[j_moulton]

Hi everyone,

I am trying to run my first case built from scratch and have run into a few problems that require some feedback. I created a 3D pipe in AutoCAD, used snappyHexMesh, and now I want to model laminar, incompressible flow through the pipe. The issues that I'm facing right now are that the Courant number blows up. I can get the solution to converge when deltat=0.5, but anything smaller than that crashes.

Here is the geometry:


checkMesh shows no problems: checkMesh

Here are the solver/schemes set up files:
fvSchemes
fvSolution

Here is the log file from running icoFoam (deltat = 0.05):
icoFoam.log
Here is some output from the terminal:

Time = 0.005

Courant Number mean: 0.000628106 max: 0.056034
smoothSolver:  Solving for Ux, Initial residual = 1, Final residual = 9.66498e-06, No Iterations 198
smoothSolver:  Solving for Uy, Initial residual = 1, Final residual = 9.72787e-06, No Iterations 208
smoothSolver:  Solving for Uz, Initial residual = 1, Final residual = 9.86911e-06, No Iterations 418
DICPCG:  Solving for p, Initial residual = 1, Final residual = 0.0312899, No Iterations 23
time step continuity errors : sum local = 0.80011, global = -0.00335702, cumulative = -0.00335702
DICPCG:  Solving for p, Initial residual = 0.0978647, Final residual = 7.77585e-07, No Iterations 39
time step continuity errors : sum local = 0.000242407, global = 4.4505e-07, cumulative = -0.00335657
ExecutionTime = 1.09 s  ClockTime = 2 s

Time = 0.01

Courant Number mean: 51.1202 max: 1136.82
smoothSolver:  Solving for Ux, Initial residual = 0.615202, Final residual = 5567.72, No Iterations 1000
smoothSolver:  Solving for Uy, Initial residual = 0.615546, Final residual = 5612.95, No Iterations 1000
smoothSolver:  Solving for Uz, Initial residual = 0.689541, Final residual = 41639.2, No Iterations 1000
DICPCG:  Solving for p, Initial residual = 1, Final residual = 0.0441477, No Iterations 80
time step continuity errors : sum local = 2.68406e+07, global = -27973.8, cumulative = -27973.8
DICPCG:  Solving for p, Initial residual = 0.000279914, Final residual = 7.89735e-07, No Iterations 86
time step continuity errors : sum local = 1.19323e+07, global = 692.433, cumulative = -27281.4
ExecutionTime = 4.9 s  ClockTime = 5 s

Time = 0.015

Courant Number mean: 9.89028e+09 max: 8.94025e+11
smoothSolver:  Solving for Ux, Initial residual = 0.849814, Final residual = 1.14434e+06, No Iterations 1000
smoothSolver:  Solving for Uy, Initial residual = 0.850397, Final residual = 26031.3, No Iterations 1000
smoothSolver:  Solving for Uz, Initial residual = 0.807674, Final residual = 4.24709e+06, No Iterations 1000
DICPCG:  Solving for p, Initial residual = 1, Final residual = 0.0325397, No Iterations 61
time step continuity errors : sum local = 2.2162e+17, global = 6.05571e+16, cumulative = 6.05571e+16
DICPCG:  Solving for p, Initial residual = 0.0024243, Final residual = 5.44929e-07, No Iterations 5
time step continuity errors : sum local = 3.98111e+24, global = 9.98492e+22, cumulative = 9.98493e+22
ExecutionTime = 8.69 s  ClockTime = 9 s

Time = 0.02

Courant Number mean: 4.77595e+26 max: 3.64552e+31
#0  Foam::error::printStack(Foam::Ostream&) at ??:?
#1  Foam::sigFpe::sigHandler(int) at ??:?
#2  ? in "/lib/x86_64-linux-gnu/libc.so.6"
#3  Foam::symGaussSeidelSmoother::smooth(Foam::word const&, Foam::Field<double>&, Foam::lduMatrix const&, Foam::Field<double> const&, Foam::FieldField<Foam::Field, double> const&, Foam::UPtrList<Foam::lduInterfaceField const> const&, unsigned char, int) at ??:?
#4  Foam::symGaussSeidelSmoother::smooth(Foam::Field<double>&, Foam::Field<double> const&, unsigned char, int) const at ??:?
#5  Foam::smoothSolver::solve(Foam::Field<double>&, Foam::Field<double> const&, unsigned char) const at ??:?
#6  ? at ??:?
#7  ? at ??:?
#8  ? at ??:?
#9  ? at ??:?
#10  __libc_start_main in "/lib/x86_64-linux-gnu/libc.so.6"
#11  ? at ??:?
Floating point exception

As you can see, the Courant number diverges in just a few iterations before producing the above error.

The full case is contained here: pipeTest2 case

Please help!

[Joshua14]

j_moulton,

icoFoam is a transient solver with a fixed time step. To accommodate Courant number issues all you can do is decrease your time step or create a coarser mesh.

Try dropping your deltaT to 0.005 or even 0.0005.

Best Regards,
Joshua