General Mechanical

General Mechanical

Spring in ansys workbench with hysteresis loop

    • Mike Andrew
      Subscriber

      Thank you very much for the people who collaborate.

      First of all, mention that I have already researched this topic quite a lot in the forum, I have already read the following posts relevant to this topic:

      https://forum.ansys.com/forums/topic/how-do-i-implement-combin40-for-springs-in-wb/

      https://forum.ansys.com/forums/topic/how-can-i-specify-a-minimum-load-to-longitudinal-spring-to-start-the-compression/

      https://forum.ansys.com/forums/topic/how-to-model-bilinear-spring/

       

      What I have to do is apply a load acceleration vs time

      To a structure that is supported on these springs, which as out, hysteresis curves of this style should be obtained.

       

       

      The load has to follow the following path to form the hysteresis form.

       

      1)

      2)

      3)

      4)

      5)

      6)

      And in this way complete the cycle, that is, when receiving loads of acceleration, force or displacement, to be able to obtain that cycle and hysteresis.

      So far I have been able to make the bilinear spring model, as explained in other posts, that is, I have been able to complete image 1 and 2 with deformation input, however, I do not know how to continue, I tried inserting damping in the code , but it doesn't work for me, I can't get stage 3 of the model, that is, when I unload the spring, it has that slope.

      I am using combin40, with the following test code

      et,_sid,COMBIN40
      !*
      Keyopt,_sid,3,3
      !GAPP=0
      K1=10000
      c=0.3
      K2=596
      FSLIDE=105000
      !*
      r,_sid,K1,C,,,FSLIDE,K2

      Without damping it would be like this.

      et,_sid,COMBIN40
      !*
      Keyopt,_sid,3,3
      K1=10000
      K2=596
      FSLIDE=105000
      !*
      r,_sid,K1,,,,FSLIDE,K2

      I am only performing the first tests with displacement input to calibrate the models, and then continue with forces and then insert it into the real model.

       

      So far I don't know what to do, any practical suggestion is welcome.

       

    • Sheldon Imaoka
      Ansys Employee

      Hi Mike,

      Instead of COMBIN40, please take a look at COMBIN39.  Even though COMBIN39 is used for multiple points, you can have a bilinear representation.  KEYOPT(1)=1 will allow for unloading on the original (initial) slope.

      Regards,
      Sheldon

Viewing 1 reply thread
  • You must be logged in to reply to this topic.