4 Replies Latest reply on Jul 7, 2016 8:14 PM by Peter Thannhauser

    Modeling organic cell patterns

    David Adam

      I would like to model a splint for a paralyzed hand. I have the hand scanned in and turned into a surface model. I have been making physical models but I want to make a 3d printed steel splint model that has and organic cell shape pattern. I know you can use textures and rendering to get that to happen visually but I want this to be in the actual physical model. I can go in and model each hole individual but I am wondering if anyone knows of a better way that looks more natural.

      Kind of like the picture I inserted here. The chair is the effect I would like on the splint on the hand. I also have modo 901 As software. Any input is welcome.
      CellChair.jpg

        • Re: Modeling organic cell patterns
          Chris Dordoni

          I don't know of any add in/plugin that can do this in SolidWorks.

           

          You could look at Rhino with Grasshopper, or Netfabb Selective Space Structures. Generally, meshes are produced as the result.

          • Re: Modeling organic cell patterns
            Jack Blake

            If i was going to take a stab in the dark at it i'd say surface the outer or inner form of your cast then make a load of sketch planes with random circles on and use the split line command to split the faces of the surfaces and delete out the circles then thicken the surface and fillet

              • Re: Modeling organic cell patterns
                Peter Thannhauser

                That's an cool looking surface & interesting problem. I'm just a SW hack, but I wonder if you could trick a design table into yielding the organic Swiss cheese matrix, and then (?somehow?) applying that to your surface. Crude idea/description/idea:

                - have a pre-made design table in Excel

                - use a random function to generate hole diameters between desired min/max boundary conditions you choose

                - the table has pre-populated X,Y coordinates that position the circle centers

                - the random diameters are populated to matrix of X,Y centers

                - SW now draws these for you automatically. Don't like it? Click randomize again!

                 

                Now, how you arrive at the pre-made center coordinate points is the tricky bit that may require experimentation. A simple Cartesian like (1,1), (1,2), (2,1), (2,2) would look very blah even with random circles. In my pic I'm visualizing centers based on intersection points of soft curvature splines. I guess a semi-orderly way of arranging the dis-orderly circles. Or maybe there is some fractal-type function you can write an equation with inside Excel. Anyway, the idea is whatever math you do inside Excel spits out design table cells & the design table auto-spits out a sketch & then (hopefully) that sketch can be the 'subtractor' to the surface.

                7-7-2016 0001.jpg

                  • Re: Modeling organic cell patterns
                    Peter Thannhauser

                    Hmm.. Examining the 'sea coral' up close with magnification makes me think simple circles will not yield anything nice looking. Scrap that idea. Unless the circles were somehow replaced by 'shapes' & even that would not look right. Maybe there is a way to digitize a large enough sample of the pic section so you could overlay that as the cutting surface? Maybe some combination of scaling, mirroring etc. could make more extended pattern that still looks random but without the manual pain of drawing a million spline curves?

                     

                    I think there has to be some fractal-type algorithm behind the shapes. The trick is how to get that expressed in a design table so it auto-draws a sketch?

                     

                    7-7-2016 0002.jpg