I talk about creating a handy home IoT system for monitoring temperature and humidity. Using ESP32 and NodeMCU with sensors, it feeds data to a Raspberry Pi for real-time climate stats. Simple, smart, and super useful for keeping my space just right.

Ah, the buzzing world of IoT! Seems like wherever you turn, there’s chatter about it – especially the buzz around affordable boards like the ESP32 and NodeMCU, equipped with built-in WiFi and Bluetooth. These tech gems are seriously bringing IoT within everyone’s grasp. Stumbled upon an irresistible deal on the SHT20 temperature and humidity sensor recently – less than a buck a piece! Naturally, I snagged a handful for every nook and cranny of my pad. Sorry wallet, but these babies are essential.

Demo Demo

Living in a place where the humidity levels play tag, and with a room housing sensitive tech gear (think gaming PC, oscilloscope, signal generators, power supplies, and a 3D printer), the paranoia is real. Even with a dehumidifier working overtime, an engineer like me craves hard numbers. Thus, was born my ambient monitoring system using NodeMCU + SHT20 as Data Acquisition Units (DAQ). Threw in a potentiometer to conveniently assign Device IDs – same code, just a twist to differentiate.

Home Usage Home Usage

On the data end? Two first-gen Raspberry Pis got enlisted. One hosts the InfluxDB (yeah, that’s a Time-Series Database for the uninitiated) while the other powers the Grafana WebUI dashboard. The setup: each DAQ sends its readings over a localized WiFi network to the InfluxDB. Grafana dips in periodically to visualize these readings. Endgame? A laptop, a connection to the WiFi AP, and voila – instant updates on climate stats from the living room to the balcony.

Architecture Architecture

P.S.: Threw in a watchdog for system stability. In case the DAQ crashes due to WiFi congestion, it’ll reboot automatically. Data regarding these reboots is also charted, offering a sneak peek into the network’s health. Sweet, right?

Give it a try ?

Github

Prerequisites

Circuit

  • Raspberry Pi *1 (Run web server and database service)
  • NodeMCU *1
  • SHT20 *1
  • Resistor 330Ω *2
  • LED *1
  • Couple wires
    Circuit Circuit

Install

Install tools for NodeMCU from Pipfile:

  • pipenv install

How to start

Make sure InfluxDB and Grafana service is already running. and there is a database in InfluxDB. Edit /env_monitor/config.json. STA_SSID, STA_PASSWORD is for which wifi device NodeMCU connect to. AP_SSID, AP_PASSWORD is for debug remotely.

  1. Setting device id by turing potentiometer.

  2. Upload code to NodeMCU: ./run_all.sh.

  3. Connect to NodeMCU through USB for debug: ./connect_device.sh. If everything goes well, you will see some information like below.

    Console Console

  4. Open Grafana webpage and import /grafana/Env_Monitor.json. Or you can design your own dashboard.

  5. Cheers \m/

Troubleshooting

  1. If InfluxDB and Grafana set on different server

    • Set admin and password: CREATE USER admin WITH PASSWORD 'password'

    Edit InfluxDB configuration file

    • Linux: /etc/influxdb/influxdb.conf
    • macOS: /usr/local/etc/influxdb.conf
    [http]
  enabled = true
  bind-address = ":8086"
  auth-enabled = true
  log-enabled = true
  write-tracing = false
  pprof-enabled = true
  pprof-auth-enabled = true
  debug-pprof-enabled = false
  ping-auth-enabled = true
    • Update configuration: sudo /usr/bin/influxd -config /etc/influxdb/influxdb.conf

  2. If there is something goes wrong on uploading code.

    • Edit code_rest.sh use --flash_size=detect 0 command.