5 Replies Latest reply on Nov 11, 2015 2:38 PM by Christopher Schaefer

    Simulation on lattice structure

    Shixuan Meng

      Has anyone ever simulate a Sparse Double Dense or honeycomb structure? seems that Solidworks underestimate these part's stiffness for my case 300 or 500 percent smaller than actual deformation. the calculation was realatively quick, but I guess Solidworks took some shortcut instead of using a lot of memory. Is there anyway to improve the accuracy? I used the smallest mesh size already.

       

      3D printer options

       

        • Re: Simulation on lattice structure
          Seckin Uslu

          If you use only solid elenments , you can use h-adaptive or p-adaptive methods

          • Re: Simulation on lattice structure
            Steven Darcey

            I prefer using the H-Adaptive, When you go  into the dialogue to "Create Mesh", you can click the drop down under Mesh Settings and type in a value. The trick here is to make sure you have at least 2 mesh elements across the thickness of your solid, to ensure that your part can create a strain through the thickness. Hope this helps!!

            Sim Mesh Settings.png

            • Re: Simulation on lattice structure
              Christopher Schaefer

              Approach 1:

              Use midplane extraction to create surfaces bodies of the lattice structure, then use shell elements.

              Downfall- many local bonded connections needed with increased error at the bonding sites between shell and solid elements. And a fair amount of investment in modeling time.

               

              Approach 2:

              Create a representative model of a portion of the lattice structure, mesh it as solids, and perform displacement-only analysis.  Be sure ot ploace a minimum of two high-qiuality saolid elements across the thickness/web direction.  Use these results to determine an equivalent stiffness and then apply that as a custom material to dummy lattice bodies in your original model. (can be solid "panels" even if they are thin so long as out-of-plane stresses are not too great)

               

              Approach 3:

              Simplify the lattice to exclude fillets or chamfers and use a very fine solid element mesh.

              Downfall- increased computational effort, especially if one is only looking for displacements.