Pull triangle mesh up to form an arc

Hello everyone,

I made an arc from a triangle mesh. Every edge of the triangle mesh has a specific length (48, 50, 52, 54 or 56). However, I also want to make a simulation of how the arc is constructed (in reality). First the mesh lays flat on the ground and has triangles with the same length of 48. Then, it has to be pulled up (with ropes for example). So some lengths will expand (to 50, 52, 54 or 56) and others will stay the same (48). Is it possible to simulate this? Kangaroo can do this I think.


PullArc.gh (48.1 KB)

Hi Robin - I had a quick look at your definition and don’t quite understand what your question is. You’ve already generated the mesh on the surface, now you want to know how it could be built in reality?

First up how many ropes are we talking about? One per node? Is there something to create the final shape eg formwork, or do you want to be able to lift/pull each piece into position using the ropes you mentioned? And how do the ‘real’ triangles grow from 48 on the ground to 52/54/56 in the air?

Hello Matthew - Yes, I know my starting flat triangle grid and the resulting arc mesh, but I want to simulate how to built it with the ropes.

I want to construct it with as little ropes as possible. A rope in every node will not be necessary I think. The mesh has to be pulled up as a whole, so not one piece separately. In reality every triangle edge is a telescoping tube, which go from 48 to 56. Except the triangle edges with a length of 48 are fixed and don’t change.

Unless your telescopic tubes are spring-loaded or something, you’ll need a lot of ropes! The transition of your mesh from flat to the desired surface is one where the mesh needs to grow radially outwards, so you’d need some kind of force in that direction (outwards for every tube).

Can you try to sketch what you think the endpoint would be? I can’t imagine a system of a few ropes being able to transform your flat mesh to the one you want. You’d be better off either with spring loaded tubes or maybe something like an inflatable bag beneath the mesh which when inflated would grow into the desired shape?

If you were to genuinely build this structure in real life I’d suggest having telescopic tubes which will lock when extended to the correct length (kinda like tent poles for large tents/gazebos). You’d then start in the middle and extend each one individually by hand. As the tubes lock into place the structure will slowly take shape. Not a fancy computational method but would be buildable like this! The last pavilion design I worked on was designed and fabricated using GH, but in the end the erection/installation was done the old fashioned by-hand way.

I was thinking of something like this:

It’s a telescoping tube which can ‘click’ to the 5 possible lengths (when pulled up with a rope). The tube can only extend in one direction. Indeed, it can also be done by hand with telescoping tubes used for tent poles. But the building time will be much longer, if you have to manually adjust every tube to the desired length. I want to have it built up as fast as possible.

Excuse me for barging in.
Is this project something new, only theoretical… or something already built somewhere else?

How can this structure be rigid?

At any time? Or is it locked later?
A triangle with a side that can extend is not rigid.

Also your nodes seems to give much freedom to the tubes, making your triangles becoming hexagons.
Does, each node, need to be “locked” after positioning?

It’s only theoretical, to see if it will work or not.
Because the tubes can only extend in one direction, it will lock automatically after it’s pulled up. The ‘stop’ will keep the tubes from sliding back.

What do you mean hexagons? Yes, every node has to be locked after positioning.

How can you prevent something like this from happening?

You can reduce the maximum angles the tubes can rotate. That can be improved.

My point about needing many ropes explained below:


How do you achieve a semicircular profile without a force pushing outwards in all directions? I think this could be achieved with one rope in the middle (like the first image) if all the tubes were spring-loaded as they would push outwards until the desired shape is met. Did you have another idea?

How about using hydraulic (or pneumatic) rams? These would push the top up and not require anything above it to pull it up.

Yeah, you’re right.

The tubes are actually a modular system and they are used to make other forms as well. That’s why there are 5 different lengths possible. If the tubes are spring loaded it will only work for this arc form but not for other forms. Another idea is to use rails (in combination with the ropes). These will push the 5 bottom points at each side inwards.

Yes, I thought about this too. Pushing it up from the ground will maybe be easier. But the principle is the same, you need some force to pull it up (ropes or hydraulic rams).