Dual Layer Mesh / Lower Layer Trihex - Upper Layer hexagonal

Hello Everybody,

I'm trying to create a definition for a double layered gridshell with a tri-hex structure as lower layer and an hexagonal layer as upper layer, just like Grimshaw's Eden Project. At the moment it looks like this:

my problem is that i'm trying to extract the red-marked points from weaverbird's Stellate Module. Is there any opportunity to subtract a list of points, from another? If i could kill the superposed vertices from both modules i could get the points. someone has any solution?

Best regards, Leo

Replies to This Discussion

Permalink Reply by Jesus Galvez on March 13, 2013 at 5:14am

Hello Leo,

I'm also working on the same thing. I've done the upper hexagonal mesh with Lunchbox's components. Problem is that the hexagons are not planar. Are your upper hexagons planar?

Attachments:

HexagonCells 5 clean.gh, 18 KB

Permalink Reply by Leo Beltramo on April 5, 2013 at 9:11am

my hexagons aren't planar. A guy from my university helped me out with an Phyton Script.

Permalink Reply by Jesus Galvez on April 5, 2013 at 12:54pm

Phyton really? But do you need looping to do the inner truss?

Permalink Reply by Giulio Piacentino on April 1, 2013 at 7:21am

Hi guys

sorry gus I was very busy teaching a class. Can I still help with this?

Giulio

Permalink Reply by Jesus Galvez on April 1, 2013 at 1:36pm

Well I'm having two issues. First is that I could not get the dome to closed on itself, the keystone is left open.

The second is that I've used Kangaroo to flatten the cells, but I can't get the Laplacian force to work together with the Planarize force. I can only get one of them to work at a time.

So after I get over these issues I'll try to do the inner frame.

I'd much appreciate if you could help though.

Permalink Reply by alex jenter on April 4, 2013 at 7:15pm

i guess if you want planar hexagons they will certainly not be regular, its impossible to apply a closed regular hexagonal tesselation to a curved surface. But if almost planar is good enough, just a random thought, isn't the inner Layer like a dual of the hex layer? After applying dual you could find the connections via proximity. Not sure if WB dual even works on this kind of geometry but it could be worth a try

cheers A

Permalink Reply by Jesus Galvez on April 5, 2013 at 12:12pm

No it doesn't matter if they are not regular. Here is a video I did of what I am doing (you'll have to download it because by default it opens in Dropbox's player which plays it with a low quality). The problem I am having, is that with the setup I have in Kangaroo, when I have toroids or irregular shapes, i.e. shapes without symmetrical axes, I start getting weird things in the picture. Plus as can be seen in the video, the setup I have in Kangaroo distorts too much the original mesh.

The force I have in Kangaroo is the Planarize component (along with a Spring force). I think adding a Laplacian force will correct the problems I am having, but I can't seem to add both at the same time (only one at a time).

Permalink Reply by Leo Beltramo on April 5, 2013 at 1:09pm

so my first question is: do u want a relaxed hexagonal structure? or do you really want a radial distribution of vertices like the one in your video? If you want to have hexagons with nearly the same length, u should work with circlepacking.

best regards, Leo

RSS