[Ahmed Khattab]

Dear Foamers,

i am running a case using simplePorousFoam, with total mesh of 900.000 cells, i think that OpenFOAM takes a long time to run the case. this is a sample for first iteration.

Time = 1

smoothSolver:  Solving for Ux, Initial residual = 1, Final residual = 6.71033e-10, No Iterations 34
smoothSolver:  Solving for Uy, Initial residual = 0.00118386, Final residual = 6.1944e-10, No Iterations 20
smoothSolver:  Solving for Uz, Initial residual = 1, Final residual = 4.55901e-10, No Iterations 32
GAMG:  Solving for p, Initial residual = 1, Final residual = 9.80576e-10, No Iterations 465
time step continuity errors : sum local = 1.5561e-06, global = 2.56885e-07, cumulative = 2.56885e-07
ExecutionTime = 114.83 s  ClockTime = 116 s

time of every step is decreasing in a slow rate, when it reached T= 5600 time became about 13 seconds. and there are residuals values.

Time = 5681

smoothSolver:  Solving for Ux, Initial residual = 2.17069e-07, Final residual = 4.33832e-10, No Iterations 10
smoothSolver:  Solving for Uy, Initial residual = 1.12953e-06, Final residual = 6.57395e-10, No Iterations 12
smoothSolver:  Solving for Uz, Initial residual = 5.88004e-07, Final residual = 3.57238e-10, No Iterations 12
GAMG:  Solving for p, Initial residual = 3.9218e-07, Final residual = 9.62315e-10, No Iterations 38
time step continuity errors : sum local = 1.74474e-07, global = 1.50349e-08, cumulative = 3.61903e-05
ExecutionTime = 35684.6 s  ClockTime = 43420 s

solver used are GAMG for pressure with tolerance 1e-09 and relative tolerance 0.0.
smooth solver used for velocity with same criteria.

[tomf]

Hi,

I think you should use a relative tolerance of 0.01 for pressure and 0.1 for velocity. There is no need to solve for the absolute tolerance in the beginning with the steady state solver.

You will see that you will have much less iterations for the linear solvers and the total run will take much less time.

Regards,
Tom