[mcfdma]

I am trying to use dynamic mesh refinement on the damBreak laminar case.

I copied the dynamicMeshDict file into the constant and included

    
correctFluxes
    (
        (phi none)
        (nHatf none)
        (rhoPhi none)
        (ghf none)
    (flux(alpha.water) none)
    );

which stopped showing an error during the computation.

I run blockMesh, setFields and interDyMFoam.

After couple of time steps, it crashes with the error. Last part of the log file

Selected 50 cells for refinement out of 50484.
Refined from 50484 to 50834 cells.
Selected 32 split points out of a possible 6061.
Unrefined from 50834 to 50610 cells.
Execution time for mesh.update() = 0.7 s
DICPCG:  Solving for pcorr, Initial residual = 1, Final residual = 9.294e-06, No Iterations 221
time step continuity errors : sum local = 3.80198e-09, global = -4.05311e-11, cumulative = 0.00084144
smoothSolver:  Solving for alpha.water, Initial residual = 0.00141145, Final residual = 9.27232e-10, No Iterations 3
Phase-1 volume fraction = 0.130176  Min(alpha1) = -1.67578e-06  Max(alpha1) = 1
MULES: Correcting alpha.water
MULES: Correcting alpha.water
Phase-1 volume fraction = 0.130176  Min(alpha1) = -2.70158e-05  Max(alpha1) = 1
DICPCG:  Solving for p_rgh, Initial residual = 0.0615071, Final residual = 0.00273732, No Iterations 2
time step continuity errors : sum local = 0.000159835, global = -9.99339e-07, cumulative = 0.000840441
DICPCG:  Solving for p_rgh, Initial residual = 0.00378723, Final residual = 0.000157494, No Iterations 9
time step continuity errors : sum local = 9.10954e-06, global = -1.33525e-07, cumulative = 0.000840307
DICPCG:  Solving for p_rgh, Initial residual = 0.000898502, Final residual = 9.3117e-08, No Iterations 193
time step continuity errors : sum local = 5.38355e-09, global = 5.51327e-12, cumulative = 0.000840307
ExecutionTime = 699.25 s  ClockTime = 700 s

Interface Courant Number mean: 0.00970562 max: 0.746663
Courant Number mean: 0.0885415 max: 0.993028
deltaT = 0.000191888
Time = 0.321025

Selected 18 cells for refinement out of 50610.
Refined from 50610 to 50736 cells.
Selected 23 split points out of a possible 6049.
Unrefined from 50736 to 50575 cells.
Execution time for mesh.update() = 0.7 s
DICPCG:  Solving for pcorr, Initial residual = 1, Final residual = 8.91003e-06, No Iterations 210
time step continuity errors : sum local = 1.37086e-09, global = 6.65906e-11, cumulative = 0.000840307
smoothSolver:  Solving for alpha.water, Initial residual = 0.00141824, Final residual = 9.28577e-10, No Iterations 3
Phase-1 volume fraction = 0.130175  Min(alpha1) = -9.66966e-07  Max(alpha1) = 1
MULES: Correcting alpha.water
MULES: Correcting alpha.water
Phase-1 volume fraction = 0.130175  Min(alpha1) = -0.000515445  Max(alpha1) = 1
DICPCG:  Solving for p_rgh, Initial residual = 0.059735, Final residual = 0.00258301, No Iterations 2
#0  Foam::error::printStack(Foam::Ostream&) at ??:?
#1  Foam::sigFpe::sigHandler(int) at ??:?
#2  Uninterpreted: 
#3   at tensorField.C:?
#4  
 at ??:?
#5  
 at ??:?
#6  
 at ??:?
#7  
 at ??:?
#8  __libc_start_main in "/lib/i386-linux-gnu/libc.so.6"
#9  
 at ??:?
Floating point exception (core dumped)

I have attached the screenshot (Left) dynamic mesh refinement - interDyMFoam (right) just interFoam
I couldn't find anything on the error.

[simrego]

Hi!


Dynamic refinement is using the hexRef8 engine, so you cut every cell into 8. Thus you cannot use in 2D case, since you cut the cells in the 3rd dimension too.
I had similar problems with it in 2D. And I think it must be because of the 2D case, since in 3D cases is working for me without any problems. Try to replace the empty patches with symmetry, or with simple walls but slip for U, and zeroGradient p, etc...
Just a guess. I didn't spent too much time with it to find the solution.

[mcfdma]

Thanks for the reply. It gave me more idea about dynamic mesh refinement. I was able to find another forum which spoke AMR.
2D adaptive Mesh Refinement

The attachment from #20 has 2D as well as axi refinement - however, I am unable to compile it.

Trying to convert into 3D case as well.

[ViktorKL]

Originally Posted by simrego

Hi!


Dynamic refinement is using the hexRef8 engine, so you cut every cell into 8. Thus you cannot use in 2D case, since you cut the cells in the 3rd dimension too.
I had similar problems with it in 2D. And I think it must be because of the 2D case, since in 3D cases is working for me without any problems. Try to replace the empty patches with symmetry, or with simple walls but slip for U, and zeroGradient p, etc...
Just a guess. I didn't spent too much time with it to find the solution.

According to simrego's answer above, applying symmetry worked for me in order to resolve the error reported above.