This is an experimental, open source (hardware & software), automated indoor countertop aeroponic factory that uses easily obtainable electronics and 3D printable & laser-cuttable parts. The goal is to build a fully automated system that will monitor humidity (DHT22) and NDVI imagery (Raspberry Pi Noir camera) and perhaps other sensors to control the watering, which utilises a COTS ultrasonic mister in an aeroponic setup.
Lighting will be provided by grow-light(s), initially using existing lights on the market with basic timing, but I plan on using light sensor(s) to control the lighting as well in the long run, and a custom grow-light that can easily be put on top.
Separating out the roots more when transferring from the seedling container and leaving a bit of soil on to help retain moisture (I still think that root development during the seedling stage in the misting chamber will improve things, as finer roots will develop. That remains to be seen. I'm growing some plants from seeds to experiment with).
Closing up mist escape routes (see the ugly masking tape in the photo below) to retain more mist in the chamber as the main problem at the moment seems to be loss of water in the plant. The chamber is still connected to the outside air through the fan port. I just had an idea about this!
I came back to the office this morning to some unhappy plants:
They obviously didn't get enough moisture during the night. I can see from looking at the log from the Pi that the humidity was reported as pretty much constant so no misting was done, which is bad:
What you're seeing here is the humidity reported by the DHT22 & the last 3 readings. I'm using an average because I've seem some spurious outlying numbers reported (going from 90%+ to 43% on within seconds), and the correct way to handle this is to discard those number and use an estimate instead, but that code doesn't exist yet.
I also suspect that the root systems of the plants are simply not developed enough to absorb sufficient moisture from the mist since they developed in soil, and I should start with seedlings instead.
Build better logging so I can see what's going on, also when I'm away form the office (i.e. a small web thing)
Increase the humidity thresholds for misting or go to a timer-based algorithm and ignore humidity altogether
Today I received the timers that I'm using for the grow lights, so I've installed v1 on my test shelves and put some basil in! I've put the timer on so the grow light is on for 12hrs per day (no idea what it should be), and the aeroponic mister detects humidity below a threshold and will activate the fan & mister.
Currently I'm a bit blind as to what the best parameters should be for humidity and lighting, but discovering these things is the whole point of the experiment!