[akidess]

It means that you are trying to solve for two variables within one equation.

[balkrishna]

Thanks. That is clear . However a formulation like below ,

fvScalarMatrix hEqn
    (
        fvm::ddt(rho, h)
      + fvm::div(phi, h)
      - fvm::laplacian(turbulence->alphaEff(), h)
     ==
        DpDt
    );

assumes constant k . How does one program without the constant k assumption ?

[RaghavendraRohith]

Hi BalKrishna

Have you found the solution for this problem. How can some body solve 2 variables here.

Regards
Rohith

[wadekar]

Dear sir,
I am getting the error after using Foam::tanh(...)

The error is: call of overloaded ‘tanh(const int&)’ is ambiguous
c[ignCell] =1-Foam::tanh(ignSite.cells()[icelli]);

The code is:

forAll(ign.sites(), i)
{
const ignitionSite& ignSite = ign.sites()[i];

if (ignSite.igniting())
{
forAll(ignSite.cells(), icelli)
{
label ignCell = ignSite.cells()[icelli];

c[ignCell] =1-Foam::tanh(ignSite.cells()[icelli]);
}
}
}

[wadekar]

Dear sir,
I am getting the error after using Foam::tanh(...)

The error is: call of overloaded ‘tanh(const int&)’ is ambiguous
c[ignCell] =1-Foam::tanh(ignSite.cells()[icelli]);

The code is:

forAll(ign.sites(), i)
{
const ignitionSite& ignSite = ign.sites()[i];

if (ignSite.igniting())
{
forAll(ignSite.cells(), icelli)
{
label ignCell = ignSite.cells()[icelli];

c[ignCell] =1-Foam::tanh(ignSite.cells()[icelli]);
}
}
}

[RaghavendraRohith]

Hi Sandip,

I hope tanh in openfoam takes only a value i.e if you define a dimensionedScalar then please define its value. See that there exists no Fields. I See Ignsites is a constant, i am not sure. If you can define it so.

Best Regards,
Rohith

[wadekar]

Thanks Raghavendra

I have tried with the interger value 5,
but still i am getting the same error:

call of overloaded ‘tanh(int)’ is ambiguous
c[ignCell] =Foam::tanh(5);

[wadekar]

Hello friends,
I am compile the below mention code and i am getting the error

error: ‘distance’ was not declared in this scope
c = Foam::tanh[distance];

 
pointField meshPoints(mesh.points());
       label i(0);
       
       forAll( meshPoints, kk)
       {   
        vector curPoint = meshPoints[kk];
        vector distance = curPoint - vector(0, 0, 0);
     
        i++;
    scalar dis = mag(distance);    
        Info<< "distance = " << dis << endl;
  
      }
      Info << i << "  Points are changed" << endl;
       
    volScalarField  c = Foam::tanh[distance];

[elham usefi]

Originally Posted by balkrishna

When i rewrite the enthalpy equation as :

fvScalarMatrix hEqn
          (
           fvm::ddt(rho, h)
           + fvm::div(phi, h)
           ==
            fvm::laplacian(k,T)
           + DpDt + heatSource
           );
//In the above , k is thermal conductivity , T is temperature , h is enthalpy .

I get the following error :

incompatible fields for operation 
    [h] == [T]

    From function checkMethod(const fvMatrix<Type>&, const fvMatrix<Type>&)
    in file /home/ifmg/OpenFOAM/OpenFOAM-1.7.0/src/finiteVolume/lnInclude/fvMatrix.C at line 1181.

How to overcome it ?

Hi i knw this post is old but i need help on this problem
in my case Cp varies in different locations, so I need to write energy equation with enthalpy. I wrote it like this:

   fvScalarMatrix EEqn
    (
        fvm::div(phi, E)
      - fvm::laplacian(k_n/cp, E)
     ==
        fvOptions(rho,E)
    );

    EEqn.relax();

    fvOptions.constrain(EEqn);

    EEqn.solve();

    fvOptions.correct(E);
    T=E/cp;
    rhok = (1.0 - beta*(T - TRef));

but I got unrealistic solution that diverges!
can u help me?

[RaghavendraRohith]

Hi,

The only error i see is that you have defined enthalpy as E=cp*T But it is
deltaE = cp*deltaT.

Hope it helps

Regards,
Rohith

[elham usefi]

Originally Posted by RaghavendraRohith

Hi,

The only error i see is that you have defined enthalpy as E=cp*T But it is
deltaE = cp*deltaT.

Hope it helps

Regards,
Rohith

HI!
thank u for ur response but I didnt get it! if we replace E with cp*T in

    (
        fvm::div(phi, E)
      - fvm::laplacian(k_n/cp, E)
     ==
        fvOptions(rho,E)
    );

the result would be like:

    (
        fvm::div(phiCp, T)
      - fvm::laplacian(k_n, T)
     ==
        fvOptions(rhoCp,T)
    );

so isnt EEqn correct?

actually first I tried TEqn like this:

    (
        fvm::div(phiCp, T)
      - fvm::laplacian(k_n, T)
     ==
        fvOptions(rhoCp,T)
    );

but I didnt get correct results!
but when I assume that cp is constant! like this:

    (
        cp*fvm::div(phi, T)
      - fvm::laplacian(k_n, T)
     ==
        fvOptions(rhoCp,T)
    );

I get good results! although the variation of cp is not very significant!
so I tried EEqn