Arduino projects

This webpage contains links to "how to" documents about using Arduinos. Because that's where my own interests lie, many of the documents arise from using Arduinos to monitor, display, and log environmental data. These documents assume some previous knowledge of how to write code for using Arduino sensors. A comprehensive introduction to using Arduinos for environmental monitoring, assuming no previous Arduino programming experience, can be found in David Brooks' book, Exploring Your Environment with Arduino Microcontrollers. The availability of sensors and other devices discussed in that book and these documents is always a moving target and there is no way to guarantee future availability.
Some of the documents referenced here refer to other documents available from this webpage. All the documents are in Microsoft Word format so you can copy code from them. All the Arduino projects described on this webpage have been developed on a Windows 10 computer using the free Arduino Integrated Development Environment (IDE). The IDE is a cross-platform application, but versions for other operating systems, including older Windows systems, may have problems with drivers, for example, or other problems for which I can offer no assistance.

Contents for Exploring Your Environment with Arduino Microcontrollers

The image shows the contents for this book. Some of the documents on this page reference the book, which is available for $25 including shipping to any U.S. address. Contact brooskdr@instesre.org. All book proceeds go to support the mission of the Institute for Earth Science Research and Education..
When errors are found in the book they will be noted here:

P 128. Line 112 of the code on this page has a } at the end of the line that shouldn't be there.

A components supplier referenced numerous times in this book, allelectronics.com, is no longer in business. Several of their former employees have started a new company that may, in time, take allelectronics’ place: www.aretronics.com.

The IZOKEE 0.96" SSD1306 I2C OLED cited in this book is apparently no longer available. Comparable SSD1306 I2C OLEDs from other vendors should work with the same code. Some of these OLEDs have split screen colors – yellow on the top few pixel rows and blue for the rest of the screen. Some have blue-on-black displays that are very attractive. These colors can’t be changed in code, and are less readable in bright sunlight than a white-on-black display.

Inexpensive UNO-compatible for testing I2C devices

The image shows an inexpensive UNO-compatible board I've set aside just for testing I2C devices with header pins. The board, which I use for many of my projects, is the Seeedstudio V4.3 UNO compatible that you can find at this URL:
https://www.seeedstudio.com/Seeeduino-V4-2-p-2517.html
Yes, the URL says "V4-2," but the product described is actually the V4.3, which as of July 2024 cost US$7.60.
You just need to upload test code for your I2C device with header pins connected to the M/F jumper wires. This board uses a USB A/microB cable. The board can be switched between 3.3 V and 5 V power/logic. Some I2C sensor modules require 3.3 V.
Also, there are several online sources for code that will test for the presence of I2C devices. They will give you the device's I2C hardware address if you need that.

Using RGB LEDs

This document shows how to use RGB LEDs, including boards using 5050 RGB LEDs. Several examples are given, with breadboard layouts and code for controlling single and multiple LEDs, including using PWM to control LED brightness.

Some Arduino data display 0ptions

This document describes several options for displaying data collected with Arduinos, ranging from simple LCD text displays to TFT displays for creating full-color graphics

Comparing PMS5003 and Sensirion SPS30 particulate sensors

This document compares data from two inexpensive Arduino-compatible airborne particulate sensors. Both devices use a single laser beam that's deflected by particulates in a fan-driven air stream flowing past the laser. Both sensors reliably track time-dependent variability in particulates, but the quantitative mass density values are considerably different.

Using piezo vibration sensors

This document shows how to use piezo vibration sensors. It suggests that rather than depending on Arduino's very coarse 10-bit 0-5 V A/D resolution, a 16-bit programmable gain ADS1115 A/D board is a much better choice for detecting small vibrations such as might be encountered from heavy traffic or construction activities.

Using an Arduino-compatible Geiger counter

This document shows how to use Arduino's interrupt-processing capabilities to look for pulses from a Geiger counter built from a kit. The code counts pulses and calculates counts per minute. This kind of code can be adapted to any situation where it's necessary to look for events that happen randomly, with only a very fast and transient indication of their occurrence.

Using Arduino UNO and Nano data logging shields

This document shows how to use data logging shields for collecting and storing data. Although it may be more "modern" to think of storing data online, there are still many good reasons for collecting, recording, and storing data locally with date/time stamps. Data logging shields make this possible. They are widely and inexpensively available online for UNO and Nano Arduino boards.

Data logging with an Arduino Mega

Although UNOs and Nanos are almost always suitable for data collection projects with multiple sensors, projects that use graphic display devices can sometimes cause program memory problems.
The software libraries required to use TFT and eInk displays take a lot of program memory. An option for users who are already comfortable using UNOs and Nanos is to use a Mega 2560. This board has eight times the program memory space as an UNO or Nano.
UNO data logging shields will fit on Mega boards, but some adjustments must be made to pin assignments, and the topic is explored in detail in this document.

Evaluating the HM3301 particulate sensor

See Comparing PMS5003 and Sensirion SPS30 particulate sensors, above. This document performs a similar comparison between PMS5003 and HM3301 particulate sensors with an I2C interface.

Introducing the Arduino UNO R4 MINIMA

This document provides an introduction to the new Arduino UNO R4 MINIMA. This board has the same physical layout as the UNO R3, and it's supposed to accept R3 code without modification – an assumption that turns out to be not entirely true, depending on what hardware you attach to the board! The R4 uses a different microprocessor that offers several advantages over the R3, including 8 times more memory for storing code compared to the R3! This board may override the need to use a more expensive UNO Mega to get more program storage space for large projects.

Introducing the Arduino Nano Every

This document provides an introduction to the Arduino Nano Every. This board has the same physical layout as the Nano, as shown below the Nano in the image, with exactly the same pinout configuation. It costs less than an "official" Nano – currently about $20 vs. $25 – but of course more than "off shore" Nanos.
The Nano Every uses a different processor than the Nano and you need to install board support for it in your IDE. It has much more memory storage space than a Nano and for many applications, it should accept Nano code with no changes. This is a huge potential advantage for projects that exceed the memory space of Nanos. Some sensor and other device software libraries may not work with this board.

Using the AS7341 10-channel multispectral sensor

My book Environnmental Monitoring with Arduino Microcontollers shows how to use two I2C multispectral sensors from Sparkfun.com -- one with 6 VIS channels and one with 6 NIR channels. (Those two sensors have the same I2C address, so a multiplexer is needed to use both of them on the same Arduino.) This document shows how to use a 10-channel multispectral sensor, the I2C AS7341 from Adafruit.com, which has 8 VIS channels, a broadband "clear" channel, and an NIR channel. The code presented in the document displays raw data from the 10 channels on a small OLED and draws a full-color bar graph showing those data.

Using the SparkFun AMG8833 8×8 thermal sensor

The AMG8833 is an interesting low-resolution 8×8 pixel "thermal camera." This document provides code for using this sensor to produce 64 16×16-pixel colored squares on a small TFT display, where the colors are based on the thermal response of each pixel on the sensor. This device will detect a human presence at least a few feet away when viewed against a relatively uniform thermal background that's cooler than a human body.
The AMG8833 breakout board is from SparkFun.com and the Arduino board is an UNO code-compatible MetroMini from Adafruit.com. There are six possible colors, ranging from black for temperatures cooler than some selected minimum, to red for the warmest sensor pixels. A small OLED displays the matrix containing the 64 color selection values.

Interpolating matrix values in an Arduino sketch

The thermal sensor document on this page shows how to use an AMG8833 8×8 pixel thermal sensor to produce a matrix of 64 16×16 pixel color squares s on a small TFT display. The visual appearance of these data can be enhanced by interpolating the original 8x8 matrix to a 15×15 matrix to produce 225 8×8 pixel pixel color squares on the TFT display.
This document shows how to write an Arduino sketch that starts with an 8×8 matrix and produces a 15×15 interpolated matrix, using two-dimensional arrays. The image here shows this interpolation done in Excel, which I used as a guide for writing the code.

A generic Arduino data logging module

This document shows how to use a small generic data logging module for Arduino 5 V boards that don't share the UNO R3/R4 or Nano physical layout, like Arduino ProMinis or Pros. (There are data logging modules that plug into the headers of UNOs and Nanos.) There's no data logging shield for Adafruit.com's MetroMini, a board that is recognized as an UNO R3 by the Arduino IDE, but which is about the same size as, but not pin-compatible with, a Nano. -

Data logging with Nanos

Based on some problems I've encountered when trying to use Nano data logging shields with projects that include a small OLED display, I wrote this document that uses the generic data logging module discussed in the previous post to log data from an MS8607 T/RH/P sensor. This is an I2C sensor that I believe is an improvement over the BME280.

A GPS system for Arduinos

This document shows how to track your position using an inexpensive GPS module. The module requires a CR1220 coin cell battery for power, which is in a holder under the pc board, below the rectangular GPS receiver module. A four-wire cable is included. An antenna, the square module, snaps onto the pc board. Code in the document shows how to extract longitude, latitude, elevation, and UTC time from GPS satellite transmissions. See the posting below about how to use thermistors with a GPS system to track and locate temperatures in "walking around" expeiments.

Adjusting emissivity for MLX90614 IR sensors
The MLX90614 IR sensor is factory-calibrated for an assumed emissivity of 1. Real surfaces have emissivities that are often close to but less than and never equal to 1. Assuming the wrong emissivity produces object temperature errors. Some applications using these sensors, including those found on this webpage, do not show how to change the assumed emissivity. Why not? Because the sample code on which those applications are based don't show the required code even though the software libraries allow the changes to be made. For the Adafruit.com MLX90614 library, the very simple fix is to define an emissivity float variable and give it a value – many tables of emissivity values are available online and 0.95 is a reasonable choice for many natural and manmade surfaces. Then, assuming that the MLX90614 methods are invoked with mlx, include this line in the setup() function:
mlx.writeEmissivity(emissivity);
after the mlx.begin() statement. See This document for more information about IR sensors and emissivity.

Monitoring surface temperatures with an IR sensor

Record high temperatures across the U.S. and Europe during the summer of 2023 motivated this project. This Document shows how to use an MLX90614 IR sensor to monitor ground temperatures in urban heat islands and other microclimates. The image shows a sensor system recording temperatures of a gravel driveway on a warm sunny day.

Building IR sensor systems for monitoring surface temperatures

The previous entry in this list described using an MLX90614 IR sensor to monitor surface temperatures. This Document gives details about building such a system. The document was developed as part of IESRE's role as a science consultant to ECoSTEM, a project at Xavier University of Louisiana funded by the National Science Foundation's Improving Undergraduate STEM Education program. There are instructions for a building an Arduino Nano-based "breadboard" thermal sensing system at an in-person workshop for secondary school students.

Correcting IR sensors for surface emissivity

IR sensors like the MLX90614 are typically calibrated for an object emissivity of 1 – a situation that's never encountered with real surfaces. This document gives emissitiites for a range of natural and manmade surfaces and two approaches to changing sensor emissivity. My preferred method is to correct temperatures "after the fact" as shown in the image.

Using spectral sensors with Adafruit MetroMini UNO-compatible

Chapter 19 in Brooks' book Monitoring Your Environmental with Arduino Microcontrollers shows how to use SparkFun spectral sensor breakouts with Adafruit's Feather M0 Express board. This Document shows an alternative approach using Adafruit's MetroMini UNO-compatible board. A small OLED is used to display data.

Arduino calculations for the Julian Date

Julian dates are used for astronomical calculations and for solar tracking applications. Because of some limitations with Arduino floating point variables, some tricks are needed to calculate a Julian Date accurately. This document shows how to do it.
The code also provides an example of how to read a string of characters representing numerical values typed in the COM port window (for this project, integers for the year, month, day, hour, minute, and second) and then split that character string into an array of integers. The same basic procedure will also work for floating point numbers.
As a bonus, the lower image shows a different calculation implemented in Excel, based on python code provided by ChatGPT, to which you could add code for including minutes and seconds. In Excel, the int() function truncates a floating point number, rather than rounding it – acting like a "floor" function.

Following the sun with Arduinos

There are interesting projects that involve tracking the sun. The best way to do this is with calculated solar positions and a servo-driven pan-tilt device. A basic need for such code is to be able to calculate the Julian date. Because of the limitations on the accuracy of Arduino floating point numbers you have to be creative about how you do this calculation – see the previous posting!
This document shows how to use servos with Arduinos and how to program a small two-servo pan-tilt device to follow the sun.

Using lux sensors to monitor sky conditions

Solar radiation reaching Earth's surface is usually measured with pyranometers. Graphs of these data show diurnal variation in solar radiation, including effects due to variable cloud cover. Lux sensors, which respond to light as perceived by human eyes, can also be used to monitor daytime sky conditions. including cloud effects. This document shows how to build an inexpensive Arduino Nano-based system to collect and log lux, temperature, and relative humidity data.

Using LiPo batteries

Lithium-Polymer (LiPo or LiPoly) batteries are widely used in low-power Arduino projects. It's important to use batteries from a reliable vendor and chargers that are specifically designed for safely charging and maintaining these batteries. This document provides a brief discussion of the hardware IESRE uses for its LiPo-powered Arduino projects.

An Arduino-friendly three-cup anemometer

Wind speed is an essential meteorological parameter. Anemometers are instruments used to collect these data. For routine monitoring activities, these devices often use three cups to collect the wind, causing a shaft to rotate. The rotation speed is related to wind speed.
This document shows how to use an inexpensive three-cup anemometer that produces a voltage (0-3.8 VDC) proportional to the wind speed up to about 70 mph. This is a completely passive device so it's very easy to use with just two Arduino connections – to an analog input pin and ground. The image shows the anemometer with a Nano and an LCD to display the output voltage when the shaft is rotating.

Calculating the Air Quality Index for airborne particulates

Measuring airborne particulates is the central science focus for the National Science Foundation-funded ECoSTEM project at Xavier University of Louisiana, for which IESRE serves as the science consultant. Inexpensive laser-based and Arduino-compatible sensors are used for this project and data are collected for PM1, PM2.5, and PM10 particulates. While the PM values themselves are of interest, it's also desirable to convert the PM2.5 and PM10 values into their corresponding AQI values as defined by the U.S. Environmental Protection Agency. These values are used by the EPA to assign health risks due to airborne particulates. The reported AQI is based on the maximum AQI associated with each of six pollutants – O₃, PM2.5, PM10, CO, NO₂, and SO₂. Especially in urban areas, the PM2.5 AQI is often the pollutant that defines the reported AQI.
This document provides Excel formulas and Arduino code for converting PM2.5 and PM10 values into their AQI values.

Monitoring airborne particulates coming from a woodstove

Airborne particulates can cause health problems if their values are too high. This document shows some particulate data collected indoors near a woodstove and converted to their AQI values as defined by the U. S. Environmental Protection Agency, using calculations as described in the previous document in this list.

Using an Adafruit 1.44" TFT display with the HUZZAH32-ESP32 Feather board

Adafruit's HUZZAH32-ESP32 board is a popular choice for WiFI/IoT applications. This document shows pin assignments for interfacing Adafruit's 1.44" full-color TFT display with this board. Those assignments should work for other Adafruit TFTs, too.

Using the ILI9225 176×220 pixel full-color TFT

There are many full-color TFT displays available from vendors like Adafruit.com and online. A less expensive but equally usable alternative is the older ILI9225 176×220 pixel TFT available from MPJA.com and other online sources. This document shows how to use the ILI9225 to create data graphs. The image shows sine and cosine curves. Code is also included for displaying temperature, humidity, barometric pressure, and Air Quality Index values for airborne particulates.
The pixel count is a little lower than newer similarly sized TFTs but it's still useful for many applications. A possibly useful feature of this SPI device is that is can be plugged directly onto the UNO header that includes power, ground and analog input pins. That feature then requires use of software SPI rather than faster hardware SPI. It also ties up analog pins A0 through A5, which will be a problem if your application needs these pins for analog output or I2C devices that use analog pins A4 and A5. By using a breadboard the pins can be wired to the usual digital pins for hardware SPI.

The MS8607 T/RH/P sensor

This document compares Adafruit's MS8607 T/RH/P sensor (shown in the image) with BME280 T/RH/P and AM2315C T/RH sensors. The BME280 is widely used in Arduiino-based weather station applications, but its relative humidity performance always degrades catastrophically in outdoor applications where relative humidity varies over a very wide range.
It's possible that the other two sensors, which are designed to promote better air flow around the sensing element, will be less prone to saturation of the hygroscopic dielectric materials used in the humidity sensor, and will also reduce self-heating of the temperature sensor that's required to convert absolute humidity to relative humidity. The MS8607 module, which allows air flow around three sides of the sensing element, should be an improvement over the BME280 module, where the sensing element is mounted directly on the PC board with no air space around it.

Introduction to using small OLED displays with Arduino Nanos

This document gives an introduction to using small monochrome OLEDs for displaying data with text or graphics. These devices work well with Nanos, for example, but large programming projects that require a graphics library will strain the Nano's programming capabilities. One solution is to use Arduino's Nano Every – a newer board that's pin- and (nearly) performance-identical to the Nano but has much more space for code. The image here shows a Nano Every project that displays PM 1, PM 2.5, and PM 10 values on three horizontal bar graphs.

Fonts for small OLEDs
I often use small OLEDs with the u8x8 text-only library. Here are some images showing three different fonts, found HERE, to be entered as the parameter into a u8x8.setFont(); statement. The font I often use with IZOKEE 0.96" OLEDs (the middle image below) is readable, but very small! The left-hand image is much easier to read if your data will fit in 4 lines, no more than 15 characters per line. The large font in the right-hand image, from a sketch that I use to test new OLEDs, is the same height as the font in the left-hand image, but wider – like a bold font. It's good only for displaying a limited amount of data.
u8x8_font_7x14_1x2_f (4 lines,15 characters each)
u8x8_font_chroma48medium8_r (A very small font!)
u8x8_font_px437wyse700a_2x2_r (4 lines, 7 characters per line)

Monitoring environmental noise levels

Noise can be a significant environmental "air pollutant," especially in urban areas where there is constant noise from traffic, construction, and other sources. This document shows how to monitor noise with an Arduino-compatible sound meter module. It incorporates a small electret microphone like the one shown in the image into a complete module that converts sounds to an analog voltage that can then be converted to decibels, the standard way to quantify noise.

Using nRF24L01+ RF modules with Arduinos

This document shows how to use nRF24L01+ RF modules operating in the license-free 2.4 GHz radio band to collect and display environmental data. These inexpensive modules are widely available online with on-board antennas for ranges of several 10s of meters indoors and screw-on "rubber duck" antennas for outdoor ranges up to 1 km under ideal open-air conditions. They are alternatives to the 915 MHz and 433 MHz RF devices described in Chapter 20 of David Brooks' book Monitoring Your Environment with Arduino Microcontrollers. The nRF24L91+ is the small pc board at the bottom of the image.

Using thermistors with Arduinos

This document shows how to use negative temperature coefficient (NTC) thermistors for monitoring temperatures. These devices are inexpensive, accurate, and easy to interface with Arduinos. They come in a variety of configurations, including the one shown in the image with 1 m stranded wire leads, to which short solid wires have been soldered for use with breadboards. See the posting above about using a GPS tracker with a thermistor for a "walking around" experiment to study heat island effects.

Using MQ-4 methane sensors with Arduinos

This document shows how to use an MQ-4 sensor to monitor methane, a major greenhouse gas. These devices produce an analog output, so they're trivially easy to interface with Arduinos like UNOs and Nanos. The analog input pins on those boards, with their 10-bit A/D resolution, produce an integer value between 0 and 1023 over 5 V. Like other MQ sensors, the MQ-4 responds to gasses other than methane and calibrating them in units of parts per million of methane is challenging and requires expensive laboratory equipment. Therefore, these devices are best suited for "sniffing out" relative methane levels in specific environments or across a range of environments. Some sources of methane include wetlands and marshes, agricultural and farming areas, landfills, and petroleum processing facilities.
There are several physically identical or similar MQ devices optimized for detecting various gasses. They all use a heated sensor element (SnO₂) so they require more power than is typical for Arduino/sensor systems. This makes outdoor use away from plug-in power sources a challenge.

Using a generic BMP-280 T/P sensor

I have written extensively about the deficiencies of BME280 T/RH/H sensors for prolonged outdoor use because of the failure of their humidity sensors. Recently I bought by mistake some HiLetGo BMP280 T/P sensors (no humidity sensor) from Amazon. If an appropriate I2C hardware address is used, these generic temperature and pressure sensors are supported by a library from Adafruit.com and this short docoument shows code for using them. Only four of the six pins are required -- VCC, GND, SCL, and SDA. Some other versions of these sensors may include only those four pins on their pc boards. Unlike similar sensor modules from Adafruit.com, which will most often work with either 3.3 V or 5 V power, these generic devices are 3.3 V power only! For testing them I have used a Seeeduino V4.2 UNO compatible with switchable 3.3/5.0 logic, but they will also work with 3.3 V power/5 V logic.

Arduino UNO/Nano SPI pin connections

Two common devices that use serial peripheral interface (SPI) connections for communicating with Arduinos are TFT displays and SD card modules for data storage. I am constantly uncertain about how to connect these devices, especially when a system needs more than one SPI device. (It's easy to connect multiple I2C devices to an Arduino system because each device has its own hardware address, but the pin connections are always the same: VCC, GND, SDA(pin A4), SCL(pin A5).) UNOs may have a second pair of I2C pins labeled SDA and SCL, just above the AREF pin in the image.) SPI connections are also available through the ISCP header at the bottom of the image, but I've never used it.
The four SPI pinouts can have different names on different devices, which can be confusing:
SCLK=CLK-->13 (fixed)
MISO=DO-->12 (fixed, not needed to write to SD card)
MOSI=DI=SDI=SDA-->11 (fixed, this SDA not the same as I2C SDA!)
SS=CS-->10 (code-selectable)
For the MOSI pin I included SDA because that's what the MOSI pinout is called on the ILI9225 TFT, an older device documented elsewhere on this web page.
It's possible to use multiple SPI devices on the same system. They all use pins 11, 12, and 13, but the chip select (SS or CS) pin must be specified in code, different for each device. That's perhaps why the "SS" pin in the image is in a different color than the other three pins and isn't available through the ISCP header. I've assigned the CS pin as digital 10 because on data logging shields for UNOs and Nanos the CS pin is hard-wired to this pin, so it can't be used for another SPI device.
The pin connections shown above are for "hardware SPI," which is what you would most commonly use for SPI devices. However, it's also possible to use "software SPI" with all pins specified in code. An example of this is the ILI9225 TFT. (See a previous entry on this web page.) Its pins are laid out so it can be plugged directly into Aruino UNO headers. The MISO connection isn't needed for displaying data and graphics to the screen:
CS-->A5
MOSI-->A2
CLK-->A1
power-->5V
GND-->GND
Software SPI may draw content on a TFT noticeably more slowly than hardware SPI, but I've used the ILI9225 in both software and hardware SPI modes and not noticed significant differences. Of course this unusual feature of the ILI9225 is useful only if you don't need access to these pins (including all the analog input pins) for anything else!
SPI connections are different for other Arduino boards, like the MEGA (see a previous entry on this web page). The pin names should be consistent, so the connections should be clear.

Using an Arduino-compatible CO₂ sensor

This document shows how to use a DFRobot CO₂ sensor. It's intended mainly for indoor monitoring, but is can also be used to "sniff out" prominent outdoor sources of this important greenhouse gas. Its I2C interface and software library makes it easy to use with Arduinos. At a cost of almost $60, it's relatively expensive for Arduino sensors, but it could be useful in classrooms and other enclosed indoor spaces to make sure that CO₂ levels remain lower than about 1000 ppm. Note that energy efficient buildings, which are highly insulated and airtight, may reduce energy consumption, but without adequate indoor/outdoor air exchange, CO₂ levels can be higher than is healthy for humans.

Data logging with the Arduino Feather M0 Express

This document shows how to use an Adafruit Feather M0 Express board with a FeatherWing data logging shield and OLED display. Although this is a relatively expensive system as Arduino systems go, there are at least four significant advantages of the M0 Express, which is about the size of a Nano, over a Nano or UNO:
1. a complete display/logging system can be put together with Adafruit FeatherWing add-on boards that minimize breadboard wiring and allow a system with several sensors to fit comfortably on a half-size breadboard;
2. a LiPo battery can be connected to the system that allows you to switch seamlessly between online USB power and offline battery power, which is ideal for projects that will be used away from a plug in power source;
3. for analog-output devices you can select 8, 10, 12, or 14 A/D resolution instead of the fixed 10-bit resolution available with UNOs and Nanos;
4. the available program memory space is huge compared to UNO/Nano boards.

A breadboard-friendly SPDT on/off switch

For battery-powered systems on a breadboard, it's very convenient to be able to have an on/off switch to control a battery that most likely will need to be connected to an Arduino board's GND and VIN pins. I recently found such a switch at adafruit.com (ID 805, $1.00), that plugs directly into a breadboard. It's a little flimsy and unstable on a breadboard, but this can be solved with some hot glue to hold it in place. Similar breadboard-friendly switches are available from spakfun.com (COM-09609, $0.95; COM-00102, $1.60)
It's easy to find other small and much cheaper SPDT switches online, but it's never clear whether their pins are on 0.1" (2.54 mm) centers for use on a breadboard. Any of these Adafruit or Sparkfun switches are reliable for their intended purpose.
For battery-operated systems, I'm now using a 6 AA battery pack (Adafruit ID 248) with a 2.1 mm power plug and a breadboard-compatible 2.1 mm jack (Adafruit ID 373). Why go to this extra expense rather than just inserting wires from a battery pack directly into breadboard tie points? Typically, these wires will to to an Arduino board's VIN and GND pins. If you're careless with these wires and insert them where they don't belong, you can short out a battery or destroy an Arduino board, as I did recently with a Nano. The on-board power-in jack and SPDT switch essentially eliminates chances for these outcomes, assuming that the power jack and switch wiring is done properly.

Display "degree" symbol (°) with the u8x8 or Adafruit GFX library

I often create projects that include a temperature in their output. It's not obvious how to display a "degree" symbol before a temperature unit, like °C, or after an angle, like 36.6°N, on a display device. THIS SHORT DOCUMENT shows hows to do it with OLED displays.

An Arduino-based microclimate monitoring system

There are several documents available from this webpage that show how to monitor air and ground temperatures – essential components of examining microclimates like those associated with urban heat islands. This document describes a project that uses an Adafruit Feather M0 Express board with a data logging shield and OLED, and thermistor, MS8607 T/RH/P, MLX90614I IR, and lux sensors to give a more complete view of conditions found in microclimates. This is a relatively expensive project by Arduino standards, but UNOs and Nanos can't handle the amount of code required. A Mega would probably work, but it's more expensive than the Feather M0 Express. This system is very compact and the M0 Express has a port for a LiPo battery that can be "hot swapped" with USB power, making this a good choice for projects that need to work when no plug-in power is available.
I believe you should consider this as an "advanced" Arduino project, as the document assumes that you already have experience working with Arduinos. The Feather M0 Express board will be unfamiliar to UNO and Nano users and you may need to reference other postings on this webpage to be successful with this project.

A LiPo battery adapter shield for UNOs

I've posted here about this item (Adafruit ID 2078) just to remind myself about its availabilty. Using UNOs away from plug-in power sources is often not easy, requiring awkward battery packs to power the UNO through its VIN pin with voltages in the 7-12 V range. Here's a better solution: a $20 UNO shield that allows you to use a LiPo battery (not included) to provide a regulated 5 V to the UNO's 5 V power pin. The shield includes an off/off switch. At the time I wrote this posting (October 2024), this item was out of stock, but hopefully that isn't a permanent situation.

Some sources for Arduinos and related products
The Arduino "ecosystem" is a truly global enterprise, and the "open source" brand philosophy means that there are many sources for Arduino boards and related hardware. Not surprisingly, there are many "off shore," often Chinese, sources, which may have quality control and delivery issues. Perhaps surprisingly, Amazon is a major source of Arduino-related products, including boards, accessories, and sensors – even what I would characterize as very obscure and even obsolete sensors.
Neither IESRE as an organization nor I as an individual have any financial or other relationship with any of these sources other than the fact that I have purchased equipment from them and found them to be reliable. Comments about these sources are based only on my own experiences and cannot be construed as an endorsement by IESRE.
• Adafruit, a leading U.S. manufacturer and distributor of top-quallity Arduino-related products, some of which aren't available elsewhere.
• Amazon, a major source of Arduino-related products from a variety of sources both domestic and "off shore." Product quality, price, and availability varies widely!
• Aretronics, a relatively new U.S. company formed by former employees of the now dissolved allelectronics.com, that has a small but growing selection of Arduino-related products.
• DFRobot, a major Chinese manufacturer of Arduino-related products. with very competitive pricing and high quality. Has U.S. distributors, including Amazon.
• DigiKey, a major U.S. electronics distributer that sells products from Adafruit.com, for example, and some off-shore suppliers of Arduino-related products. Buying off-shore products from a U.S. distributor can eliminate issues with potentially long shipping times.
• Elegoo, a Chinese manufacturer and supplier of Arduino-related products, with very competitive pricing and dependable quality control, especially for Arduino boards and "starter kits." Best to go through Amazon.
• HiLetGo, a Chinese manufacturer and distributor of Arduino-related products, with very competitive pricing and dependable quality control, especially for Arduino boards. Best to go through Amazon.
• MPJA Electronics, a U.S. distributor of older and/or surplus electronics, including a growing range of Arduino-related products. They have very good customer support.
• SeedStudio, a major Chinese supplier of Arduino-related products, with very competitive pricing and high quality. Has U.S. distributors, including Amazon.
• SparkFun, a leading U.S. manufacturer and distributor of top-quality Arduino-related products, some of which aren't available elsewhere.