I’ve used an infrared thermometer at work for a while now and find it a great tool in many situations. However I felt that there is still room for improvement. I’d like to see better sampling modes, live trend graph, and even USB data logging. There are some tools that have some of these features, but they are quite expensive… and I’d like to build my own anyway so I can customise it to suit my requirements.

So another project was born, and first thing for me was to find a sensor module. 🙂

The infrared temperature sensor

The sensor module I’m using is from Freetronics, an Australian hobby electronics company that make lots of Arduino compatible modules. I bought mine from my local Jaycar store.

Product page:

• Freetronics IR Temperature Sensor Module
• Quickstart Guide

The sensing area is a circle of 1:1 ratio with the distance. For example if I measured the surface of a wall 1m in front of me, the measurement area would be a 1m diameter circle. I would’ve preferred a narrower measurement area, but this is still ok so I’ll work with what is available.

Troubleshooting

Unfortunately the particular module I purchased wasn’t working for me out of the box. I wasn’t sure if I was doing something wrong, so I hooked up my DSO (digital storage oscilloscope) for a closer look:

This is a very simple serial signal consisting of a data line and clock line. You can see from the DSO captures above neither are doing anything despite the acquire line being driven from the Arduino to request data. I contacted Freetronics and they suggested I may have a faulty module and they will replace it for me.

Luckily the next module worked perfectly! 🙂

The first prototype

I quickly built up the first prototype using an Arduino Nano and a monochrome OLED display from Adafruit. Here it is on a breadboard:

Here’s a closer look at the infrared thermometer module:

This is running the first version of my software which has a few bugs, and is a work in progress. The large number is the current measurement with a live trend graph underneath. One bug that is visible in the photos is the incorrect rolling-average / smoothing value shown in small text in the middle of the graph axis (the other two values are the graph minimum and maximum range).

I plan to bug fix this software and add more features like sample memory and other view modes. I also have a compact enclosure I’d like to use which I will detail in a future blog post.