[Suya_94]

Hello All,


I am using overPimpleDyMFoam solver of OpenFOAM-ESI (v1812), where I deform the overset grid based on displacementLaplacian solver. I used the "oscillatingDisplacement" function in my pointDisplacement file with a given amplitude and omega. The oscillating body is a 2D square.


For anyone familiar with the overset functionality, based on the overset mesh deformation, my hole, interpolated and calculated cells will update at the new timestep. However, I am noticing that for my case, the time-step at which the cells update, I get this error:


Courant Number mean: 0.0138782082 max: 1.0227297
Time = 0.04

nonePBiCGStab: Solving for cellDisplacementx, Initial residual = 0, Final residual = 0, No Iterations 0
nonePBiCGStab: Solving for cellDisplacementy, Initial residual = 0.336278329, Final residual = 4.0984428e-07, No Iterations 4
inverseDistance : detected 2 mesh regions
zone:0 nCells:2401 voxels63 63 1) bb-2.50049995e-05 -2.50049995e-05 -2.50049995e-05) (20.000025 15.000025 0.500025005)
zone:1 nCells:1568 voxels63 63 1) bb4.99999432 5.49999432 -5.67890835e-06) (9.00000568 9.50000568 0.500005679)
Overset analysis : nCells : 3969
calculated : 3773
interpolated : 180 (interpolated from local:180 mixed local/remote:0 remote:0)
hole : 16

Mesh Courant Number mean: 0.000747932642 max: 0.621333533
PIMPLE: iteration 1


--> FOAM FATAL ERROR:
attempt to access element 0 from zero sized list

From function void Foam::UList<T>::checkIndex(Foam::label) const [with T = int; Foam::label = int]
in file /home/suyash/Downloads/OpenFOAM-v1812/src/OpenFOAM/lnInclude/UListI.H at line 157.

FOAM aborting

#0 Foam::error:rintStack(Foam::Ostream&) at ~/Downloads/OpenFOAM-v1812/src/OSspecific/POSIX/printStack.C:234
#1 Foam::error::abort() at ~/Downloads/OpenFOAM-v1812/src/OpenFOAM/lnInclude/error.C:282
#2 Foam::Ostream& Foam:perator<< <Foam::error>(Foam::Ostream&, Foam::errorManip<Foam::error>) in ~/Downloads/OpenFOAM-v1812/platforms/linux64GccDPInt32Debug/bin/overPimpleDyMFoam
#3 Foam::UList<int>::checkIndex(int) const in ~/Downloads/OpenFOAM-v1812/platforms/linux64GccDPInt32Debug/bin/overPimpleDyMFoam
#4 Foam::UList<int>:perator[](int) const in ~/Downloads/OpenFOAM-v1812/platforms/linux64GccDPInt32Debug/bin/overPimpleDyMFoam
#5 void Foam::dynamicOversetFvMesh::addInterpolation<Foam: :Vector<double> >(Foam::fvMatrix<Foam::Vector<double> >&) const at ~/Downloads/OpenFOAM-v1812/src/overset/dynamicOversetFvMesh/dynamicOversetFvMeshTemplates.C:383
#6 Foam::SolverPerformance<Foam::Vector<double> > Foam::dynamicOversetFvMesh::solve<Foam::Vector<dou ble> >(Foam::fvMatrix<Foam::Vector<double> >&, Foam::dictionary const&) const at ~/Downloads/OpenFOAM-v1812/src/overset/dynamicOversetFvMesh/dynamicOversetFvMeshTemplates.C:475
#7 Foam::dynamicOversetFvMesh::solve(Foam::fvMatrix<F oam::Vector<double> >&, Foam::dictionary const&) const at ~/Downloads/OpenFOAM-v1812/src/overset/dynamicOversetFvMesh/dynamicOversetFvMesh.H:282
#8 ? in ~/Downloads/OpenFOAM-v1812/platforms/linux64GccDPInt32Debug/bin/overPimpleDyMFoam
#9 ? in ~/Downloads/OpenFOAM-v1812/platforms/linux64GccDPInt32Debug/bin/overPimpleDyMFoam
#10 ? in ~/Downloads/OpenFOAM-v1812/platforms/linux64GccDPInt32Debug/bin/overPimpleDyMFoam
#11 ? in ~/Downloads/OpenFOAM-v1812/platforms/linux64GccDPInt32Debug/bin/overPimpleDyMFoam
#12 __libc_start_main in /lib/x86_64-linux-gnu/libc.so.6
#13 ? in ~/Downloads/OpenFOAM-v1812/platforms/linux64GccDPInt32Debug/bin/overPimpleDyMFoam
Aborted (core dumped)



I read few other posts where it was hinted that it could be associated with situation where an emptyPatch fields are accessed. However, I am not able to figure out why it happens when a mesh deformation solver (displacementLaplacian) is used in conjunction with overset solver.



I have been using overPimpleDyMFoam solver for many years, although with multiSolidBodyMotionSolver that has tabulated6DoFMotion functionality. I never noticed this error here.



Could it be associated with a probable bug in the code, or if anyone else encountered it?.



For further specifics. In overPimpleDyMFoam.C, it occurs as the momentum equation is solved (UEqn.H file, Line 21)


solve(UEqn == -cellMask*fvc::grad(p));


Please let me know if anyone has any suggestion to point out the issue or help in fixing it.


Thanks

[Suya_94]

Hello,


The issue is probably related to some minor bug in the code for v1812. I tried the same case with version released ahead of v1812, and the issue seems resolved.