RADEX MR107 Linux Reader

Read radon (Bq/m3), temperature, and humidity from the Quarta-Rad RADEX MR107 radon monitor on Linux — no proprietary software needed.

Includes SQLite storage, web dashboard with historical charts, CSV export, and Homey Pro integration.

Why?

The RADEX MR107 only comes with Windows software (Radex Data Center). This tool reads the sensor directly via USB serial, making it possible to integrate radon monitoring into Linux-based home automation systems like Homey Pro, Home Assistant, or any MQTT-based setup.

Features

• Real-time readings — radon (Bq/m3), temperature (C), humidity (%), plus device averages/min/max
• SQLite database — persistent local storage of all readings
• Web dashboard — dark-themed dashboard with Chart.js graphs, auto-refresh, color-coded radon levels
• CSV export — download all historical data for analysis in Excel/Google Sheets
• Homey Pro integration — push readings to Homey Logic variables via local API
• MQTT support — pipe JSON output to any MQTT broker
• Cold start detection — identifies when the device is still warming up
• Systemd-ready — run as a background service with automatic restart

Requirements

• Python 3.6+
• pyserial (pip install pyserial)
• RADEX MR107 connected via USB
• Linux (tested on Debian 13 and Ubuntu 24.04)

Installation

git clone https://github.com/Frimurare/radex-mr107-reader.git
cd radex-mr107-reader
pip install pyserial

USB Permissions

The MR107 appears as /dev/ttyACM0. Add your user to the dialout group:

sudo usermod -aG dialout $USER
# Log out and back in for the change to take effect

Verify the device is connected:

lsusb | grep ABBA
# Should show: ID abba:a004 QUARTA-RAD MR107

Usage

Single Reading

python3 mr107_reader.py

Output:

[2026-04-04 16:35:00] Radon: 218.0 Bq/m3 | Temp: 21.0C | Humidity: 35.8% | Avg: 195.2 Max: 312.0 Min: 87.0

JSON Output

python3 mr107_reader.py --json

{
  "radon_bq": 217.9,
  "temperature_c": 20.9,
  "humidity_pct": 35.8,
  "radon_avg": 195.2,
  "radon_max": 312.0,
  "radon_min": 87.0,
  "countdown_s": 3421,
  "cold_start": false,
  "timestamp": "2026-04-04 16:35:00"
}

Continuous Monitoring with Database

python3 mr107_reader.py --loop 60 --db radon.db

This polls the MR107 every 60 seconds and stores each reading in a SQLite database.

All Options

Option	Description	Default
--port	Serial port	/dev/ttyACM0
--loop N	Poll every N seconds (0 = single read)	0
--json	Output as JSON	off
--db PATH	SQLite database path	none
--homey IP	Homey Pro IP address	none
--homey-token TOKEN	Homey API bearer token	none
--homey-device ID	Homey Virtual Device ID	none

Web Dashboard

A built-in web dashboard for viewing radon statistics in your browser.

python3 dashboard.py --db radon.db --port 8088

Then open http://YOUR_IP:8088 in a browser.

Dashboard Features

• Live sensor cards — radon, temperature, humidity, device average
• Color-coded radon levels — green (<100), yellow (100-200), red (>200 Bq/m3)
• Emoji indicators — 😊 safe, 😬 elevated, 🧟 high
• Historical charts — switchable between 6h, 24h, 3d, 7d, 30d
• Temperature & humidity — dual-axis trend chart
• Statistics — total readings, averages, min/max
• CSV export — download all data as a CSV file
• Database info — shows database size and total readings
• Auto-refresh — updates every 60 seconds
• WHO & Swedish guidelines — reference lines at 100 and 200 Bq/m3

Dashboard Options

Option	Description	Default
--db PATH	SQLite database path	radon.db
--port N	Web server port	8088
--bind ADDR	Bind address	0.0.0.0

Running as a System Service

Create two systemd services for 24/7 monitoring:

Reader Service

sudo tee /etc/systemd/system/radon-reader.service << 'EOF'
[Unit]
Description=RADEX MR107 Radon Reader
After=network.target

[Service]
Type=simple
User=YOUR_USER
ExecStart=/usr/bin/python3 /path/to/mr107_reader.py --loop 60 --db /path/to/data/radon.db
Restart=always
RestartSec=10

[Install]
WantedBy=multi-user.target
EOF

Dashboard Service

sudo tee /etc/systemd/system/radon-dashboard.service << 'EOF'
[Unit]
Description=RADEX MR107 Radon Dashboard
After=network.target radon-reader.service

[Service]
Type=simple
User=YOUR_USER
ExecStart=/usr/bin/python3 /path/to/dashboard.py --db /path/to/data/radon.db --port 8088
Restart=always
RestartSec=5

[Install]
WantedBy=multi-user.target
EOF

Enable and Start

sudo systemctl daemon-reload
sudo systemctl enable radon-reader radon-dashboard
sudo systemctl start radon-reader radon-dashboard

# Check status
sudo systemctl status radon-reader
sudo systemctl status radon-dashboard

Homey Pro Integration

Push radon readings to your Homey Pro (2023 or later) via the local REST API. No custom Homey app needed — it uses built-in Logic variables that you can use in Flows.

Step 1: Get Your Homey API Token

1. Go to https://tools.developer.homey.app/
2. Log in with your Athom account
3. Select your Homey
4. Copy the API token (Bearer token)

Step 2: Start the Reader with Homey

python3 mr107_reader.py --loop 60 --db radon.db \
    --homey 192.168.1.50 \
    --homey-token YOUR_API_TOKEN_HERE

Replace 192.168.1.50 with your Homey's IP address (find it in the Homey app under Settings > General).

Step 3: Verify in Homey

Open the Homey app and go to More > Logic. You should see three variables updating automatically:

Variable	Description
radon_bq	Current radon level in Bq/m3
radon_temperature	Temperature in C
radon_humidity	Humidity in %

Step 4: Create Flows for Alerts

Example: Get a push notification when radon exceeds 200 Bq/m3 (Swedish action level):

1. Open Homey app > Flows > Create new Flow
2. When... > Logic > "A numeric variable changed" > select radon_bq
3. And... > Logic > "Number is greater than" > 200
4. Then... > Notifications > "Send push notification" > "Warning: Radon is above 200 Bq/m3!"

Other useful Flow ideas:

• Alert when radon > 100 Bq/m3 (WHO guideline)
• Alert when radon > 300 Bq/m3 (critical level)
• Daily summary at 08:00 with current radon level
• Turn on ventilation (via smart plug) when radon is high

Systemd with Homey

Add the Homey flags to your systemd service:

ExecStart=/usr/bin/python3 /path/to/mr107_reader.py --loop 60 --db /path/to/data/radon.db --homey 192.168.1.50 --homey-token YOUR_API_TOKEN_HERE

Virtual Device Mode — Homey Dashboard Tiles (Recommended)

To get proper sensor tiles in Homey (with icons, graphs in Insights, and dashboard support), use a Virtual Device. This gives you a real "Radon Monitor" device in Homey with radon, temperature, and humidity readings.

How it works: Homey's Virtual Devices have measure_* capabilities that are read-only — they can't be set directly via the REST API. The reader works around this by creating triggerable Flows that use the "Set a virtual sensor value" action card, then updates the value and triggers the flow automatically each polling cycle.

Step 1: Install Virtual Devices

Open the Homey app > Apps > Search "Virtual Devices" (by Arjan Kranenburg) > Install.

Step 2: Create the Virtual Device

1. Homey app > Devices > + (add device)
2. Select Virtual Devices > Device
3. Name it (e.g., "Radon Monitor MR107")
4. Add these capabilities:
   • measure_radon (Radon) — if available
   • measure_temperature (Temperature)
   • measure_moisture (Humidity/Moisture)
5. Place it in the correct zone (e.g., "Basement")
6. Save

Step 3: Find the Device ID

You need the device ID to pass to the reader. Find it via the API:

curl -s "http://YOUR_HOMEY_IP/api/manager/devices/device/" \
    -H "Authorization: Bearer YOUR_TOKEN" | \
    python3 -c "import sys,json; data=json.load(sys.stdin);
[print(f'{did}: {d[\"name\"]}') for did,d in data.items() if 'virtual' in d.get('driverUri','')]"

Step 4: Start the Reader with Virtual Device

python3 mr107_reader.py --loop 60 --db radon.db \
    --homey 192.168.1.50 \
    --homey-token YOUR_TOKEN \
    --homey-device YOUR_DEVICE_ID_HERE

The reader will:

1. Update Logic variables (for use in Flows/conditions)
2. Auto-create triggerable Flows named "MR107 Set measure_radon" etc.
3. Update each flow's value and trigger it every polling cycle
4. The virtual device receives the new values and stores them in Homey Insights

Step 5: Add to Homey Dashboard

1. Open Homey app > Dashboards
2. Add your Radon Monitor device as a widget
3. You now have live radon/temp/humidity tiles in your Homey dashboard

Step 6: View History in Insights

Go to Homey app > Insights > select your Radon Monitor device. Homey stores historical data and shows graphs for each capability.

Systemd with Virtual Device

ExecStart=/usr/bin/python3 /path/to/mr107_reader.py --loop 60 --db /path/to/data/radon.db --homey 192.168.1.50 --homey-token YOUR_TOKEN --homey-device YOUR_DEVICE_ID

MQTT Integration

Pipe JSON output to any MQTT broker:

# Publish to MQTT every 5 minutes
python3 mr107_reader.py --json --loop 300 | while read line; do
    mosquitto_pub -h mqtt-broker -t "home/radon/mr107" -m "$line"
done

Home Assistant via MQTT

Add to your Home Assistant configuration.yaml:

mqtt:
  sensor:
    - name: "Radon Level"
      state_topic: "home/radon/mr107"
      value_template: "{{ value_json.radon_bq }}"
      unit_of_measurement: "Bq/m3"
    - name: "Radon Temperature"
      state_topic: "home/radon/mr107"
      value_template: "{{ value_json.temperature_c }}"
      unit_of_measurement: "C"
    - name: "Radon Humidity"
      state_topic: "home/radon/mr107"
      value_template: "{{ value_json.humidity_pct }}"
      unit_of_measurement: "%"

Dashboard API Endpoints

The dashboard exposes a simple REST API:

Endpoint	Description
GET /	Dashboard HTML page
GET /api/latest	Latest reading as JSON
GET /api/readings?hours=24	Readings for the last N hours
GET /api/stats	Overall statistics
GET /api/dbinfo	Database size and row count
GET /api/export.csv	Download all readings as CSV

Radon Guidelines

Level	Guideline
< 100 Bq/m3	WHO recommended maximum
< 200 Bq/m3	Swedish action level for existing buildings
< 200 Bq/m3	Swedish limit for new construction
> 200 Bq/m3	Action recommended — improve ventilation
> 400 Bq/m3	Urgent action needed

Protocol

The USB protocol was reverse-engineered by intercepting the proprietary Radex Data Center software's communication with the device using Frida dynamic instrumentation.

Device Info

• USB VID:PID: 0xABBA:0xA004
• Interface: CDC ACM (appears as /dev/ttyACM0)
• Baud rate: 9600, 8N1

Communication

A single 18-byte command polls all sensor data:

TX (18 bytes):

7B FF 20 00 06 00 83 04 00 00 DA FB 04 08 0C 00 EF F7
|------- header (12) --------|  |--- command (6) ---|

RX (70 bytes):

Offset	Type	Value
24-25	uint16 LE	Cold start flag (0 = warming up)
28-31	float LE	Current radon (Bq/m3)
32-35	float LE	Temperature (C)
36-39	float LE	Humidity (%)
52-55	float LE	Average radon
56-59	float LE	Max radon
60-63	float LE	Min radon
64-65	uint16 LE	Countdown timer (seconds)

All float values are IEEE 754 single-precision, little-endian.

Internal Memory

The MR107 stores up to 1000 data points internally (one reading per 4-hour cycle = ~166 days of history). The protocol for downloading stored history has not yet been reverse-engineered — contributions welcome!

Troubleshooting

Problem	Solution
Permission denied: /dev/ttyACM0	Add user to dialout group: sudo usermod -aG dialout $USER then log out/in
No such file: /dev/ttyACM0	Check USB connection: lsusb | grep ABBA
Failed to read MR107	Device may be in cold start (wait 1-2 minutes after power on)
Cold start - warming up	Normal after power on — the MR107 needs time to begin measuring
Dashboard shows no data	Start mr107_reader.py with --db first to populate the database

Disclaimer

This tool was created through clean-room reverse engineering of the USB protocol. It is not affiliated with or endorsed by Quarta-Rad. Use at your own risk.

License

MIT

Credits

Protocol reverse-engineered using Frida dynamic instrumentation. Built by Ulf Holmstrom / Manvarg AB, with Claude Code. Additional protocol reference: Maniak003/RADEX_MR107_Monitor