Where Roots Grow, People Glow.

What SmartFertilizer Is

SmartFertilizer is currently in its pilot version. The team is validating its engineering, performance, and field readiness through practical testing and measured results.

A farmer cannot see underground. Moisture, and eventually nutrient conditions, vary across a single field — but decisions about irrigation and fertilization are usually made for the whole field at once, based on surface appearance, schedule, or experience.

SmartFertilizer is a low-cost, self-powered node that takes that measurement where the crop actually is. It reads soil moisture, evaluates the reading against a configurable threshold, and runs entirely on energy it harvests itself.

We are not building a new soil sensor. We are building a different way of deploying soil sensing: cheap enough to install in numbers, independent enough to install without wiring, and designed around a decision rather than a dashboard.

What We Have Built

The current prototype is a working, tested unit:

• ESP32-based control unit — handles sensing, threshold evaluation and power state management
• Capacitive soil-moisture sensing — chosen over resistive probes for corrosion resistance and longer field life
• Solar energy harvesting — the primary power source
• Rechargeable energy storage — carries the node through night and low-light periods
• Power conversion and management — regulated supply to the sensing and control stages
• Threshold and decision logic — the reading is evaluated, not just recorded
• Modular architecture — additional sensors can be added without redesigning the core unit
• Field readings — taken from real soil samples, not simulated

How the Energy Path Works

sunlight → energy management → storage → sensing → measurement → sleep

The node does not need to run continuously. Soil moisture does not change on the timescale of seconds, so continuous operation would spend energy for no additional information.

Instead the system operates on a duty cycle:

wake → measure → evaluate → sleep

This is the difference between attaching a solar panel to a device and designing around an energy budget. The measurement interval, not the panel size, is what makes long-term autonomous operation possible.

Experimental: Kinetic Energy Harvesting

Alongside the solar system we are testing a secondary, experimental harvesting subsystem using a flexible piezoelectric element driven by wind flutter and rain impact.

This is a research direction, not a production feature. Solar is the primary and proven energy source for the current prototype. We are investigating whether kinetic harvesting can meaningfully supplement it during extended low-light periods.

Why Off-Grid Matters

A single sensor near a farmhouse can be plugged in. The value of soil sensing, though, comes from having many measurement points — ten, twenty, fifty across a farm — because that is the only way to see variation.

At that point, running power to every device stops being practical. Cabling cost, trenching, maintenance and failure points scale with the number of nodes.

Energy autonomy is therefore not a feature of SmartFertilizer. It is the precondition for the thing SmartFertilizer is trying to become.

The Experiment

The question we set out to answer was not "does the sensor produce a number."

Can SmartFertilizer distinguish meaningfully different soil conditions?

We prepared three soil samples — dry, intermediate, and recently irrigated — and took repeated readings from each.

The point is not that the sensor produces a number. The point is whether that number can change a decision.

Results pending — this section will be populated with measured readings once testing is complete. We do not populate it with estimated values.

What We Are Building Toward

Each stage depends on the results of the one before it.

1. 1 Measure — Soil moisture and environmental conditions at a single node.
2. 2 Understand — Extended sensing: soil electrical conductivity, pH, temperature, improved moisture accuracy — and eventually nutrient sensing, where technically validated. Combined with crop-specific thresholds.
3. 3 Map — Multiple nodes across one field, identifying unusually dry or wet regions.
4. 4 Assist — Indicating where irrigation or fertilization attention is most needed.
5. 5 Integrate — Possible connection to existing irrigation and fertigation infrastructure.

A Note on the Name

SmartFertilizer describes where this system is going, and the decision it is built to support.

To be precise about the current stage: the prototype does not dispense fertilizer, and does not measure nutrient content. It helps determine whether soil conditions are appropriate before irrigation and fertilization decisions are made. Nutrient sensing and any dosing capability belong to later phases, and only where we can validate them.

Intended Impact

• Less unnecessary irrigation → lower water demand
• Better-timed fertilization → reduced risk of unnecessary input use
• Locally harvested power → deployment without grid infrastructure
• More spatial information → intervention where it is actually needed

What We Still Need to Validate

• Water saved per growing cycle
• Energy autonomy across seasons and weather conditions
• Sensor reliability and drift over time
• Number of nodes required per unit area
• The relationship between our readings and real agronomic decisions

We publish this list deliberately. The difference between a hypothesis and a result is the whole substance of the project.

What Makes This Different

Not a new soil sensor. A different way of deploying soil intelligence.

Low-cost · off-grid · modular · field-oriented · built and tested locally · designed around decisions rather than dashboards.

We don't want farmers to receive more data. We want them to face fewer uncertain decisions.