#connectiveenvironments

Stereo Amplification and Dual Screen Routing

The stereo amplifier that we purchased (the LEPY 2020A) was key to the success of the oscilloscope system. Prior to purchasing an amplifier, the waveform was so small the display appeared as simply a quivering line on each axis. The amplifier increased the amplitude of the signal enough to provide each axis with a waveform capable of taking up the entire screen. In addition to this, the amplifier came with a built in voltage transformer. It seems as though this transformer as provided the coil within the television with the proper amount of signal, therefore solving the problem that plagued many of our previous iterations. Because the television was being fed the proper amount of voltage, the system didn’t fail or automatically should down. 

With the issue of the television and waveform displays solved, we moved on to the problem of how to make this installation into a dual screen set up. We took advantage of some home studio equipment we had available. Using an audio interface (a Focusrite Scarlett 18i20), we were capable of developing a series of routing paths that would split input signals, meaning each microphone could be sent to multiple sets up televisions at once (up to 8 in fact.) We then purchased a specialty microphone cable that converted a standard XLR jack into a stereo RCA output. Each stereo amplifier requires a stereo RCA input, so this cable was paramount in transferring the standard microphone jack into something capable of being received by the amplifiers.  

Within the audio interface’s software, we assigned these specialty microphone cable as outputs 1+2, and a monitor speaker for each TV as outputs 3+4. We then instructed each microphone signal to be sent to their corresponding specialty cables, as well as sending each microphone to the speakers. Because the specialty cable resulted in a stereo RCA jack, we then routed ONE of each stereo pair to either amplifier. The resultant signal path gave each television a display of both microphone inputs. 

a diagram showing the wire we used to split the XLR audio signal from the microphone into left and right RCA signals.

diagram of the full system with pictures of each element

User Studies and Results

With all systems solved and ready for installation, we ran a series of user studies in order to observe how participants would interact with the interface. We specifically wanted to see how a diluted sense of each other’s presence might affect a user’s interaction with the opposite person. Would such a scenario enhance two strangers attachment do one another? What might occur when two people familiar with one another? We wanted to see what trends we could observe based upon the relationships of the two people, and also see how much cooperation was fostering through interacting with the interface. 

In addition to letting the two participants hear both of their voices while speaking, we also engaged in an alternative version of the experiment that prohibited the users from hearing one another. This involving cutting the microphone input to the speaker, pumping white noise to each user, and giving each person a pair of ear plugs. We observed the differences between a purely visual stimulus and one with both audio and visual cues. 

While the studies were occurring we also recorded each interaction on both microphones and with a video camera on each participant. With the recordings handy, we can pump the original audio back through the interface, effectively creating a simulation of their initial dialogue. 

After each pair finished their session, we gave them each a questionnaire in order to gauge their reaction to the experiment. The questions focused on their own feelings during their experiment, their engagement and connection to their partner, and also if they felt the interface enabled or inhibited a meaningful interaction. 

Discussion of Results 

After observing and recording the user studies, we began to notice several trends develop within the experiments. Many reacted positively to their partners role in the interaction. Each group had a different dynamic, but in general pairs (who often did not know each other prior to the installation) would sing together, hum together, or make wacky noises together. They observed the visual output of their voices, augmenting the sounds they made to play with the shapes that appeared on scream. Some would coordinate efforts, each producing a different noise to facilitate a unique shape. Others attempted to match notes in order to make a “perfect circle” or to at least make the waveform stand still for a brief moment. 

There was a minority of participants who seemed to feel mildly uncomfortable with the experience, choosing to talk in short bursts and observe their individual axes move horizontally or vertically, never fully giving themselves to the interface. Yet it should be noted that there was never any disinterest. Those who felt uneasy engaging with their partner still interacted with the interface and expressed enjoyment in the outcome. Still others began to engage before laughing too hard to continue, not an ideal but still very satisfying outcome. 

In general, observing one's partner through the abstracted lens of diluted physical presence, meaning a distillation of an individual into merely one or two perceptible variables, seems to have an effect on the user's inhibitions and sense of humility. In a way the installation acting as an ice-breaker of sorts. Here are some quotes from the questionnaire that reinforce that notion: 

What effect did the interface have on two strangers interacting with one another? 

“I could communicate with someone that I didn’t know through something that I wasn’t familiar with, but it was actually working.”

“It was just an instant moment of collaboration.”

“I feel as if I went on a blind date spontaneously, the TV allows for a level of intimacy that you might not have ever expected with the person you’re doing the experiment with.”

“I didn’t know my partner, but the oscilloscope allowed us to have fun together and be silly in a way that we wouldn’t have been able to without the oscilloscope.”

How did the interface contribute to the interaction with the other user?

“Well, it was the limitation of interaction that really defined the connection.”

“Took away the human element. It felt like the future of communication.”

“It was kind of like a battle to see who could get the longest or highest line between the two of us. But it was cool to see the mix of the same pitch and volume.”

“It made me question his intentions or emotions.”

“Technology typically doesn’t require more than one person to activate, so the act of collaboration makes more a dependence we tend not to expect/encounter.”

The connective compass ignores typical visual feedback describing a person’s direction. Alternatively, providing vibration patterns or cues as the tactile feedback to explain the person’s orientation. 

During our design process we worked with 2 of each of the following parts as we built our prototype. SparkFun 6 Degrees of Freedom Breakout - LSM303C, SparkFun GPS Breakout - Chip Antenna, SAM-M8Q (Qwiic), Vibration Motor, SparkFun Haptic Motor Driver - DRV2605L, ELEGOO 170 tie-Points Mini Breadboard kit for Arduino, MELIFE 2 Pack ESP32 - ESP-32S Development Board, & Insignia - 5000 mAh Portable Charger. Which ultimately led us to the final wiring design shown in the Fritzing Diagram below.

The next process included designing an enclosure that would efficiently house the electronics and the wiring shown above. We worked through this process with more of an efficiency approach than a aesthetically pleasing one. This allowed for the following our first compass prototype. See an exploded axonometric drawing below describing our first iteration of the compass. Moving forward we will work to refine this design so that the materiality and the 

Future work would first focus on further refinement of the compass’ design. During our initial design process we were more concentrated on the technical components and assembly, which ultimately drove our physical design. As the design continues to develop, we would need to ask ourselves the following questions; how doest his compass feel?; what is the materiality of the compass?; what is the ideal size of this object?. Furthermore, we must consider the emotional factors as we refine our design. These decisions will be beneficial to how potential users experience and interact with the compasses. 

Additionally we may consider incorporating user tests with our compasses. In these user studies we would need to consider testing three different group; two companions, two friends, and two strangers. The feedback form these types of users would allow for us to determine how successful each group interacted with the compasses. Observing the results from the tests would allow us to identity which areas in design we would need to focus on refining. 

Another factor we could consider testing is the physical location of where the user tests are based. Since the compasses focus on orientation of a person, it would be interesting to see how users react when tested at varying geographical distances. One test could be done at a small scale on a campus while the other test could be done at a larger scale in a city. 

In order to create an oscilloscope from an old CRT TV we must understand CRT TVs. The thing we are after is the electron ray (figure 1) and the deflection coils (figure 2). Normally the TV would send electronic information to the deflection coils that pull on the electron ray as it shoots towards the screen, mapping each pixel. In our case, we want to take the wires that would transmit the location information of the pixel and wire them instead to audio signals coming from various places, like microphones, synths, or music. The result should be the waveforms created by the music visualised on the screen of the TVs.

fig. 1

fig. 2

fig. 3

To start this process we bought two tvs off of craigslist and immediately ripped them apart and cut up their insides! After sifting through a bunch of youtube channels, we found a way to discharge the TV with a wire and a screwdriver(figure 3). To discharge the CRT TV we needed to attach the wire to ground and the screwdriver to ground port at the top of the cathode ray tube(figure 4). We needed to ground the TVs before touching any of the fun stuff inside, for the first time we did this very tentatively, scared of the 20,000 volts the cathode ray could hold!!! After a couple successful attempts we felt more comfortable and were able to start the oscilloscope making process.

fig. 4

We cut the four wires, tested their continuity to see what wires were connected, and soldered them to the left and right of a ¼” audio inputs. For the first test we turned it on without connecting it to any signals, our goal here was to achieve a dot in the middle of the screen. And the tv turned on and gave us the dot but it was very small, so we decided to adjust the screen knob in the back of the TV to turn up the amount of light produced by the electron beam. Turns out (haha) turning the screen knob while the TV is on changes the voltage through the system and fries the flyback transformer (figure 5). So the first TV we tested died the day she joined the team RIP: Abigail. 

fig.5

Note: we named all the TVs in alphabetical order to keep track and to give them some support when they need it or for funerals. So A=Abigail B= Betsy and so forth…

The next TV we tested was a newer smaller TV named Betsy. She was more promising so when we soldered the ¼” audio input, it produced a point in the middle. But Betsy was strange. We started to input audio with a synth and she began by responding very slightly to the notes on the synth. As we continued, she outputted a high pitch sound that started to play over the synth. The light in the middle got brighter and brighter, the sound from the TV got louder and the sounds from the synth got softer and softer. Then something really weird happened, the synth stopped outputting different notes, only the high pitch noise that the TV made, even after we unplugged it from the TV the keyboard would only make one sound at one volume. So we stopped using Betsy and gave the synth a couple days rest. 

So we need more TVs and a better method. We called around and turns out many people wanted to get rid of their old CRT TVs, so with a little organizing, two cars, and three sets of arms, we picked up 6 more TV in one day. We also got some speaker that would relay signals and protect any input source and allow us to hook up the wires directly from a phone or computer.

In the subsequent test on Carter and Darrel ran into similar issues. A problem we ran into was to do with the year the TVs were made. Basically there is some kind of safety mechanism in the TVs from 2000 on that reads the voltage going from the motherboard to the electron ray that will shut off the electron beam if there is a difference in voltages between the sides of the coils. So we only found this out after having the TV work until we tried to change their inputs. We could produce a horizontal line (cutting vertical deflection coil wires) and a vertical line (cutting horizontal deflection coil wires), but as soon as we put any signal through the cut wires the TVs would power off.

Horizontal beam created on Eddie (the fuzz around the bottom of the line is the created from the audio signal

Vertical beam created on Eddie

The first signs of success we had was with the fifth TV, Eddie, an older TV without care for annoying safety mechanism. We were able to use an audio interface to wire signals to one of the directions in the TV. But the only thing we were able to produce was small variations in either direction with the sound that we had. But as the signal continued, the beam seemed to split and made a similar high pitched to Abigail and Carter. The next day when we brought it to class to show off our success, it failed to even turn on, so it probably died some time after the previous nights tests.

The last thing we changed was the audio interface, LEPY, that had two inputs for microphones, two left and right outputs for a speaker system, and a voltage adapter. We also tested on some of the older TV, Fritz and Gertrude, but we did the same process as all the TV beforehand and able to finally achieve the oscilloscope goal!!! And we tried it on some willing participants that had fun with the interface!

Heavily inspired by this piece, we began thinking of ways we could use the human voice to create a visual composition between two individuals. By using just the X and Y inputs, each voice could control one axis, effectively rendering the interaction of two human voices as a single visual abstraction. 

Luckily, old CRT televisions from the 1990’s use very similar technology to an oscilloscope. We can modify these televisions to control the X and Y coordinates of their screens using audio inputs such as microphones, instruments, or anything else that produces noise. The ultimate goal is to create an interface for two individuals to use their own voices to create a collaborative composition, thus facilitating human interaction through their own vocal cords. 

Below is a link to Andrew Duff’s live show. Watch a few minutes of it to see these visual pieces become animated, and also see how they relate to the audio signal. 

This is important because one of the factors we want to identify is not just the GPS location but also the orientation. The north direction reads as “0.0″ and as you rotate the device clockwise 90 degrees the reading moves to “90.0″ to indicate the east direction. Here is a video showing the device working and the output data being displayed on the screen. These readings will be key as we work towards connecting the two devices later this week. 

Once we received all of the materials and parts for our project we were tasked with modifying a socket.io web page that Dimitris had provided to us. We implemented the GPS coordinates from the SparkFun GPS Breakout - Chip Antenna, SAM-M8Q (Qwiic) to this socket.io web page.