EM_CONTROL_MAGNET syntax issues

- April 2, 2026 at 9:55 pm
  
  troy.baker
  Subscriber
  
  I recently noticed something I didn't understand with the EM_CONTROL_MAGNET keyword. On page 385 of the keyword manual (https://lsdyna.ansys.com/wp-content/uploads/2025/04/LS-DYNA_Manual_Vol_III_R16.pdf) it mentions there are two inputs, both mcomp to turn on the keyword, and ncycm to set the number of cycles. So it would be written like this:
  *EM_CONTROL_MAGNET
  $# mcomp ncycm
  1 100
  However, when I use lsprepost to make a keyword file and include the EM_CONTROL_MAGNET keyword, there is no ncycm command, only the mcomp, and it is automatically written as:
  *EM_CONTROL_MAGNET
  $# mcomp
  1
  
  Moreover, with the syntax written by LSPP, the simulation will crash stating "forrtl: severe (164): Program Exception - integer divide by zero" after assembling Fem-Bem matricies, and building P mat, Q mat, and W mat through 100 percent. This may be a problem with the syntax, or maybe the rebuilding of magnetic vector of a deforming magnet happens too often and there is a threshold somewhere between NCYCM being undefined and 100 that allows the simulation to complete. If it helps, here is the EM chunk of my .k file
  
  *EM_PERMANENT_MAGNET
  $# id pid mtype north south hc
  1 1 1 2 1 21033
  *EM_CONTROL
  $# emsol numls - dimtype nperio - ncylfem ncylbem
  1 100 0 0 1 1
  *EM_CONTROL_MAGNET
  $# mcomp
  1
  *EM_CONTROL_TIMESTEP
  $# tstype dtcons lcid factor tsmin tsmas rlcsf mecats
  3 0.0 0 1.0 1.0E-4 0.25 25 0
  *EM_CONTROL_COUPLING
  $# thcpl smcpl thlcid smlcid thcplfl smcplfl
  1 2 0 0 0 0
  *EM_SOLVER_FEMBEM_MONOLITHIC
  $# mtype stype abstol reltol maxit
  0 1 1.0E-6 1.0E-4 100
  *EM_MAT_002
  $# mid mtype sigma eosid murel eosmu deatht
  1 4 0.0 0 1.1 0 1.0E28
  *EM_EXTERNAL_FIELD
  $# fieldid ftype fdef lcidx lcidy lcidz
  1 1 1 4 2 4
- April 3, 2026 at 7:42 am
  
  ErKo
  Ansys Employee
  
  Hi
  Some keywords are not always fully implemented and done in LSPP so that could be the case.
  
  Please feel free to add the smallest of parts/meshes/ em example here and users/members can perhaps look at it (just say like a cube with max 20 elements or so).
  
  Also state version used NCPU…etc.
  
  All the best
  
  Erko
  - April 3, 2026 at 9:45 am
    
    troy.baker
    Subscriber
    
    Ciao,
    I did not realize that all keywords were not fully integrated with lspp, I thought that was only an issue with the workbench environment.
    My body is 0.06x0.01x0.005 with mesh a uniform cube sizing 0.001x0.001x0.001 (a mm scale body in SI units). I am running Dyna 2025 R2 V4.13.2 with NCPI set to 1 and using SMP double-precision
    - April 3, 2026 at 10:05 am
      
      ErKo
      Ansys Employee
      
      If you add to LSDYNA Magnets snapping — Welcome to LS-DYNA Examples the below it works fine:
      
      see the help manual for ncycm (second variable below):
      
      *EM_CONTROL_MAGENT
      $# mcomp
      1 1
      Erko
    - April 3, 2026 at 10:55 am
      
      troy.baker
      Subscriber
      
      $# LS-DYNA Keyword file created by LS-PrePost(R) 2025 R2 (v4.13.2) - 10Jun2025
      $# Created on Apr-3-2026 (12:29:59)
      *KEYWORD
      *DEFINE_CURVE_TITLE
      Gravity
      $# lcid sidr sfa sfo offa offo dattyp lcint
      1 0 1.0 1.0 0.0 0.0 0 0
      $# a1 o1
      0.0 0.0
      0.003 3.2
      0.006 6.5
      0.009 9.8
      3.6 9.8
      *DEFINE_CURVE_TITLE
      Field
      $# lcid sidr sfa sfo offa offo dattyp lcint
      2 0 1.0 1.0 0.0 0.0 0 0
      $# a1 o1
      0.0 0.0
      0.1 -0.01
      1.2 -0.03
      1.7 -0.05
      3.6 -0.05
      *DEFINE_CURVE_TITLE
      Zero's
      $# lcid sidr sfa sfo offa offo dattyp lcint
      4 0 1.0 1.0 0.0 0.0 0 0
      $# a1 o1
      0.0 0.0
      3.6 0.0
      *LOAD_BODY_Y
      $# lcid sf lciddr xc yc zc cid
      1 1.0 0 0.0 0.0 0.0 0
      *CONTROL_IMPLICIT_AUTO
      $# iauto iteopt itewin dtmin dtmax dtexp kfail kcycle
      1 10 5 1.0E-8 1.0E-5 0.0 0 0
      *CONTROL_IMPLICIT_GENERAL
      $# imflag dt0 imform nsbs igs cnstn form zero_v
      1 1.0E-5 2 1 2 0 0 0
      *CONTROL_TERMINATION
      $# endtim endcyc dtmin endeng endmas nosol
      3.5 0 0.0 0.0 20.0 0
      *MAT_ELASTIC
      $# mid ro e pr da db not used not used
      1 1600.0 1318200 0.49 0.0 0.0 0 0
      *EM_PERMANENT_MAGNET
      $# id pid mtype north south hc
      1 1 1 2 1 21033
      *EM_CONTROL
      $# emsol numls - dimtype nperio - ncylfem ncylbem
      1 100 0 0 1 1
      $#-------------------------------
      $# This Part Here
      $#-------------------------------
      *EM_CONTROL_MAGNET
      $# mcomp
      1
      $#*EM_CONTROL_MAGNET
      $# mcomp ncycm
      $# 1 100
      $#-------------------------------
      *EM_CONTROL_TIMESTEP
      $# tstype dtcons lcid factor tsmin tsmas rlcsf mecats
      3 0.0 0 1.0 1.0E-4 0.25 25 0
      *EM_CONTROL_COUPLING
      $# thcpl smcpl thlcid smlcid thcplfl smcplfl
      1 2 0 0 0 0
      *EM_SOLVER_FEMBEM_MONOLITHIC
      $# mtype stype abstol reltol maxit
      0 1 1.0E-6 1.0E-4 100
      *EM_MAT_002
      $# mid mtype sigma eosid murel eosmu deatht
      1 4 0.0 0 1.1 0 1.0E28
      *EM_EXTERNAL_FIELD
      $# fieldid ftype fdef lcidx lcidy lcidz
      1 1 1 4 2 4
      *EM_OUTPUT
      $ matS matF solS solF mesh
      2 2 2 2 0
      *EM_DATABASE_PARTDATA
      $ outlv dtout
      1 1.16
      *NODE
      $# nid x y z tc rc
      1 0.0 0.0 0.0 0 0
      2 0.012 0.0 0.0 0 0
      3 0.024 0.0 0.0 0 0
      4 0.036 0.0 0.0 0 0
      5 0.048 0.0 0.0 0 0
      6 0.06 0.0 0.0 0 0
      7 0.0 0.0025 0.0 0 0
      8 0.012 0.0025 0.0 0 0
      9 0.024 0.0025 0.0 0 0
      10 0.036 0.0025 0.0 0 0
      11 0.048 0.0025 0.0 0 0
      12 0.06 0.0025 0.0 0 0
      13 0.0 0.005 0.0 0 0
      14 0.012 0.005 0.0 0 0
      15 0.024 0.005 0.0 0 0
      16 0.036 0.005 0.0 0 0
      17 0.048 0.005 0.0 0 0
      18 0.06 0.005 0.0 0 0
      19 0.0 0.0 0.005 0 0
      20 0.012 0.0 0.005 0 0
      21 0.024 0.0 0.005 0 0
      22 0.036 0.0 0.005 0 0
      23 0.048 0.0 0.005 0 0
      24 0.06 0.0 0.005 0 0
      25 0.0 0.0025 0.005 0 0
      26 0.012 0.0025 0.005 0 0
      27 0.024 0.0025 0.005 0 0
      28 0.036 0.0025 0.005 0 0
      29 0.048 0.0025 0.005 0 0
      30 0.06 0.0025 0.005 0 0
      31 0.0 0.005 0.005 0 0
      32 0.012 0.005 0.005 0 0
      33 0.024 0.005 0.005 0 0
      34 0.036 0.005 0.005 0 0
      35 0.048 0.005 0.005 0 0
      36 0.06 0.005 0.005 0 0
      37 0.0 0.0 0.01 0 0
      38 0.012 0.0 0.01 0 0
      39 0.024 0.0 0.01 0 0
      40 0.036 0.0 0.01 0 0
      41 0.048 0.0 0.01 0 0
      42 0.06 0.0 0.01 0 0
      43 0.0 0.0025 0.01 0 0
      44 0.012 0.0025 0.01 0 0
      45 0.024 0.0025 0.01 0 0
      46 0.036 0.0025 0.01 0 0
      47 0.048 0.0025 0.01 0 0
      48 0.06 0.0025 0.01 0 0
      49 0.0 0.005 0.01 0 0
      50 0.012 0.005 0.01 0 0
      51 0.024 0.005 0.01 0 0
      52 0.036 0.005 0.01 0 0
      53 0.048 0.005 0.01 0 0
      54 0.06 0.005 0.01 0 0
      *PART
      $# title
      beam
      $# pid secid mid eosid hgid grav adpopt tmid
      1 1 1 0 0 0 0 0
      *SECTION_SOLID
      $# secid elform aet - - - cohoff gaskett
      1 1 0 0.0 0.0
      *ELEMENT_SOLID
      $# eid pid n1 n2 n3 n4 n5 n6 n7 n8
      1 1 1 2 8 7 19 20 26 25
      2 1 2 3 9 8 20 21 27 26
      3 1 3 4 10 9 21 22 28 27
      4 1 4 5 11 10 22 23 29 28
      5 1 5 6 12 11 23 24 30 29
      6 1 7 8 14 13 25 26 32 31
      7 1 8 9 15 14 26 27 33 32
      8 1 9 10 16 15 27 28 34 33
      9 1 10 11 17 16 28 29 35 34
      10 1 11 12 18 17 29 30 36 35
      11 1 19 20 26 25 37 38 44 43
      12 1 20 21 27 26 38 39 45 44
      13 1 21 22 28 27 39 40 46 45
      14 1 22 23 29 28 40 41 47 46
      15 1 23 24 30 29 41 42 48 47
      16 1 25 26 32 31 43 44 50 49
      17 1 26 27 33 32 44 45 51 50
      18 1 27 28 34 33 45 46 52 51
      19 1 28 29 35 34 46 47 53 52
      20 1 29 30 36 35 47 48 54 53
      *SET_NODE_LIST_TITLE
      NODESET(SPC) 1
      $# sid da1 da2 da3 da4 solver its -
      1 0.0 0.0 0.0 0.0 1
      $# nid1 nid2 nid3 nid4 nid5 nid6 nid7 nid8
      1 7 13 19 25 31 37 43
      49 0 0 0 0 0 0 0
      *SET_SEGMENT_TITLE
      South
      $# sid da1 da2 da3 da4 solver its -
      1 0.0 0.0 0.0 0.0MECH 0
      $# n1 n2 n3 n4 a1 a2 a3 a4
      7 1 19 25 0.0 0.0 0.0 0.0
      13 7 25 31 0.0 0.0 0.0 0.0
      25 19 37 43 0.0 0.0 0.0 0.0
      31 25 43 49 0.0 0.0 0.0 0.0
      *SET_SEGMENT_TITLE
      North
      $# sid da1 da2 da3 da4 solver its -
      2 0.0 0.0 0.0 0.0MECH 0
      $# n1 n2 n3 n4 a1 a2 a3 a4
      6 12 30 24 0.0 0.0 0.0 0.0
      12 18 36 30 0.0 0.0 0.0 0.0
      24 30 48 42 0.0 0.0 0.0 0.0
      30 36 54 48 0.0 0.0 0.0 0.0
      *BOUNDARY_SPC_SET
      $# nsid cid dofx dofy dofz dofrx dofry dofrz
      1 0 1 1 1 1 1 1
      *END
    - April 3, 2026 at 11:15 am
      
      ErKo
      Ansys Employee
      
      The above example runs fine with:
      
      *EM_CONTROL_MAGENT
      $# mcomp
      1 1
      
      But this keyword is never used I saw so you do not need (magnetization or scalar potential follows the elements so does not need to really be recomputed even if part bends)
      
      Used official LS-Dyna rel 16.1.1-20 (you stated the LSPP rel. that is not important, it is the solver below that is):
      
      | Version : mpp d R16 |
      | Revision: R16.1.1-20-g0c90cad538
      
      (I would not suggest using any other rel. for em analysis – only official rel R16 as the one stated above)
      All the best
      
      Erko
    - April 3, 2026 at 1:12 pm
      
      troy.baker
      Subscriber
      
      My institution has licenses for Ansys 2025r2, and probably wont update that until 2026r2. Currently only 2026r1 is released as far as I know. Is there that much of a gap between the EM keywords between the release 16.1.1 and the release that would come installed with a full 2025 r2 license?
      Solver output files state:
      
      | LS-DYNA, A Program for Nonlinear Dynamic |
      | Analysis of Structures in Three Dimensions |
      | Date : 05/14/2025 Time: 11:18:04 |
      | Version : smp d R14 |
      | Revision: R14.1.1-16-gfadfb49b17
      
      So i believe that is release 14.1, which would be the 2023 version. I previously used the 2023 R2 license and updated it a few months ago, maybe it only updated files related to workbench and not Dyna. Although the LSrun files are stored in Program Files/ANSYSInc/v252 where the rest of the Ansys 2025 version files should be kept instead of v232, where the 2023 version files should be. Is there a specific way to run an update for LS dyna under the license key i have without having to uninstall, clear all files, and reinstall the rest of my Ansys programs?
      
      I ask becasue once using the two argument based EM_CONTROL_MAGNET keyword, I run into error with
      *EM_OUTPUT_FORCES
      $# iout outdt lcoff sf
      1 0.0 0 1.0
      
      where it says the line relating to arguments is incorrect:
      
      *** Error 10246 (KEY+246)
      line contains improperly formatted data
      reading *EM_OUTPUT
      At line# 78 of file
      C:\Users\user\filepath\Testing.k
      --------------------------------------------------------------------------
      1 0.0 0 1.0
      --------------------------------------------------------------------------
      It references EM_OUTPUT but then the commands under the line for EM_OUTPUT_FORCES, so maybe the older release doesn't recognize the output_forces keyword
- April 3, 2026 at 2:01 pm
  
  ErKo
  Ansys Employee
  
  Do not use WB LS-Dyna (not supported only resistive heat solver is) for this and do not use R14.
  
  For EM always use official LS-Dyna rel 16.1.1-20 (you stated the LSPP rel. that is not important, it is the solver below that is):
  
  | Version : mpp d R16 |
  | Revision: R16.1.1-20-g0c90cad538
  
  (I would not suggest using any other rel. for em analysis – only official rel R16 as the one stated above)
  
  You can talk to your Uni and they can download and install R16.
  LSDYNA Download / Install Overview
  See here for credentials
  LS-Dyna License Manger – Network License Registration
  
  All the best – that is it for me – happy easter
  and good luck
  
  Erko
  - April 15, 2026 at 8:34 am
    
    troy.baker
    Subscriber
    
    Sorry for the delayed response, it took me a bit to organize updating the solver with R16 and my university license. I was focused on that instead of what you said when you mentioned
    "But this keyword is never used I saw so you do not need (magnetization or scalar potential follows the elements so does not need to really be recomputed even if part bends)"
    
    So by setting the magnetization in em-permanent-magnet to be segments of the mesh body, the solver automatically matches the magnetization vectors to that axis between those two segments, even when it starts bending. For my own reference, are there other versions of the em-permanent-magnet argument that would need to recompute the magnetization vector, or has the em-control-magnet keyword become obsolete for this scenario?

Viewing 2 reply threads

You must be logged in to reply to this topic.

LS Dyna

EM_CONTROL_MAGNET syntax issues

Edit Discussion