TAGGED: contact, element-distortion, shell
-
-
September 21, 2023 at 1:22 pmNoahSubscriber
Hello,
I am simulating the deformation of a metallic tube in the low velocity range.
Unfortunately, large distortions occur at the shell elements (see figure).Â
These stretch to infinity and the voltage drops to 0.
These distortions occur during self-contact.
Attempts to counteract this by applying Hourglass were not really successful.A short excerpt from the K-File:
*CONTROL_HOURGLASS
$#   ihq     qh Â
     8    0.03
*CONTROL_SHELL
$#  wrpang   esort   irnxx   istupd   theory    bwc   miter    proj
   20.0     1     -1     0     16     2     1     0
$# rotascl   intgrd   lamsht   cstyp6   thshel   Â
    1.0     0     0     1     0
*CONTACT_AUTOMATIC_SINGLE_SURFACE_ID
$#   cid                                 title
     1Self_Contact
$#  surfa   surfb  surfatyp  surfbtyp  saboxid  sbboxid    sapr    sbpr
    13     0     3     0     0     0     0     0
$#    fs     fd     dc     vc    vdc   penchk     bt     dt
    0.2    0.18    0.0    0.0    5.0     0    0.01.00000E20
$#   sfsa    sfsb    sast    sbst   sfsat   sfsbt    fsf    vsf
    2.0    1.0    0.0    0.0    1.0    1.0    1.0    1.0
$#   soft   sofscl   lcidab   maxpar   sbopt   depth   bsort   frcfrq
     1    0.2     0   1.025    3.0     2    100     1
$#  penmax   thkopt   shlthk   snlog    isym   i2d3d   sldthk   sldstf
    0.0     1     2     0     0     0    0.0    0.0
$#   igap   ignore   dprfac   dtstif   edgek   unused   flangl  cid_rcf
     1     0    0.0    0.0    0.0         0.0     0
*CONTACT_AUTOMATIC_SURFACE_TO_SURFACE_ID
$#   cid                                 title
     2dropmass
$#  surfa   surfb  surfatyp  surfbtyp  saboxid  sbboxid    sapr    sbpr
    13     10     3     3     0     0     2     0
$#    fs     fd     dc     vc    vdc   penchk     bt     dt
    0.2    0.18    0.0    0.0          0    0.01.00000E20
$#   sfsa    sfsb    sast    sbst   sfsat   sfsbt    fsf    vsf
    1.0    1.0         0.0    1.0    1.0    1.0    1.0
$#   soft   sofscl   lcidab   maxpar   sbopt   depth   bsort   frcfrq
     2    0.1     0   1.025    1.0     2     10     1
$#  penmax   thkopt   shlthk   snlog    isym   i2d3d   sldthk   sldstf
    0.0     0     0     0     0     0    0.0    0.0
$#   igap   ignore   dprfac   dtstif   edgek   unused   flangl  cid_rcf
     1     0    0.0    0.0    0.0         0.0     0
*CONTACT_AUTOMATIC_SURFACE_TO_SURFACE_ID
$#   cid                                 title
     3bottom
$#  surfa   surfb  surfatyp  surfbtyp  saboxid  sbboxid    sapr    sbpr
    13     11     3     3     0     0     0     0
$#    fs     fd     dc     vc    vdc   penchk     bt     dt
    0.2    0.18    0.0    0.0          0    0.01.00000E20
$#   sfsa    sfsb    sast    sbst   sfsat   sfsbt    fsf    vsf
    1.0    1.0         0.0    1.0    1.0    1.0    1.0
$#   soft   sofscl   lcidab   maxpar   sbopt   depth   bsort   frcfrq
     2    0.1     0   1.025    1.0     2     10     1
$#  penmax   thkopt   shlthk   snlog    isym   i2d3d   sldthk   sldstf
    0.0     0     0     0     0     0    0.0    0.0
$#   igap   ignore   dprfac   dtstif   edgek   unused   flangl  cid_rcf
     1     0    0.0    0.0    0.0         0.0     0Â
Maybe someone has an idea why this happens?
-
September 23, 2023 at 5:08 amUshnish BasuAnsys Employee
While we cannot comment on a specific model, I would suggest in general to make sure you don't have overlapping contacts. You may also find these recommendations helpful:
https://www.dynasupport.com/howtos/material/negative-volume-in-soft-materials
-
September 25, 2023 at 10:24 pmNoahSubscriber
I have tried several methods. So far, I have only been able to solve the problem by modeling the floor geometry as a rigid wall. So it seems to be due to overlapping contact formulations. Is there any other possible solution for overlapping contacts?
Greetings
-
September 25, 2023 at 10:49 pmUshnish BasuAnsys Employee
You can either use one single-surface contact, or define your contacts carefully so the contact surface pairs don't overlap
-
- The topic ‘Shell Element with large distortions’ is closed to new replies.
- LS-DYNA Installation Issues with Student Workbench 2024 R2
- LS-Dyna CESE SMP d vs MPP d solver
- CESE solver – Ignition mechanism
- Cross-coupled stiffness elements in LS-DYNA
- Mathematical model generation stuck at 10%
- About combine different unconnected body into one part
- CONTROL_REFERENCE_CONFIGURATION
- Tiebreak using Segment set for contact b/w 20 noded Hexahedral elements
- shape memory alloy material in LS-DYNA
- CESE combustion model
-
1236
-
543
-
523
-
225
-
209
© 2024 Copyright ANSYS, Inc. All rights reserved.