Just another Beyond Bits and Atoms Blogs site
TuneTrain Team Post-in-Progress
Jenny Moryan, Jamie Diy, Darri Stephens, Nicole Roach
5/26/11
After realizing that our original idea for a tangible music wall was already being done on the SLATE toolkit, we decided to re-evaluate what our final project would be. We began a brainstorming session to determine what learning goals related to music we most wanted to tackle. A few possibilities were discussed: rhythm, reading music, composition/creative expression, artistic expression, and telling a story (narrative with mood music).
We threw around several ideas:
- Attaching music to brush strokes and paint color… The user would hear the music as s/he painted. The music associated to the strokes/color would be recorded, effectively allowing for the finished piece to have its own soundtrack.
- Using tabletop recognition to read musical puzzle pieces… A user could arrange the puzzle pieces into a pattern which would align with a music staff to compose a tune.
- Tying musical tones and graphic line designs to tiles… A user would arrange these tiles in a 4 x 4 array to create a graphic line composition which would then play a tune. Conversely, the user could create the tune first, and then flip the piece to see what collage of tiles s/he created.
Though there were aspects of these ideas we liked, we felt our learning goal could be stronger, and we didn’t feel like any of these ideas took advantage of the affordances of a tangible tool.
Our last brainstorming session led us to our current idea, which focuses on teaching rhythm. Inspired by an idea of an online game, we decided that we could create music pieces that fit on a track to help students understand rhythm. Below is our initial sketch of our project idea.
After our initial brainstorming session, we began to figure out how to implement our design. The first thing we tried to create were the music note blocks. At first we wanted to create these pieces out of something with a more rounded shape, so we tried to create one out of clay that would be used for forming molds.
After playing with the button for a while, we decided that it would depress from something like acrylic and that this would fit in better with our design.
We then began sketching the basic train track in Corel. We initially thought we would include bridges connecting our outer circle to the inner circle which will contain the Arduino board, as seen in the image below.
But after some consideration, we determined that the button connected to the train would need to be connected to the circuit board contained in the inner circle, so the inner circle will turn as it moves the train around the track. We also worked to design the actual pieces that will be our music notes on the train track. Below are some of our sketches and list of items we plan to make.
We began cutting the initial versions of our prototype out of cardboard yesterday and figured out a lot about what worked and what needed to change in our design as we tested elements together.
As we build the train, we are working to determine if the width of the train and the axles for the wheels are wide enough to allow the train to turn on the curved path that we are creating for it. We tested this today with initial cardboard versions of the train and train track as seen below.
We also began testing the button on the Arduino board. We hope to get it to emit a sound out of the computer tomorrow.
5/27/11
During our lab time on Friday, we began considering other elements to include in our project such as recording and playing back sound. We would like to allow students to play rhythms from multiple tracks over each other and see how they sound.
Having already made cardboard prototypes of our track, the musical pieces, and the train, our team felt like we were progressing in good shape. On Friday afternoon, we sat down to discuss our next steps in order to keep moving forward when we discovered that we had overlooked a crippling issue. In short, we already had tackled the issue of keeping the Arduino’s wires from tangling as our train was going to move around the track, moving forward by rotating a center circle via a motor; yet somehow we had neglected to consider how that big, thick, white USB cord attached to a computer would (or more realistically, would not) move with the train. And that big, thick, white USB cord was to supply not only power but programming capabilities allowing us to provide TuneTrain’s feedback/learning objectives.
After hours of debate, we considered and discarded the following options:
· Hang the USB cord from above like a mobile… too cumbersome.
· Use Arduino shields and a battery pack… too heavy for the motor.
· Hook the non-Servo motor to a DC power source… doesn’t address the programming issue.
· Use a WAV-shield and not hook up to a computer… literally, weighty issues- shields and battery pack.
· Use a magnetic ball to rotate the Arduino board and components… still doesn’t address the USB cord.
· Use some ball-bearing, lazy susan-type of device to distribute the weight… limits our programming options, including recording possibilities.
We kept returning to the fact that we didn’t want to be limited; we wanted access to the computer so that we could play with sound files, allow for feedback, and possibly incorporate recording into TuneTrain. Finally, we realized that all the problems stemmed from the fact that our train needed to revolve. In the end, we made a tough decision. Jimmy had suggested that we move the track, not hte train. Although it may not be as satisfying from a kid’s perspective to not be moving a train, the decision allows for more functionality and more affordabilities for TuneTrain. The focus is more directed toward creating a musical “track” (including easier access to placing pieces on the track), and the overall design is more structurally sound. Back to the drawing board!
5/28/11
On Saturday, we met to continue working/brainstorming about the issues we had with turning our train, but not turning everything else. Jimmy had suggested that we try to turn the track instead of the train (essentially creating a record player), but we were hopeful that we could still create a moving train on a stable track. Jamie came into lab with an idea for using a conductive spool that would allow a conductive string rotate around it while maintaining contact with the stationary object. Below is an image of the model Jamie created.
After some discussion with Jamie’s boyfriend Randy, we had a new idea for taking the switch apart and making a conductive tape ring that runs around the track and connecting it to a wire that feeds into the Arduino. Jamie and Jenny ran through trying to create this on our cardboard train and it appears to work. So we will try this as our first method for closing the switch.
Meanwhile, Nicole began figuring out processing with the Arduino. She was able to connect an IR sensor which would show the variable input on the screen and make a switch activate a song (“We Are the Champions”).
Jenny worked on cutting the pieces out of acrylic which took some time because the laser cutter kept going offline and it took a lot of time to get it back online and working with the computer.
On Sunday, Jenny cut the track out of acrylic and also made music notes made out of vinyl for the music pieces. Since we ran out of acrylic glue, she couldn’t finish connecting the backs to the pieces. Here is an image of how it is looking so far.
5/30/11
On Monday, Darri worked to build our train out of acrylic. Even though we had already tested this with cardboard, adjustments needed to be made to allow for enough room for the button in the front of the train and for the new holes we decided to make for the button connections to reach the golden strips on the edges. Below is an image of the train as it was being built.
Over the weekend, Nicole had experimented with different instruments, and on Monday she learned to create notes making sine waves. These don’t sound exactly like any particular instrument but have a better sound quality for what we need as the train travels around the track.
Jenny built the stands for the train. She and Jamie then worked to create the golden border for the track which will have wire contacts attached to the Arduino board. First, Jenny drew the arcs based on the track in Corel. We then went to the vinyl cutter to try and cut them out. We managed to get several “motor errors” before we realized we could not use it effectively. We tried to simply engrave the pattern so we could cut it out, but we could not get the vinyl cutter to work consistently. We ended up cutting out arcs in scrap acrylic and using that to trace our pattern. Below is an image of the carnage of gold paper.
Nicole created the software component for TuneTrain. She had experimented with playing sound files of different instruments, but delays in the playback were causing her some difficulty. So instead, she learned to create notes making sine waves. This approach worked out much better, allowing for simple real-time sound generation. Eventually Nicole added functionality for recording songs and saving them to the computer. She created an appealing computer interface so that now users can access their songs from the TuneTrain application and even layer tracks in real-time and during playback.
But it turned out nicely on the board.
5/31/11
Next, we created a stand for the DC motor and battery pack to sit under the inner circle. We allowed for the battery pack to be removed, but we glued the DC motor into the stand for stability, and then glued the inner acrylic circle to the motor via a wheel. The stand ended up being a tad too high, so we created what are essentially “platform shoes” for the outer track to make the two pieces level. We also carefully glued the gold strips to the edges of the track, making sure that the glue was only on the outer edges to ensure a continuous electrical connection. We had tested for this earlier, since we were concerned that the Super Glue may impede the electrical connection. We realized that as long as there was some portion of the gold conductive track was without glue, the connection was maintained. We layered the conductive track pieces in the same direction of the train’s movement. After making the conductive track, we also built “skis” to attach to the touch sensor’s wires in order to increase surface area and ensure a steady connection. The shape was meant to keep the wires from getting caught as they glided around the conductive track. Yet when we tested the train, the skis’ edges just got caught on the note pieces’ edges. Below are images of the conductive tracks and first iteration of wires’ contact points, aka, skis:
Since the mini springs did not allow for the desired flexibility needed in the train’ movement, the final design we decided on for connective pieces were sideways V which allowed for a natural spring that maintained contact around the entire conductive track. Once all these connections were in place, we were ready to test our train with the motor.
We also realized that we need to slow down the motor with a potentiometer.
Simultaneously, we began thinking about a curriculum that could accompany the TuneTrain. We knew that we wanted to reinforce the idea of rhythmic pattern and creative composition. Although we were wed to the idea that there is no right or wrong when it comes to artistic expression, we did want children to begin to hear when music sound rhythmically harmonious or when they heard some sort of dissonance. Since our intended users most likely are too young to understand all the mathematical implications of beats of four and eight, we figured that they could at the very least hear when a pattern did or didn’t fit into the track’s loop of thirty-two beats. We decided to create activity cards that would provide scaffolding as they explored this musical terrain: Copy a Track, Complete a Track, Color a Track, Mystery Tracks, Compose. In the activity cards, Copy a Track, we encouraged our learners to play with patterns; the cards prompt learners to emulate the pattern on the track so that they could hear what this pattern looks like. They then are encouraged to make changes to the pattern and listen to the difference. In another set of cards, Complete a Track, learners are given an incomplete pattern to finish. They have to rely on their visual-spatial skills to recreate the entire pattern on the TuneTrack. Then they are asked to listen to the pattern to determine if it is complete or not. If not, they are encouraged to change the pattern to sound more rhythmic (i.e., comprised of beats of four). In Color a Track, children are encouraged to use the TuneTrain to create their own compositions. The cards allowed for children to record their patterns visually as well as audibly in the computer’s TuneTrain library (as stipulated in the Compose cards). Lastly, Mystery Tracks asks children to listen to a prerecorded tune from the library and try to recreate it by sound alone.
At this point in the project, all of our pieces independently were working successfully. The train looked adorable and fit well on the track; its wheels moved and adjusted as they moved around the note pieces. The DC motor was functioning via a two-battery pack and was attached to the inner circle; and the train, attached via a fitting on the inner circle, rotated as well. The touch sensors wires connected to the conductive tracks and maintained a steady electrical connection. The note pieces had all been adjusted and sanded so that they snap fit perfectly into any sort of arrangement on the track. The Arduino functioned perfectly and Nicole’s programming allowed for the train to play a single note when the touch sensor was depressed. We tried a quick test before retiring for the day, and all the pieces seemingly came together with a few train toots! The train was moving too quickly around the track, but we were thrilled with the initial results.
6/2/11
The next day, we experimented with a potentiometer to adjust the speed, but soon we realized the large potentiometer would only work at very low levels. Instead, we chose to use a regular resistor to control the speed of the train. We also realized that as the train (connected to the inner circle via an arm) needed to maintain a consistent revolution, we had to align the inner and outer pieces perfectly. Therefore, we built a base platform for the whole apparatus. Encouraged by our previous day’s accomplishments, we were ready to put all the pieces together permanently. Yet as we began attaching the motor, we found that the train had problems moving at a consistent speed around the track. One of our initial solutions was to remove the front wheels because we thought they may be causing too much resistance as the train turned. Despite this change, the train still had an inconsistent speed around the track.
Since our toy is teaching rhythm, a consistent speed was vital to meeting our learning goals. Not only was the train moving at an inconsistent speed, but the train itself was bouncing, meaning that contact with the touch sensor and with the conductive track was inconsistent. We noticed that the inner circle itself was not quite level. We questioned and toyed with the idea that the weight of the train was tilting the inner circle. We tried weighing down the train; we tried adjusting the arm; we tried counterbalancing the weight of the train on the inner rotating piece. We systematically tried to test and rule out each possible variable. We even took off the back wheels, allowing for the train to move more freely only by way of the arm. Although this change helped with the bouncing issue, the speed of the train still was inconsistent. No matter what we did, after a few rotations, we were seeing the same problems again and again. All of our initial elations were dissipating…
Then, while trying to adjust the center piece, we accidentally broke the inner circle off of the motor. While seemingly devastating, we actually realized that the motor itself was “jumping” and causing the inconsistencies in the rotation. We then found a GoGo Board motor to use instead. We reattached the inner circle to a new GoGo Board wheel (taking even greater care to make sure it was centered and level) and remade and reattached the train arm’s fitting (the broken bed on the laser cutter had resulted in some angled cuts).
After testing this solution out for some time, we realized the GoGo Board motor was also inconsistent. Steve later told us that all DC motors would have this issue. Steve then helped us connect a Servo motor using another Arduino board to maintain a steady speed as the train moved around the track. With the train finally moving at a steady speed, we were ready to see the TuneTrain in action.