Hello everyone,
I am new to use Precice and Calculix and with average user experience with openfoam.
I am trying to run my first FSI case, which is basically a round thin cylinder with inlet and multiple outlet on cyliunder surface.
I did follow some tutorials on Youtube to setup everything and all looks all right, excep that after i run the case both solid and fluid
I got an error in solid calculix simulation as follows:
no convergence
iteration 8
Using up to 1 cpu(s) for the symmetric stiffness/mass contributions.
Factoring the system of equations using the symmetric spooles solver
Using up to 1 cpu(s) for spooles.
Using up to 1 cpu(s) for the stress calculation.
Using up to 1 cpu(s) for the energy calculation.
average force= 113562.383880
time avg. forc= 39145.738939
largest residual force= 84220750.994938 in node 2018 and dof 3
largest increment of disp= 1.136432e-01
largest correction to disp= 5.589215e-02 in node 5999 and dof 2
no convergence
iteration 9
Using up to 1 cpu(s) for the symmetric stiffness/mass contributions.
Factoring the system of equations using the symmetric spooles solver
Using up to 1 cpu(s) for spooles.
Using up to 1 cpu(s) for the stress calculation.
Using up to 1 cpu(s) for the energy calculation.
average force= 122386.020614
time avg. forc= 42086.951184
largest residual force= 106961463.006889 in node 469 and dof 3
largest increment of disp= 1.136432e-01
largest correction to disp= 3.390082e-02 in node 6332 and dof 1
[CLILT-PF4EAT0N:03930] *** Process received signal ***
[CLILT-PF4EAT0N:03930] Signal: Segmentation fault (11)
[CLILT-PF4EAT0N:03930] Signal code: Address not mapped (1)
[CLILT-PF4EAT0N:03930] Failing at address: 0x7ffd0000002d
i looked into the error and i thought it is related to Yung modulus and possion which lead probably to high residual forces somewhere on the surface which leat to exist the simulation
so i played with the Yung modulus value and it has some affect on when the simulation stop
low yung modluls make the simulation stop at the early beggining in the first time step
I increased yung modulus very high and i got the simulation exist in the time step 2
the cylinder is really thin 1mm thickness and it should be elastic to some limit
I also thought the problem could be that Calculix is running as single core so i tried to update the run.sh with more than one core but it failed and didnt work, and probably it is not the isssue.
I also thought that problem could be related to some control limits about force residual but i never find any residual controls value written anywhere in the tube.inp file which is as follows:
*INCLUDE, INPUT=all.msh
*INCLUDE, INPUT=inlet.nam
*INCLUDE, INPUT=outlet.nam
*INCLUDE, INPUT=interface.nam
*MATERIAL, Name=EL
*ELASTIC
26000000000000.0, 0.1
*DENSITY
1000.0
*SHELL SECTION, Elset=Eall, Material=EL, OFFSET=0.5
0.0001
*STEP, NLGEOM, INC=1000000
*DYNAMIC,DIRECT
0.0001,0.01
*RESTART,WRITE,FREQUENCY=1
*BOUNDARY
Ninlet,1,1,0
Ninlet,2,2,0
Ninlet,3,3,0
Noutlet,3,3,0
*CLOAD
Ninterface,1,0.0
Ninterface,2,0.0
Ninterface,3,0.0
*NODE FILE
U
*EL FILE
S, E
*END STEP
I also read about memory allocation and multithread mpi problem on some threads but have no idea are they related to my case or no
any ideas or help how to solve the problem, your help is really appreciated as i am totally lost at the moment.