Adding catch pan

After Carlo made the parts for the box I spent a solid day helping him and the other artists sand the thing. Finally got the thing put together and the legs on. I added a catch pan with some others in the group, filled it with cat litter, and we put oil absorbent pads on top to shield the electronics that will eventually be housed within the box.

Videos of large extended cores

Tall cores made of screws and drill bits and last a piece of cut and spiraled cast iron.

 This core was eventually ground down into a smooth cone and cut with spirals in the hopes that these more gradual contours would allow the ferrofluid to climb. The cone didnt work any better.

Experimenting with taller cores

didnt have much luck here. Used a huge drill bit, both wound our own transformer wire around it and used the transformer cores.

 This is me after using the bench grinder on some cast iron bar stock to make our own 3 inch long spiral magnet core. Didnt work. only fit four transformer cores on. I expected more though based on our working module.

Some photos of how I made electromagnets using screws and wall transformers

current progress on smart surfaces ferrofluid

The last project which focused on creating small modular unit to replicate the behavior of fish in a shoal was unsuccessful. The experiments I did with covering space on a window were very similar to another groups efforts and were not conclusive that the effect would be interesting or efficient. The rest of the group also became unconfident that the track and rail idea would work. The track device we prepared for our presentation was very incomplete and was both autonomically and mechanically unsuccessful. Given these failures we had a soul searching session with the professors and tried to find some inspiration and some common ground within the group. Eventually we settled on a medium to work with, ferrofluid, and I think with this parameter we have begun making forward progress again. It is still not easy, some of the group member still occasionally look back to old projects and their advantages rather than working with this idea. With some reinforcement and momentum within the group I think the backtracking will stop and we will get into a creative groove again.

We have begun getting potential core shapes together, made an acrylic box, attempted two types of ferrofluid and are beginning to get an overall picture of what this may look like.

The ferrofluid with toner isn't looking good, were giving up on that.

The ferrofluid made by etching steel wool then precipitating it into a ferrofluid with kerosene and ammonia was attempted but unsuccessful. This is likely due to the lack of any readily available oleic acid or ascorbic acid at 8 o'clock. The acid is necessary in the method to prevent clumping of the microscopic iron particles. We tried it with crushed up vitamin c, it didn't work. I will get some oleic acid at Ann Arbor Chemical tomorrow and we will try again. For now though we are buying a liter of professionlly made ferrofluid, it will cost us 200 dollars to get in time.

As for the cores we have some large bolts and nails that work when within a part of a common appliance transformer, the high coil density side. I will explain tomorrow with pictures, but basically I smashed a transformer, removed the core and took out the solenoids.

I would like to see more action to the cores, static cores will not do for my smart surface. I am now considering moving the solenoid along or withing the core, using shielding to mute certain areas on the core selectively, or using flexible core materials (either springs or bendable metal) to react to the weight of the climbing ferrofluid.

Our old biomimetic story is represented in this project by the density, movement, and translation of the tips in the magnetized fluid. many small tips can be made to come together into one large tip by weakening the magnetic field, as many small fish come together to make a shoal.

more video

attempts at making the surface unfold using static charge and electroadhesion

stupid mylar balloon. The poor aluminum layer made it impossible to get electro adhesion

better mylar but the plastic is too hard to get into.

For the rest of the videos I got a less stiff, easier to contact mylar. I just up the voltage and observe the effects.

In the end I determine that this idea is too dangerous and difficult to supply. More problems come up when considering supplying hundreds of panels at different times and under arduino control.

more videos in next post

Final smart surfaces project decided

Shortly after my last post we did choose an idea, the flock of robots that cover a surface. Ultimately the 3v3 split was resolved by me switching sides. There were some slight changes made regarding the application of the surface. First of all the surface was imagined on a window both to create shade and possibly harvest light for power. Secondly they would need to function as part of a greater whole not as independent pieces with independent purposes. The last thursday we were attempting to deliver a mechanism by which these things could move on a surface, our first a attempt at a prototype. We began with a complex design of a row of fish hooks on a screen which would grab in sequence like a centipede. We decided however to attempt something simpler, a solid shape traveling on a rod. This design despite its apparent simplicity encountered a host of problems.
The circuit we were using to alternate the motors direction, built from scratch (my bad), did not work. The arduinos would not allow us to load programs, a result of having pins connected to the inputs during the attempt at loading, for several hours. Basically thats it, but that put us far enough behind that no arduino was working for the presentation.
For this week we decided to persue any and all ideas for the movement of the parts and to split up into whatever size groups to test our ideas.

Some images of the scroll like device that I hoped to use as a 2" by 2" module to cover a window

The scroll poles are joined by gears at the base. When one is spun they spread over the window surface. The actuation could be caused by servos. Many of these could be arrayed on a window and be reactive to the presence of people or intense light.