Page 1 of 1

Converge on 1st iteration, then Nan in second?

Posted: 29 Jan 2013, 04:03
by einatlev
Hello
What are possible reasons that my model seems to converge OK in the first Navier-Stokes iteration, but then starts right away with NaN residuals in the second?

Here's the output:
....
: 25 0.1305E-05 0.1305E-05
: 26 0.5092E-06 0.5092E-06
: 27 0.8435E-07 0.8435E-07
: 28 0.1305E-07 0.1305E-07
: 29 0.2200E-06 0.2200E-06
: 30 0.2651E-08 0.2651E-08
: 30 0.2651E-08 0.2651E-08
ComputeChange: NS (ITER=1) (NRM,RELC): ( 12.751305 2.0000000 ) :: navier-stokes
FlowSolve: iter: 1 Assembly: (s) 176.00 176.00
FlowSolve: iter: 1 Solve: (s) 227.97 227.97
FlowSolve: Result Norm : 12.751304541532681
FlowSolve: Relative Change : 2.0000000000000000
FlowSolve:
FlowSolve:
FlowSolve: -------------------------------------
FlowSolve: NAVIER-STOKES ITERATION 2
FlowSolve: -------------------------------------
FlowSolve:
FlowSolve: Starting Assembly...
FlowSolve: Assembly:
: .................... 11%
: .................... 23%
: .................... 34%
: .................... 45%
: .................... 57%
: .................... 68%
: .................... 79%
: .................... 91%
: ...............Assembly done
DefUtils::DefaultDirichletBCs: Setting Dirichlet boundary conditions
DefUtils::DefaultDirichletBCs: Dirichlet boundary conditions set
FlowSolve: Dirichlet conditions done
CRS_IncompleteLU: ILU(2) (Real), Starting Factorization:
CRS_IncompleteLU: ILU(2) (Real), NOF nonzeros: 184618001
CRS_IncompleteLU: ILU(2) (Real), filling (%) : 480
CRS_IncompleteLU: ILU(2) (Real), Factorization ready at (s): 40.90
: 1 NaN NaN
: 2 NaN NaN
....
Thanks!
Einat

Re: Converge on 1st iteration, then Nan in second?

Posted: 14 Sep 2018, 04:59
by jtapas3761
Hi Einat,

Did you find any solution to this problem? I am also facing the same situation and have no idea why this happens !