Overview
Contents
The roots
Starting off
It all began with an interest in learning how to use Arduinos. After a few years of working on robot foundry's design team (and many years prior as a student!) building an assortment of simple-circuit powered robots, my desire to implement microcontrollers and programming to a larger-scale personal project had been growing. I'd procured a few years of coding experience at the time, but had never before gone about integrating that experience with hardware. So, I decided to buy myself a small knock-off eBay arduino, and chose to learn how to program it though a project-based approach (as usual). While some people go about learning the language with simple LEDs, buttons, and perhaps a servo or two, as usual I was more interested in a productive and larger scale introductory project. After much brainstorming, I stumbled upon the idea of somehow automating the care of a plant. The idea would be to somehow control the lights and water automatically and dynamically, so that the plant could grow for long periods of time without any human involvement. It seemed like a fun project, and so I dove in, beginning with the most fun decision in the process: choosing what to grow. At the time, it was just a small educational project, but since then it's scaled farther than I ever anticipated.
Moldy Strawberries
As I began my research, I realized that I was actually more enticed by the concept of the project (mass-garden-automation) than the actual implementation of the Arduino. Learning a custom and slightly esoteric programming language wasn't something that particularly interested me, but having infinite access to edible plants was very enticing. Though I'd start out the project with a simple Arduino-based system, later on, as will be explained, I moved away from the microcontroller and chose a larger-scale redundant system for the plant automation system. For the research process, I started out by exploring fast growing, easily maintainable plants. I really wanted to grow avocados, but they take a very long time to grow and generally are full sized trees. There is, however, a specific variety of avocado called wartz avocados that are smaller, but one issue that quickly came to my attention is that once they reach maturity, which could take a very long time, the avocados themselves take a while to regenerate between harvests. Greens like spinach seemed to be popular choices for the type of system I was planning on making, but I wanted a crop that I could easily preserve over the long term, and unfortunately greens lose their texture when frozen. Berries seemed to be an appealing choice, given their quick germination and grow time, and so I eventually settled on strawberries.
Ultimately I settled on strawberries specifically, since I found that there are fairly common stackable grow towers readily available. The nice thing about the towers is that I could scale them to perfectly fit my closet, and could easily add more towers as needed. I ordered some generic strawberry seeds, and got to work cleaning out my closet. I lined the floor with a plastic garbage bag to prevent moisture from damaging the wood floor, and bought a Govee brand water leak detector in case of any emergency spills. I sectioned off the right half of the closet for the tower and then set up a large bucket with fertilizer-water and affixed the arduino to the inside wall of the closet with velcro. Parsing together some code snippets I found online I figured out enough Arduino to set up a system for automatically dispersing water when the soil dried up using a soil moisture sensor, and covered the closet walls in tin foil with the hope of maximizing the amount of light hitting the plants. I purchased a small relay, a device that lets a low power device like an Arduino toggle an external power connection to higher power devices, and ran the positive wire of an extension cord across it, so I could control the lights automatically with the same system. Cutting the extension cord felt a bit scary, but it seemed to work so I didn't overthink it.
With everything complete and operational, the seeds did indeed germinate as anticipated, but the watering system didn't go according to plan. As time progressed, it didn't water often enough, and when I tinkered with the watering threshold it tended to over-water and cause spillage. To compensate, I ended up needing to manually water from time to time, which defeated the whole point of the project. So, I ended up pivoting to timer-based-watering, which worked, though it was slightly less precise.
For the grow lights, I started out with a cheap Amazon grow light, but that didn't last long. After only a few days in the PSU melted itself, but since I had planned for the worst, it simply tripped my GFCI outlet (I ran an extension cord from a specially-protected laundry machine outlet, which has auto-shutoff for this sort of thing) and shut itself off. The replacement lights I bought were definitely better, but still weren't bright enough for the flourishing strawberries, so I ended up again purchasing higher quality more powerful bar lights. This third choice worked well, but it heated up my closet to an ambient 85 and drew 90 watts of power. The expense was notable (probably around $10-15 a month of extra power usage), but less than the quantity of strawberries I was expecting the plants to yield.
Success-ish
Within a few weeks the plants were off to a great start: they began branching out and runners (technically they're called stolons), which are baby plants attached to the mother plant, started shooting out of the lower parts of the stems. Cutting them off turned out to be important, since they took significant energy away from the plant's fruit production, and it quickly became a daily chore. This is definitely something I'd take into account for future plants, since it was quite tedious. On a few occasions I placed pots next to the plant to clone it, for strawberry-plant-gifts. Flowers eventually appeared on the plants, and then the first few fruits sprung out. I began using a basic "miracle gro" fertilizer at this stage, worried that the soil would be depleted of nutrients with the new fruit. The strawberries were indeed juicy, bright red, and aesthetically perfect.
Mold
The success unfortunately didn't last much more than a few weeks, which is when the mold started. A slight white powder type mold began appearing on the undersides of leaves, and it quickly spread to the entire plant. It seemed that it was "powdery mildew" that I was dealing with. I added a small dehumidifier from Amazon, and drilled a hole in the water tank to redirect the collected water back into the plant watering reservoir, which actually helped reduce the need to water the plant as often as an added bonus. I bought Neem oil as a natural fungicide to protect the edibility of the fruit, but it was futile at stopping the mold and only did so much as deter flies. In such a moist, enclosed setting, with plants growing in dirt, fungus was inevitable, but it got quite bad. Unfortunately I had to cave to a powerful toxic synthetic fungicide, which almost overnight completely ridded the plants of fungus, but had effectively made them ornamental for months to come. In the meantime I began planning for the next season, contriving a solution to prevent the fungus before it were to start. To me, the culprit was clear: it was the soil. I'd have to go dirt-less.
Hydroponic Transition
The Transition
Through the previous strawberry endeavor, I learned a lot. Keeping the environment clean is key, and consistency maximizes yield. Before setting up the next crop, I moved the strawberries out onto a deck, and completely bleached out the closet to kill all residual mold. While the Arduino was fun to tinker with, it'd be in the best interest of the plants to go with a more robust and consistent system this next time around. For the upcoming crop, I chose to implement hydroponics, a technique wherein the plants grow with their roots directly in nutrient water or mist, eliminating the need for soil. Hydroponics is really neat since the plants grow on average 30% faster and use up to 10x less water. I also opted for preventative biofungicide, which utilizes naturally occurring soil bacteria that 'eat' fungus: a fungicide that's not only non-toxic to humans, but scientifically proven effective. For the seeds, I bought generic basil seeds, which I'd soon realize was highly unideal. And, for the germination medium, I went with rockwool (instead of a make-shift paper towel like I did for the strawberries). Rockwool is a fluffy stuffing-like material, used most frequently in insulation, but also commonly botanically.
Assembled Grow Tower
For my first attempt, I decided to go with an Amazon basil grow rack, since it claimed to provide an all-in-one system that could assemble quickly and easily. A bit weary of leakage, I carefully put it together, and placed my basil seeds into the included grow pods. For the first few weeks, while figuring out ideal nutrient ratios and water acidity, the basil did successfully began growing. However, I quickly ran into a roadblock: the small tubes were being jammed by the ever-growing plants' roots. The larger they got, the more unwieldy the system became, and the less water they received. Moreover, cleaning the system was virtually impossible without removing all the plants and disassembling the tubes, since the roots were jamming themselves in the tubes. Unfortunately, I needed to change gears, but, on the brighter side, I did establish hydroponic basil as a plant with potential.
Custom Vertical Tower
The first idea for the overhaul was to build a vertical hydroponic grow tower. In the setup, a pump would turn on every 10 minutes for 10 minutes, carrying water to the top of the system, and letting it drip on and drench the roots. Vertical systems like the one pictured don't require an additional air system, since the roots are constantly exposed to direct air. Another key advantage is space efficiency: whereas in typical systems the plants height is limited because of bushing tendencies, with a vertical system the space of my closet could be maximized by having multiple towers of the height of my closet. Instead of buying a kit this time, I made my own vertical system by drilling holes along a large PVC tube, and then placing smaller segments of tube within the holes. Unfortunately the product didn't work out, since my pump was so powerful as to propel water out and along the edges of the tube, wetting the outside of the central tower and fostering algae. Additionally, the tower itself was particularly difficult to keep upright, even amidst many ropes and wires, and since this setup involves water dripping down the central tube, it was extremely loud. It was disappointing, but I wasn't defeated. It was time for another method.
Deep Water Culture
My first success: deep water culture (DWC) hydroponics. In DWC hydroponics, plant roots are suspended in a nutrient-rich water solution, with airstones constantly pumping air into the water, allowing the roots to maximally absorb the nutrients and minerals. The system itself is extremely simple on paper, but requires managing large reservoirs of liquid—ensuring that they hover around the right acidity level, that there's the right fertilizer amount, and that they stay fungus-free. The approach seemed much more feasible than any of my prior systems, since I'd just need a few bins and grow baskets. Agog to proceed, I got to work planning.
Setup
To start, I purchased three large storage containers from Lowes. While food safety is obviously of importance, food grade bins are astronomically expensive, so I settled on bins with food-safe plastic composition, even though they weren't technically rated food grade. My plan was to use a super large 3" drill bit and some 3" grow baskets filled with pebbles to hold the plants in place. The air pumps I used were initially small bubblers, but, after later learning that 1 wat of air pumping power per liter of liquid is the general practice in hydroponics, went with two commercial air pumps. Since they were extremely loud, I also threw in some vibration dampening pads that cut down on the sound a bit, and placed a dollar-store fan above the air pumps to prevent overheating.
Progress
- Baby plants
- Early growth
- Pruning basil
The system was all set up, so now all that was left was to wait. The plants germinated flawlessly, and were quickly ready to be placed into the pebble-stuffed grow cups. The tanks were full and fertilized, and the air pumps and lights' timers had been configured properly in advance. The new DWC setup worked like magic, requiring extremely low maintenance and yielding extremely high loads of basil. Unlike my strawberries, there was no soil involved, so the setup was significantly cleaner and more maintainable. Something that I quickly learned is that basil requires meticulous pruning to maximize yield, which explains why the basil from my previous tower-approach grew as vertical stalks. By cutting between the leaf stalks, I was able to get the plant to branch out into a bush, rather than the tall stalks I had in my older systems. With such rapid growth, the harvests quickly began rolling in. It was finally time to start producing pesto.
Pesto Production
- Pesto-packed freezer
- A harvest
- Bowl of pesto
While indeed the basil tasted great as a salad mix-in and pasta topper, pesto was the inevitable real deal. It was one of the primary reasons I chose basil: as the plants grew beyond my appetite, it'd be super easy to freeze the final product. For the first few rounds of pesto, I kept it super simple, just blending together basil, salt, and a bit of olive oil and garlic. As the batches went on, I started getting more creative, experimenting with adding walnuts and almonds, along with weirder ingredients like kale and hot peppers. Each week I reaped around 2 cups of pesto, which I froze in ice cube molds and then zip locked, so that I could microwave one cube at a time as needed. Over the weeks, the freezer filled with pesto, and, even two years later we still have a seemingly endless supply.
Basily Takeaways
The deep water culture system seemed to be the best route, yielding the greatest amount of basil while being extremely easy to maintain. I decided to keep up the system for a few years, slowly building off of prior years by improving the setup. I switched to special basil seeds, which have been bred to be resistant to the most common basil illness downy mildew. During my first few years I worried that the basil had mites, but shortly learned that they were in fact the buds of roots. I also spraypainted the bins with a layer of opaque black, followed by two layers of white, to prevent light from penetrating the water and causing algae. I also upgraded my air stone system to new, larger disc airstones, which provide a much steadier and higher quality flow of air. I also installed a fan in the ceiling of the closet, along with two smaller fans on the walls, aimed at the plants and machinery. Since the fans use lots of electricity, they are set on timers to turn off an hour after the lights, and on when the lights turn on. And finally, I've switched my rope out paracord, which is much more durable, easier to work with, and overall sleeker.
The New Shishitos
The choice
With my optimized system already assembled and ready for the next year's plant, I've entered research mode once more. Basil was great because of how rapidly it grew, but, equally important, that its product (pesto) could be frozen. Surfing through the vast array of potential hydroponic candidates, including tomatos, bell peppers, lettice, spinach, and more, I eventually settled on Shishito peppers. They were the perfect match: tasty, roast-able, and freezable. And, most importantly, unlike many other types of peppers, they're particularly fast growing.
Progress
Since germination, the peppers have been growing rapidly, day after day. After some research, I learned that in the past I'd been using way less fertilizer than I could have been, so I've increased my concentration. Additionally, I quickly learned that the new, cheaper tubing that I purchased this year was completely unsuitable for my purposes, as the acidic water quickly ate away at the plastic, and the tubing began to leak. My check valves (valves that you put on a tube that force water to only travel in one direction) worked, preventing the water from entering the expensive electric air pumps, but still allowed the water to leak onto my floor. Luckily I'd installed some water sensors on the floor, and was able to quickly dry the floor. Since this incident, I moved the air pumps to a higher location (even though the check valves worked, I was scared of the possibility of them not working, and having the water pumps higher up would make it impossible for the water to backflow into the pumps). I switched to a much higher quality PVC clear tubing, and have had no issues since.
At this point there are no actual peppers on the plants. The fertilizer I've been using is optimized for foliage growth, so I'm fairly certain that that has been why the peppers haven't begun to grow. Soon I'll be switching to a different fertilizer, which is optimized for fruit growth, which should kickstart flowering. After a bit of research, I've also come to find that peppers require pollination in order to produce fruit, so, while there is indeed a notable wind current in the closet due to the fans, it may not be enough, and I may need to learn how to manually pollinate the plants. Since this is an ongoing project, I'll be sure to keep updating this post as the peppers grow (and hopefully eventually make it into my freezer [and stomach!]).