Interfering Gyroid Boxes Millipede

Hi everyone.

I guess this question is quite difficult and requires some advanced mathematical knowledge.

What I have below is two gyroid boxes, each with a different density. (I did not make the Millipede script, I have found it on the forum.)

My question is, how could I make the two boxes flow into each other, in order to make them work structurally as one piece and to be able to perform a structural analysis on the obtained gyroid 'beam'.

Could anyone help me out please? Thanks.

Attachments:

Interfering Gyroid Boxes.gh, 16 KB

Replies to This Discussion

Permalink Reply by Bob Mackay on August 1, 2016 at 10:39am

I think joining two such meshes together would be impossible. What I would do is take a single density volume of gyroid beam and define within it a frustum of a pyramid which can then be distorted to give your 1x1x2 beam.

Permalink Reply by Andrei on August 3, 2016 at 5:23am

Erik,

Can you be a bit more specific on what you are trying to achieve?

A smooth transition with a change in cell size? - Bob Mackay's suggestion is the right place to start

Alternatively for this and possibly neater, you could change the implicit equation you are using to approximate the gyroid in such a way that the size of the cells changes along the X axis or as desired. Then feed the result into the Millipede Marching cubes algorithm and you will get a clean mesh.

Keep the two distinct volumes and locally join them in the middle? - Then I would look closer at the intersection and find a strategy to weld vertices together.

Permalink Reply by Erik Pelicaen on August 4, 2016 at 7:46am

Thank you both for your replies.

The final goal would be to create a perforated wall structure with different densities at different locations (e.g. higher density where there are higher stresses), by using the gyroid algorithm.

However, manipulating the gyroid approximation formula in the way I want to (by not distorting the holes) seems like a difficult thing to do.

How would you change cell sizes? It would indeed be great to be able to work with it like Voronoi cells.

Permalink Reply by Andrei on August 4, 2016 at 12:05pm

Erik,

In that case I suggest to start working with a 3d polyhedral grid structure corresponding to the triply periodic cells. This grid will be easier to deform and manipulate in regard to an attractor. At the end, you can do a volumetric mapping of the gyroid cell inside the deformed polyhedra.

Best regards

Andrei

Permalink Reply by Erik Pelicaen on August 5, 2016 at 3:15am

Ok thank you Andrei, I will keep that in mind.

RSS