Hi vertical farmers,
We will use our third post to justify our automatic nutrient dispenser implementation and assumptions.
On hydroponic systems the plants are fed uniquely via the nutritive solution. This cultivation mode is much less tolerant to faults than a cultivation on soil that works as a buffer where the plant can find nutrients not supplied via the irrigation.
For optimum growth the nutritive solution needs to be dynamically adapted to the needs of the growing culture and its vegetative stage (i.e., flowering, fruit,...).
Every plant requires the known NPK macro-nutrients (nitrogen, phosphorous and potassium) and several micro-nutrients like calcium, magnesium, iron, sulfur, boron just to name a few. Adding to the complexity, some of the nutrients required by the plants can not be mixed in concentrated solutions as they will react, causing some of the chemicals to precipitate and become inactive (typically nitrates and sulfates). This can be avoided mixing one after another with a large amount of water.
Most professional growers buy dry fertilizers and make their own nutritive solutions. This requires a higher levels of understanding on the chemistry of ingredients and the requirements of the culture. Also those fertilizers are not easy to find in small quantities.
Most suppliers of hydroponic systems offer stock solutions of 2 or 3 components with a balanced mix of macro and micro-nutrients advising on the correct mixing proportions for leafy vegetables, small fruits or larger fruits on flowering or fructification stage.
We will use one of those commercial solutions with 2 compatible components A and B that claim to be a good starting point to grow a great variety of plants with good success rates.
This approach has advantages because is a ready solution to start growing adapted to hydroponics and based on the built experience of a third party and disadvantages as it is not optimized for a specific culture and are usually rather expensive.
We mentioned already two nutritive concentrated solutions plus water. Additionally we will be using a fourth element for the pH correction. This is an important parameter as it determines the absorption rate of the nutrients by the plants. If not fine tuned the plants may not be able to grow even if all the needed nutrients are available in the solution. Extreme values of pH can lead to irreversible damage on the plants.
As the solution pH will change based on the nutrients concentration on the solution this must be monitored periodically (depending on the consumption nutrient profile of the plant) and every time a nutrient is added to the solution.
Also the Electrical Conductivity (EC) is an important parameter to measure as it gives a good indication of the strength of the solution. The higher concentration of nutrients the higher the EC value will be.
Typical reference values for pH and EC of lettuce and strawberries can be found on the next table as an example.
| Plant | pH | EC [mS/cm] |
|---|---|---|
| Lettuce | 5.5-6.5 | 0.8-1.2 |
| Strawberries | 5.5-6.5 | 1.8-2.2 |
As it can be seen from the table the strawberry requires a stronger nutritive solution than the lettuce.
As booth pH and EC have a strong dependence with the temperature it´s important to measure it to perform a measurement compensation.
We will be using the portable meter HI9813-5 and combined measurement probe (HI1285-5) from Hanna Instruments for precise temperature, pH and EC measurements as a reference ground through. This will require periodic probe calibrations with reference solutions.
We will integrate and calibrate a cheaper pH sensor probe (see picture bellow) and a NTC resistor to monitor the pH of the nutritive solution and check its performance as an alternative affordable solution.
An overall schema of the automatic nutrient dispenser implementation can be found on the next figure.
Our automatic nutrient dispenser will use the two nutrient solutions(A+B), the water and a pH down(acid) solution to keep the solution on the main reservoir on reference levels of pH and EC.
The nutritive solution and oxygenation will be achieved via air pump and porous stone that will generate the turbulence required.
Level sensors on each reservoir will trigger alarms when critical low levels are reached and a refill is required.
One nutrient doser prototype was assembled to test the behavior and implementation feasibility of the proposed system as shown on the next pictures.
As the electrovalves are on the way we simulated its behavior with manual valves. The buffer recipient is a transparent tube of 10 mm inner diameter that allows a rather decent resolution of about one milliliter per linear centimeter.
We will implement a level sensor with five different elements spaced about one centimeter each to be able to sense and fill different volumes of liquids in the tube. We are still struggling with its implementation as either magnetic or optic sensor options will have their advantages and disadvantages. For instance on our first trial with a magnetic sensor the float element could not sustain the weight of the magnet. Also our attempt with optical sensor (IR Led + photo-diode) revealed some susceptibility to the ambient light.
As an example, the planned implementation for a magnetic level sensor is presented on the next image.
On the next post we will finish the details on the automatic nutrient dispenser and hopefully provide a working demo.
As always, feel free to put any questions or share your thoughts and suggestions on our design and methodologies.
Keep in touch!
