[nikwin]

Dear all,

I'm new to OF and trying to implement a Poisson solver to calculate a pressure field from a velocity field. For the simplest case, inviscid and incompressible that would be, laplace(p) = -div(div(UU)).

After reading U and setting pP (pressure to be derived) initially to p, using the cavity case in IcoFoam as a test case, the Poisson equation is implemented as

fvScalarMatrix pPEqn
(
fvm::laplacian(pP) == -fvc::div(fvc::div(phi, U))
);

pPEqn.setReference(pPRefCell,pPRefValue);
pPEqn.solve();

and solved with the GAMG solver. The problem is that I'm getting strange results. Is there anything wrong with the implementation? What is the difference with the following implementation, (which does not want to work)

fvScalarMatrix pPEqn
(
-fvc::div(div(phi, U))
);

solve(pPEqn == fvm::laplacian(pP));

Is there a direct way of setting the boundary conditions for pP to fixed values, something similar to boundaryField().types()?

Thanks
/NW

[ivan_cozza]

Hi Niklas,
did you success to implement your Poisson solver?
I'm managing to implement exactly the same solver, to get a pressure field from an incompressible velocity field. Do you obtain good results?
Thank you,
Ivan