Physical Computing - Week 6 (24-10-08)

11/3/2024

Witch's Brew Cocktail - A Visual an Audio Experience

For our P-comp midterm project, Julia and I were interested in exploring light readings through water and how ink droplets can obscure light readings to alter a sound output.

The final project became a collaborative interaction that allows a few people to add onto each other's input, whether it was to add different ink colors or to stir the mixture. Ultimately changing the volume of the sound output as the water becomes more saturated, creating the final "Witch's Brew Cocktail" (to embrace the halloween theme).

Project Timeline

Bill of Materials

Testing (10/04/2024)

This project roots from Julia's inspiration of ink swirls in water and how the ink eventually emulsifies.
After some experimentation, we decided on acrylic ink (left image below) as it maintained a higher level of saturation as it dropped into water and presented clearer swirl shapes in comparison to the other pigment sources such as gouache paint (middle image below) and acrylic paint (right image below).

Acrylic ink dripped into water

Diluted gouache paint dripped in water

Diluted acrylic paint dripped into water

Additionally, we tested with different vessel shapes to study the visual presentation of the ink.

The bowl (top image on left) offered more width and focused target for the ink droplet to sink towards, but did not provide enough height for sufficient visual "sink time". In other words, the ink dropped and reached the bottom too quickly.
The cup (center image on left) allowed enough height for us to visualize the ink swirl effect, but the volume is still limited.
The soup container (bottom image on left) gives ample volume and height for the ink to swirl but it sinks in an unpredictable manner which requires multiple ink drops to create enough change in light readings. This can cause quick over saturation and shorten the possible interaction time.

To work with something we already had, we opted for a cylindrical vase that I had at home which is similarly shaped as the soup container.

Design Development (10/07/2024 - 10/11/2024)

Before proceeding with fabrication we met with residents Josh (10/9) and Kai (10/11) to get advise on designing with lighting and tips on fabrication. Probably one of the smartest decisions we made for this project and really helped us descope the project down to a more manageable and forgiving design.

As you can see below, our design went from very organic shapes (sketches part 1 left image) to a tiered design with exposed components (sketches part 1 right image), and finally we settled for a rectangular structure with a nested form inside to house the phototransistors (sketches part 2 right image).

This led us to our paper models.
Starting with a 6" wide x 6" long x 6" high housing structure, we experimented with 2 internal variables:

The distance between the top opening and phototransistor (left image below). We worked with a 4" high tube, 3" high tube, and 1.75" high tube.
The hole opening size (right image below). We tested a small hole with a 0.5" diameter, perforated holes, and a large hole with a 1.5" diameter.

Ultimately, the average readings were all very similar and the variation in noise seemed to be consistent, so we opted for the combination that provides us with the brightest readings (1.75" tube + 1.5" dia cutout), which allowed for more change in values as the water became more obscured.

Readings from 1.75" tube + 1.5" dia cutout

Readings from 4" tube + 1.5" dia cutout

More Testing, but with code (10/14/2024 - 10/15/2024)

Before moving on to fabrication, I still had to figure out the code for our sound output. There was a lot of discoveries while learning the mozzi library. This portion of the project honestly feels like a fever dream and I'm still not entirely sure how this works.
To summarize my understanding

Sounds are made of waves
Waves are mathematical (trigonometry)
Code is made up of a lot of math

Somehow, I figured out the volume control aspect of the mozzi sample code (see video on left), but still wanted to replace this buzzing chord with an ambient sound for the project.

Fabrication and Troubleshooting (10/21/2024 - 10/22/2024)

This is arguably my favorite part of the project.
In our final design, we opted for a fully enclosed white box that is reminiscent of a museum pedestal.

There are two cutouts on the side for the speakers and a small mouse hole cutout on the back for the cable passthrough which connects to the arduino.
The case is made of laser cut white opaque acrylic.
A translucent cup secured to the underside of the top panel holds the phototransistors in place.
The breadboard is secured to the back acrylic panel via magnets.
The speakers are screwed onto nuts which are super glued to the side acrylic panels.

During this evening, students in the other classes were working late as their midterm projects were due the next day, so I decided to join them and continue tweaking the code in hopes that I could channel the "final crunch" energy and make more progress in creating an oscillating ambient sound.

Guess what? it worked!! I was able to combine two mozzi example codes and make minor changes to accept a volume multiplier.

The next step was to review whether the ink drops can replicate what the hand does during the testing and fabrication phases. Of course, it did not. The hand is able to move at different speeds and cover the phototransistor with much more flexibility for a more dynamic sound experience, so just the black ink alone wasn't going to cut it.
We experimented with different water levels in the vase, but the shallow water only sped up the change in light reading rather than creating a suspenseful change of state. We ran out of steam at this point.

Final Testing and Tweaking The Code (10/25/2025)

After a few days away from the project and some decent sleep, we met back together to make some final changes to the project before presenting to the class on 10/29. We made minor tweaks to the code so that the audio experience was not so subtle that it would be buried by the visual aspect.
Julia also suggested that we reintroduce different ink colors to see how it would change the light readings, which was truly a brilliant idea.
It turns out, the black ink was so heavy and opaque, it created visual dramatic effect as it sunk to the bottom of the vessel and substantially altered the sound state due to minimal immersion.
The magenta ink would immerse into the water easily which meant it lowered the light readings in small increments.
The blue ink swirled beautifully in the water but would not fully immerse into the water, partially settling at the bottom of the vessel, a nice average between the magenta ink and black ink.

Final Presentation

video to be uploaded soon!