!blog - corey menscher

So I got some amazing responses on my project. Everyone loved it, especially kids. Most of the reactions revolved around the stunning nature of viewing in infinite number of LEDs, but it was all positive.

The best feedback I got, however, was the fact that there was a short line to view my piece when the floor started to get crowded. Most viewers figured out the interactivity pretty quickly, but I usually had to explain it to them once. If I were to tweak the piece at all, I would try to make the touch sensors a bit more obvious. One of my original ideas was to cut the paper touch pads in the shape of a hand, but I decided against. In hindsight I think it would have made the interaction obvious, as most people tried using just their fingers to press on the rubber nubs, rather than using their whole hands.

I feel honored to have been part of the show. Here are some photos of the end result, as well as some of people viewing the piece…

Infinity Mirrored Room - LEDs Forever Photoset on Flickr

“Infinity Mirrored Room - LEDs Forever” is complete! Not only that, but I made it into the 2007 Winter Show! I’m so psyched. It’s quite an honor to have a project from a foundation course accepted into the show. I just hope people like it! Here’s a video of the final product…

So I’ve spent, I dunno, about 8 cumulative hours trying to get three QProx QT113 capacitive sensors to work as proximity sensors.  These are sensors that detect the capacitance of a human touch and return a binary value to indicate said touch.  (It’s exactly the same tech as the iPod touch wheel.)  They supposedly can be “tuned”, however, to read the capacitance of human presence within a few centimeters.  I need about 8cm or so.  That way when someone walks up to the box and leans in, it can sense their presence and cycle the pattern that is displayed, as well as to detect the presence of a user’s hands near the sides of the box.  Currently, it reads about a distance of about 2cm.  It’s not bad, as there are plenty of people who seem to enjoy the box’s visuals so much that they want to climb inside, but it doesn’t switch for everybody.  I’ve tried nearly everything that the datasheet recommends…from increasing the capacitance at the sensor pins via parallel capacitors, to altering the material and size of the electrode.  Nothing seems to help.  It’s driving me nuts!

So I have the enclosure mostly done. The mirrors need to be tightened up and i need to figure out a good way to attach them to the interior, but it’s lookin’ good regardless! I’ve also decided to put the LED panel at the bottom of the piece. It’s a departure from the original, but it provides for a solid base. Besides, it is safer for the panel and looks just as good!

So I’ve begun programming some initial patterns for use in my PComp final project, and boy is it fun! I’ve already made about 8 test patterns, and have some more ideas. Dan Shiffman’s library is relatively straightforward to use. I think I’m only going to be limited by my imagination at this point!

For my Physical Computing final project, I will be creating an update to a piece by artist Yayoi Kusama called “Infinity Mirrored Room - Love Forever”. The piece was first created in 1966, and was remade in 1994. I saw it in August at the Whitney Museum’s “Summer of Love” Exhibit…which was a collection of from the 60’s and 70’s psychedelic movement.

The original IMR-LF is an approximately five foot tall rectangular box. In the center of the top portion of the box is a several inch wide slot that allows the viewer to peer into the interior. Inside, the viewer finds a small room, about two feet high, covered in mirrors on all sides except for the ceiling. The ceiling is studded with a pattern of multicolored lamps that flicker in succession to provide an illusion of infinite space. Two planes of the lights travel in all directions, by which the viewer is easily mesmerized. At the back of the room, the viewer sees his own eyes reflected back…as if his body has been transported to this infinite field. I love the play on perspective and illusion that a small space can contain an infinite one, and that the viewer can be visually drawn into this new world.

When I was viewing the piece and watching museum-goers experience it, I noticed that there was really no interactivity. People would queue in a line, walk up to the viewing portal, and look inside.

While contemplating my final project, I remembered this piece and how I was struck by its low tech approach. It uses small lightbulbs and some basic timers to light the bulbs. I asked myself how it could be reconstructed to be a more “21st Century” piece by using LED lighting and a microcontroller. The answer: Infinity Mirrored Room - LEDs Forever.

My intention is to build a slightly scaled-down version of IMR-LF. I will use an arduino, some IR rangefinder sensors, and Processing to drive a Color Kinetics iColor Tile. The iColor Tile is a 24″ x 24″ array of 144 RGB LED’s. The LED’s can be driven over ethernet by Processing using a UDP library written by Daniel Shiffman. I will construct a roughly 24″x24″x24″ cube without a top, over which the iColor Tile will be placed. Inside, the cube will be mirrored on all sides, like the original IMRLF, and a viewing port will be cut in one side. Using a 12″ Powerbook, the iColor Tile will project a series of patterns, which will be reflected throughout the infinite space inside. The sensors and Arduino will be used to trigger changes in the pattern…speed, direction, and probably the pattern itself. Here’s a SketchUp 3D mockup:

There are two major aspects to the project…constructing the enclosure and programming the visuals and sensor functionality. I’ve begun a few Processing apps to create interesting visual patterns, but they will need to be refined and integrated with Shiffman’s library. When I am happy with visuals and have the communication with the LED panel working, I will focus on building the enclosure.

I’m very excited to begin work on the project. I’ve never attempted anything like this before!

UPDATE: The piece was was a tremendous success at the 2007 Winter Show! I have posted some photos of the final product at Flickr.com.

We completed our prototype of the TuneDroppa. It consists of a cheap demo umbrella (the real one will be transparent and of a “bubble” design), one contact mic, one blue LED, and our breadboard/arduino setup. We also added a 3 watt amplifier and a 4″ speaker to amplify the tone that will be triggered by raindrops. The tone is currently being generated by the Arduino, but we’re hoping to use some 74C14’s as oscillators eventually. This is just a proof of concept, and it seems to be going well!

So my Physical Computing midterm group (Taylor, Celina, Alex, and me) has decided on a project for the midterm. We’re not sure of the name yet, so for now I’ll just call it “Soundbrella”.

The basic ideas is to place contact mics on the panels of an umbrella. When a raindrop hits a panel, the contact mic will presumably detect it. A tone will be triggered and an LED (white?) will light up. A small speaker will be mounted to the inside of the umbrella, preferably above the holder, and the LED’s will (hopefully) dangle down from the frame. We anticipate using a clear “bubble” umbrella (http://umbrellastand.com/bubbleumbrella.html?gclid=CKqHs5bu8I4CFQIuHgodunTbJw) to provide a more immersive experience for the user, and for better acoustics. We also think it will be cool for the LED’s to be seen from a distance walking down the street.

I asked around for some sound generation options. Our original concept was to use an MP3 player of some sort, but Gian Pablo Villamil recommended using 74c14’s. They’re cheap, and they can be used with AD5206 digital potentiometers. He has previously worked on a project using a similar configuration, so he could become a very useful resource! I did find a 40 second programmable IC (http://store.qkits.com/moreinfo.cfm/ISD2540) which could also be used, but I’m not sure how to program it. I also found a “Polyphonic Ringtone IC” which contains a small MIDI synth (http://www.winbond-usa.com/en/content/view/166/1573/), but I doubt that it would work in this application, and we’d have to request a sample…which is apparently kind of hit-or-miss.

My duty at this point is to determine the materials requirements, except for the Umbrella itself…which is Celina’s job. After some research, I have determined that we will need the following:

6 - contact microphones - Alex says he can get these easily, and there are typically six fabric panels in an umbrella. We may need to add more, depending on the sensitivity per panel.

1 - arduino - we’ve all got one

2 - AD5206 Digital Potentiometers - this is a chip that contains 6 electronically-controlled potentiometers. (http://www.analog.com/en/prod/0,,761_797_AD5206,00.html)

1 - 74C14 Hex Schmitt Triggers - there are six triggers on the chip, and we can use one per contact mic to produce a square wave. Each wave’s frequency will be triggered by a digital pot. (http://www.datasheetarchive.com/preview/466481.html)

6 - LED’s (White) - these will be controlled by one of the digital pots. Hopefully we can do some slick stuff like fading them out. (see http://www.arduino.cc/en/Tutorial/SPIDigitalPot)

1 - audio amplifier - I already own a 3 watt amp kit, but that may be overkill.

1 - speaker - I also own a 1 watt speaker, as well as some small piezo speakers I scavenged from a couple of audio greeting cards.

Time to get to work!

So it’s been a while since I posted about the Physical Computing labs, but I wanted to document my progress with my attempt at creating something a bit more interesting than a potentiometer dimming an LED.

Lab 3 had us using a variable resistor to perform an action. So, I took my digital chopsticks and turned the aluminum-tip switches into a variable resistor! I used a voltage divider circuit to test the voltage of the item being held by the chopsticks. Thus, if the chopsticks were being used to hold metal, the “holding” LED would both be triggered…but only if it reached a certain threshold. The purpose is to test the theory that sushi has a certain level of conducivity, and I may be able to set the threshold to the level of sushi to trigger another action…

SERVOS! Lab 4 had us using a servo with the variable resistor. I thought about what I could do, and it seems that the servo would be a perfect solution for an automated soy sauce dispenser:

I had a tough time with this assignment…mostly due to the loop() method. My desired action was to have the servo perform a full cycle of its motion (pour then reset) once triggered, but I found it extremely difficult to keep it from stopping mid-pour or resetting prematurely if the Arduino did not detect a high enough voltage at the tops…such as when you let go of a conductive item. I eventually got it to work (mostly), as you can see.

How Users use an Elevator

- User approaches elevator and presses one of two call buttons…one for vertical ascent, and one for vertical descent. The buttons are usually positioned on a panel relative to their respective direction, but they may also be labelled with icons (arrows). The button typically illuminates to let the user know that a car is being routed.

- The user waits until a car is available to be routed to their current floor. An algorithm in determines the resting position of the cars (ground floor, midway, top floor) unless all cars are in use.

- When the car arrives, the user enters and selects the desired floor by pressing a numerically labelled button. The button then illuminates to indicate the selection. The doors contain proximity sensors to determine whether the user’s position will block the doors from closing. Once the doors are clear, and sufficient time has elapsed to wait for additional users to enter the car, the doors begin to close. If another user attempts to catch the elevator from from afar, the user presses the “close doors” button repeatedly.

- The car travels in the desired direction while constantly “listening” for other users on floors whose intended destination intersects with the current travel direction. If one is found, the car stops and lets the additional user on board. The new user presses the desired floor button, whether or not the button has been previously selected…which is indicated by the button being illuminated.

- When the car arrives on a user’s desired floor, the car stops and the doors open. The doors remain open until the proximity sensor determines they are not being blocked and continues along the current path, has reached the minimum or maximum floor, or is re-routed to a floor with another waiting user if all users for this session have been dropped off on their floors.