Although the impulse to solve a problem in a complex system is usually to use a microcontroller and some sensors to solve the problem automatically, the interaction with the real world is often much more difficult than it seems. For example, measuring soil moisture seems to be an easy way to ensure that plants get the right amount of moisture, but soil is a challenging environment for electronic products, and this solution usually leads to more problems than it solves. [Kevin] Pay attention to this problem with the soil moisture sensor and set out to solve it A simpler but indirect way to electronically monitor plants.
Instead of relying on soil conductivity to test the soil moisture content, he developed an alternative method to determine whether plants need watering only by continuously weighing. His hypothesis is that when water is available to breathe from plants or evaporate from the soil, the weight of plants that need water will decrease. This means that using a reliable sensor like a load cell to measure weight instead of an unreliable sensor like a soil moisture sensor will produce more reliable data, which he can use to automatically water the plants.
[Kevin]The construction is based on ESP32 and commercial load cells, both of which are built into the bottom of the plant pot. The design hides all electronic devices in a comfortable shell, and can also communicate related information wirelessly. However, the real story here is not the new use of the ESP32 chip, but the out-of-the-box problem solution that solves this problem by using atypical sensors. However, this is not to say that you can never use other sensors to directly monitor your garden and automate its health.