[rcastilla]

Hi,

I am doing a very large simulation with pimpleDyMFoam. The simulation has to stop after some time and then it is resumed with startFrom=latestTime. The odd behaviour is that the residuals (specially for cellMotionU) are much bigger for the same time after resuming than in the original simulation. The consequence is that pressure rises and finally simulation diverges and crashes.

I wonder if this behaviour is normal, or I am missing something important, and how can it be avoided. Has the previous timestep mesh velocity to be stored as it happens with U (U0)?

Just for the sake of illustration, these are the residuals for the first simulation:

DICPCG:  Solving for cellMotionUx, Initial residual = 3.014665e-05, Final residual = 7.21103e-06, No Iterations 1
DICPCG:  Solving for cellMotionUy, Initial residual = 2.564917e-05, Final residual = 6.150698e-06, No Iterations 1
DICPCG:  Solving for cellMotionUz, Initial residual = 5.70637e-05, Final residual = 9.442849e-06, No Iterations 3
AMI: Creating addressing and weights between 735136 source faces and 142500 target faces
AMI: Patch source sum(weights) min/max/average = 0.999999, 1.284449, 1.000723
AMI: Patch target sum(weights) min/max/average = 0, 1.000187, 0.1921055
AMI: Creating addressing and weights between 12818 source faces and 142500 target faces
AMI: Patch source sum(weights) min/max/average = 0.999999, 1.000001, 1
AMI: Patch target sum(weights) min/max/average = 0, 1.00001, 0.09399662
GAMG:  Solving for pcorr, Initial residual = 1, Final residual = 7.70526e-06, No Iterations 34
GAMG:  Solving for pcorr, Initial residual = 0.007424773, Final residual = 9.022955e-06, No Iterations 7
GAMG:  Solving for pcorr, Initial residual = 0.0007156682, Final residual = 7.523588e-06, No Iterations 4
GAMG:  Solving for pcorr, Initial residual = 0.0001249183, Final residual = 8.536026e-06, No Iterations 2
time step continuity errors : sum local = 1.287284e-11, global = -4.058237e-14, cumulative = 2.063061e-08
PIMPLE: iteration 1
smoothSolver:  Solving for Ux, Initial residual = 0.000446605, Final residual = 2.503277e-06, No Iterations 2
smoothSolver:  Solving for Uy, Initial residual = 0.0003534157, Final residual = 1.543715e-06, No Iterations 2
smoothSolver:  Solving for Uz, Initial residual = 0.0002096194, Final residual = 3.72596e-06, No Iterations 1
GAMG:  Solving for p, Initial residual = 0.01010752, Final residual = 4.525623e-06, No Iterations 15
GAMG:  Solving for p, Initial residual = 0.0008452819, Final residual = 6.474103e-06, No Iterations 4
GAMG:  Solving for p, Initial residual = 0.0001524509, Final residual = 6.10551e-06, No Iterations 2
GAMG:  Solving for p, Initial residual = 3.652736e-05, Final residual = 6.288872e-06, No Iterations 1
...

And this is for the same timestep, but after resuming (same time, same deltaTime)

DICPCG:  Solving for cellMotionUx, Initial residual = 0.9875883, Final residual = 9.365624e-06, No Iterations 45
DICPCG:  Solving for cellMotionUy, Initial residual = 0.8734927, Final residual = 8.890372e-06, No Iterations 45
DICPCG:  Solving for cellMotionUz, Initial residual = 0.8945875, Final residual = 9.334527e-06, No Iterations 42
AMI: Creating addressing and weights between 735136 source faces and 142500 target faces
AMI: Patch source sum(weights) min/max/average = 0.999999, 1.284449, 1.000723
AMI: Patch target sum(weights) min/max/average = 0, 1.000187, 0.1921055
AMI: Creating addressing and weights between 12818 source faces and 142500 target faces
AMI: Patch source sum(weights) min/max/average = 0.999999, 1.000001, 1
AMI: Patch target sum(weights) min/max/average = 0, 1.00001, 0.09399653
GAMG:  Solving for pcorr, Initial residual = 1, Final residual = 8.244299e-06, No Iterations 36
GAMG:  Solving for pcorr, Initial residual = 0.004357187, Final residual = 8.350664e-06, No Iterations 6
GAMG:  Solving for pcorr, Initial residual = 0.0003333046, Final residual = 7.19278e-06, No Iterations 3
GAMG:  Solving for pcorr, Initial residual = 5.795673e-05, Final residual = 4.281988e-06, No Iterations 2
time step continuity errors : sum local = 1.799065e-11, global = 1.149708e-13, cumulative = 1.149708e-13
PIMPLE: iteration 1
smoothSolver:  Solving for Ux, Initial residual = 0.007882085, Final residual = 1.447875e-06, No Iterations 1
smoothSolver:  Solving for Uy, Initial residual = 0.00347453, Final residual = 7.595343e-07, No Iterations 1
smoothSolver:  Solving for Uz, Initial residual = 0.05136888, Final residual = 8.64171e-06, No Iterations 1
GAMG:  Solving for p, Initial residual = 0.01248666, Final residual = 5.58013e-06, No Iterations 27
GAMG:  Solving for p, Initial residual = 0.002796536, Final residual = 5.829347e-06, No Iterations 25
GAMG:  Solving for p, Initial residual = 0.001070345, Final residual = 8.3443e-06, No Iterations 19
GAMG:  Solving for p, Initial residual = 0.0004921974, Final residual = 8.892418e-06, No Iterations 18
...

Hope somebody knows the answer to this mistery (at least for me).

With best regards

Robert

[dickcruz]

Hey Robert,

Did you run the simulation in parallel?

[rcastilla]

Hi, dickcruz,

Yes. Do you think it has something to do?

Robert

[dickcruz]

Well, it may not be the issue, but you might want to check that you're actually copying the latest timestep into your processor directories.