Kangaroo Problem- wrong surface output

Hello People,

I would like to have an output profile as shown in red but somehow kangaroo is having another output, can't figure out what´s wrong ?

Anyone please

Attachments:

1.jpg, 143 KB
kangaroo.gh, 6 KB
rhino file.3dm, 131 KB

Replies to This Discussion

Permalink Reply by Hyungsoo Kim on November 13, 2016 at 11:57am

Have a look at the attached.

Attachments:

Shell_FormFinding.gh, 17 KB

Permalink Reply by SA on November 13, 2016 at 3:19pm

Man , thank you so much

This is exactly what I was looking for

Permalink Reply by SA on November 13, 2016 at 3:37pm

Sorry was too promt to reply, guess my desired form is still NOT acheived or is it coz the centre anchor point is too High. The surface below the anchor point should have the same curvature as the corresponding surface above the anchor point or is it no so In practical, confused ;-(

Permalink Reply by peter fotiadis on November 14, 2016 at 1:35am

You may(?) find this attached useful (??).

Actually ... hmm ... not (due to code used instead of native stuff) but ... anyway.

Note: K1 099 on duty since every K2 2.1.2 related solution of mine these days is carried over solely via C#.

BTW: In fact a realistic solution should provide means to individually modify the nodes on the seed polyline but that's only doable via code. But alternatively you can reference some polyline and modify the nodes via gumball

BTW: Rhino file provided is for having some fun with instance definitions:

Attachments:

Polyline_to_Kangaroo_v3.gh, 170 KB
Polyline_to_Kangaroo_v3.3dm, 168 KB

Permalink Reply by SA on November 15, 2016 at 2:08am

Hey Peter,

Thanks a lot for the reply and a detailed definition, hufffff that took me sometime to go through it.

but errrr... if I may say so, its a little deviation from my question or may be the q itself wasnt clear.

In the definition and model I sent one can easily see anchor points and the output is varying as described in the section from what it would be if its made in fabric,

SO do you think taking the middle anchor points will solve the problem or like you said this approach is all togther wrong to the right solution ?

PS not so good with GH so dont know how to remove just the anchor points in the middle (the center ones only) and may be then it will have a free flow as described in the section with red pen.

Attachments:

1.jpg, 449 KB
2.jpg, 508 KB

Permalink Reply by peter fotiadis on November 15, 2016 at 3:20am

A membrane that is attached "linearly" to some sort of rigid members (as your images suggest) is a recipe for disaster (fabric tolerances etc etc). Try to avoid that route of thinking at any cost > you'll pay a very heavy price when attempting to do the thing in real-life. Unless of course you work in close collaboration with some top maker (BirdAir) ... but even so ... expect some tears.

Anyway there's a lot of bad installations of that type around (but as usual failures are kept hidden from publicity).

If the goal is to use some "side" envelope as well (thus ... on first sight ... the linear attachment appears "handy") there's other solutions that can assure proper/smooth tensile conditions and a proper "perimetric" join with the envelope.

BTW: Unless there's a cone (big/small, smallish) around (like the option in my example) there's absolutely no reason to use any "middle" anchor point. BTW: we NEVER use a single middle point so to speak ... even it appears so. In any case if the thing is small forget middle anchor points.

Permalink Reply by peter fotiadis on November 15, 2016 at 3:38am

BTW: this approach is what I would highly recommend for a "star" like deployment.

And this is what the linear attachment means