Difficulty in manipulating pin-connected lattice structure accurately

Hello:

I have been trying for over a week now to accurately manipulate my rhino model, as if it were acting in the real built world, with tolerances and all. I have scoured the net for tutorials or examples of how to rotate rows of components around an ever-changing axis. The screenshots should help clarify this. I've also included two definitions that are basically a mess of components, just as a way of illustarting how I've been thinking about tacking this problem. I have switched between treating the problem as a surface manipulateable issue creating boxes and frames, and treating it as a simple nested rotation. I've gotten nowhere with either, and am wondering if I'm looking at this all wrong or overthinking it. I would like to be able to create a definition that functions based on my rotating individual rows of the units in one direction or the other, and be able to do the same for each string of them like I have done manually in Rhino in the curved wall screenshot. Any guidance is much appreciated, being a novice in GH.

Attachments:

joint.jpg, 1.2 MB
email1.jpg, 291 KB
newrotatingdef.ghx, 216 KB

Replies to This Discussion

Permalink Reply by taz on October 20, 2009 at 5:39pm

Kuser,

What you are attempting to do is not necessarily a simple problem. One thing that GH doesn't do very well is dimensional constraints. This is something that animation software does much better (IK handles and such) or even Solid Works (mechanical components).

That being said I don't think what you're trying to do is impossible in GH, but all your dimensional constraints need to be described mathematically. This can get tricky.

You might want to search for discussions about rigid origami or bar linkages. I think if you can create a lattice with the correct geometric reference points, then you could copy and orient your component onto the lattice.

The one thing I can suggest to get started would be to begin with a simpler 2D condition and then modify your definition to work for the more complicated 3D case.

-taz

Permalink Reply by Kuser on October 25, 2009 at 12:13pm

Thanks for the insight Taz. I have heard that maybe Solidworks is the way to go. I am interested in getting GH to do it, since theoretically it seems capable of it. I'll take your advice and simplify the problem to the essential 2D mathematical movements and see where it ends up. Thanks again for the help.

RSS