algorithmic modeling for Rhino
Im a mechanical engineering student and right now and researching about elastic and plastic deformation in 3D models printed in strained elastic clothes (as lycra or spandex).
In order to achieve high performing models, I'm simulating its behaviour through grasshopper, but I haven't achieved much. My goal is to give the simulator a shape (a circle, an square or an X for example) and receive the volume that it would produce, as how the cloth and the 3D printing will be deformed altogether, from the simulation (something like this: https://www.youtube.com/watch?v=a2fst73hA2A). I tried to do it with kangaroo physics and Kangaroo 2, but I only got to extend and deform with a "pre-made" force, like a distributed force through a line with the shape I wanted, sinusoidal for example (screenshots of the baking attached). Then I changed to Karamba, but I'm really lost since is the first time I'm using it.
In first place, I would like to undestand how to edit the beams' transectional area and its mechanical properties (I'm working with ABS and PLA plastic for the beams). In second place, I haven't been able to get together the cloth contraction thats generated when the cloth is released and the deformation of the beams that are joined to it.
Im attaching pictures of the physical prototypes too.
Thanks a lot.
a very interesting project!
I attached a definition how it could be done: 190724_ShellBeamContraction.gh
It gives however only the structure's response under the assumption of small displacements.
I couldn't make the large deformation calculation work. I think there is a numerical problem caused by the thin membrane: Karamba3D calculates it as a shell. The bending stiffness is very small as compared to the in-plane stiffness which results in an ill-conditioned stiffness matrix.
Thanks a lot Clemens!
I'll give it a try.
What would you do to use any figure to the model and it would automatically deform it? Right now I'm working with a figure much more complex than an X, and based on the definition you made, I haven't been able to reach the expected deformation.
in which respect do the results you get deviate from what you expect?
I don't get to deform a square for example. I tried changing the crossed lines wich you used by a square, following the same logic, but it doesn't deform but stills flat. Could you help me with that?
please find attached an example with a square cross section: 190804_ShellBeamContractionSquareCroSec.gh
Thanks a lot for your reply it was actually useful to understand more about "shaping" the beam. Anyway, I'm trying to do this: a deflection due tensile strenght in a rectangular shape. As you can see, it sags in a very particular way, which I'm trying to simulate.
It deviates from what I get in the simulation tho, but I still don't understand why.
The shape you get is probably partly the result of buckling with large displacements. In theory it could be analyzed using Karamba3D's 'AnalyzeNonlin'-component in combination with the arc-length method. In practice the component does not converge reliably in situations like this since it is still work in progress.