I am working on a GH file to develop a paneling system for a tensile fabric structure we plan to use as lost formwork for a light weight paper structure. We have been using very elastic fabric to create dynamic forms, but the elasticity allows for too much deflection once continued layers of pulp (liquid paper) are added. After a certain number of layers, the fabric stretches too far and begins to tear the paper layers. Our solution is to use a rigid fabric (burlap) and join it into our desired form prior to applying tension. So now we need a way to find the most efficient paneling system to be used with our rigid fabric formwork. I have been able to use kangaroo to simulate our tensile forms, but I am having a hard time figuring out how to flatten the mesh. If I could flatten the mesh, then I can manually draw the panels. I was able to create a nurbs surface and smash it, but then I get hundreds of surfaces from the UV grid. In my eyes the UV grid is not the most efficient way to panelize the formwork. Does anyone have any suggestions on how to flatten the mesh without the restriction of the UV grid?
I am still having some issues with this process. I am unable to Unroll the surfaces because they are doubly curved. Also, when I Smash the surfaces they deform by around a 14% surface area increase because they are not developable. This might be an issue I can resolve by reorganizing the UV grid in grasshopper. Regardless of developability, this should suffice for now. Thanks again.
Thanks Anders! The Stripper/Unroller tutorials by Piker are great. I think I could make something work with Stripper, but your first suggestion about making NURBS patches from a subdivided quad-mesh seems even more promising. Do you have any more links to share that might help me get started on this process?
One small adjustment I’d suggest to this is taking the strips diagonal relative to the isocurve directions.
This way they are closer to following the curvature direction, so the strips will be less twisted and should unroll with less distortion
@njwheele another way to experiment may be to intentionally use non-geodesic strips taken from the base mesh, since you’re not necessarily after precision, and may want to see how the paneling scheme influences your material form-finding. You can imagine breaking the mesh down into many different patterns - nested squares, L-shaped interlocking strips, etc.
The sewn-up results would no-doubt vary from your digital model, but if that’s welcomed, you could use this process to quickly model patterns and approximate the cut profiles.
We started working with more organic forms, so structuring the mesh for unrolling panels has become a little more complicated. I have been able to draw polygons based from the same point and loft between them to create meshes with holes in the center (see EX 1&2). But, the base curves I have been using to create this new form are parabolas. Is there a way to find points along the parabola based off the points along the circle (or vise versa)? I need a set of points that allow for a more direct path from the parabola to the center of the circle, and create a panel that is closer to a trapezoid with the top and bottom edge being parallel.
Also, I found out that the fabric I am using (burlap) cuts nicely on the laser cutter. The laser cutter bed is 24"x48", so I’m hoping to find a way to keep the unrolled strips under the width and length confines of the laser cutter bed. For size reference, this structure will being around 12’ tall.