[philippebv]

Hi,

Daniel, it's awesome that you got results with DES and DDES for
channelOodles. What does it look like? Is the RANS region increased in
DDES? For IDDES, I personally have no interest in it since it's for
WMLES. But I would be glad to see if it helps with log layer mismatch.

For the boundary conditions, I have had a few concerns about them, and
I'm not quite sure that I impose nuTilda and nuSgs correctly. I ran
cases using, at the inlet, a Dirichlet condition for nuTilda and a
Neumann condition for nuSgs and it worked fine:

FoamFile
{
version 2.0;
format ascii;

root "";
case "";
instance "";
local "";

class volScalarField;
object nuTilda;
}

// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* * * //

dimensions [0 2 -1 0 0 0 0];

internalField uniform 0.0000693426;

boundaryField
{
inlet
{
type fixedValue;
value uniform 0.0000693426; // nu_t/nu = 1
}

outlet
{
type inletOutlet;
inletValue uniform 0;
value uniform 0;
}

wall
{
type fixedValue;
value uniform 0;
}
}

FoamFile
{
version 2.0;
format ascii;

root "";
case "";
instance "";
local "";

class volScalarField;
object nuSgs;
}

// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* * * //

dimensions [0 2 -1 0 0 0 0];

internalField uniform 1.5e-5;

boundaryField
{
inlet
{
type zeroGradient;
}

outlet
{
type zeroGradient;
}

wall
{
type zeroGradient;
}
}

Best regards,

Philippe

[gerado]

hi Hrv,
I am struggling since one week to understand why nuTilda have to be set in some case where the SpalartAllmaras modell is not used.
After a deep look in the src of all turbulence models ,I have founded ,that nuTilda is only defined in SpalartAllmaras model
SpalartAllmaras: nuTilda = nut/nu;
KEpsilon: nut = Cmu*k^2 /epsilon;

why do I have to set nuTilda when for example the KEpsilon model is used. why all turbulence model have to solve the nuTilda equation?


kind regards


gerard

[lin]

Originally Posted by gerado

hi Hrv,
I am struggling since one week to understand why nuTilda have to be set in some case where the SpalartAllmaras modell is not used.
After a deep look in the src of all turbulence models ,I have founded ,that nuTilda is only defined in SpalartAllmaras model
SpalartAllmaras: nuTilda = nut/nu;
KEpsilon: nut = Cmu*k^2 /epsilon;

why do I have to set nuTilda when for example the KEpsilon model is used. why all turbulence model have to solve the nuTilda equation?


kind regards


gerard

I guess Hrv could not answer your question.At least I could not reproduce this problem.I do not need to solve nuTilda when using kepsilon model.

[aldo.iannetti]

Hi
can you explane me the meaning of this error?


LHS and RHS of + have different dimensions
dimensions : [0 2 -2 0 0 0 0] + [1 -1 -2 0 0 0 0]

I got it in launching interDyMFoam in OF 1.5-dev,
It is a bug? how to fix it? I read I have (maybe) to recompile the solver, can you help me?


Thanks
Aldo

[matejfor]

That one I know!

the error clearly statest, your LHS and RHS have different dimensions, namely:
LHS has dimension of: m^2/s^2
and RHS has dimenstions of: kg/m/s^2. You don't need to recompile anything, only to check the dimensions of variables in your 0 directory. Once you set this properly, everything will run fine. There's also possibility to turn-off the dimension check, but it's not a good idea at all.

good luck
matej

[goma]

Hi at all

I have a similar problem. I want to solve a simpleFoam case with SpalartAllmaras. Can someone help me with this error?!

incompatible dimensions for operation
[U[0 1 -2 0 0 0 0] ] == [-grad(p)[1 -2 -2 0 0 0 0] ]

If I set up the case with an comressible solver (rhoPisoFoam) the error is similar.

thx