LS Dyna

LS Dyna

Topics relate to LS-DYNA, Autodyn, Explicit STR and more

Negative Jacobian – Fail

TAGGED: ,

    • Eray Kayar
      Subscriber

      Can you please advice me any further steps to overcome negative Jacobian fail? The following steps were tried to handle.

      The soft body impact FE model configured using SPH impactor and shell (UD facing and backing skins) & solid (honeycomb core) elements of a wing. The UD elements are formed using *MAT54/55 (Enhanced damage).

      • ELFORM changed to reduce integration. Shell element formulation is converted to 2 from 16 to reduce number of integration points when I got the negative Jacobian fail (The termination rows are added to this message). *SECTION_SHELL cards are deleted after creation of *PART_COMPOSITE. Then, the section is changed on this card.
      • Hourglassing formulation type & parameter
      • Number of plies increased
      • TFAIL, DFAILM, DFAILS, EFS parameters
      • Model clean-up
      • Decreasing timestep & remove mass scaling
      • Solver: 2022 R1

      Unit system: [t,mm,N,s]

      FE Model

      SHELL ELEMENT QUALITY CHECK SUMMARY:
      --------------------------------------------------------------------------------
      Quality Name        Min. val       Max. val       Allowable    #Violated(%)
      --------------------------------------------------------------------------------
      Min side length     4.12           5.44           3            0(0%)
      Max side length     4.77           8.5            30           0(0%)
      Aspect Ratio        1              1.85           10           0(0%)
      Warpage             0              0.128          10           0(0%)
      Min Quad Angle      88.3           90             45           0(0%)
      Max Quad Angle      90             91.7           135          0(0%)
      Min Tria Angle      0              0              30           0(0%)
      Max Tria Angle      0              0              120          0(0%)
      Taper               0              0.0143         0.7          0(0%)
      Skew                0              1.27           45           0(0%)
      Jacobian            0.911          1              0.6          0(0%)
      Char. length        4.12           5.46           1            0(0%)
      Area                20.6           45.6           5            0(0%)
      Feature angle       0              85             30           8192(4.55%)
      TimeStep            3.98e-07       5.26e-07       1e-07        0(0%)
      --------------------------------------------------------------------------------
      #QUADS(%):180224(100%), #TRIAS(%):0(0%), #TOTAL OF FAILED(%):8192(4.55%)
      --------------------------------------------------------------------------------

      Fail message from .d3hsp file:

       *** Warning 40098 (SOL+98)
           ALE time step size is less than dt2ms at cycle 20
           Mass scaling time step size of
           element id: 456411 has a dt: 5.8372E-07
           element id: 350107 has a dt: 5.8372E-07
           element id: 354939 has a dt: 5.8372E-07
           element id: 359771 has a dt: 5.8372E-07
           element id: 369435 has a dt: 5.8372E-07
           element id: 374267 has a dt: 5.8372E-07
           element id: 379099 has a dt: 5.8372E-07
           element id: 388763 has a dt: 5.8372E-07
           element id: 393595 has a dt: 5.8372E-07
           element id: 398427 has a dt: 5.8372E-07
       
       *** Warning 40101 (SOL+101)
           ALE time step size is less than dt2ms.
           Further output has been suppressed beyond cycle 21
       f a i l e d   e l e m e n t   r e p o r t  time=  1.1307E-05
           number of failed solid  elements = 0
           number of failed beam   elements = 0
           number of failed shell  elements = 1
           number of failed tshell elements = 0
           shell element 782030  failed at time  1.1307E-05

      Regards,

    • Andreas Koutras
      Ansys Employee

      Hello,

      On CONTROL_SHELL, could you set:

      NFAIL1=1 (or NFAIL4 for fully integrated shells)

      BWC=1

      PROJ=1

      Also please set INN=4 on CONTROL_ACCURACY.

      Let us know how it goes with these settings.

    • Eray Kayar
      Subscriber

       

      Hello Andreas,

      Thank you for reply. Unfortunately, the problem did not resolved by this advice.

      Regards,

      Eray

       

Viewing 2 reply threads
  • The topic ‘Negative Jacobian – Fail’ is closed to new replies.