Y branches for parametric bike frame model

I am in the process of using Grasshopper to parametrically model a bike frame. A bike frame clearly has branching tubes. How can I parametrically model an intersection between three tubes in a smooth, organic way? I had hoped to use/learn about the t-splines plugin to do this but I now know this plugin has been canned. Is there an equivalent alternative to t-splines for Rhino or can somebody basically advise how I could achieve smoothly intersecting tubes with Grasshopper please?

Edwin

Replies to This Discussion

Permalink Reply by David Rutten on December 21, 2016 at 6:36am

You can use Weaverbird and Sub-D meshes, which is at least an approach that should carry over reasonably well to upcoming Sub-D surfaces in Rhino6. To a degree you can already test Rhino6 SubD stuff if you sign up for the Rhino Wip via Serengeti.

Permalink Reply by Edwin Stokes on December 21, 2016 at 7:29am

Thanks for your advice. I'll download and start learning how to use Weaverbird. I may come back to you asking for a bit more help, if OK...

Permalink Reply by David Rutten on December 21, 2016 at 4:55pm

This is Rhino6 WIP.

Permalink Reply by David Rutten on December 22, 2016 at 3:45am

It's not quite the same T-Splines as I understand it. Sub-Ds in Rhino6 are new types of objects based on nurbs-patches. It doesn't actually subdivide into meshes (unless you want it to), rather it immediately creates the analytic limit surface using nurbs patches. It's fast, it's exact and it's compliant with sub-d standards widely used elsewhere.

Permalink Reply by David Rutten on December 22, 2016 at 3:49am

I find it somewhat disappointing that a SubD surface based on a square control grid doesn't result in a perfect circular cross-section. I don't actually know if this is mathematically even doable, but even if it was we couldn't take this route because it would make Rhino subds behave differently from other subds. Still, to my mind, that almost makes them useless for anything except organic stuff.

Permalink Reply by Laurent DELRIEU on December 21, 2016 at 1:56pm