Question about the fvm::SuSp, and fvm::Sp
When I'm programming with the OpenFOAM, I got confused with the difference between fvm::SuSp, and fvm::Sp, could anybody tell me the difference? As far as I know, fvm::SuSp(a,b) will calculate a*b implicitly or explicitly according to the sign of a, however, when a is positive, is fvm::Sp explicit or implicit? fvm::Sp(a,b) will calculate a*b implicitly anyway.
The second question is related to a phenomenon I found when I'm solving the continuity equation, the exact form should be :
ddt(alpha) + div(phia, alpha) - alpha * div(phia) ==0
This supposed to be solving ddt(alpha) + U_a*grad(alpha) =0, and the implementation I made in the code is : (alpha is a scalar, volume fraction)
+ fvm::div(phia, alpha, scheme)
- fvm::Sp(fvc::div(phia), alpha, schemer)
where phia = fvc::interpolation(Ua) *mesh.Sf()
Interestingly, after this, I got that alpha does not change with time anymore, which means that the equation reduced to ddt(alpha) =0, which is clearly not true if we take the definition of fvm::Sp().
However, my worry is that, in the momentum equation, they are solving :
+ fvm::div(phia, Ua, "div(phia,Ua)")
- fvm::Sp(fvc::div(phia), Ua)
I'm wondering if this fvm::Sp( ) makes a difference between the scalar source and vector source. Thanks!!!!
Any idea would be appreciated!!
Sorry for the mistake I made in the continuity equation
I made a mistake, I think the exact form for continuity equation should be:
ddt(alpha) + div(alpha*phia) =0
not : ddt(alpha) + U_a*grad(alpha) =0
So, forget about the continuity equation I described, however, the question about the difference between fvm::SuSp() and fvm::Sp() is still valid, Thanks!
A bit late, but however...
A bit late, but however...this page on the wiki helps a lot in clarifying!
|All times are GMT -4. The time now is 00:59.|