Fork me on GitHub

Project Notes

#440 Logging Voltage Measurements To Adafruit IO

Using Adafruit IO and an Ethernet shield to collect analog measurements to the cloud with MQTT.



I wanted a simple way to log the voltage of the LEAP#439 QX5252 Solar Night Light, rechargeable battery to get a sense of the overnight discharge and daily recharge pattern.

It didn’t need to be particularly accurate, but I did want to get a fairly continuous log at least every 5 minutes or so. Sounds like a perfect case for some Arduino-powered data logging!

I could just capture some data locally to an SD card, but sending the data to the “cloud” is much more appealing, as it means I can access and do some analysis at any time without needing to be local to the device.

There are many established and roll-your-own solutions for this problem, but I’ve chosen to use this as an opportunity to experiment with Adafruit IO.

Adafruit IO is a publically-hosted MQTT service that of course offers a free tier for experimentation, and as you would expect Adafruit has libraries for Arduino, ESP8266, ESP32 and Python environments. It also appears to be a pretty faithful implementation of the MQTT specification, so other MQTT libraries should work with it too.

MQTT is a pub-sub framework and can be used for bi-directional data transfer and control, but I’ll only be using a fraction of its capabilities (I just want to do some data collection in the cloud).

Adafruit IO has some basic dashboarding automatically built-in, which is enough to visualise the collected data. But they also have integration options for doing something with the data once collected, such as trigger IFTTT workflows.

About MQTT

MQTT (Message Queuing Telemetry Transport) is an ISO standard for pub-sub over TCP/IP, designed for light-weight clients (like an Arduino) or where bandwidth is limited. Development of MQTT is coordinated by the community at

MQTT protocol defines a binary control packet structure. There are 15 control packet types (0x0 and 0xF are reserved/forbidden):

Control Packet Payload Value Description
CONNECT Required 0x1 Client request to connect to Server
CONNACK None 0x2 Connect acknowledgment
PUBLISH Optional 0x3 Publish message
PUBACK None 0x4 Publish acknowledgment
PUBREC None 0x5 Publish received (assured delivery part 1)
PUBREL None 0x6 Publish release (assured delivery part 2)
PUBCOMP None 0x7 Publish complete (assured delivery part 3)
SUBSCRIBE Required 0x8 Client subscribe request
SUBACK Required 0x9 Subscribe acknowledgment
UNSUBSCRIBE Required 0xA Unsubscribe request
UNSUBACK None 0xB Unsubscribe acknowledgment
PINGREQ None 0xC PING request
PINGRESP None 0xD PING response
DISCONNECT None 0xE Client is disconnecting


The code is a simple variation of the mqtt_ethernet.ino example from the Adafruit MQTT Library. Some variations:

  • logs an analog reading from A0
  • does not subscribe to any commands
  • configuration is extracted in a separate header file that can be excluded from the repository (so keys are kept private)

Resolution (or lack thereof)

The voltage readings I’m expecting will be in the 1-1.3V range. Direct sampling with a 5V Arduino Uno’s analog pin sacrifices a great deal of resolution - basically I’ll have a 4.9mV steps.

That’s not great, and could be improved with some off-board amplification or higher resolution ADC, but in this case I prefer to accept the lack of precision for the benefit of not needing any additional circuit components.

Setting up Adafruit IO

It’s pretty simple! Create an Adafruit account and use it for access to the Adafruit IO console. The “View AIO Key” link provides the username and key details needed to configure an MQTT client.

The AIO username, key and feed name are defined in settings.h. Since this includes “secret” details, it is not checked into the repository. Before compiling the source, copy settings.h.template to settings.h and fill in the required information.

Adafruit IO Features

The Adafruit IO cloud service has 4 key components (on top of the MQTT infrastructure):

  • Feeds - the “data bins” used to collect data/messages
  • Dashboards - web tool for building simple interfaces to present data and controls
  • Triggers - setup actions that can fire on a schedule or based on feed data values
  • Services - a selection of third-party applications with pre-built integrations that can be enabled


With the script running, data will automatically be collected in the nominated feed. The feed page provides direct access to the raw data (including export features), and also has basic charting to get a sense of the data:



It is also very easy to create a custom dashboard that includes feed data:


Dashboards are probably much more interesting with more data feeds and control interfaces!


Simple event-based or scheduled triggers are easy to setup. When fired, triggers can publish a message or call a webhook. In the paid version they can also send email.

I setup a low-voltage trigger that calls a webhook:


For testing purposes, I pointed it to a Request Bin instance. Here’s what an incoming event looks like:


The event is a nice JSON structure, although I haven’t been able to find a specification for this yet (the Triggers API doc appears a little deficient at the moment):



Since I purposely avoided any external circuitry, this is about as simple as it gets: tap the battery voltage with A0 and make sure the DUT has a common ground:



Power Problems

After running some successful tests with the Arduino plugged into my Mac USB, I switched to an external 12V/2A power supply in order to leave it running. Within 15 minutes the Ethernet shield had locked up with an error.

After a little investigation, I’m suspecting the problem is with the 5V regulator of my (cloned) Arduino skewing low (~4.7V) and perhaps not delivering reliable current. Something to dig into and solve later…

For now, I’ve avoided the problem by simply powering the unit with a regulated 5V supply direct to the 5V pin of the Arduino. No more lock-ups.


Adafruit IO is pretty amazing! It is extremely easy to get something simple up and running fast. It doesn’t over-reach in terms of its features (or price), so I can see this being a great prototyping platform for relatively straight-forward control and data collection. But this also means you can probably imagine much more ambitious projects that require more than Adafruit IO has to offer.

For my initial problem - log some data about the state of a rechargeable battery over a couple of days - it was perfect. Literally took me 10 minutes from signing up at Adafruit to when I had data streaming into a feed. Doing more research on how it worked and writing these notes took far longer;-)

Credits and References

Project Source on GitHub Project Gallery Return to the LEAP Catalog

This page is a web-friendly rendering of my project notes shared in the LEAP GitHub repository.

LEAP is just my personal collection of projects. Two main themes have emerged in recent years, sometimes combined:

  • electronics - usually involving an Arduino or other microprocessor in one way or another. Some are full-blown projects, while many are trivial breadboard experiments, intended to learn and explore something interesting
  • scale modelling - I caught the bug after deciding to build a Harrier during covid to demonstrate an electronic jet engine simulation. Let the fun begin..
To be honest, I haven't quite figured out if these two interests belong in the same GitHub repo or not. But for now - they are all here!

Projects are often inspired by things found wild on the net, or ideas from the many great electronics and scale modelling podcasts and YouTube channels. Feel free to borrow liberally, and if you spot any issues do let me know (or send a PR!). See the individual projects for credits where due.