LS Dyna

LS Dyna

Topics related to LS-DYNA, Autodyn, Explicit STR and more.

*INITIAL_SHELL_STRESS with *ELEMENT_SHELL_OFFSET

    • Jérôme RENAULT
      Subscriber

      Hello,

      Further to my question regarding access to the AWG group pages, I'm getting the input files presenting a possible workflow for obtaining the initial stress state of a rotating engine part (described in § 5.1.1 and § 5.1.2 of the document LS‐DYNA® Aerospace Working Group "Modeling Guidelines Document" version 24.1).

      I'm doing the same thing with my model, but I'm unable to obtain a steady state for my rotating part at the end of the second step. The main difference is that my model includes *ELEMENT_SHELL_OFFSET elements. So, I simply modified my input file by replacing the *ELEMENT_SHELL_OFFSET elements with *ELEMENT_SHELL elements, and the behavior seems much better, except for the triangular elements, which still don't reach the steady state.

      Two questions at this point:

      1. Is it possible to use a prestress analysis following the workflow presented in the AWG document with offset shell elements and enabling additional options ? I need to use shell offset element as I will work with composite parts modelled with ACP.
      2. How do I handle the problem with triangular elements? (My part is a cone, which will be difficult to mesh with only quad elements, but not impossible, I admit.)

       

      Best regards,

       

      Jérôme

    • Jérôme RENAULT
      Subscriber

      To add some information to the topic ...

      I try the second procedure which involves the dynamic relaxation. I get the same issue with offseted elements. Nevertheless, once applied on non offseted elements, the results is acceptable with the triangle shaped elements (no stress concentration and/or discontinuities around) compared with the first procedure.

      By the way, the two questions of the first messages are still open.

       

      Best regards,

       

      Jérôme

    • igandiko
      Ansys Employee

       

      Hello, is it possible to share your files, perhaps through a support case? Have you tried running the latest version of the solver? Please check if adding drilling stiffness stabilization using DRCPSID in *control_shell helps (https://lsdyna.ansys.com/wp-content/uploads/2025/02/drilling-rotation-constraint-for-shell-elements-in-implicit-and-explicit-analyses.pdf). 

      Note that NLOC affects the stiffness of built-up section (overlayed elements). “[NLOC] Affects stiffness of a “built up” section. For example, two shells overlayed using the same nodal connectivity and NLOC=0 will have stiffness equal to the straight sum of the 2 shell stiffnesses. If they’re offset using NLOC=1 and NLOC=-1 respectively, the bending stiffness is increased in accordance with a net section thickness of thick1 + thick2.”

      If dynamic relaxation with non-offset triangular elements provide reasonable results, please check the difference in results at t=0 with DR phase vs. implicit pre-stress analysis. Since the pre-stress analysis is implicit, I would follow the implicit guildelines and also turn on implicit accuracy flag. Note that you can also run implicit dynamic relaxation in the 2nd procedure for comparison. 

       

       
    • Jérôme RENAULT
      Subscriber

      Hello Igandiko,

      I've tried your various suggestions (except a test with the last version, as I can only run 202R2 on my server) ... without success. 

      It's possible to share my files; please let me know how to proceed (maybe with customer support space page ?)

       

      Best regards,

       

      Jérôme

      • igandiko
        Ansys Employee

        If you're able to create a support case on Ansys customer support space, that would be ideal. Please let me know if you can.

        If you cannot create a support case, then please email the files to support@ansys.com (along with this forum link) and request them to forward the query to LS-DYNA team. Thanks! 

        • Jérôme RENAULT
          Subscriber

           

          Hello Igandiko,

          No problem, I open the case 00437219

           

          Best regards,

           

          Jérôme

           

    • Jérôme RENAULT
      Subscriber

      At the same time, I'm trying to find a solution on my own. Regarding point 2 and the triangular elements, activating the OSU option in *CONTROL_ACCURACY is a good solution. The triangular elements no longer evolve during the rotation phase until an external load is applied, of course.

       

      Best regards,

       

      Jérôme

Viewing 4 reply threads
  • You must be logged in to reply to this topic.