CFD Online Discussion Forums - solving pressure equation

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)

- OpenFOAM Running, Solving & CFD (https://www.cfd-online.com/Forums/openfoam-solving/)

- - solving pressure equation (https://www.cfd-online.com/Forums/openfoam-solving/76233-solving-pressure-equation.html)

solving pressure equation

Hi all,

I am simulating wake vortices in ground effect and I'm using the icoFOAM application. I would like to know why it solve twice for p at each time step. I have no error message or warning just the following output when I'm solving.

Code:

Courant Number mean: 0.0258412 max: 0.99935

deltaT = 0.000909091

Time = 0.000909091



DILUPBiCG:  Solving for Ux, Initial residual = 0.00117738, Final residual = 8.40907e-08, No Iterations 4

DILUPBiCG:  Solving for Uy, Initial residual = 0.00100752, Final residual = 6.02822e-07, No Iterations 3

DICPCG:  Solving for p, Initial residual = 0.360239, Final residual = 9.95302e-07, No Iterations 982

time step continuity errors : sum local = 1.82785e-10, global = 4.3745e-21, cumulative = 4.3745e-21

DICPCG:  Solving for p, Initial residual = 0.00983221, Final residual = 9.57613e-07, No Iterations 872

time step continuity errors : sum local = 2.375e-10, global = -6.8944e-20, cumulative = -6.45695e-20

ExecutionTime = 40.92 s  ClockTime = 42 s

This increase a lot the duration of the simulation. Can anybody explain to me why it is like that.

Thank you,

Pascal

Hi Pascal,

"Solving for p ..." shows up twice because in icoFoam's standard setting the number of pressure corrections in the PISO loop is set to two. If you wish, you can change this at the bottom of the system/fvSolution file where it reads

Code:

PISO

{

    nCorrectors     2;

    nNonOrthogonalCorrectors 0;

    pRefCell        0;

    pRefValue       0;

}

But I would leave it at 2, or even increase it for less orthogonal meshes ...

While your at it, you should do some reading on how PISO works in Ferziger/Peric or the original paper by Issa and Olivera...

Thank you for all the information

Pascal