 2D, LES is not applicable.

PS:
in the polyMesh directory i dont have the <u>blockMeshDict</u> file. Well, it doesn't matter, I can still run the case, although a little difficult for me to understand the whole case.

ehsan March 2, 2015 13:15

Could anyone help me to understand smooth delta functionality on cubRootVolume detla? I checked the source codes but I did not understand.


All times are GMT -4. The time now is 21:12.