This project visualizes the dynamic movement of water through a grid of LEDs and translates the amplitude of sound into simulated gusts of wind. A LED matrix is connected to an accelerometer to detect a level datum in the hopes of simulating a boat floating in water. It is attached to a rounded base with embedded earth magnets that is pulled by electromagnets. These magnets are controlled by 4 mic sensors measuring the amplitude of sound. In a sense it’s a boat traveling in a sea of light, propelled by the amplitude of sound.

The concept of this boat in water started with a table design for the ITP 4th floor lounge. It had to be lightweight and be able to dock on the ceiling grid when not in used. Fiberglass as the surface material was an exciting choice. With this material, a thin translucent skin allows for lighting to be on the underbelly. This gives dual functionality in the table design, when it’s on the floor it’s a table, when it’s docked on the ceiling it becomes a chandelier. There were plans to combine this table and my pComp finals together. For the pComp side, I wanted to interact with the table-lighting through sound. Where one can call it’s chandelier with cellphone/mobile app interface to turn on or have built-in mic sensors to detect the amplitude of surrounding sound and have reverberating animation inside this table-light.

(really rough rendering…)

From this development the project had to be scaled back due to time constraints. The basic material(fiberglass) and interaction of the lighting was kept. It then took on the form of a boat lined with LEDs simulating a water level with sound triggering the rocking of the base as if it was a gust of wind.

The Form:

From Internet

An illustrator drawing maps out a close guesstimation to real proportions of row boats. Ribs at one inch increments helps to create exact 3D measurements.

Then it’s laser cut and taped to layers of blue foam. It’s then carved and smoothed with a layer of spackle.

The inner and outer form above.
Unfortunately… this first attempt failed miserably… The urethane resin used for the fiberglassing from Smooth-on, Crystal Clear 202. is very sensitive to water and temperature. Failure to follow its finicky conditions will result in the resin not curing, sticky, and/or forming milky white foam. On top of it all… I failed to match the appropriate mold release which rendered the mold useless…

Total failure… lost about 60 hours of work…

So a different process for round two. Failure does have its advantages and allowed me to effectively weight the cost of time relative to its benefits. Needless to say it didn’t take 60 hours more to recover.
First lesson learned, using a more yielding resin (drawback being it cures in amber clear color instead of crystal clear, but really… it’s all the same…)

If possible use materials that doesn’t take time to cure/dry. Before, each coat of Spackle took an hour to dry, then there was sanding time…
Modeling clay works super well in this application. However it was still difficult to match the mold release.

Fiberglass Round 2!




This is about 8 layers of fiberglass and epoxy.
Behind the scene.

Demold

Testing the LEDs
These are 3mm ultra bright LEDs from Newark.com. First testing the brightness relative to the distance from the fiberglass surface. In the end,  sanding the tips of each LED help defuse the light and helps increase the resolution of the spacing of light.



The upper shell is the outer form of the row-boat. The bottom shell is the inner form which holds the LEDs in place at 1/2 inches away from the outer surface. The markings are to be drilled and each LED fits snuggly in place.
This is the beginning of assembling the LED Matrix. The Anodes are soldered in columns and Cathodes are soldered in Rows. Careful attention to keep all the anodes and cathodes separated.




This matrix is controlled by turning on the anodes and cathodes of that particular LED with a LED driver. Two TLC5940 LED drivers are daisy chained together to provide the 29 columns of this LED matrix.

Testing the circuit, everything lights up!

Running one of the library codes to test the sequence of the pins. Confirmed daisy chaining success!

Next, play with the code to turn on only the LEDs you want to.

For the base:
Creating a rounded base so to allow for rocking. It needed to be symmetrical and one way to achieve that is to spin it. Pipe fixtures are mounted to the material and a fancy dust collecting system was used to contain the blue foam.

Epoxy is coated on the blue foam surface to give it strength and spray painted.
Small earth magnets are embedded around the base. These magnets will be pulled by electromagnets installed under the table. 

Here’s a test with the electromagnets working with the accelerometer.

The overall form so far

Things left to do:
Making final connections.
Clean up the form.
Spray paint the bottom.
Coding the water animation.
Connect the electromagnet to mic sensor.
Final package with table.

*Many thanks to my classmate and friend Deqing for his brillant technical knowledge!!! Thanks!!!!!!*