Inflated fabric coming apart K2

Grasshopper Kangaroo
1

Hi

i am trying to simulate an inflated fabric cover using K2.

i am not able to “stitch” all different piece of fabric…i tought that coincident points = stitched point but for some reason that i don’t know some of the stitching lines are not connected and panels are coming apart when subjected to pressure.

image
image1031×707 274 KB

is there a “sewing goal” meaning some sort of goal the keep connected coincident points from a group of adjacent meshes ?

20260907 FABRIC COVER INFLATION.gh (13.7 KB)

Is there a way to weld together adiacent meshes ?

thanks

2

Coincident points need to be within tolerance.

You can set it to 0.01 and it should work.

image
image1173×441 61.9 KB

image
image1477×1368 346 KB

3

many thanks !!! , they seemed to be coincident to me but eventually they weren’t you are right ,

i have also tried the MagnetSnap goal and adjusted the goals strenght and it seems to be more realistic., i think it means that points are close enough to be recognized as concident..doens’t it ?

image
image1119×725 321 KB

the bad side is that the deformed mesh topology is messed up

@martinsiegrist do you have any suggestion on a better way to fix it other than rising tolerance?

20260907 FABRIC COVER INFLATION 1.gh (18.3 KB)

i saw that there is a coincident goal but it works only with 2 points

thanks

4

What’s messed up with the topology?

5

sorry i didn’t post further details

when i run the simulation i notice some weird behavior , some edges seems to disappear

image
image1040×501 263 KB

6

Combine and clean is another way…

Btw you can also create the mesh on a plane and use the anchor target option

20260907 FABRIC COVER INFLATION_planar_target.gh (134.3 KB)

image
image1920×1237 243 KB

7

very smart apporach , thanks @martinsiegrist

can i ask you a further question ? I would like to add diagonals lines and add them to lenght goal in K2 to better simulate fabric behavior, i tried to deconstruct mesh and extracr faces corners but i am not able to get it to work.

would you take a look to attached definition and see what is wrong? thanks

20260907 FABRIC COVER INFLATION_planar_target- diagonal lines.gh (127.3 KB)

8

Kangaroo has a component to Diagonalize and another one which is called warp weft.

To make this out of fabric you’d also have to split the entire thing into manufacturable pieces…

9

thanks @martinsiegrist , it took me a while to get it to “work” but i still struggle to have a realistic behavior.

at first I tried to feed all lines ( edges+diagonals ) in to the same lenght goal ( same strenght ) but diag lines were still streching , especially in the corners.

i then tried to use 2 lenght goal with different strenght on diagonal in the hope to reduce stretch at corners but things got worse.

image
image850×768 313 KB

in the inner anchor points mesh deform and stretch quite realistically

image
image668×533 44 KB

but corners dont behave as expected

image
image443×505 53.5 KB

if you have any suggestion i appreciate. thanks

20260907 FABRIC COVER INFLATION_planar_target- diagonal lines 1.gh (128.5 KB)

10

More tension along the edge.

How are you actually going to inflate the fabric?

11

i tried to connect naked edges to same lenght goal as diagonals..those have strenght 2 orders of magnitude the panels edges..i still think the behavior is not realistic at all.

20260907 FABRIC COVER INFLATION_planar_target- diagonal lines 2.gh (131.2 KB)

well , i am not going to make this , the trial was just meant to simulate real ripstop fabric ( at least under a looking point of view) the same i use for kites , to arrange to strut was easier that making a kite or a sail model for istance.

in reality the bias properties are not 2 order of magnitude the warp weft properties ( the real behavior is actually the opposite ) ..i.e. it is correct to have alongation bias wise but not as much as we see in these examples

image
image775×624 201 KB

any suggestion ? maybe K2 is not the proper tool ?