View Issue Details

IDProjectCategoryView StatusLast Update
0000243OpenFOAMBugpublic2011-07-08 17:10
Reporteruser220Assigned Tohenry  
PrioritynormalSeverityminorReproducibilityN/A
Status resolvedResolutionfixed 
Summary0000243: Possible bug in SpalartAllmaras::DnuTildaEff()
DescriptionIs there perhaps a bug in the RANS Spalart-Allmaras turbulence models (both the incompressible and compressible versions) in the computation of DnuTildaEff()? Take the incompressible version as an example:

tmp<volScalarField> SpalartAllmaras::DnuTildaEff() const
{
    return tmp<volScalarField>
    (
        new volScalarField("DnuTildaEff", nuTilda_/sigmaNut_ + nu())
    );
}

From the SA equations one would expect that also nu() to be divided by sigmaNut_, i.e.

tmp<volScalarField> SpalartAllmaras::DnuTildaEff() const
{
    return tmp<volScalarField>
    (
        new volScalarField("DnuTildaEff", (nuTilda_ + nu())/sigmaNut_)
    );
}
TagsNo tags attached.

Activities

henry

2011-07-08 15:51

manager   ~0000526

It would not be correct to scale the laminar viscosity which relates to the laminar part of the stress and not the turbulence modeling.

user220

2011-07-08 16:05

  ~0000527

Ok, but according to eq (267) in the Ashford thesis the entire diffusion term should be scaled: $\frac{1}{\sigma} \nabla \cdot \left( \nu + \tilde{nu} \right) \nabla \tilde{nu} \right)$. This is indeed also how you implement it in the LES version:

 tmp<fvScalarMatrix> nuTildaEqn
    (
        fvm::ddt(nuTilda_)
      + fvm::div(phi(), nuTilda_)
      - fvm::laplacian
        (
            (nuTilda_ + nu())/sigmaNut_,
            nuTilda_,
            "laplacian(DnuTildaEff,nuTilda)"
        )
      - Cb2_/sigmaNut_*magSqr(fvc::grad(nuTilda_))
     ==
        Cb1_*STilda*nuTilda_
      - fvm::Sp(Cw1_*fw(STilda, dTilda)*nuTilda_/sqr(dTilda), nuTilda_)
    );

Cheers
Claes

henry

2011-07-08 16:16

manager   ~0000528

Sorry, the laminar nu for the laminar stress should not be scaled but can be scaled for the nuTilda equation according to the modeling. I will change the code to be consistent and push the fix. Thanks for reporting.

henry

2011-07-08 17:10

manager   ~0000529

Resolved by commit e8c6db8899c73bdb039eeb2cc6f8e6808b0ccbbb

Issue History

Date Modified Username Field Change
2011-07-08 15:45 user220 New Issue
2011-07-08 15:51 henry Note Added: 0000526
2011-07-08 15:51 henry Status new => closed
2011-07-08 15:51 henry Assigned To => henry
2011-07-08 15:51 henry Resolution open => fixed
2011-07-08 16:05 user220 Note Added: 0000527
2011-07-08 16:05 user220 Status closed => feedback
2011-07-08 16:05 user220 Resolution fixed => reopened
2011-07-08 16:16 henry Note Added: 0000528
2011-07-08 17:10 henry Note Added: 0000529
2011-07-08 17:10 henry Status feedback => resolved
2011-07-08 17:10 henry Resolution reopened => fixed