LS Dyna

LS Dyna

CONSTRAINED_BEAM_IN_SOLID

Tagged: 

    • Alireza
      Subscriber
      HinI am trying to simulate a solid component which has more than 150,000 beam elements in it. I used beam (truss element) for the beams and solid Hexa 8 and Tet 10 for the solid part. I have the following questions:n1) Does the CONSTRAINED_BEAM_IN_SOLID support tet10 element? In my case when I used tet10 and coupled them with the truss elements seems like the LS-DYNA cannot solve it? However, I tried it with tet4, hexa8, and hexa 20 and it works.n2) The CONSTRAINED_BEAM_IN_SOLID according to its formulation seems like only couple the velocity and acceleration so as a result the deformation of the solid element reinforced with the truss elements is controlled by the truss elements however seems like the resulted stresses in the truss and solid elements are totally different and separate and they cannot be reported when I use the contour to show the for example von Mises stress. What could be the problem here?n3) I cannot see the von Mises stress in the truss element when I go to the truss element section, so I was wondering what could be the problem?n4) For the simulation like this do you think CONSTRAINED_LAGRANGE_IN_SOLID would be a better choice? Because when I used that I did not get a suitable coupling. I set the ncoup as 2 which I guess does not make any difference in the results.nThanksn
    • Ushnish Basu
      Ansys Employee
      Hello,nCurrently CONSTRAINED_BEAM_IN_SOLID does not support 10-node tets, but truss element are supported at least if the PENALTY option is used. nTruss element is a resultant-type element so does not compute stresses. To get stresses, use a type 1 beam. Please see the Theory Manual for details.n The algorithm of CBIS is to distribute the mass*acc and mass*vel from the beam elements to the solid elements through the shape functions of the solid elements. Then the displacement of the beam nodes is interpolated from the solid nodes. (This conserves energy and momentum between the beam nodes and the solid nodes)nCONSTRAINED_LAGRANGE_IN_SOLID is not advised because of the reasons mentioned in the manual under CBISnIt is recommended to use the PENALTY option and NCOUP=0. The PENALTY method better conserves energy and momentum, and performs better for transient problems.nRegardsnn
    • Alireza
      Subscriber
      Hi,nI distributed many beam (type 1) (Beam diameter 50 micron, beam length 100 micron, beam stiffness 600MPa neoHookean) elements into a solid hexahedral elements (8-noded, Solid stiffness 0.8 MPa neoHookean), used constrained beam in solid with NCOUP=0. Then I applied velocity boundary to the solid matrix and read the X and Y stresses in the solid Matrix. I have the following question:n Although the beams are distributed in the X direction, when I read the X and Y stresses they are 100% the same? But the beams are in X direction so the resulted stresses in the X direction must be higher! I guess there is something wrong here! Then I removed the beams however the stresses in the X and Y directions in the solid matrix have not been changed. Seems like fibers have no effect on the resulted stresses and strains of the solid matrix. If we coupled them together wo we expect to see directional stiffness in the model but seems like fibers do not play any role in the resulted stresses and strains. Would it be possible to let me know what would be the problem here?nBestn
Viewing 2 reply threads
  • You must be logged in to reply to this topic.