Gratitude, post veneer warping

This is an interactive display that reflects and encourages gratitude. It is comprised of a wall-mounted wood panel, lit from behind. Capacitive touch sensing allows users to seed the panel with moments of gratitude that inspire later moments of contemplation for each other. Materials include laser cut pecan veneer, sanded opaque white acrylic, and warm white LEDs. Capacitive touch sensing is handled by the Arduino CapSense library, and TLC5940s drive the LEDs. Unfortunately, finding an adhesive compatible with both the paper-backed veneer and a plexi has been a struggle. As you can see in this image, although the veneer looked beautifully smooth at first, it has warped and come away from the backing in a few places. It’s likely that the only solution for this problem is to use an expoxy to coat the entire face, sealing the veneer and the plexi within. I’d like to stay away from that option, so I’m considering laser cutting a piece of MDF to match the veneer, and then using a layer of paper behind it to diffuse the light.

The user scenario goes something like this: while thinking of someone or something she is grateful for, the user touches the center of any spiral on the panel. A brief pulse of light will shine through the perforations to let her know her input has been noted. After a delay of several hours, a light will appear in the center of the same spiral. Slowly, the whole of the spiral will begin to glow, becoming brighter and brighter, inviting a user (the same user as before or another) to take a moment to contemplate someone or something he or she is grateful for (in my experience, contemplating one thing I am grateful for often reminds me of other reasons I feel gratitude). As the spiral begins to fade, the user has the opportunity to note another moment of gratitude by touching another spiral. In this way, I hope to allow multiple users to inspire a sense of gratitude in each other.

Inside the face of Gratitude (sensors/LED modules)

Arduino Duemilanoves and TLC5940s

First prototype of capacitive touch sensing module

Capacitive touch sensing

Jellyfish movie controlled by heart rate from Caroline Brown on Vimeo.

HRMI + video playback code

Sound Level and Heart Rate

Red is the sound level, which I still need to smooth out a bit. Green is my heart rate. Note that the y values are increasing rather than decreasing as they head toward the bottom of the image. This image is taken from a file I recorded while I was waiting for my sweetheart, who had been away all weekend. The sound level increased when he walked through the door and we said hello. As you can see, my heart rate went through the roof, too. Mushy, right?

Code is here.

SparkFun Polar Heart Rate Monitor Interface

For Rest of You this week, Dan asked us to collect data from two sensors and send the data to a file. He suggested that one sensor  be facing inward (e.g. galvanic skin response, heart rate, body temperature) and one be facing outward to detect environmental changes (e.g. light, noise, air temperature). I decided to monitor my heart rate while sleeping and combine that with sensing noise in the room. Often, around three in the morning or later, very loud voices float up from the street and through our open window. I rarely wake up, but my fiancé does. I wanted to see if I notice the noise on some level.

I borrowed a Polar Heart Rate Monitor and Heart Rate Monitor Interface (HRMI) from Dan. The monitor and interface couldn’t be easier to use– the HRMI is set up to detect the heart rate monitor right out of the box,  and SparkFun has great documentation on their HRMI, as well as sample software, so I walked through the steps to get set up fairly quickly. I combined the basics of Dan Julio’s hrmi_simple Processing sketch (included in the hrmi_demos.zip file) with the sample createWriter sketch on the Processing site to send my heart rate values to a text file. Once I had the heart rate side of things working, I used Dan Shiffman’s Sound Threshold with Sonia sketch from Learning Processing to figure out how to detect sound levels. I combined the sound sensing code with my heart rate code, and eventually wound up with this sketch.

I set up my laptop and the HRMI next to my bed, put on the heart rate sensor, and went to bed. There were a few times during the night that signal dropped out, so I lost my heart rate reading, but I was able to get some pretty good results. Here’s a snapshot of what the output file looks like:

HRMI Plus Sound Readings

The lowest rate I observed for my heart while I was awake was about 64, but while I was sleeping, it dropped down into the low fifties. As you can see in the snapshot above, however, when the noise level increased to .2 instead of the .1 where it seemed to hover most of the night, my heart rate increased back up to the low sixties. There’s no way to know, though, whether I was reacting to noise, or if the noise and increase in heart rate were due to my own shifting around in my sleep.

Here’s a snapshot of another parallel spike in values:

HRMI Plus Sound-- parallel spike in values

This spike in values occurred right before I woke up and stopped the program, around 4:30 in the morning, so I wonder if in this case it was a noise that raised my heart rate and woke me.

There are a few improvements I’d like to make to this sketch. First, I’d like to add a time stamp to the output. Secondly, I’d like to graph the values together. I’d also like to find a way to secure the heart rate sensor a little bit better so that it doesn’t shift in my sleep and risk sending false readings.

Graphing Analog Input

For the first week of Rest of You, Dan asked us to brush up on our physical computing skills by reading input from two sensors through Arduino and displaying the readings in Processing. My skills were pretty rusty, so first I reviewed the Analog In and Serial Out labs. I wanted to see readings from a thermistor and a photocell over time (sort of a rough way of sensing proximity), so  I graphed readings from both using Tom Igoe’s Serial Graphing sketch.  If you click on the image above, you’ll see parallel changes in the readings based on when I held my hand over the sensors.