Fully automated remote control and monitoring of agricultural growing systems
Following up on Evan’s earlier work on this project, I have started designing v2 of our system. The primary focus of my designs is to make the system more self-contained and lower-maintenance. To that end, the design is centered around the ubiquitous 5-gallon plastic bucket. The large reservoir means fewer water changes and a nice, stable base (my system will be outside on a balcony, so wind is a concern).
I also wanted to refine the pump and bottle attachments so that they were sturdier and could be assembled quickly with basic tools. The key parts developed so far are:
Water Distribution Manifold
Three bottles fit on the bucket efficiently and I needed a way to distribute the water. This manifold connects the pump at the bottom of the bucket with three outlets on top of the lid. All of the plumbing will use the same 1”-8 thread and nut. The coarse thread is a good fit for 3D printing and is interchangeable with metal parts. If we were to manufacture kits, standard bolts and nuts could be used to test fit and clean up the printed parts. The manifold can be downloaded here: http://www.thingiverse.com/thing:869643
This fitting attaches a 2L bottle to a flat plate. The hole in the bottom is sized for a threaded fitting, though I won’t be using any in this project (the water will just flow out the hole into the bucket). The bottle holder can be downloaded here: http://www.thingiverse.com/thing:878942
The same ring that Evan used in the previous update will be used again in this version. The three rings will need to be held apart by a spacer which I still need to design. It can be downloaded here: http://www.thingiverse.com/thing:837700
I used this basket (http://www.thingiverse.com/thing:738791), as well. It really doesn’t print very well on an FDM machine like our RepRap Mendelmax, but it works and will be covered in expanded clay, anyway, so the imperfections won’t be visible.
The system will look pretty much like this when it is done. Evan's plants died since his last post, and we are hoping that the larger amount of water (5gal vs 1gal) in this new system will make the nutrient levels easier to manage.
The first step to this project is to get a simple hydroponics system up and running. This will let us get a feel for hydroponic growing and and what parts we might want to add sensors and control to.
After looking at a few systems online, I really liked the idea of using half a two liter bottle as the growing pot. I figured it would be fairly easy to attach the bottle to the reservoir lid using a threaded ring. I found this two-liter bottle cap start on Thingiverse that did the trick. With it, I was able to attach the bottles to the lids of a 6 quart rectangular container. Hydroton clay pebbles are being used for the growing medium. To keep the hydroton from rolling out, I printed this basket which fits perfectly at the bottom of the two-liter bottle.
The next challenge was finding a good way to fix the 5/16 ID hose to the top of the bottle. I originally tried to punch a couple holes and zip tie it, however the plastic was too flimsy. So I designed a rim that would fit on top of the bottles and has holes to tie the hose in position, pointed down at the plant.
Originally, I chose this self priming diaphragm pump. It was cheap and used a DC motor which would make it easy to control the flow. However, it was much too loud. So, I found another 12V submersible DC pump that could be placed right in the reservoir. This made the system super quiet and had a bit better flow. However this pump does not take well to variable voltage adjustment to control flow. For on/off control, the pump is hooked into a MOSFET switch and controlled by a Teleimperium microcontroller board. For now, I have a simple program that will run the pump for 15 minutes every hour. There is also a simple air pump to oxygenate the water, and a one wire temperature sensor.
I still need to program the controller to read the temperature and upload it to a server. Also planned is using a solid state relay to control the lighting. To monitor water quality, I am using a cheap EC meter along with a pH kit. For this crop of strawberries, the target pH is 5.5 to 6.5 and the EC should not exceed 1000ppm. Fox Farm liquid concentrate is the fertilizer being used, at a concentration of 2Tbs per gallon.
The next big challenge will be to develop a continuous EC meter that can be hooked into the Teleimperium so that data can be collected and graphed with Teleceptor. Automated pH measurement would be nice, but there may be issues with leaving a probe in the water at all times. We will have to research options.
Hydroponics is the act of growing plants without the use of soil. When growing plants in soil, water serves to make nutrients in the soil soluble and provides oxygenation for the plant's roots. In Hydroponics, however, the nutrients are instead made available directly in the water. In the most abstract sense, a hydroponics system requires a nutrient rich water supply, a means of oxygenating the water supply, and a means to transport the water to the plants. Since nutrients can be expensive, typical hydroponics systems are recirculating – the runoff from transporting water to the plants returns to the original water source. In the following, we discuss the main parameters that should be considered in a hydroponics system. This description is not intended to be complete, but rather gives an overview of the parameters that are most likely to affect a hydroponics system.
Water chemistry refers to the makeup of a water source. In hydroponics, we chiefly care about three factors: pH, concentration of Macro-nutrients, and oxygenation. pH is important due to nutrient lockout. The various nutrients within a water source are more easily absorbed by a plant's roots based on the pH of the water source – too basic or too acidic, and the plant will not be able to absorb the nutrients from the water, even if they are present. The desired pH range will vary depending on the plants in the hydroponics system, but typical values are between 5.5 and 6.5.
Macro-nutrients refer to Nitrogen, Phosphorus, and Potassium (fertilizers typically list these as relative concentrations: 10-10-10 means the nutrients are balanced) – plants use these three in greater concentrations than other nutrients like Iron and Magnesium, which are called Micro-nutrients. The particular concentrations your system will need depends heavily on the plants you plan on growing. Leafy vegetables like lettuce and cucumbers need a higher concentration of Nitrogen, whereas others (like tomatoes) require a more balanced mixture. In any case, hydroponics systems use special nutrient mixes designed for hydroponics – regular fertilizers are not water soluble enough for use in hydroponics, and will cause buildups that can be difficult to get rid of. As the plants use up the nutrients in the water, more fertilizer will need to be added. The common way of detecting when it is time to add more is with an Electrical Conductivity (EC) Meter. These meters measure how conductive a solution is. This is important for hydroponics because we typically use distilled water (or de-ionized water) and add nutrient solution. Since the nutrient solution is the only source of ions, EC is a reasonable measure of nutrient concentration. EC is not a direct measure of nutrient concentration, however, and if other sources of ions make their way into the water source, then EC will not be a very accurate measure of nutrient concentration.
Finally, oxygenation is a critical factor in the water source. If the water has low dissolved oxygen, the roots of a plant can 'drown'. In container gardening, this manifests as over watering, and the symptoms are very similar to under watering. Since hydroponics systems lack soil to retain water, the major concern is under oxygenated water. Luckily, we don't need to worry about over oxygenating the water. For the majority of setups, a small air pump with an air stone is sufficient – check the manufacturer's suggested gallon rating when choosing an air pump. Oxygen can also be introduced when the runoff returns to the water source if there is a waterfall (or the water free falls through air).
Watering cycles do not seem to be well researched, but the common wisdom states a 15-on-45-off cycle. This can be done all day, or the cycle can be reduced at night. However, certain hydroponics setups require constant water flow – Nutrient Film Technique (NFT) is an example of constant water flow. For more traditional systems, the cycling allows the water to drain completely from the roots...Read more »
We have been working with time series databases and graphing for many years. This has been primary for network equipment using software such as cricket and cacti as well a writing our own software known as grasshopper and portcullis. One of the main problems with all of these programs is the configuration overhead of not being able to quickly and new devices to the system. Another issues is if data storage should use a averaging round-robin database like rrdtool and whisper or save every datapoint to a SQL database. We started working on new graphing software known as teleceptor which has a simple json based configuration as well as the ability to use both whisper and sqllite for datapoints. Data can then be analysed by downloading the full data stored in the SQL database or quickly generate long terms graphs from the whisper database.
As well as working on software we have also been working on hardware, our goal is to make an easy to deploy node that is capable of being connect to many different types of sensors and actuators as well a being able to be remotely located using wireless and/or RS485. Our latest attempt is the Teleimperium.
We are big believers in open source thus all our software and hardware design are available on Github. Our goal for this summer is to take the technology we have been developing as well as create some new technology in order to create a fully automated remote growing system. As NMSU Alumni we understand the importance of agriculture and hope to further improve growing techniques by integrating modern real time data analysis as sophisticated control systems.