11 Replies Latest reply on Apr 18, 2018 12:32 PM by Bill McEachern

    Metal (sheet) composite

    Will Griffiths

      Hi,

       

      I'd like to compare deflections for a range of metal sheet thicknesses (say 0.3mm, 0.5mm, 0.8mm) that are bonded together to give a total thickness of around 5mm. The length and width of the sheets are 925mm x 470mm. From the available tutorials I'm not clear on the best way to do this, please could someone advise.

       

      Thanks

       

      Will

        • Re: Metal (sheet) composite
          Ryan Dark

          Hi Will,

          If the layers are using all the same material then you could assume that the object is just 5 mm thick and use a Shell Mesh to run the analysis (GoEngineer - Simulation: Sheet Metal and Shell Mesh)?  If the layers are not using the same material, and you have access to Simulation Premium you could make use of Composite Shells (SOLIDWORKS Simulation: Composite Shells) to define the various thicknesses and materials of each layer to also run this analysis.

            • Re: Metal (sheet) composite
              Will Griffiths

              Thanks Ryan.

               

              Am I missing something, since I will be bonding the sheets (all to be 1060 Alloy), I would expect the assembly to be stiffer than a solid sheet of the equivalent thickness..... Would the composite shells option still be most appropriate?

                • Re: Metal (sheet) composite
                  Ryan Dark

                  Bonded layers of aluminum sheet would only be stiffer if the bond material itself had a higher elastic modulus than the aluminum and makes up an appreciable thickness of the composite you are making.

                   

                  If, like you describe, you are using all one material, then I might expect the composite of layers to be about the same stiffness as one uniform sheet of the same overall thickness.

                    • Re: Metal (sheet) composite
                      Will Griffiths

                      Thanks Ryan, of course, this makes complete sense. I'm rather rusty on the structural side!

                       

                      In practice, my actual design comprises four 1060 alloy sheets (0.8mm, 1.5mm, 0.8mm, 0.8mm) bonded together. The 1.5mm sheet features several large cut outs (a distance from the perimeter), which will affect things. Would Composite Shells still be the best way? I have seen video tutorials that use a split-line to split metal sheets and ensure sufficient meshing.

                      • Re: Metal (sheet) composite
                        Simon Yang

                        "If, like you describe, you are using all one material, then I might expect the composite of layers to be about the same stiffness as one uniform sheet of the same overall thickness."

                        "In practice, either using a shell with total thickness or composite shells with individual thicknesses would yield the same results.  The choice on which to go with is up to you."

                        ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

                        Above experience (statement) is correct if the sheet metal (material) is Isotropic; however, for some sheet metal materials, such as aluminum, titanium, etc. it may be not right.

                        As the anisotropic effect of metallic materials, in practice, it can be got that result, such as 1+1+1 > 3, for aluminum, titanium.

                          • Re: Metal (sheet) composite
                            Will Griffiths

                            Thanks Simon - So I'm clear, the material's grain could/should have an effect, but is this something that actually can be reflected in SW Simulation?

                              • Re: Metal (sheet) composite
                                Simon Yang

                                So sorry, I do not know how to do FEA for anisotropic sheet metal.

                                There is a similar question regarding "Anisotropic Modeling", from Mark Kooy @7-Jun-2016 7:53 PM. But no answer.

                                Here is a message from SolidWorks help:

                                Elasticity Models

                                The behavior of a material is said to be elastic when its displacement is linearly proportional to the applied load, and returns to its undeformed state when the load is removed.

                                For elastic materials, stress is directly proportional to the strain as indicated by: 

                                {σ} = [D] {ε -  εt}

                                [D]  = elastic (isotropic, anisotropic, or orthotropic) material or material stiffness matrix

                                {σ} = the total stress vector

                                {ε} = the total strain vector

                                t} = thermal strain vector

                                The above rule is an approximation that is valid as long as the strains are small for certain types of materials. The above equation is known as constitutive relation. While they take this simple form for linear elastic materials, the constitutive relations can be quite complex for nonlinear materials.

                                Looks SolidWorks Simulation can do it by using customer material, maybe try it.

                                Wish SolidWorks Simulation experts give some clue.

                                  • Re: Metal (sheet) composite
                                    Bill McEachern

                                    My two cents worth: I would use the laminated shell element but you could put in a stack of solids to do the same thing using an orthotropic model for each material and put in the orientations. In my experience the variation in properties is not likely to be large in metals from rolling or extruding so the response change is likely to be small - proportional the variation in properties. But you may have some specialized material that has a large change or the small change in response is what you are after. If the small response matters then you may have to put in the adhesive layers as well and that would be best done with the laminated shell element to avoid meshing issues.