algorithmic modeling for Rhino

Hi Everyone, 

I'd like to create a vertical louver mobile, but didn't work out. When I rotate then on facade, the rotation are not corect. Someone knows how make it works?

Thank you, 

Views: 1788


Replies to This Discussion

There may be better ways to orient those base planes for the rectangular stations used by 'Sweep 1'.  It can be frustrating.  I struggled a bit on this one myself, but I knew from experience that sweep would work on this problem, provided those end rectangles are "square" to the curve, on planes tangent to the curve and parallel to the base surface.  Until that's done, I would disable the slow 'Swp1' component.

on planes tangent to the curve

P.S.  I mis-spoke there.  Sweep station curves (rectangles in this case) are on planes perpendicular to the curve's tangent at each point.  'PFrames' in the latest effort ( ), re-aligned "parallel to the base surface".

Well, I couldn't let this go...  It nagged at me, so I tried a different approach.

Of the four edges on each subsurface, I was using the first one for the rail - 'List Item' w/ i=0.  Instead, I made 'i' an integer slider, 0 to 3, to choose any edge as the pivot axis (vertical or horizontal).  Then I used 'PFrames' on that edge and figured out a way to align them using the surface normal vector at each point ('Srf CP' and 'EvalSrf').  I had to use the base surface for that instead of the subsurfaces because surface normals are "messed up" at the edges.

Now it's very easy to have as many stations for 'Sweep 1' as you want, which makes it possible to have full length louvers without torsion wreaking havoc.  I replaced my edge slider with a feature to easily switch between Vertical and Horizontal, changing the UV dimensions as well as which edge (left or top) is used for the sweep rail.

There is a flaw in "louver width", causing the louvers to overlap slightly.  I understand why, and as I write this, an idea of how to fix it occurs to me...  But I'll leave that alone for now.

I left my "Tree/List Viewer" tool in place this time so you can see how I use it.  Three copies in this case, the second and third controlled by 'path idx' and 'list idx' sliders in the first one.  Yellow is the 'PFrame' tree, cyan is the surface normal frame and black is the aligned frame ('Align' output).  Being able to visually inspect geometry like that is really important for me in understanding GH!

There is still room for improvement...



Hi Joseph,

I adapt your algorithm because what I want is subdivide the surface and the result be like image attached. This way it's a good idea? because with what I did, the angle louver have problem again ;/

Is it possible subdivide surface on another way?

Thank you


I'm really done on this one, Camila, sorry.  






  • Add Photos
  • View All

© 2020   Created by Scott Davidson.   Powered by

Badges  |  Report an Issue  |  Terms of Service