These videos show two out of a series of experiments that I run last week, one successful and one unsuccessful. I am trying to push my research forward by programming six modules to self-assemble passively–that is without adding energy to the system. I am testing the 3.* modules for this experiments, programmed with a mid-outer ring configuration. I haven’t tested any other configurations yet. Such modules have more configuration possibilities than the 2.* modules, and wider surface areas that allow them to float on top of the water carrying the magnets. In approximately 20% of the experiments, the self-assembly process was successful and the final shape was the expected snowflake.

In the unsuccessful experiments, I identified the following design problems:

1. The shape of the modules prevent them from assembling in some cases, this is especially true for clockwise rotations. S./ Fillet vertices.
2. Two convex positioned magnets–that should repel each other–can sometimes stick together sideways. S/ Review magnets configuration.
3. When there are more than two modules in the water, the magnetic field can make two of them to stick together in an undesired way. However, this problem is inexistent when there are only two modules in the water. S./ This shouldn’t be a problem for a large tessellation with all the slots programmed.

The next steps should include:

1. Inner, mid-inner and outer rings configurations for the 3.* prototype
2. Mixed configurations for the 3.* prototype
3. Design the 4.* prototype  the problems shown by the 3.* prototype
4. Exploring other tessellations
5. Measure (area? weight? dim? etc?)

To think about

1. Float-ability and position for a self-healing tesselation
2. Blueprints and representation methods
3. 3D
4. Modular architectures