Applying pattern without distortion on a curvature surface

Replies to This Discussion

Permalink Reply by peter fotiadis on March 19, 2016 at 2:37am

Is this Academic?

Other than that:

1. What are the materials of choice? (especially the red "connector": is it some elastic thingy bolted both sides?). Is FEA involved on this kiosk? (in order to avoid the obvious, he he).

2. Either attempt to get planar modules (via Kangaroo1/2). For instance K2 can quite effectively achieve planarity on quads > then your modules are rather easy to "derive" on a per quad basis.

3. Or (more complex in manufacturing) compose your modules out of planar "trapezes" (8 per module = 16 per double combo). This gives you far more freedom at the cost of assembling them properly. If (theoretically) these are made via some "large" scale 3d printing that is not a big issue ... but you don't have this option anyway.

BTW: within a scale of 1 to 10 rate your experience with GH.

Permalink Reply by KMK on March 19, 2016 at 2:49am

Yes it is academic. I am a student.

The material is going to be plywood. And the red connector is going to be steel. We are also planning to bend it by a machine. We are going to bolt it both sides. I am new in Grasshopper so I am not very experienced. If I can not produce the code I am going to draw it one by one :D 

Hexagons or octagons from trapazes is actually what I discussed with the instructors but Since I am very new on grasshopper I couldn't imagine the way I am going to produce the code.

Do you have any adivise or example of code ? 

Permalink Reply by peter fotiadis on March 19, 2016 at 4:16am

Well...

1. I would strongly advise NOT using steel as connectors (pain and treas on sight). Steel requires precision bending far far and away from what this little demo is about. If instructors told you that ... tell them that real-life and Academic thinking are a "bit" different. Use reinforced (sandwich: in the middle there's a nylon "tissue") elastic material that allows the tolerance liberties that are critical for effortlessly assembling that thingy. For having a rough idea on these > go to the automotive shops and ask to see the flexible belt/strap that drives the engine auxiliary systems (like the alternator, for instance) - or even the camshafts on certain modern engines. "Similar" materials are available in sheets as well in various thicknesses. 

2. For getting an idea about planarity matters ... see this attached that I've made for some fellow user using Kangaroo2. Pay attention to the C# that reports the "final" state of things: Academic planarity and real-life planarity are NOT the same thing. Note the critical role of the elastic connector as well (steel doesn't understand an iota from "liberties" on that matter).

3. For going from planar quads to planar octagons ... well ... I'll prepare a small demo later on.

Other than that this is NOT a task for a novice ... not to mention connectivity data [freaky trees, he he] that are obviously required for the assembly or layouts for making the modules .. but if you are persistent enough... well ...

Attachments:

• YAZ_kinda_demo_V2A.gh, 148 KB
• YAZ_kinda_demo_V1.3dm, 172 KB

Permalink Reply by peter fotiadis on March 19, 2016 at 4:46am

And this is rather more "entry" point on that rabbit hole matter (the planarity, that is).

Using Kangaroo2 and Lunchbox for making the modules.

Attachments:

• Planarity_on_Polygons_V1.gh, 136 KB

Permalink Reply by KMK on March 19, 2016 at 5:00am

thanks for your help. I will tell the instructors that material you mentioned. By the way ı found a rhino plug in called Panelling tools. Do you know if it is gonna work ? I am also going to work on the codes you send me a lot.. Thanks

Permalink Reply by peter fotiadis on March 19, 2016 at 6:07am

Well ... the "trad" way to do planar stuff (on meshes) is the Evolute Tools Pro thingy ("cheapo" as well). But our Daniel (a very smart fella I confess) delivered the goods with K2 so ... stick to Plan A.

Added a 5 minutes LOL C# code on the previous example: doing planar hex Breps out of non planar (worst case scenario, that is) hex cells.

Show this to the instructors and ask them: Boffins ... can you think of an adjustable "joint" (hint: a ball pivot one) that could allow us to hold all these in place? [and "by-pass" planarity: I do hope that Daniel doesn't read this, he he].

Attachments:

• Planarity_on_Polygons_V2.gh, 137 KB

Permalink Reply by Laurent DELRIEU on March 19, 2016 at 6:05am

I didn't look at Peter definitions (C# ? ;
). Here simple def for putting constant shapes (circle and square) on a surface. The next step will be to connect them and generate shape of connector. I will consider that holes are all the same on your panels. So the connectors will be uniques. Position of panels could surely be relaxed with Kangaroo. But At the moment I gave you 2 strategies to put panels. 

Attachments:

• flat_panel.gh, 20 KB

Permalink Reply by Laurent DELRIEU on March 19, 2016 at 2:34pm

I looked at the positions of the bending of connectors and they are not all between the panels. So don't look at my proposition !

Permalink Reply by peter fotiadis on March 20, 2016 at 4:15am

Well ... Laurent a quick question:

If you are familiar with K2 matters (if not don't read any further):

I'm fixing some (internal) stuff (only C# code including the K2 part) for planarizing any polygon ("on" surface Lists [open/closed]). In some occasions the Clamp Angle goal thingy is used. 

For a poly this goal accepts Line pairs (say: a given segment and the next one in the same segment order for all polys) and the upper/lower desired limit allowed with regard the derived angle.

I count the beans (the angle) in the trad matter: always in anticlckwise order with respect a fitted Plane (to poly vertices).

Apparently Daniel has other things in mind and LOL stuff occurs.

Notify if you have time to check the C# on that angle matter/goal (if yes provide your e-mail).

Permalink Reply by Nik Willmore on March 19, 2016 at 7:13am

Older start of an answer here, via Kangaroo2 goal to keep polygons flat:

http://www.grasshopper3d.com/forum/topics/non-twisted-box-or-module...