Weather and air quality station: Part 1

2025-11-01 • 505 words • 3 min read • #embedded #raspberry-pi #weather_station

Battery-Powered Weather & Air Quality Station

Raspberry Pi Pico 2 W + Inky Impression 7.3” + Bosch BME690 + Bosch BMV080 + Meteoblue API
Firmware written in C for the Pico’s RISC-V core

Project Overview

This project builds a standalone weather and air quality monitoring station. The Raspberry Pi Pico 2 W collects measurements from two Bosch sensors (BME690 and BMV080), pulls forecast data from the Meteoblue API over Wi-Fi, and displays results on a 7.3” e-paper display (Inky Impression). The device will be battery powered and enclosed in a plexiglass case to protect components while ensuring airflow.

System Architecture

Data flow:

Sensors (BME690 + BMV080)
  -> Data Processing + Rolling Averages
    -> Wi-Fi Forecast Fetch (Meteoblue JSON)
      -> UI Construction (framebuffer)
        -> E-Paper Display Refresh

Components:

Raspberry Pi Pico 2 W (RISC-V)
BME690: temperature, humidity, pressure, gas (IAQ)
BMV080: particulate matter (PM1, PM2.5, PM10)
Inky Impression 7.3” e-paper over SPI
HTTPS networking using lwIP + mbedTLS
Lightweight JSON parsing in C (ex: jsmn)

Design:

Event-driven firmware with deep sleep between tasks
Forecast caching for offline operation
Minimal UI optimized for e-paper refresh constraints
All core firmware written in C

Display and UI Strategy

Refresh e-paper only on a schedule or when values change
Off-screen framebuffer rendering
UI features:
- Temperature and humidity
- PM2.5 concentration
- Short forecast summary
- Battery voltage
- Timestamp for last update

Suggested refresh interval: 20 to 60 minutes

Power System

Battery: 3 AA cells in series

Electrical notes:

NiMH: approx. 3.6 V nominal, 4.2 V fresh
Safe to connect directly to Pico VSYS input (1.8 to 5.5 V range)

Series rules:

Voltage increases (3x)
Capacity stays 2000 mAh
Energy increases to approx. 7.2 Wh

Wiring:

3xAA holder (+) -> switch -> optional protection -> VSYS (Pico)
3xAA holder (-) -------------------------------> GND (Pico)

Battery level can be read via VSYS/3 ADC pin.

Recommended Duty Cycles

BME690: 2–5 min interval
BMV080: 10–60 min interval
Wi-Fi forecast: 30–60 min interval
E-paper refresh: 20–60 min interval

Estimated Runtime

Assumptions:

7.2 Wh battery energy
85% regulator efficiency
Approx. 6.12 Wh usable at 3.3 V
Average 3.3 V load depending on duty cycle

Mode	Avg Current	Runtime Hours	Runtime Days
Air Quality Focused (BMV080: 5 min)	2.50 mA	~742	~30.9
Balanced (BMV080: 10 min)	1.36 mA	~1364	~56.8
Battery Saver (BMV080: 60 min)	0.38 mA	~4880	~203

Major power contributors:

BMV080 sensor duty cycle
Wi-Fi bursts
Display refresh cycles

Enclosure Notes

Plexiglass case with airflow channels for sensors
Prevent heat transfer from MCU and display into sensors
Secure e-paper ribbon cable
Provide easy access for battery replacement

Development and Testing Plan

Bring up sensors and verify I2C
Implement sensor data processing and UI rendering
Add Wi-Fi and Meteoblue JSON support
Measure power usage and adjust intervals
Assemble enclosure and validate environmental behavior

Error messaging:

Wi-Fi unavailable (using cached forecast)
Sensor communication error
Low battery (auto-reduce update frequency)

Future Work

Final wiring/PCB design
Battery gauge improvements
Optional data logging
Potential partial refresh optimization

Critical Path