Inflating bubbles with bulges

Good morning community!
I am working on my own project on a vault that in some points of its edges are passable and that in certain points that follow a pattern swell more in the way of protuberances to accommodate a larger volume in its interior.

The project is intended to look like this volumetric study by the Spanish architectural studio AMID CERO 9.
I have tried to recreate with kangaroo the following model:
(I have marked the model in red. The image may look a bit wrong).

20220529_VATICANO-GAY_PlasticAnchor@DanielPikerPOST_v#000.3dm (1.8 MB) (28.1 KB)

@DanielPiker I hope you or someone else can help me because it is quite specific as I am looking for these specific shapes. The final image of the inflated mesh in plan and elevation should look like the following picture

Another problem I have with my meshes is the following inflation error, especially in the vertices of my perimeter. I show it below.

All the generated meshes that are based on VORONOI polygons as shown in the graphics I have made and subdivided them in 3Ds Max because with TRIREMESH and QUADREMESH I can’t get the edges of each polygon to match well. Below I attach my attempt to do it with these algorithms. Doing it with 3ds Max is hard work so is it possible to automate it? Thanks. (17.6 KB)

1 Like (22.0 KB)

1 Like

Thanks @HS_Kim but what I’m trying to do is that the curves that are anchors are not anchors in the ground plane at the beginning, but only once the mesh is inflated, that is to say that they form bulges once the mesh is inflated and I don’t know if this is done with PLASTIC ANCHORS and if so I don’t know how to use them properly.

I’m not quite understand what you want, anyway, you can move some of your anchor points to your desired targets something like this… (20.6 KB)

4 Likes (21.5 KB)

A few small improvements I’d suggest here-

If you want the seams to act like tighter cables in the surface, causing creases, you can do this more simply by getting these edges from the ‘C’ output of the remeshing and assign them different Length goals.

‘Volume’ is often a bit easier to work with than ‘Pressure’. The reason being that pressure maintains the same force per unit area, and when the membrane resistance is low, this can lead to a runaway effect where the shape expands, so the area increases, meaning more force, so greater expansion etc. and it explodes.
‘Volume’ instead makes the pressure dependent on the contained volume, avoiding this runaway effect.

I also added a ‘SphereCollide’ to stop the mesh passing through itself when it expands a large amount. It isn’t always needed, so you can sometimes disable this part to speed things up. For some of the shapes in your original image, the different ‘bulbs’ of the shape look like they are pressing against each other, and for these the self-collision will come in handy.

This also reminds me of Andrew Kudless’s work - see some previous discussion on this here

1 Like

Thank you very much @DanielPiker and @HS_Kim for your contributions! This is more or less what I’m looking for, I’m attaching an image where the black lines would be those wires on the surface.
I also attach an image of a Smiljan Radic project which is what I want to generate in case it serves better to illustrate, that kind of lines, wires, on the swollen surface.

Regarding your explanation I have several questions:
1º Which is better to use TriRemesh or QuadRemesh, to remesh the surface since in both there is the possibility of putting “GUIDES CURVES” [QuadRemesh] and “FEATURES” [Triremesh]. Which one adapts better to those curves that will be later the wires in the swollen surface?
2º_“Volume”, great I was hesitating to use it because I have seen in other inflation posts that you used this component instead of pressure. Thanks.
3º_“Lenght(Line)” Cables on the inflated surface. The only way to get these lines which is the output “C=Creases” of the TriRemesh is with this component. I know that in 3DsMAx when subdividing a mesh it is possible to select lines that will have “Crease” other than 0.0, but in grasshopper I don’t know very well how to select those lines from the mesh.
4º_“SphereCollide” understood, but “Self-Collision”, how to use it?
Forgive me for asking so many questions but these complex ways of inflating with different resistances of cable lines is quite new to me.

Here’s another one using weaverbird… (17.5 KB)


Thank you very much again for all the contributions made. You don’t know how grateful I am, but I have two other important and silly problems.
@DanielPiker I don’t know why in the next grasshopper file I contribute I always get the inflated in negative. I have changed it a thousand times:
-The direction of the curves.
-The normal of the network surface.
-The crv creators of the U and V of the network surface.
and I have not been able to change the direction of the inflation and I don’t know why?
@HS_Kim regarding your solution I love it, but would it be possible to make the facetted dome but on some polygonal curves randomly arranged in space?
20220604_FACETTED_DOME.3dm (283.1 KB) (21.6 KB)

Try to use negative strength value of Volumegoal. (30.5 KB)

1 Like

Thank you very much @HS_Kim . I think I was working so many hours yesterday that I wasn’t practical and didn’t apply a negative to solve the problem.
I just wanted to know why?
The truth is that I don’t really understand why sometimes you need large values for VOLUME and 1.00 for strength and other times with the same size of surfaces you need such small values of both values. Let’s see if @DanielPiker could explain it to me or just try it with the slider.
One last thing, I liked your FACETTED DOME so much that I’m trying to make it but it always has to be a surface of revolution with a circular base, right? because with a polygonal base like mine I can’t do it. I leave you my attempt to use the WeaverBird component to see if it is possible to use it in these surfaces.

20220604_FACETTED_DOME.3dm (246.7 KB) (27.3 KB)

There are other ways and you could do a similar job differently as long as your base surface has a rectangular topology… (15.1 KB)


IT’S GREAT!!! Thank you so much @HS_Kim ! That’s what I was looking for!
I didn’t know the SMORPH component.
I have a number of questions about this definition, if you could answer them I would learn:
1º_Can this method of creating a voronoi in any POLYGONAL based SRF NETWORK be done???
2º_THE RECTANGLE BOUNDARY on which the VORONOI is created must always have the DIMENSIONS of the NetworkSRF?
3º_The inputs of P0 and P1 of SMORPH are the origin of coordinates because we have created the VORONOI in the XY plane?
Let’s see if I can apply your definition to other POLYGONAL SRFs.
Thank you very much again!!! (I hope I won’t have any more problems). :sweat_smile:

Sorry @HS_Kim for asking you again, but I’m not sure what the inflation depends on to get it right.
If it depends on the CRVs, on the NETWORK SRF, on the MESH, on the position of all these with respect to the origin of coordinates. Is it better to inflate everything near the origin?
Well, if you could give me a hand, another one, in this last polygonal curve I would appreciate it.

20220605_FACETTED_DOME_v#03.3dm (303.0 KB) (26.9 KB) (28.1 KB)