Monday, January 11, 2016

Home - Experimenting with soilless growing

I've been accumulating parts and now I finally have enough to start growing food indoors.
My goals are to learn more about hydroponics and growing plants without soil, to validate some assumptions that support my designs and of course to grow some food that I can eat.

I will post updates periodically. I may even bother to take pictures, but no promises. As of right now I am just under $200 into the project and hoping to get a variety of crops for no more than another $100. I could have bought a lot of vegetables at the store for that much cash, so hopefully this works out.

Read on for details.

My first attempt (strawberries) failed due to operator error. (Fungus caused by poor drainage.)

Product links below are through Amazon and are for illustration only. These are not sponsored links. I strongly recommend doing your own shopping / research since products and prices change frequently.
Hopefully it doesn't need saying, but this project requires wiring and is dangerous if you don't know what you are doing. Please don't electrocute yourself or anyone else.
I am using a very DIY approach. If you are not interested in learning about plants and electronics or troubleshooting cobbled-together systems, go for a retail growing machine. They are expensive and their supplies are expensive but they seem to be reliable and should be fairly easy to use.

The base of the system is an O-Grow bloomhouse we picked up for about $30. It has a light steel tube frame, four 18x24 wire shelves and a zippered plastic cover. This is designed to start plants in early spring, but I will be using it year-round if possible. I wish I could afford Metro shelves.

I have it near a window, but there is definitely not enough light. I am supplementing with LED light strips; these are intended for theatre, mood lighting and entertainment so they are less than $10 per 5-meter roll. I have four rolls of deep-red, one blue and one white, or about 360 watts. I used SUPERNIGHT brand; they are working well so far. That should be enough for 1.64m² of vegetables or greens, way more than I have space to grow right now. Still, the tape is cuttable and has solder pads, so I can make circuits to fit individual shelves. When I bought mine there were silicone-sealed versions for a bit more money; if I were to do it again I would buy that type instead since they should last longer in a humid environment.

These are powered right now by a 360w 12v switching power supply for $24. It's also a SUPERNIGHT unit. It does not come with any cords, so you will need to make your own power cord to match your local outlet. The power supply is controlled by a GE digital timer for $14. The timer manages to be less confusing to program than a VCR, but not by much.

The lights are attached to the shelves with velcro plant ties. These can also be used to wrangle foliage; comes as a long roll so pieces can be cut to size as needed. I had to cut little divots into the sides of the velcro to get it to fit between the LED emitters on the tape, but it holds everything together nicely.

Throw in some various bits and bobs (soldering iron, fine tips, a multi-arm gripper with loupe, silver solder and a reel of heavy-gauge speaker wire) and I'm ready to get dangerous.

I used a couple of old cookie sheets (scratched and rusty) as trays. It's only enough for one shelf; depending on how this goes I may invest in better trays but I don't know yet what depth(s) I will want.

My first attempt was bare-root strawberries. I planted them in store-brand potting soil in 4" plastic pots. This was a terrible idea. They sprouted wonderfully and grew like wildfire until some kind of fungus devoured them. I tried using a garlic-based antifungal mixed up in my kitchen (recipe supplied by internet rumors), but ended up killing the plants (or perhaps just hastening their demise).
Long story short, strawberries like sandy soil and good drainage. Potting soil will cause them to rot.

The next attempt will use Leca / Hydroton / Plant!t clay pellets. I also picked up some 1" rockwool plugs for planting seeds. I want to grow radishes, but I think that would work best by sticking the seeds between two sheets of tissue paper (not paper towel or toilet paper, actual tissue), laying the sheet on a small bed of media and putting a thin layer of media over the top. Should be interesting to see if radishes grow into a proper shape when grown in these big irregular pellets. I'll be trying strawberries again as well.

I do not have a pump yet, so anything requiring circulation is out for now. I'll have to try some kind of raft / wick arrangement with the clay pellets, so hopefully that works out ok. It is known to work well with lettuce so I will try some black-seeded Simpson in rockwool.

I'm probably going to need several more things to really get going, but hopefully not too much more money. I have cats; we buy litter that comes in large plastic buckets and I have been saving those for the last two years. They look to be ideal for trying passive hydroponics just by cutting holes in the lids and using net pots. They should also work well for holding larger plants like tomatoes; I will try rigging lights off the side of my shelving. Other than that the big missing piece is a pump; I have a second timer so I can automate a flood and drain cycle, but I'd like to try NFT as well.

I may still need to try coir or another fine-grained media for root crops like radish. If I have a chance I'll get a few pounds of in-shell peanuts, save the shells and try making my own media.

Updates will be few and far between but I will report on any interesting developments.


  1. CELSS (closed ecological life support systems) seem to be hard.

    Folks sometimes point to the Biosphere fiasco in attempts to debunk this notion. But I believe (hope) simpler, robust ecologies can be designed. Please post updates on your efforts.

    I guess a big CELSS problem is dealing with waste. A sewage system that returns discarded CHON to the system is desirable. In a small enclosed environment that can be a challenge.

  2. I would like to point towards aquaponics, which not only results in plants, but also fish. Its sometimes called bacteria farming, as its main component are bacteria, which can live in biofilters, that process the waste products of fish (and potentially humans) into something the plants can use as fertilizer. You only need to feed the fish and get nice plant life. You can also feed some of the things we don't eat of the plants to the fish.

    Gravel should be used instead of soil, through any hydroponic grow style works well, like floating plants in a fish tank, with a biofilter for the bacteria.

    For the most part it is, other than straight hydroponics, mostly a self regulating progress, though you have to keep an eye on the ammonia levels of the water.

  3. Thanks for the comments.

    I would argue that Biosphere 1 was not a serious attempt. The people involved were not scientists and had no evidence to suggest the system would actually work. It was an unfortunate incident, one that set back research on the subject for many years. It seems like NASA in partnership with certain universities is the only organization doing controlled environment ag studies these days, or at least the only one publishing articles open-access.

    There are two approaches for recycling waste; as mentioned by Warringer the most common is bacterial remediation. Microbes convert the nitrogen-bearing compounds into nitrate (amino acids to urea / ammonia to nitrite to nitrate) and the rest ends up mainly as CO2, methane, hydrogen gas and water. This includes practices like composting as well as aquaculture. In fact, algae processing of human waste (first using Chlorella and then Spirulina once it was {re}discovered) has been suggested often in the context of life support systems.
    The other approach is chemical. Most waste treatment plants use a chemical pass (flocculation) to separate wastes, then a filtration step to produce clean water. The waste stream is then processed by bacteria to recover nutrients (especially phosphorus). I'm assuming a community-scale waste processor would use supercritical water oxidation as the first step and distillation plus remineralization as the second step for producing clean water. Residues would be CO2 and ash for the most part, though durable oxides would also accumulate.

    I've kept fish, both bettas and mildly saline community tanks. I had live plants but not enough to draw down all of the nitrogen generated by the fish. The tap water where I lived at the time had nitrates in the 60-100 ppm range (depending on rainfall), so nitrogen levels were challenging to manage. I certainly don't miss the almost monthly warnings about choliform bacteria. I left my tanks behind when I moved and do not really want to take on that level of effort and responsibility again.
    Even so, keeping ammonia levels to a minimum is easy if you avoid sudden changes to temperature or chemical composition of the water and always keep a refuge for bacteria. I was terrible about cleaning the substrate (sand in one case, gravel in the rest) so there was always a ready population of bacteria available. For those who are more rigorous with their tank cleanings, do not clean or replace your filter at the same time as a thorough substrate cleaning.

  4. My long-term habitat plan calls for multiple approaches. First of all, an aquaculture system is definitely part of the plan; direct recycling of nutrients in fish waste would minimize energy needs for wastewater treatment. Any biomass suitable for making plastic feedstock (ethanol) would be fermented and distilled; high-protein waste would be used as feed and the rest would be passed on to the reactor. Any biomass not suitable as food or plastic feedstock is fed either to animals directly or to insects which are then turned into animal feed. Waste from that process (as well as human waste and nonmetal refuse) is passed through a SCWO reactor. The reactor's waste outputs are CO2 and concentrated mineral contaminants. The minerals from that process are used as nutrient solution for Spirulina; anything not bio-available is extracted through chemical means and converted to soluble nutrients. The Spirulina is fed to humans, animals and insects alike as a supplement or can be autolyzed into a nutrient solution suitable for hydroponic use. CO2 from the reactor is piped into hydroponics areas to maintain an elevated CO2 level in the 1000 ppm range. Multiple nutrient cycles are closed in this way, not just C-N-O but also N-P-K and traces, while being self-balancing. An excess of one particular nutrient would accumulate at the Spirulina step without allowing toxic levels to reach food crops; metabolites and byproducts are destroyed in the reactor and cannot accumulate.

    I do not plan to reach that level of complexity in my own meager experiments unless I were to become unexpectedly wealthy or employed in the field. Neither is likely. The most I can hope for is that my cobbled-together yield figures might be supported or refuted by actual observation, which will tell me whether the error bars are so wide as to preclude meaningful predictions.