top of page

Gavin Hughes' STEAM Honors Design and Development Senior Capstone Project

Jill Luoma-Overstreet - HDD Final Poster.png

Final expo poster

Professional review slideshow

30 second trailer

PXL_20220520_151040388.MP.jpg

Hey, that's me, famously tall person Gavin Hughes, I don't have any nice pictures of myself, so have one where I look happy instead.

Food deserts(3).png

Poster for professional review night

IMG-9079(2).jpg

Now you know what I meant by pictures where I look happy

One of the first things we did was go some research into things we think we could do that would improve the world, click the link for a kinda bad slideshow on my research into that, which will only exist for a couple more weeks because I'm not allowed to transfer ownership to a non-ahs account

https://docs.google.com/presentation/d/1DLiobQeOTMbrYYbE_tTMk4TJZg6RpZ4yPe_K9FKeuok/edit?usp=sharing

I defined my problem as "Healthy and sustainable food is inaccessible to many people", and found aquaponics as a potential solution

Once I decided on that and Hunter decided to group me together with Joseph and Lucas due to the similarities in our problems (immediately after telling the class how group projects always fail no less), we began research and planning and while I have consulted some other information, these sites were the ones that most influenced our initial ideas

Grow beds

https://dengarden.com/gardening/indoor-hydroponic-garden

https://www.nosoilsolutions.com/how-to-grow-hydroponic-bell-peppers/

https://whyfarmit.com/hydroponic-peppers-101/

https://whyfarmit.com/deep-water-culture/

https://whyfarmit.com/ebb-and-flow-hydroponics/

https://www.growertoday.com/best-ph-for-hydroponic-tomatoes/

https://twopeasinacondo.com/growing-guides/a-guide-to-growing-tomatoes-indoors/#step-4-light-and-placement

https://www.aquariumsource.com/aquarium-water-test-kits/

 

Fish

https://university.upstartfarmers.com/blog/best-temperature-for-tilapia

https://lakewaytilapia.com/How_To_Raise_Tilapia.php

https://www.ncrac.org/files/biblio/ncrac114.pdf

https://www.aquaponicsusa.com/products/aquaponics-tilapia-food.html

 

Sensor package

http://cdn.sparkfun.com/datasheets/Sensors/Temp/DS18B20.pdf 

https://randomnerdtutorials.com/guide-for-ds18b20-temperature-sensor-with-arduino/

I also did some research into food deserts , how they're created, and their impacts on the people who live in them

 In 2015, 23.5 million Americans lived in food deserts https://www.dosomething.org/us/facts/11-facts-about-food-deserts 

Just transporting food accounts for ~14% of the the food sector’s carbon output https://www.oregon.gov/deq/FilterDocs/PEF-FoodTransportation-ExecutiveSummary.pdf 

While exact numbers are complicated, an efficiently run fish farm could provide meat at 1/6th the carbon impact of traditional beef production https://www.greeneatz.com/foods-carbon-footprint.html https://thefishsite.com/articles/assessing-the-carbon-footprint-of-aquaculture 

People who live in food deserts have higher rates of obesity and diabetes, a fact that is worsened because food deserts frequently also have no or poor access to medical treatment https://www.medicalnewstoday.com/articles/what-are-food-deserts#health-impact 

While the last mile problem is more related to delivering goods to consumers, it still impacts large institutions. Shipping goods to central hubs is easier and cheaper than distributing it to all the storefronts. Alternatively, completely decentralizing would negate the issue entirely. 

(Context https://en.wikipedia.org/wiki/Last_mile_(transportation) )

2.PNG
1.PNG
3.PNG

Some of our very first prototype drawings, trying to figure out the basic form factor

Only a few weeks into the project we hit our first big hurdle, the PVC. The entire planning phase we had planned to take PVC pipe, slice it in half, and then just have open beds to put the plants in, we spent a good chunk of time getting everything right and in place, but when it came to glue the corner pieces of pipe, we learned that PVC cement requires pressure to bond, which meant that our half-pipes could never form a watertight seal. We tried several different ways of joining it before eventually being forced to just buy fresh pipe and keep it whole this time, only cutting small holes for the 3d printed baskets to go in

PXL_20220531_171522310.jpg
PXL_20220531_171440989.jpg

Top left is our only successful joining using the half pipe out of around ten attempts, top right is a tool we created to try and provide extra support and pressure to the pipe from the inside to see if we could get it to seal. Bottom left is the jig we designed to help cut the same size hole every time, bottom right is an example of that once completed

PXL_20220531_171535990.jpg
PXL_20220531_171445535.jpg
By the night of the showcase, we had just managed to hand up the final pipes and glue most of them into their final angles. Another more minor issue we ran into was that our hanging solution, using the J hook you can see in the top left picture did not provide us enough stability to allow for reliable water flow, so at the last moment we had to come up with a new, more stable, but less easily changed system of support pillars from underneath. We had working lights and a pump but no code for the relays so we just turned them on and off with the switch for demonstrative effect.
PXL_20220531_171505419.MP.jpg

How did we destroy the relays you might ask? Well, we were trying to power the arduinos off our power strip using only one plug instead of needing three, I was running a test with one of the boards plugged into my computer, while also being powered by the wall outlet. I thought something smelled funny so I stopped, got Lucas to come over and open the relay box, and sure enough, magic smoke comes wafting out. So we had to replace the relays and I made new protoboards that didn't include the now unnecessary wires for power from the wall.

Once we finished the showcase, I had time to work on more of the digital issues. I had already spent weeks getting the waterproof thermometers to work correctly, which was one of the most frustrating experiences of my life. But now the relays switched on and off on a set time interval. The top protoboard is for the thermometers, the bottom for the relays. The top one stopped working after I added that led and the relay board had to be replaced after we destroyed the relays. The paper is just some psudo code I had to write to figure out what I needed the relays to do.

There was a fun issue I had with the relay code, I was trying to get it so the lights would stay on for 16 hours while the pump cycled on once a hour, but for some reason, the lights wouldn't turn on. I could get them on in other pieces of test code, usually with some weird delay or flicker, until it finally struck me. I had a line in one of the test codes that I thought would set all the relays to off, this way I could just standardize which way everything was starting at, however, when I ran that code, the lights would flicker on for a moment. I had been using the command "HIGH", thinking it turned the relays to the on position, and "LOW" turned them to the off position, but it was actually the other way around. The lights weren't turning on because the code literally said for them to stat off for 16 hours. I felt a little silly after that one, but now I know I guess.

Here's the link to the github repository should anyone be so inclined, it's got everything

https://github.com/STEAMAquaponicsProject/STEAMAquaponicsProject

I was pretty insistent from early on that if we wanted plants to present by the expo, we needed to get started as early as possible. We had seeds delivered in the first or second of things, but Hunter said we should wait, and so I only planted them once I got back from spring break. This meant that by the time we needed them to be brought to the school, the plants that had sprouted large enough to be moved (mostly sage) were still pretty small, and none of them ended up surviving transplant. We ended up getting some parsley seedlings, which caused it's own problems because I hadn't finalized the relay code by that point and it was a Friday, so we didn't know if the transplanted plants would survive without light or water for the weekend. Most of them didn't, we had to prune away somewhere between half and two thirds, but the ones that did make it thrived surprisingly well. Well enough that they actually caused a new problem, the roots created enough drag in the pipes that they started overflowing.

PXL_20220531_234839927.jpg

The brave survivors now have a loving home in my back yard, they look a little rough but I did transplant them a few hours ago, I'm sure they'll be fine

This is what happens when I an impossible task. I haven't the faintest idea how to insert a stl file into wix, but I have found an arguably worse way to present the image.

There aren't really any wacky stories with the baskets, Lucas designed them in an afternoon, we printed them in big batches over a couple days. They're surprisingly durable and get the job one.

The only other super big thing that happened was expo night

 Lucas in the foreground I'm in back with the cool hat and extremely formal solid color flannel

Copy of DSC_6638.JPG
Copy of DSC_6634.JPG

Here's Joseph, probably telling very truthful things to those nice people.

Copy of DSC_6523.JPG
Copy of DSC_6554.JPG

Here's me on the right, explaining the data collection system with my laptop on top a pile of sweatshirts because the table we worked on was a little higher than the one during the expo and we didn't plan for the cable to be too short. One the left I'm accepting the "Gritty engineer" award for our team. It is vital no one explain to an adult why this is funny.

Copy of DSC_6750.JPG

Contributors

Joseph Morphew
Lucas Liabraaten

Hard working teammates, honestly seems like a bit of an insult to include them. We all worked our butts off to get this where it is now.

Mr. Hunter
Ms. Jill
Aaron Barth
Danner Kretchmar
Brandon Orr
Erin Johnson

Worked to get us the materials and help we needed, even when things started to get hectic by the end

Aaron helped so many times with the code and he also just checked up on us, asked how it was going. Kretchmar had many useful insights into keeping fish, even though we didn't end up going that route

My professional reviewers, both provided critical feedback, but also just helped me gain confidence in our system and the motivation to improve it for the expo

Copy of DSC_6662.JPG



Our goal was to address food deserts by providing a more sustainable, lower cost way to make produce and protein. Coming out of the showcase was to have alive plants, running water, and computer-controlled pumps/lights. We recognized that fish wasn't practical given time and ethical constraints but moved to future proof the system to allow for them should the next team want to try. We had to adapt after the failure of the half-pipe idea.
   The people who came to our table saw a shelf, glowing purple, filled with pipes with little plants sticking out of them. The lights were hung up by zip ties and the pipes were supported by bits of pipe, but the tops of the plants were still brushing the lights. There was a laptop connected to the side of it because we didn't have enough time for a self-contained digital readout. The water flushed back into an empty tank, we obviously didn't have any fish, and although none of them would know, the necessary addition of a settling tank. Everything looks yellow if you look at the lights for a little too long, but I'm not sure what we can do about that one.

Thoughts and considerations

    I've already mentioned the biggest challenges I faced. Getting the waterproof thermometer hardware and software to work right took weeks. Learning how to read and use C++ from scratch, using the incorrect technical diagrams, and scraping together every hint from the reviews on its retail page. Having to go back and add the ability for it to read more thermometers from one pin was harrowing, not because it was actually that hard, a few days at most, but because of how it brought back the memories from the first time I had to work with them. But, made it through to the end, even if I did have to redo the protoboard three times.
   Similarly, getting the pipes to work. Not even just the half-pipe fiasco, but once we went to whole pipe, we still needed to glue every corner at a specific angle, or else the water wouldn't flow. It was nerve-wracking. If we messed up towards the end it would have ruined over a week of work and a lot of good pipe, which we would have had to then get more of, trim to size, cut the basket holes, and then finally reglue the critical angles. We actually did mess up a couple of times, putting the angles on backwards, we had enough spare lying around that we managed, but we were down to our last piece by the end. It wasn't just mentally stressful, PVC cement is pretty unpleasant stuff, it stinks and the primer will stain anything it touches, including my hands, clothes, and the floor outside the makerspace. But, almost none of the pipes leak and the angles are all mostly correct, so it turned out alright.
   The last major struggle was probably that issue I had with the relay code. It seems silly now, but the fact that the HIGH and LOW commands were switched from what I assumed they were cost me at least three days, by then most of the critical stuff was done, but that doesn't mean I wouldn't have liked those three days to work on quality of life stuff like that temperature readout. Oh well, such is life. If we started fresh tomorrow we could probably have a seriously improved design in three months for way less money. But we're not. we're moving on. That's how it goes sometimes. Even if it moving on is a challenge, we all have to eventually.
   But it's not like it was all a waste. It was super cool to teach myself C++ and gain a lot of confidence in my ability. I'm not even good but I know what I can do and it's a lot better than what I could do 7 months ago, that's for sure. Likewise, I now have a decent amount of practical electrical knowledge. I probably shouldn't be wiring anyone's house but that's still not the sort of thing I ever thought I'd get to learn in high school. Lastly, it was just nice to have a solid team. We've all known each other for years. We've worked on projects in the past and have pretty compatible personalities. Just imagining what it would have been like to work with someone I'm bad with for 6 months. Even if I initially liked them, I can't imagine there would be very many positive feelings by the end. But that didn't happen, everyone put in as much as they could. We split the work, if someone needed help we helped and if someone wanted to do it themself, we split up to cover more ground. We decided things as a group but specialized in our tasks. We created our own terminology for the random things we needed words for that we had never come across before in normal life. It was just an all-around great time and crew.
   I'm not the most pro-social team-spirit guy, but I will say I'm a lot more inclined towards any similar I might come across in the future. While I have my issues with the STEAM program, I also think its taught (intentionally or no) me a lot of things. I've learned technical skills, but I've also learned how to teach myself technical skills when what we learned in class doesn't help. Self-reliance is crucial, but so is the ability to ask peers questions. Those things aren't inherent to tinkering with raspberry pis but I think it's probably a lower barrier to entry than a wet lab or anthropological institute, so I'll let it slide. I joined the STEAM program because I went to Anoka middle but was just outside the bus range for Anoka high, I didn't want to leave all my friends behind, and you got bussing if you were in STEAM. But I got in, and even if the class roster is full of... interesting people I've also met some of my favorite people, so there were probably a lot worse choices I could have made.

bottom of page