You're In Control (Urine Control)
Photo Documentation - Fabrication
MAS863 Final Project 12/16/02
Hayes Solos Raffle and Dan Maynes-Aminzade
My landlord owns a demolition company. When I asked him if he had a urinal, he laughed and said maybe. A couple of weeks later, I picked up this beauty from him. It needed quite a bit of refurbishing, but it cleaned up nicely.
The initial research problem was to sense the position of a stream of water on the back of the urinal. Our first approaches employed various pressure sensing apparatus, none of which was sensitive enough to respond to a water stream. Later approaches mimicked past media lab projects like Ping Pong Plus and the Tapper Window that triangulate between 4 sensors to determine position. Such strategies assume that the signal will take time to travel from point of impact to the sensors, or get weaker as it travels from the point of impact to the individual sensors.
In one scenario, we mounted sensors directly to the back of the urinal. This didn't work because the porcelain did not dampen the signal well over distance. In the photo above, I am mounting piezoelectric sensors to the back of a custom formed acrylic sheet that we anchored to the urinal. We expected the signal to be weaker if it was farther from a sensor because the acrylic would absorb some of the sound waves as they travelled through it. Unfortunately, the waves' change in amplitude with distance from the sensors was not predictable. This may have been because we were actually measuring displacement of the entire acrylic sheet, which is dependent on how it is mounted. Alternately, the sound waves may have been bouncing off the edges of the sheet as they travelled from point of impact to the sensors, giving unpredictable results. Whatever the cause, we abandoned this strategy as the deadline approached, and I decided to work with the sensors.
Piezo sensors create a clear signal when they bend. While triangulating a steady-state analog signal requires a membrane that predictably dampens amplitude over distance, our final approach sought the opposite: a membrane with discrete local deformations. This final working prototype employed a flexible mylar membrane (5 mil drafting vellum, painted white), with local mechanical isolation. Foam tape mechanically isolates areas of the mylar from one another, and local sensors measure deformations of the membrane in response to a water stream. When I applied the mylar construction to the urinal, a compound curved surface, the membrane had uneven tension over its surface. This resulted in slightly uneven sensor outputs because tighter areas deformed less in response to the water stream. This inconsistency was addressed by custom-tuning the following amplifying circuits to deliver clear signals to the microprocessor.
The amplifying circuit used an op amp (with gain ranging from 10-100) and an envelope follower to filter the signal. The signals were processed by a 16F877 PIC chip as digital inputs.
The impact of a stream of water created a signal that would break the 2.5 volt threshhold neccessary to send the PIC chip's digital input high. We chose to use digital inputs because reading them is fast and easy; analyzing 16 analog inputs would be quite slow, and we knew we could create a good game design with this simple sensor behavior.
Ultimately, our electronics were housed on a large breadboard that was mounted behind the urinal. The complete circuit includes 16 1.5" piezoceramic buzzers (our sensors), 16 amplifying circuits, a 16F877 PIC chip, a serial line driver for the PIC to talk to a PC, and a 5V power regulator to keep everything running smoothly. PIC code is here.
We mounted the urinal to a sheetrock wall to create a convincing interactive experience. Since the flush-valve was not functional, we routed the sensor wires from the urinal basin through the chrome-plated plumbing fixtures to the breadboard, which is behind the wall. Once we had all of the electronics connected and had the serial link and game working, we mounted a vaio directly to the back of the wall with gaffer's tape.
The white wall and placard made reference to art installations, where the urinal has a rich history. The video game is our interpretation of the classic carnival game "whack-a-mole." Position on the back of the urinal corresponds to position on the screen. The player attempted to hit hamsters as they jumped from one hole in the ground into another hole in the ground. A successful hit turned the hamster yellow, made it scream and spin out of contol, and rewarded the player with ten points. The parabolic paths of the hamsters concealed the grid-like arrangment of sensors, resulting in a fluid transition between input and output. The game was programmed in C++ on the Windows 2000 operating system. Download the video game code here.
In order to allow both men and women to participate in the demonstration, I created a customized game controller. It is a play on nintendo-style game controllers, plumbing equipment, and strap-on dildo harnesses. The oversized phallic nozzle is powered by two water reservoirs located to suggest oversized ovaries, making it oddly hermaphroditic. The controller consists of a nylon belt, a formed acrylic pelvic plate, water bottles, tubing, and a flexible garden hose nozzle. It is worn around the waist and the bottles are gripped and squeezed to pressurize a stream of water.
Surprisingly, our first test was successful, and the system ran reliably for the entire 3-hour demonstration.
While using a urinal was novel for women, most got the hang of it quickly and had fun with the new experience.
Shiny and white... While this prototype was an effective proof-of-concept, we look forward to developing You're In Control for actual restroom use. While urinating outdoors is playful for most men, bathroom sanitation requires a serious focus and conformity. We hope that further developments of You're In Control will help encourage restroom cleanliness while reintroducing play to the act of peeing.
You're In Control: Video
documentation
You're In Control: Home Page