🌱 open-ube-root-chamber

Automated aeroponic sprouting chamber for ube (Dioscorea alata) mini-setts.

Built during Passport Turtles Live Jams — open coding sessions where engineers worldwide improve real hardware running in a Manila pilot lab. If you're here, you're either a member, curious, or you grow ube. All three are welcome.

---

The Problem We're Solving

Traditional ube propagation buries tuber cuttings in soil. They rot. Overwatering, soil pathogens, and inconsistent humidity wipe out 30–50% of a batch before anything sprouts. The labor cost of daily manual checking makes small-scale production uneconomical.

This system suspends ube mini-setts in a sterile mesh tray and mists their exposed roots with an oxygenated nutrient solution on a precise automated cycle — eliminating the rot vector entirely and reducing human handling to near-zero.

Metric	Soil Method	open-root-chamber
Sprouting success rate	~60%	~90%+
Labor per cycle	Daily watering	Near-zero
Contamination vector	Soil pathogens	Eliminated
Cost per unit	~$0.30	~$0.08 operational
Hardware cost	—	~$50–80 per unit

---

How It Works

An ESP32 microcontroller reads a DHT22 humidity and temperature sensor every 2 seconds. A deadband controller fires relay-driven ultrasonic misters and heat mats to hold the chamber at exactly 85–90% humidity and 28–30°C — the precise window ube setts need to break dormancy without rotting. A Raspberry Pi camera above the tray runs a computer vision pipeline that measures root length per vessel and flags contamination before it spreads.

---

Repository Structure

open-ube-root-chamber/
├── firmware-esp32/         # ESP32 environment control (C++ / Arduino)
│   ├── main.ino            # Main control loop (mister + heat mat)
│   └── pid_controller.h    # Reusable deadband controller
├── edge-vision/            # Root vigor detection (Python / OpenCV)
│   ├── root_vigor.py
│   └── requirements.txt
├── hardware-configs/       # BOM, wiring, nutrient solution recipe
│   └── chamber_specs.md
├── simulation/             # Terminal simulation of the full system
│   └── simulate.py
└── docs/                   # Wiring diagrams, calibration guides

---

Quick Start

Step 1 — Run the simulation first (no hardware needed)

Before touching any hardware, validate the system in your terminal:

pip install rich
python simulation/simulate.py

Watch a 30-day sprouting cycle play out live — sensors, relay states, vessel tray, and serial log all updating in real time. Use this to understand the system and tune setpoints before flashing.

Useful flags:

# Simulate a dry room (Manila dry season)
python simulation/simulate.py --ambient-humidity 65 --ambient-temp 23

# Simulate high contamination risk
python simulation/simulate.py --contamination-risk 0.002

# Fast batch run with no live UI (same as GitHub Actions)
python simulation/simulate.py --days 30 --speed 999 --batch

If the simulation shows the mister cycling too aggressively or the heat mat never turning off, adjust HUMIDITY_TARGET, TEMPERATURE_TARGET, or DEADBAND at the top of simulate.py — then mirror those values in firmware-esp32/pid_controller.h before flashing. The simulation and firmware must stay in sync.

---

Step 2 — Wire the hardware

Component	Spec	Approx Cost
ESP32 dev board	ESP32-WROOM-32	$5–8
DHT22 sensor	AM2302	$3–5
2-channel relay	5V, optocoupled	$2–4
Ultrasonic mister	5V, 113kHz disc	$4–6
Heat mat	10W seedling mat	$8–12

Pin map:

GPIO	Connected To
4	DHT22 data
16	Relay CH1 → mister
17	Relay CH2 → heat mat
2	Onboard LED (heartbeat)

Full wiring diagram and nutrient solution recipe in hardware-configs/chamber_specs.md.

---

Step 3 — Flash the ESP32

Install Arduino IDE from arduino.cc/en/software

Add ESP32 board support — Arduino IDE → Preferences → Additional Board Manager URLs:

https://raw.githubusercontent.com/espressif/arduino-esp32/gh-pages/package_esp32_index.json

Then: Tools → Board → Boards Manager → search esp32 → Install

Install libraries via Tools → Manage Libraries:

• DHT sensor library by Adafruit
• Adafruit Unified Sensor

Flash:

• Open firmware-esp32/main.ino
• Select board: ESP32 Dev Module
• Select your COM port
• Click Upload

---

Step 4 — Read the Serial Monitor

Open Tools → Serial Monitor, baud rate 115200. You will see:

[open-root-chamber] Firmware started.
[LOG] Humidity: 76.3% | Temperature: 25.1°C | Mister: ON | Heat Mat: ON
[MIST] Relay ON — humidity 76.3% (target 87.5%)
[MIST] Relay OFF — humidity 88.2%

---

Step 5 — Run edge vision on a Raspberry Pi

cd edge-vision
pip install -r requirements.txt
python root_vigor.py

Place a ruler in the camera frame first and set PIXELS_PER_CM in root_vigor.py to calibrate. Adjust ROOT_HSV_LOWER / ROOT_HSV_UPPER if root detection is poor under your lighting.

---

Calibration Notes

If your ambient room humidity is far from the 87.5% target (Manila dry season can drop to 55%), the mister may run continuously and oversaturate the chamber. Reduce DEADBAND to 1.0 in pid_controller.h for tighter control. Monitor via Serial for the first 24 hours after flashing to confirm the relay is cycling, not stuck ON or OFF.

---

Contributing

See CONTRIBUTING.md for the Live Jam schedule and how to claim a vessel assignment. Engineers can own a specific vessel in the Manila lab — watch your code optimizations reduce mortality on your actual batch via the live camera feed.

---

License

MIT — build it, fork it, grow ube with it.