ESP32 Accelerometer Monitor

This project implements a high-frequency vibration monitoring system using an Adafruit ESP32-S3 Feather and an Adafruit LIS331HH triaxial accelerometer. It features secure MQTT communication over TLS, CBOR data serialization, a web-based configuration portal, and power-efficient operation via deep sleep.

Hardware Requirements

• Microcontroller: Adafruit ESP32-S3 Feather (with 4MB Flash / 2MB PSRAM).
• Sensor: Adafruit LIS331HH (High-G Accelerometer) or any compatible I2C sensor.
• Status Indicator: On-board NeoPixel (GPIO 17 for Adafruit ESP32-S3).

Core Functionality (main.cpp)

The main.cpp script follows a structured execution flow designed for reliability and ease of use:

1. Configuration: Loads parameters from data/config.json. If configuration fails or required fields are missing, it launches the Web Configuration Portal.
2. Web Configuration Portal (AP Mode):
   • Creates a WiFi Access Point named DIMITRI-[ClientID].
   • Features a Captive Portal (automatically opens the config page).
   • Modern, themed (Dark/Light/High-Contrast) web interface for updating all parameters (WiFi, MQTT, Sensor, NTP, Sleep, etc.) without reflashing.
3. Connectivity:
   • Connects to WiFi and syncs time via NTP (essential for TLS).
   • Establishes a secure MQTT connection over TLS using data/ca.pem.
4. Data Acquisition:
   • Captures a burst of high-frequency samples (e.g., 1000Hz) using the LIS331HH.
   • Uses esp_timer_get_time() for microsecond-precise sampling intervals.
5. Dynamic Threshold Gating:
   • Calculates the RMS magnitude of the vibration burst.
   • Compares it against mag_rms_threshold (configurable via web/JSON).
   • Skips transmission if vibration is too low, saving energy and MQQT bandwidth.
6. CBOR Serialization & Transmission:
   • Packs raw data and metadata into CBOR format.
   • Publishes data in blobs (X, Y, Z, and timing) to the configured MQTT topic.
7. Deep Sleep: Powers down the device for a set interval (sleep.seconds) to preserve battery.

Status LED (NeoPixel)

• Solid Green: Normal operation (Init/Acquisition).
• Blinking Green: Successful transmission, entering sleep.
• Blinking Yellow: Data below threshold, skipping transmission.
• Blinking Purple: Web Configuration Portal Active.
• Blinking Red: Error state (1: WiFi, 2: NTP, 3: MQTT, 4: Sensor, 5: Config).

Partition Table

The project uses a custom partition scheme partitions_adafruit_no_ota.csv to maximize space:

• nvs: Non-volatile storage for WiFi data (20KB).
• app0: Main application firmware (approx. 1.4MB).
• spiffs/littlefs: Filesystem for config.json and ca.pem (approx. 1.3MB).

Web Configuration Portal

If the device cannot connect or find config.json, it enters "Provisioning Mode":

1. Connect your phone/laptop to the WiFi DIMITRI-esp32... (Password: dimitri1234).
2. A configuration page should pop up automatically. If not, go to 192.168.4.1.
3. Fill in the fields and click Save & Restart.

Configuration (config.json)

Alternatively, you can manually upload [config_example.json] renamed to config.json in the data/ folder:

{
  "wifi": { "ssid": "...", "password": "..." },
  "mqtt": { "host": "...", "username": "...", "password": "...", "topic": "..." },
  "sensor": { "i2c_addr": 24, "range_g": 24 },
  "acq": { "n_samples": 500, "fs_hz": 1000, "mag_rms_threshold": 10.78 },
  "sleep": { "seconds": 300 }
}

How to Upload

This project uses PlatformIO.

1. Preparation: Ensure data/config.json and data/ca.pem are ready.
2. Upload Firmware: pio run -t upload.
3. Upload Filesystem: pio run -t uploadfs (Crucial for the web portal and config).