Experiments in Communication

We printed the first circuit board for In Touch. Everyone at ITP was very impressed, which was kind of nice, and the board came out really pretty, especially since the substrate is white(ish).

the process:
after printing (with a laser printer) the traces on glossy paper, press with iron for ~3 minutes, to transfer the toner to the copper clad

soak in (soapy) warm water till the paper peels off on its own

use a sharpie to fill in broken traces, and a dentist's tool (or anything pointy i imagine) to remove excess toner/paper

dip in ferric chloride for a while till only the traces are left. then rinse with water and use acetone to clean the toner off of the copper traces

drill the holes for the components using really small drill bits carefully cause they break easily!

Posted by kyveli at 12:41 PM | Permalink | Comments (0) | TrackBacks (0)

We have a title for our project, as well as a lot of the parts of it working. The xports are finally talking to each other through UDP (turns out it was not that difficult -- we just had to change the settings and make sure we weren't making mistakes). In the configuration settings each of the xports is set to "CC" connect mode (as opposed to D4) and the remote IP and port for each one is the IP and port of the other one. We are using port 10001 for one of our xports and 10002 for the other one. So now we can open two terminal windows, connect to each xport serially, type, and what we typed appears in the other xport's terminal window, jst like magic!

Last week we got the communication protocol part down. We incorporated into our code what our arduino should send once one of the IR sensors senses a touch (which means that its LED has to change color) as well as what it should do should it be sent something serially. Then we tested this through the Arduino window's serial monitor (we only had one xport at the time), and after tweaking our code a bit we were able to make our prototype respond to these messages.

We have sketched out the board on eagle, jut need to play around with the connections a bit because the circuit is somewhat complicated. After we're finished with that, hopefully we can go on to etch the boards and then do a ton of drilling and soldering...

all 32 LEDs that will make up one of the 2 screens

Posted by kyveli at 10:49 AM | Permalink | Comments (0) | TrackBacks (0)

With some help from Ben Chao, Andy and I were able to get the LED driver to work, and randomly change the color of a bunch of tri-color LEDs. This made us veeery happy.

We are about to start working on the shift-in registers, and on writing some code to get the palette LED to set the color for the LED of the screen. On Wednesday I am going to Material Connection (as an assignment for another class) so I will look into material we can use for the screen.

Then the next step will be to network the screens to each other.

Posted by kyveli at 06:14 PM | Permalink | Comments (0) | TrackBacks (0)

I am working with Andy to make a pair of networked LED panels that people can use to communicate. It will be a dedicated, always on, persistently online channel of communication between two people who own the panels. The interaction will be through touch, so you will touch each LED to change its state. I imagine the panel hanging on someone's wall, or on their bedside table.

The screen will be low resolution, primarily because of time constraints, but we will use full-color LEDs to give the users an different, more abstract means of expression. We were torn between making a higher resolution screen of single-colored LEDs (which would be on/off, or PWM to produce a grayscale image) or making a smaller (and lower res) screen of full-color LEDs, but we decided to go for the colors.

Posted by kyveli at 07:38 PM | Permalink | Comments (0) | TrackBacks (0)

I teamed up with Andy, Mandy and Allistar, and we thought about different possible interfaces for a pong controller (a moustache that you would touch was pretty high on our list). Then we decided to put the interface aside for a while, work on the technical part of the assignment. It took us a little while, and some advice by someone who has taken Networked Objetcs before about the correct order of hitting reset and "x" to get to the Xport's configuration menu, but once we accomplished that things went pretty smoothly.

On a trip to Ricky's to get mustaches, we came across this amazing set of two pink guns with holster, that we hacked (with aluminum foil) so that the trigger would be a switch. It was really satisfying to use as a controller, and once you got the hang of it, you could be pretty precise with the movement of the bar. Here's a photo of our finished pong controller in action:

Posted by kyveli at 04:43 PM | Permalink | Comments (0) | TrackBacks (0)

I have a mac, and the weather widget is one of my favorite things about it. I have multiple ones open, to monitor the weather in places I could be (Athens (my home city), New York, London (many friends of mine live there)). It's like a game, and my goal is to be in the place with the warmest temperature. I am losing right now.

I was away this week, so I took the opportunity to talk to my family about what their feelings are about the weather and what they might want from a thermometer. They are definitely not as addicted to checking the temperature every second as I am, but my dad likes to go sailing so he is interested in knowing what the wind conditions will be, and he also wants to be able to access forecasts from the sailboat (ideally through his cellphone).

For our ambient thermometer, we decided that we want something that helps you know exactly how the weather is (or will be), without using abstract values (degrees, percentages, etc), but by creating a tangible sample of the weather that you can instantly sample with your hand.
We discussed a lot what the shape of the device should be, what the interaction will be and how the interface will look. Our notes are on the wiki.

Posted by kyveli at 04:42 PM | Permalink | Comments (0) | TrackBacks (0)

For the first Networked Objects assignment we had to choose a thing, an action, and a response from a list and improvise. I choose "caress", "cup" and "color" and made a cup that changes color when caressed.
The circuit is really simple. There are two photo cells on the surface of the cup, each of which is connected to one of the G and B pins of an RGB LED. As you cover each of the photocells with your fingers, the color of the LED changes from turquoise to a bluer or a greener tint. It also dims if you cover both of the photocells at once.

Here is a video of the cup in action (double-click to play):

Posted by kyveli at 08:42 AM | Permalink | Comments (0) | TrackBacks (0)