firebase post demo

In this tutorial, we are going to send data as strings to Firebase cloud platform using ARIES v3.0 board and ESP8266 NodeMCU. The data which we send as string from the program will get updated in firebase. This is a test program to know whether the firebase and ARIES v3.0 board has established the connection properly or not.

ESP8266 NodeMCU

NodeMCU is an open-source Lua-based firmware and development board specially targeted for IoT-based Applications. It includes firmware that runs on the ESP8266 Wi-Fi SoC from Espressif Systems, and hardware that is based on the ESP-12 module.

ESP8266 is a highly integrated chip designed for the needs of a new connected world. It offers a complete and self-contained Wi-Fi networking solution allowing it to either host the application or to offload all Wi-Fi networking functions from another application processor. Its high degree of on-chip integration allows for minimal external circuitry. It is designed to occupy a minimal PCB area.

Prerequisites

Components Required

  • ARIES v3.0 Board
  • USB type C to USB type A cable
  • Micro USB type B to USB type A cable
  • NodeMCU module
  • Jumper Wires

Connection Diagram

NodeMCU ESP8266 WiFi ModuleARIES v3 Board
VCC+3V3
GNDGND
D7 (RXD)TX1
D8 (TXD)RX1
Connection pins for NodeMCU with ARIES v3.0

Libraries Required

Arduino libraries extend the standard Arduino API and are created by the community. They simplify complex code. Each Arduino IDE version includes a library manager for easy installation. The libraries needed for this project are:

  • ArduinoJson (Version (5.13.2)
  • VEGA_FirebaseArduino (Version 1.0.1)

To install the libraries directly from library manager required for this project.

Go to Tools -> Manage Libraries -> and search with the library names mentioned above

Click on Install

Procedure

Refer the connection Diagram and connect NodeMCU module to ARIES v3.

Next we need to configure Firebase cloud by following the below steps:

Steps to do in Firebase cloud platform:

  • click on “Go to Console
  • Click on “Add Project” and give any project name for your project ( Example project name – randomData) and click on Continue
  • Click on Continue -> Continue
  • Select default account for Firebase and Click on “Create Project” and wait for project to be created

Click on “Continue” and you can see your project created

  • Next we need to create the database for our project.
  • Click on Build -> Real time Database -> Create Database
  • Click on Next -> Start in Test Mode
  • Note: If you selected start in Locked mode then you need to change the rules of database to false for working, after creating the data base.
  • You can see the url required for this project on the top of your database. Copy this url for future use
  • Go to Project Overview -> Project Settings -> Service accounts -> Database Secrets, to copy the Database key
  • Click on show and copy the database secret key
  • Turn on your mobile hotspot before uploading the code
  • Now, Open Files -> Examples -> Under Examples for ARIES v3 -> UART-> Firebase -> SendString_NodeMCUtoFirebase
  • Change the below settings with your corresponding credentials in the SendString_NodeMCUtoFirebase file
  • For the example provided, the settings before and after change is shown in the below figure

Note 1:

Don’t forget to remove the “https://” and the last “/” while pasting the url, to get correct output

Note 2:

The Firebase Arduino library has a reference to a fingerprint of the Firebase SSL certificate. Sometimes this fingerprint may not match the current fingerprint.This fingerprint is in  FirebaseHttpClient.h (typically in Home/Arduino/libraries/firebase-arduino/FirebaseHttpClient.h in Ubuntu).

To find and change the current fingerprint:

  1. Go to https://www.grc.com/fingerprints.htm
  2. Enter “test.firebaseio.com”
  3. Record the fingerprint (e.g. it is currently 6A:7D:A1:11:7F:F3:C2:45:0C:F7:22:E7:A2:35:EA:61:12:55:83:48
  4. Open Home/Arduino/libraries/firebase-arduino/FirebaseHttpClient.h
  5. Replace value of kFirebaseFingerprint [] with the fingerprint (without colons)
  6. Recompile

e.g. the following fingerprint works at this point of time:

static const char kFirebaseFingerprint[] =
  "6A:7D:A1:11:7F:F3:C2:45:0C:F7:22:E7:A2:35:EA:61:12:55:83:48";   //2023-09

For more details refer the following link.
  • After changing above settings, connect NodeMCU board to your laptop using micro USB type B to USB type A cable.
  • Go to Tools -> Board -> ESP8266 Boards(2.7.4) -> NodeMCU 0.9 (ESP-12 Module)
  • Go to Tools -> Port -> Select appropriate port
  • Now Upload the code to your NodeMCUESP8266 module
  • Open the Serial Monitor in 9600 baud rate, Once the code is uploaded you can see the “Connected to VEGA” message on serial monitor along with status as shown below.
  • You can also check the firebase real time database created. if the connection is established properly you can see the below default message printing on the database.
  • Now you can disconnect the USB cable from NodeMCU module.
  • After disconnecting USB cable from NodeMCU module, follow the below steps to upload code to ARIES v3 board.
  • Go to Tools -> Board -> VEGA Processor ARIES Boards -> Select ARIES v3
  • Open Files -> Examples -> Under Examples for ARIES v3 -> UART-> Firebase -> SendString_AriestoNodeMCU

Connect Aries board to your computer via the USB port of a Laptop/Desktop/PC. We have to use a USB type C to USB type A cable. The cable should be connected to UART-0 port of the ARIES v3 board, and the Laptop/Desktop/PC should be preinstalled with Arduino IDE and VEGA ARIES boards of latest version.

  • After connecting ARIES v3 board to your computer, Make sure you have selected ARIES v3 Board from Tools -> Board -> VEGA Processor: ARIES Boards -> ARIES v3
  • Select Programmer as VEGA XMODEM from Tools -> Programmer -> VEGA XMODEM
  • Also select appropriate port from Tools -> Port -> COM* (ARIES Board)
  • Now upload the code to your ARIES v3 board by clicking on Upload button

Output

  • Once the code is uploded to ARIES v3 board, open the Serial Monitor at 115200 baud rate, and you can see the “Start”,”Next” and “Done” comments while each string is getting updated in Firebase. Open firebase to view the strings sent.

Demonstration

By mydhily

Leave a Reply

Your email address will not be published. Required fields are marked *