enviro-sensors

Enviro custom scripts

Standalone MicroPython scripts for Pimoroni Enviro boards. Replaces the stock enviro library with a unified entrypoint (main.py) that loads board-specific sensor code based on config.model. Keeps the same Pimoroni UF2 dependency for sensor drivers.

The generated sensor data is stored locally and uploaded periodically. Once uploaded, the JSON can be processed by ./data-vis in order to render some simple HTML charts.

Supported boards

Board	Module	Sensors
Indoor	board_indoor.py	BME688 (temp, humidity, pressure, gas/AQI), BH1745 (light, colour temperature)
Weather	board_weather.py	BME280 (temp, humidity, pressure), LTR-559 (light), wind vane, anemometer, rain gauge
Urban	board_urban.py	BME280 (temp, humidity, pressure), PMS5003I (PM1, PM2.5, PM10), MEMS microphone (noise)

Set model in config.py to "indoor", "weather", or "urban".

Enviro firmware

Tested with version 0.2.0 of the Pimoroni firmware (likely to work fine with later versions).

Follow the guide here: https://github.com/pimoroni/enviro/blob/main/documentation/upgrading-firmware.md

Utilising the releases such as pimoroni-enviro-v1.22.2-micropython-enviro-v0.2.0.uf2.

Once the device has been flashed, connect to it via USB and remove all default files and folders, replacing them with all of the .py files from this repository. Modify the config.py as needed, then run main.py (main.py will also automatically be executed on boot).

How it works

Each script runs a loop:

1. Clock sync — fetches time from pool.ntp.org via NTP, writes to the PCF85063A RTC. Re-syncs every resync_frequency hours
2. Read sensors — takes a single reading from all onboard sensors. USB temperature compensation is applied when running on USB power
3. Save locally — appends a CSV row to readings/<date>.csv. Column headings stored once in readings/columns.txt
4. Cache for upload — writes a JSON file to uploads/ containing the reading, device nickname, model, UID, current power mode (usb/batt), and current free disk percentage
5. Upload — when cached file count reaches upload_frequency, connects to WiFi and POSTs each JSON to upload_url. Failed uploads are retried next cycle
6. Sleep — on battery: sets an RTC alarm and powers off (board re-powers on alarm). On USB: time.sleep() until the next reading

Power management

• VSYS_EN latch — the power rail is latched (Pin(2, Pin.OUT, value=True)) as the very first action on boot, before any imports or delays. On battery wake from an RTC alarm, a hold capacitor keeps the board alive only briefly — if the latch happens too late, the board powers off before the script can run
• Dynamic vbus_present — the USB power pin is read via a cached Pin object at the start of every loop cycle, so the device switches between USB and battery behaviour dynamically — including correct sleep mode and temperature compensation
• I2C bus reset — 16 SCL toggles run at startup to recover from stuck transactions after an unclean reset (e.g. USB disconnect mid-operation)

Disk space management

When free space drops below 10%, the script attempts to upload cached files. If uploads also fail, it thins the cache by removing every other JSON file — halving the count while preserving coverage across the full time range. Repeated thinning degrades resolution gracefully rather than losing a contiguous block of history.

Upload payload fields

Each cached/uploaded JSON payload includes:

• nickname, model, uid, timestamp
• readings (sensor values)
• power_mode (usb or batt)
• free_space (current filesystem free space percentage)

Worker and Storage

Create a Cloudflare Worker using the content of worker.js, bind the Worker to an R2 bucket (setting the variable name enviro_r2 to the name of the bucket). The upload_url value in config.py should be the Worker's public URL (optionally, add HTTP auth to the Worker, and add the credentials to config.py).

Configuration

All settings live in config.py:

Setting	Description
model	Board type: "indoor", "weather", or "urban"
wifi_ssid / wifi_password	WiFi credentials
wifi_country	ISO 3166-1 alpha-2 code for regulatory domain (e.g. GB, US)
nickname	Device name included in upload payloads
reading_frequency	Minutes between readings (aligned to clock grid)
upload_url	HTTP endpoint for JSON uploads
upload_frequency	Number of cached readings before triggering upload
http_username / http_password	Optional HTTP Basic Auth
resync_frequency	Hours between NTP re-syncs
usb_power_temperature_offset	°C subtracted from temp when on USB power
silent_mode	True to disable all LEDs (activity + warning)

Supporting files

File	Purpose
main.py	Unified entrypoint — shared logic (power, WiFi, NTP, CSV, upload, sleep)
helpers.py	Stateless utilities (datetime, file ops, disk space)
board_indoor.py	Indoor sensor init + read
board_weather.py	Weather sensor init + read (incl. wind/rain)
board_urban.py	Urban sensor init + read (incl. PMS5003I, mic)
config.py	Device configuration (model, WiFi, upload, schedule)
logging.py	Minimal file + stdout logger with 4 KB rotation
worker.js	Cloudflare Worker that receives JSON payloads and stores them in R2
battery.md	Battery life estimates

Requirements

• Pimoroni custom MicroPython UF2 (provides breakout_bme68x, breakout_bme280, breakout_bh1745, breakout_ltr559, pimoroni_i2c, pcf85063a)
• Pico W with Enviro Indoor, Weather, or Urban board
• Cloudflare Worker with an R2 binding