CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM Programming & Development (http://www.cfd-online.com/Forums/openfoam-programming-development/)
-   -   understanding the term: fvm::Sp(fvc::div(phi), h) (http://www.cfd-online.com/Forums/openfoam-programming-development/68926-understanding-term-fvm-sp-fvc-div-phi-h.html)

dominik_christ October 6, 2009 11:34

understanding the term: fvm::Sp(fvc::div(phi), h)
 
Hello everyone,

when having a closer look at the energy equation as it is used by solvers for reacting cases, I cannot figure out where the term
- fvm::Sp(fvc::div(phi), h)
comes from. All I could find out is that it is a source term of some kind but when I try to derive an energy equation for enthalpy I do not get such a source term.

Could anybody please enlighten me? :-)

Thanks in advance!

Regards
Dominik

chegdan April 28, 2011 11:37

Quote:

Originally Posted by dominik_christ (Post 231649)
Hello everyone,

when having a closer look at the energy equation as it is used by solvers for reacting cases, I cannot figure out where the term
- fvm::Sp(fvc::div(phi), h)
comes from. All I could find out is that it is a source term of some kind but when I try to derive an energy equation for enthalpy I do not get such a source term.

Could anybody please enlighten me? :-)

Thanks in advance!

Regards
Dominik


Dominik,

I know this is an old thread, but I'm sure others run into it and have the same question. The origin of this term Sp(div(phi),h) comes from the expansion of the div(U,h) term in the transport equation.

div(U,h) = h*div(U) + U&grad(h).

In a completely converged domain the div(U) -> 0.

However, sometimes there is incomplete convergence and there is some generation (or consumption) that will throw off the energy balance. Have a look at

http://www.cfd-online.com/Forums/ope...silon-eqn.html

and

http://www.cfd-online.com/Forums/ope...culations.html

for maybe a little explanation.

Dan

kaifu May 7, 2011 05:53

Hi Dan, you mentioned that

Quote:

Originally Posted by chegdan (Post 305445)
In a completely converged domain the div(U) -> 0.

which I cannot understand. For example, in an incompressible flow we may have div(U) = 0 everywhere. But how could you apply it to all the other cases? And what is the definition of U in you opinion? For example in two phase flow, will it be the velocity of each component or a mixed velocity? Or maybe there's no physics principle for div(U) = 0 at all? Is it a kind of restriction from mathematics? If possible, can you give me some references? Thx

// Kai

kaifu May 7, 2011 08:14

Hi Foamers

again... about the div U =0... 'cause it's obviously correct in the simple single-phase incompressible flow.

However if it comes to two phase flow, where there is a phase change term on RHS of eq. for instance in alphaEqn.H, it looks like
Code:

            ddt(alpha)
        + div(phi, alpha)         
          ==
          Gamma/rhoa

It is a kind of 1st order differential equations. alpha is bounded only if we have div(phi)=0, as the equation required. It seems that we cannot find any proof of div(phi)=0 directly from physics. All we can find come from the mathematical requirement. Any comments?

//Kai


All times are GMT -4. The time now is 14:59.