I already described how to create a PlatformIO project and install libraries in the first part. My project is called “MyNodeMCU”. The structure is shown below:

This project is a slightly modified Cloud4RPi example.

I decided to store the device token and Wi-Fi credentials in the configuration file instead of code.

The platform.ini file looks as follows:

[platformio]
default_envs = nodemcuv2

[env:nodemcuv2]
platform = espressif8266
framework = arduino
board = nodemcuv2

Libraries installation is quite simple. You can do it from the IDE’s graphical interface, or by adding required library names to the lib_deps section of the platform.ini file:

; ...
lib_deps =
    cloud4rpi-esp-arduino
    Adafruit Unified Sensor
    DHT sensor library
build_flags =
    -D MQTT_MAX_PACKET_SIZE=1024
    -D MQTT_MAX_TRANSFER_SIZE=128
    -D CLOUD4RPI_DEBUG=0
    -D SSID_NAME=\"__YOUR_WIFI__\"
    -D SSID_PASSWORD=\"__YOUR_WIFI_PASS__\"
    -D CLOUD4RPI_TOKEN=\"__YOUR_DEVICE_TOKEN__\"

 Added libraries will be automatically installed into a project’s subfolder.

The main.cpp header looks as follows:

#include <arduino.h>
#include <esp8266wifi.h>
#include <cloud4rpi.h>
#include "DHT.h"

Adafruit provides a DHTtester.ino example of a sensor connection.

This code initializes a sensor and defines a structure to store the measurement result (in case it was successful):

#define DHTPIN 2      // Digital pin connected to the DHT sensor
#define DHTTYPE DHT11 // DHT 11
// ...
DHT dht(DHTPIN, DHTTYPE);
dht.begin();
// ...
struct DHT_Result {
    float h;
    float t;
};
DHT_Result dhtResult;

Once we have that data, the next step is to send it to the Cloud4RPi service.

The Cloud4RPi for Arduino page describes the library API, which is a set of methods used to:

-create, read and update variables, -send variable values into the cloud using the MQTT protocol.

The library supports three variable types: Bool, Numeric, and String.

The library workflow starts with creating an API instance using the Device Token from the cloud4rpi.io website (refer to the article’s part 1 for details).

#if defined(CLOUD4RPI_TOKEN)
    Cloud4RPi c4r(CLOUD4RPI_TOKEN);
#else
    Cloud4RPi c4r("!!!_NO_DEVICE_TOKEN_!!!");
#endif

Then, declare variables for DHT11 readings:

c4r.declareNumericVariable("DHT11_Temp");
c4r.declareNumericVariable("DHT11_Hum");

 Then, get data from the sensor, save them into variables and publish the data to Cloud4RPi:

c4r.setVariable("DHT11_Temp", dhtResult.t);
c4r.setVariable("DHT11_Hum", dhtResult.h);
c4r.publishData();

Temperature and humidity does not change quickly, so sending more than one value per 5 minutes is not required.

Cloud4RPi supports diagnostic data along with variable values. I used uptime, Wi-Fi signal strength, and IP address as diagnostic data:

c4r.declareDiagVariable("IP_Address");
c4r.declareDiagVariable("RSSI"); // WiFi signal strength
c4r.declareDiagVariable("Uptime");

Note: The millis function I use to obtain uptime resets to zero every ~50 days. Which is more than enough for my project.

The following code sets diagnostic variable values:

c4r.setDiagVariable("RSSI", (String)WiFi.RSSI() + " dBm");
c4r.setDiagVariable("IP_Address", WiFi.localIP().toString());
c4r.setDiagVariable("Uptime", uptimeHumanReadable(currentMillis));
c4r.publishDiag();

The uptimeHumanReadable function converts milliseconds to a convenient form:

String uptimeHumanReadable(unsigned long milliseconds) {
    static char uptimeStr[32];
    unsigned long secs = milliseconds / 1000;
    unsigned long mins = secs / 60;
    unsigned int hours = mins / 60;
    unsigned int days = hours / 24;
    secs -= mins * 60;
    mins -= hours * 60;
    hours -= days * 24;
    sprintf(uptimeStr,"%d days %2.2d:%2.2d:%2.2d", (byte)days, (byte)hours, (byte)mins, (byte)secs);
    return String(uptimeStr);
}

The function outputs a string like this 5 days 10:23:14 instead of a strange big number.

After compiling the created code and flashing it into NodeMCU, the device connects to a cloud service and starts sending data.

You can increase logging verbosity by setting the CLOUD4RPI_DEBUG preprocessor variable to 1 (add -D CLOUD4RPI_DEBUG=1 to build_flags section in platform.ini file).

Next, open the cloud4rpi.io site and notice the new device online. Open it to see all variable values received from the device: sensor and diagnostics.

At this step, the data connection to the cloud is operational. Now, let’s configure the visual representation of the data.

I used the Dashboard configuration UI to create the following dashboard:

The dashboard is shareable, so I instantly share it with my friend.

The full project’s code is available in the gist.

That’s all for now!

Questions and suggestions are welcome in the comments.