SunCalc is a gardening tool designed to help busy people better understand the light levels in their home and yard. Academically, it was an experiment in hardware prototyping and documentation of process. The documentation website can be found at http://suncalc.weebly.com. We also created an Instructables guide, which won 3rd place in the Instructables 2016 Circuits Contest.
Group members:Gayatri Gaekwad, Pierce McBride, Lyric Liu
My roles:Brainstorming, Research, Hardware Prototyping, Soldering
SunCalc is designed for busy people who are unfamilar with the sunlight levels around their homes and do not have time to check and calculate hourly light levels. SunCalc watches over the spot it is placed in and tells users how much daily sunlight is present, measured according to the sunlight categories you see listed on seed packets (Full Sun, Partial Sun/Shade, Full Shade). This allows the user to know what kind of plants might be suitable for a particular area of their garden, or to troubleshoot why pre-existing plants might not be faring so well in a certain location. My role was predominantly researching about plants and sunlight levels, helping choose electrical components, and soldering the prototype.
Designing for various types of plant species
The largest challenge of this project was to design a helpful gardening tool for various skills levels and a broad spectrum of both indoor and outdoor plants. Despite covering a large range of scenarios, we also needed our product to be simple and easy to decode with a single glance.
To accomodate all these factors, we focused heavily on the way the device would represent how much light it recieves. After asking questions such as "How much light does a plant need?" and "How do sunlight needs differ among plant species?" we arrived at our solution: The sunlight metric listed on the back of seed packets. These instructions are accessible and used by all ranges of expertise, and could easily be encoded into our design.
As a result, we mapped our 6 LEDs to an hour of sunlight each, with each set of two representing a category: Full Shade (defined as 1-2 hours of direct sunlight), Partial Sun/Shade (3-4 hour of direct sunlight), and Full Sun (5-6+ hours). Anything above 6 hours is still considered Full Sun and has no special terminology, so lights were not added beyond this.
Researching and choosing compatible hardware components was also a challenge for this project. Due to tight time constraints and limited prototyping experience, we wanted our initial hardware order to leave our options open to any later conceptual revisions. We chose a Flora board due to its abundance of available pins and ordered a high capacity battery to ensure our device could stay powered for well over 24 hours.
We considered opting for a solar panel rather than battery, which is a strong alternative from a conceptual standpoint. However, with the deadline looming and shipping times to consider, we went with a safer choice due to not knowing for certain how much power a solar panel can provide or how much power our particular device would require.
We also ordered a WiFi adapter in the hopes of using Adafruit I/O to graph and track sunlight measurements over time. In the ideal version of our product, the adapter would allow users to check these easy graphs whenever their busy schedules allow. However, due to time constraints and power concerns, this was not used in the pictured final prototype.
The inspiration for the concept came from looking at various existing DIY guides, as per the class instructions. We knew we wanted to focus on a gardening aid, and saw many examples of water measurement tools for plants but few for sunlight. Hence, we chose to create a sunlight calculator in order to contribute a device that has not been documented quite as much.
Next we began to research what plants need in terms of sunlight, and common ways that might be measured. We settled on the seed packet terminology as a standard and easily relevent way to measure light. Discovering that Full Sun, the highest level listed on seed packets, cooresponds to 6+ hours of direct sunlight, we decided to have a maximum of 6 lights on the device.
Sketches & Mockups
From here, we began to sketch and 3D model our prototype. We iterated on the sketches and case design to adjust for which parts needed to fit inside the device.
Due to technical difficulty and time constraints we removed the WiFi adapter, which was originally intended to provide users with graphs and records of long-term sunlight data via Adafruit I/O. The port it was supposed to go in was repurposed for recharging the battery. Aside from the issue of connecting to WiFi, however, the device is otherwise able to store long term data, just unfortunately unable to display it to the user in the final prototype we created.
We did a single test print of the case (green) prior to the final case (white.) The test case came out rough, but was useful for planning the arrangement of componenents during soldering.
We tested and calibrated the sunlight meansurement capabilities of the UV sensor by taking it outside and keeping track of its console outputs. After calibrating it to take a UV reading every 15 minutes and determine if the sunlight levels are high enough to count that as an hour of direct light, we sealed the components in the final case.
Given more time, there are a number of issues I would like to address. If this were to become a commercial product, the safety of the lithium battery in the sunlight all day might be a issue. Additionally, the way we installed the switch for the battery did not fully work as intended, possibly due to the Flora board protecting itself from batteries suddenly disconnecting in this way. More research into the full details of the electrical components would be a good idea. We had discussed using solar panels instead of a battery, but due to the short timescale we did not get a chance to test whether minimal levels of light could reliably power the LEDs for 24 hours. Conceptually, however, this would be an obvious choice to explore in the future.
As mentioned above, I also believe that connecting the device to Adafruit I/O would be a great way to improve it. Logging sunlight statistics over time would be very valuable, and ease the required attention from the user, who must check it every day with the current prototype. We reserved a port for a wifi adapter for this very reason, but ultimately decided to use it for recharging the lithium battery instead.
Overall, this was a very fun and rewarding project. I could imagine using the SunCalc myself. There are many things I would like to improve, but for a short-scale project in which most group members were new to electrical prototyping I feel that this was a success.