Eco Grow

ESP32-powered smart compost box: 5 sensors monitor decomposition in real time, auto-control moisture & mixing, and stream live data to a dashboard — turning food waste into fertilizer.

Comment

Every day, restaurants and households throw away massive amounts of food waste with no system to recover its value. We learned that one-third of all food produced globally is wasted, and most of it ends up in landfills where it releases methane — a greenhouse gas 80x more potent than CO₂. Traditional composting exists, but it's completely blind: no one knows if the temperature is right, if the moisture is too low, or if the decomposition is actually working.

We asked ourselves: what if the compost bin could think?

That question became Compost IQ.

** What it does

Compost IQ is a smart IoT composting system that monitors and controls organic food waste decomposition in real time.

A Smart Compost Box equipped with 5 sensors continuously tracks:

  • 🌡️ Temperature (DHT22) — optimal range: 40–65°C
  • 💧 Moisture Level — keeps the compost hydrated automatically
  • 🧪 Gas / Odor (MQ-135) — detects abnormal decomposition gases
  • ⚗️ pH Acidity — ensures the right chemical balance for bacteria
  • ⚖️ Biomass Weight (Load Cell + HX711) — tracks decomposition progress

Two smart actuators respond automatically:

  • A Water Pump activates when moisture drops below 40%
  • A Mixing Motor stirs the compost every 6 hours

All data streams live to a professional web dashboard built in React, showing real-time charts, alerts, a Decomposition Score (0–100%), and full actuator control with Auto/Manual modes.

The system supports Dark & Light mode and Arabic/English language switching — designed to be accessible globally.

** How we built it

Hardware Layer: We connected 5 sensors to an ESP32 microcontroller using GPIO pins (Digital for DHT22, Analog for Moisture/Gas/pH, DT+SCK for HX711). Two relay modules handle the pump and motor. The ESP32 sends a JSON payload to our backend every 60 seconds over Wi-Fi.

Backend Layer: A Node.js + Express server (Port 3001) receives sensor data via POST /api/sensors, validates it, checks thresholds, saves to SQLite for historical storage, syncs to Firebase Realtime Database for cloud persistence, and pushes live updates to the frontend via Socket.IO — all in under 100ms per cycle.

Frontend Layer: Built in React + Vite + Tailwind CSS with Zustand for state management. Recharts powers the analytics charts. A custom simulation engine runs realistic sensor data in dev mode, so the full dashboard works without any physical hardware connected.

Intelligence Layer: 8 threshold rules fire warnings and danger alerts across all 5 sensors. Auto-actuator logic runs inside the alert service. A composite Decomposition Score formula tracks bio-progress:

What's next for Eco Grow

Built With

Share this project:

Updates