For this project, I wanted to explore concepts I hadn't tried before (technology outside of capacitive sensors). I also wanted to create something purposeful that could solve a problem. Before I chose my final project idea, though, I had three ideas.

I decided that my final project, named "Dēng Lóng" (meaning lantern in Mandarin, while dēng means light and lóng means dragon), would be a mix of the carbon monoxide sensor and the carbon dioxide sensor lamp with a twist (the second and third project ideas above). I liked the use of the carbon monoxide sensor, as well as the concept of a light that would notify the user when gas levels were high. And instead of the lamp, I'll design my own casing that takes inspiration from Chinese lanterns. A successful project will have a functional CO sensor as an input, LEDs as an output, a clean casing for electronics, and an accompanying app that uses a bluetooth module. If there is additional time, there will be an additional I2C LCD screen that lets the user know the safety of their immediate environment.

For specific parts of the project, various tutorials and projects were very helpful. I used an Instructables tutorial by BlueMatter as reference for the wiring of the bluetooth module and my MITAppInventor app. I used a Hackster.io tutorial by Ingo Lohs as reference for the wiring and code of the carbon monoxide sensor.

Overall, I had to be thoughtful in the design of the project: the sensor had to have uninterrupted access to the user's immediate environment, and I wanted everything to look as clean as possible.

Now, onto the creation of Climate Lantern!

I created a clearer sketch of my design so I could have a better reference for CAD designing. I focused on tabs and clearance values, but many of the larger dimension values are subject to change. Note: the +1 and +2 values are not in any specific units, just in relation to the clearance parametric value I set.

I then worked on making all the sides of the lantern in Fusion360. Here are the pieces, as well as the parameters I worked with:

Meanwhile, I also designed a hook for the casing (which unfortunately broke after printing when it fell on the ground). I later redesigned the hook to make it look like the red hook in later pictures.

Afterwards, I wanted to paint on it to make it look cooler. I used watered down acrylic on the wood to stain it a little red. I also painted visible brown tab sections red, as well as a dragon on the bottom part that continued around the base of the lantern! The casing's done! (The last picture shows what it looks like when lit up).

Starting with the sensor, I followed this Instructables tutorial to figure out how to hook it up to an Arduino. After struggling with the code from the same tutorial for several days (while overheating and shorting a sensor), I found this tutorial from Hackster.io that provided the proper code. The MQ-7 sensor works in heating cycles; it was important that the code was able to change the voltage output and the reading timing. A picture of the setup is below on the left. The actual setup was rather simple, with ground, 5V, and A0 as the analog pin. I then tested the sensor with a lit match inside of a glass cup. I would light the match right before the reading timings, and I got good results! At normal levels, the sensor readings hovered around 1-3. When there was significant smoke, the sensor would spike to 15-16.

I then moved onto the LED's. I hadn't worked with them before, so it took a couple days to fully understand how they work and should be wired. I observed this image of Nil Peguero's project shown below (to the left) and sketched out how I saw the lights should be wired (to the right).

I then made a small test board with 3 LEDs that could easily be extended into a circle shape to test things out. I designed the board in EagleCAD and milled it out. The file can be found in the folder toward the bottom of this page, but here's how the board turned out. I soldered on 3 LEDs and 3 0.1 uF capacitors. I hooked it up to an Arduino (with ground, 5V power, and A0 as the analog pin).

I programmed it with the test code in the file in the folder, and it worked! At first the white light was a little too harsh, so I adjusted the numbers to what it is in the file now in order to make it a bit more yellow.

I then moved onto the bluetooth module! While I was waiting for the board module to come to the lab, I created the MITAppInventor app based on the aforementioned Instructables tutorial. Here are pictures of it:

After trying for almost a month, I realized that I couldn't get my app to work. I went with my original project plan of just the I2C LCD display, MQ7 sensor, and LED's.

I started putting the previously tested modules together, now that I knew the code I was going to use for each and the wiring layouts. I sketched together what my final board would look like with an ATtiny1614.

I designed the board above on Eagle, milled it, and soldered it. Here's what it looked like:

Because the jtag2updi programmer I used in Fab Academy couldn't be found, I programmed my board with my friend Teddy Warner's programmer! After it didn't work, I decided to test each individual section. After each section worked, I redesigned the same board more carefully (interestingly, though, all the traces were the same). This is what it looked like!

After it worked, I made another version of the board with the LED's on the backside. I made sure the design could sit under a rectangle of the casing. I used this video to learn how to make double sided boards in EAGLE. This is how it turned out after I soldered it! I first put in the vias, then soldered everything down starting with the LED side.

I combined all my test codes into one Arduino file (the _____.ino file), and it worked! Here it is working:

I then imagined what the electronics would look like inside of the casing. I then edited the casing dimensions as mentioned in the previous section, and I was done with electronics!

One more look at the final result:

All the files are in this Google Drive folder! All final files are in the final files folder.

Overall, I loved working on this project. I experienced numerous challenges throughout this process (many of which are not on this webpage but all of which are on the Google Doc embedded toward the top of this page), from various boards and modules not working to having to constantly readjust the casing design. I learned so much too. I learned about how the MQ7 sensor works and how heating and cooling is not only important but can also be controlled through code. I learned how to paint on wood, and how to make acrylic more opaque for diffusion purposes. But I also learned how the better apply my previous knowledge of project timelines and expectations to real life.

This project, more than any before, I actually met overall time goals---finishing electronics before Spring break and finishing the casing after coming back. I had to consistently show up in the lab; 2 hours each day early on led to me being able to finish on time. I thus also learned how I work as an engineer. I love spending more time on the design end of a project, laying out what I want to see and how I want to make my project. Carrying out the project becomes much easier this way. I've learned that I love designing and that I love piecing together a project like a puzzle until the day it's due. That's what I did this time too.

In the future, I hope to better design a hook so that my project can be hung. I hope to add more gas sensors too, to make my Deng Long lantern more useful and applicable in numerous circumstances. I think the project itself right now is rather cohesive, so if I were to add more to my project, I would have to re-put electronics in a separate casing, building upon the ideas of the project right now.