The following scientific principles were broadly applicable in Edison Co.'s design process and were used by every prototype.
Edison's design had to be based off of the research they have done for sleep inertia, because to find a solution it first takes an understanding of the problem in order to fix it. The science behind sleep inertia involves the homeostatic and circadian processes that happens in the human body causing cognitive impairment. As the body is suddenly woken, these processes can be interrupted, causing the cognitive impairment commonly known as sleep inertia. The jarring sounds inherent to traditional alarm clocks play roulette with the body's delicate cycles and balance, and can contribute to sleep inertia. A slow wake up helps preserve the body's natural state without disruption to these systems. With this in mind, a product which uses sound and light in a non-jarring, gentle way is vital.
Coding was used for activation of the prototype system in testing a proof of concept for the product. Coding is the team's bread and butter, allowing them to have all of their individual components in working order for the proof of concept design. For sound stimuli, a Python script hosted off of a Raspberry Pi was used. As for light based stimuli, no code was necessary for the first prototype, as the first iteration used a pre-built system not ideal for the final product. For the second prototype and beyond, the team hopes to build a software from the ground-up which will control both lights and sound with one easy to use system. Like with all code, Edison Co's code could always be improved, and is their primary tool used to solve the problems stated in the Sleep Science section.
The Engineering Principle used in this prototype was documentation using the PLTW Engineering Notebooks, allowing for "official" documentation and notetaking on the continuation of the project. The Engineering Notebook was a vital tool for project organization.
A Kanban board is a common Agile methodology to help teams keep on-track and informed. Cards, which can be physical or digital, are moved through several categories on a board (i.e. development, design, testing) as the project progresses.
Edison Co. was tasked to use a Kanban board for their project management needs. Specifically, the tool Trello was employed. Members regularly checked the board, moving cards along as they completed tasks. Due to the fact that every Edison Co. member had an account with Trello, the company was able to add their user profiles to a task when it was assigned to them. The checklist and user story print tools also became valuable.
Below are several example captures of the Edison Co. Kanban, showing progress as tasks move through the board.
For the sound stimuli, an mp4 file of a custom alarm tone was run through a program designed by Parker to get the desired effect for the product. A Raspberry Pi 3 and 4 were used with a basic OS to program the speaker to play the mp4 file at a set time. The fading effect was achieved using the mp4 file, which faded in. This audio file was custom-made using royalty free, stitched together in video editing software.
The team needed to incorporate light and sound stimuli that would become progressively more potent as time passed. Both types of stimuli would need to triggered in a way such that the overall effect is a gradual wake-up. For the light stimuli in Edison Co's first prototype, it could only be turned on and off at a certain time through a smart plug discussed in more depth later. Using the Wemo app was not difficult, but it took some testing to get to know the inner-workings of the app. For example, to set a light timer, one had to confirm two separate prompts, otherwise the alarm would not trigger. This was not ideal, though it worked for a simple proof of concept.
Overall, the proof of concept was put together using scraps that Edisons Co's team already had on hand, such as an old Bluetooth speaker that Ely kept but hadn't used for a while. The same can apply to Parker's Raspberry Pi's, as well as the WeMo brand smart plug, all of which had been out of use for a while before being used in the proof of concept. Other things such as the paper lamps and lightbulb were all bought on Amazon and shipped specifically for the project.
The first major issue of the testing stage was troubleshooting the Bluetooth capabilities to Raspberry Pi OS in order for the prototype to play sound out of the team's Bluetooth speaker. Although Bluetooth was something that was incorporated into the OS, the ability to have sound be played out through a Bluetooth speaker required specific software to work properly. Parker troubleshooted this issue, and solved the problems that were stopping the code from working. In general, Parker had to pick up a lot of knowledge on how to operate a Unix-based system like Raspbian.
The second issue was scheduling a wakeup time, which was difficult to program on the basic Raspberry Pi OS. However, Parker made use of the Unix-native job scheduler "Cron" and was able to use it to trigger certain events at certain times on certain days of the week. It took some time for Parker to nail down and understand the Cron syntax, but it became a powerful tool once understood. Using Parker's code, Edison Co. was able to program the speaker to set a desired time for wake up. Then, the team used the WeMo app to set the time for the lamp to turn on. After that, the prototype functioned, and would turn on the lamp and the speaker at the same time, allowing for the desired effect.
Prototype I was largely focused on whether the design idea would be feasible. Thus, the team had to think logically about what format would be appropriate to display their data. In the end, the procedural and algorithmic approach of a flow chart was chosen. This way, one could see evolution in the functionality of the prototype as troubleshooting progressed.
Looking to centralize the system, Edison Co. looked to the Raspberry Pi to control the light-based stimuli. This would involve finding a way to dim and brighten the light using the Raspberry Pi. Fundamentally, this task sounds simple, but it is important to consider that the Raspberry Pi outputs a digital signal, which is either on or off. To combat this issue, the group used a technique known as "pulse-width modulation" (PWM), which is often implemented when there is a need to convert a digital signal to analog. PWM works by turning a digital signal on and off rapidly, therefore simulating a constant, analog signal which is lower than if the digital had been left completely on. By increasing the time length of the "off" periods, the signal can be reduced, allowing Edison Co. to brighten the light. Figure 1 shows a PWM signal dimming.
The Raspberry Pi provides a set of GPIO, or "general purpose input/output," pins for the user to either provide sensor input or electrical output. Additionally, there are several ground and constant 5V/3.3V pins located amongst the GPIO. Edison Co. used the GPIO pins strictly as output to brighten a test LED. The GPIO layout of a Raspberry Pi 3B (which Edison Co. was using) is provided below in Figure 2. To control the pins, Edison Co. employed the RPi.GPIO Python module, which was relatively straight-forward to learn.
Since the team had not yet received the dimmer they had ordered, Edison Co. decided to use a basic LED/resistor breadboard circuit. In a way, Edison Co.'s updated program which now used PWM to slowly brighten a light was the actual second prototype, and so a breadboard circuit seemed like the best way to test whether it would actually work or not. It only took a few minutes of research for the Edison Co. team members to refresh on breadboard use, though finding the proper resistor was a bit more of a challenge.
To determine whether PWM would work, Edison Co. used basic tactile output (LED brightness) to gather whether the prototype script had worked, so no unique STEM principles were applied in data collection for Prototype II.
Building Prototype III required a lot of electronics work. Many connections had to be cut and soldered. Luckily, while Parker, Jace, and Ely were rather inexperienced with the skill, David was skilled in this regard. He did a lot of the soldering in the prototype, and also gave both Parker and Ely demos of exactly how to do it, as in Figure 3.
The box housing for all electronics of Prototype III was 3D printed. Ely had a solid understanding of CAD and was able to develop the file, but Edison Co. needed additional help actually printing what they had designed. They turned to peer Kiwi Hamley for help in this process.
Edison Co. did work with several tools in the shop, including the scroll and band saw, in order to develop Prototype III. The scroll saw was used to cut out a slot for the GPIO pins, whereas Ely used the band saw to access some interior electronics.
Upon testing, Edison Co. took notice of several exposed pins and leads connected to the dimmer board. Each one of these vulnerabilities could potentially carry high-voltage, as David pointed out. By recognizing and understanding these issues, Edison Co. was able to address them safely.
Edison Co. made use of a spreadsheet to record their results for Prototype III, in contrast to the flowchart of Prototype I. Spreadsheet see regular use across STEM fields, namely engineering and software engineering, as they provide a wrapper for data that is easy to analyze, organize, and view.
The alarm lamp may seem simple at first, but there are a lot of individual parts that all need to work together. First off, the lamp is plugged into a Bluetooth smart outlet called a WeMo. The WeMo smart plug uses Bluetooth and a proprietary app to set a schedule for the lamp to turn on and off. There is also a Bluetooth speaker that hangs from the paper lamp which is called the Dot. The Dot is controlled by a Bluetooth-enable Raspberry Pi, which is in turn connected to a Wi-Fi router. The user's computer is connected to the Wi-Fi router, and the user can control the Raspberry Pi remotely through the router. The Raspberry Pi runs the code to make the alarm turn on and off at the right time.
One good way to validate our prototype is through Edison Co's cohort Ms. Mason, who has a background in coding and computer science and who has also worked with industry professionals in the past. With her knowledge in the subject, it is possible for her to approve Edison Co's use of CAD prototypes, and any other use of hardware, software, and further prototypes. She would be great for confirmation of how the prototype does and can give ideas for further prototypes later down the line. Other ways of validating Edison Co's prototype include the opinions of the team's peers, as they constitute a wide range of needs/interests, and can provide unbiased opinions from the perspective of a prospective consumer/user. They can provide insight on user quality and the effectiveness on prototypes and can provide criticism and future prototypes.
Hirzel, Timothy. “Basics of PWM (Pulse Width Modulation) | Arduino Documentation.” docs.arduino.cc, May 12, 2022. https://docs.arduino.cc/learn/microcontrollers/analog-output.
[Figure 2] Asassy. (n.d.). GPIO. In sparkfun. https://learn.sparkfun.com/tutorials/introduction-to-the-raspberry-pi-gpio-and-physical-computing/gpio-pins-overview
[Figure 4] Doyle, E. (2022a). 3D Print Progress.
[Figure 3] Doyle, E. (2022b). David Soldering.
[Figure 5] Doyle, E. (2022c). David Works on Scroll Saw.
[Figure 1] Eighthave. (2007). Pwm duty cycle. In Wikimedia Commons.
[Figure 6] Krug, D. (2022). Ely Using Band Saw.