My grandfather was a farmer all his life. He farmed in the mountain lands of Peru where not much rain comes down in the winter. The farmers here must share the water from a great pond a couple of miles away from the plantation area. To assure that there is enough water for the hundred plots of farmland, there is a system followed where the water travels through concrete canals made with trap doors that direct the water flow to access certain areas depending on which doors are open allowing the flow of water. This system works so a plot of land receives water 1-2 times a week depending on the size of the land. Typically you have your "time of the week" to take advantage of this water. For example, my grandfather would have 6 am-6:30 am on Tuesday and 11:30 am-12 pm on Fridays to have water reach his land. Although this is not a lot of water, when managed correctly the plants receive enough water to grow properly. Experienced farmers know how to navigate around the water provided but there is still enough water to drown your plants through these watering cycles for a rookie farmer.
Having the equipment to measure the moisture level of the soil used would be essential to those who don't properly know how watered their plants are. In situations like the one in Peru, it is ideal to take advantage of all the water you receive during your cycle but knowing the exact amount of water needed could be essential to maximize the growth of your crops. Overwatering could spoil or drown the crops. Having a system that measures the moisture levels of the soil knowing exactly how watered your crops are would help farmers better understand their plantations and the necessity of water. It could help a farmer decide to increase the amount of water used during their cycle or potentially skip a cycle session to not overwater the crops. Having data of the soil you are depending on your crops to producewould be revolutionary to the agricultural industry. Essentially, they would be taking control of the moisture levels their crops are receiving and knowing when it would be ideal to water again compared to relying on a weekly cycle schedule like Peru. Local farmers would benefit from this innovative technique of water control by eliminating the risk of drowning or spoiling their plants due to lack of or extensive hydration to the plants.Project Overview
Our mission is to reduce the amount of worry that people tend to endure while wondering when to water their plants. After much research, we figured out that the best sensor to measure the moisture levels within the soil will last around 6 months. The sensor that best fit our needs for this project was a capacitive moisture sensor. This project will help people water their plants at the appropriate time or if they simply forget to water the plant after a busy day. When the particle that processes data determines that the plant is in need of water, a notification will appear on your phone telling/reminding you that your plant needs to be watered. More details about how our Plant Moisture Project works can be found below.Working Towards a Solution
Choosing the appropriate Sensor
During our research, we looked at the available sensors that measure the moisture levels in the soil. The two sensors that we narrowed our options down to were the capacitive and resistive moisture sensors. We chose to go with the capacitive moisture sensors since they are less likely to experience corrosion than the resistive sensors. The capacitive sensors can also regulate voltage, and operate from ~3.0-5.0V, which is within the range that our Particle Argons can operate.
Calibration and Data Collection
For our sensor calibration, we calibrated the sensor based on the limit of the dryness that the soil can reach before water is required for the plant. Normally, the sensor gives a reading of around 2400mV, so we chose to set the limit at 1750mV for the plant to be watered. The voltage reading on the sensor goes down as the soil gets dryer, and once it reaches the threshold limit, a signal will be sent out telling the user that the plant needs water. Data collection for this project works by taking in a bit count of 3798. The measurement that is received from the particle is divided by the bit count to get one ratio. Then this ratio is multiplied by the supply voltage of 3.0 V to get a sensor reading that is measured in millivolts. This data from the particle can be further processed and used to get a reading and determine if the plant needs to be watered. It should be pointed out that if you try to build and test this project on your own, you will more than likely have different sensor values than the ones in this project, due to the different sensors that can measure soil moisture and the specific type of soil that you use. This difference in the values may require you to either water the plant more or less. In case you want to let your plant dry out before you water it so it doesn't attract pests or grow fungi on it, you can change the indicator level.
Communication between the Particle Argons
The first Particle Argon (Argon 1) initiates data collection and data processing through sending/publishing an event to the particle cloud, which can be seen either on a laptop or on a phone that has the Particle app installed. The message instructions is to turn the other two Particle's LED lights on. Once the other Particle's received this message, the LED lights should turn on. The second Particle Argon (Argon 2) has the capacitive soil sensor, which we picked out, attached to it, and it begins to collect readings from the soil. This Particle Argon will publish the data that it receives to the particle cloud, which will be seen by all three Particle Argons. The third and final Particle Argon (Argon 3) will wait for the data from Argon 2 to be received, and once it has received the data, Argon 3 will begin summing up the data. Once the third Particle Argon receives enough data, it will find the average soil moisture based on the data that it collected. Depending on how much moisture is still left in the soil, Argon 3 will publish several messages through the particle cloud. The first message will show whether the plant is in need of more water, while the second message will toggle the LED lights on and off to tell Argon 2 to stop collecting data. Through the use of graphing software, the data that is collected from Argon 2 will be put into an Excel spreadsheet, which will provide a "live" graph displaying the voltage reading (in mV) versus the time that it was collecting data (in minutes). Through that same software, when Argon 3 publishes the event which shows whether or not to water the plant, a text message will be sent to a phone, which is connected to all 3 Particle Argons, indicating whether the plant needs to be watered or not.