This project originated from the urgent need to transition to remote learning in March 2020 due to the COVID-19 pandemic. The sudden shift to remote instruction provided only a week’s notice, leaving little time to adapt the curriculum. At that time, I resorted to using computer simulations, particularly the PhET Interactive Simulations,1 which I frequently incorporated as teaching aids during lectures. However, it became apparent that simulations alone could not adequately replace hands-on laboratory experiences. Recognizing the importance of practical laboratory work, I aimed to provide my students with authentic lab experiences using tangible equipment. This approach aligns with the 2014 “AAPT Recommendations for the Undergraduate Physics Laboratory Curriculum,”2 which emphasizes the significance of conducting real experiments. As I prepared to teach General Physics I (classical mechanics) in the upcoming fall semester, I recognized the need to promptly begin planning and preparation to ensure the quality and effectiveness of the laboratory component.
I had the crazy idea to use the sensor board hacked from a Sphero Mini Robot3 (Fig. 1). The inertial measurement chip (IMU) on the Sphero Mini sensor board is the six-axis Bosch BMI055,4 commonly used in commercial virtual reality headsets. Students would use the IMU (Fig. 2) for “at-home” physics labs I would develop. The graphical user interface (GUI) of the Sphero EDU5 app would allow students to record and log data. The user-friendly GUI turned out to be a significant highlight of the student experience.
The Sphero robot, shown as packaged and shipped from the manufacturer, along with the sensor board, after removal from the robot.
The Sphero robot, shown as packaged and shipped from the manufacturer, along with the sensor board, after removal from the robot.
In the summer of 2020, I convinced Sphero to donate 44 Sphero Minis to the Furman Physics department. My primitive early experiments with the sensor board can be seen in my Furman Physics YouTube Channel6 video “At Home Physics Labs with Sphero Sensor Board.”7 A video summary of the entire project is available at “Furman Sphero Mini Sensor Board Project.”8 As an aside, while this was not a conscious decision during the development phase, the project lends itself to inquiry-based laboratory instruction.9,10
Introduction
You can purchase a Sphero Mini for $50, but you may find a used one on eBay for much less. You may find one abandoned in a kid’s toy box. The process by which I hacked the sensor board requires only 10 minutes with a small Phillips screwdriver. Details are available at Physics Sensor Data Acquisition Tool Hacked From a Sphero Mini.11
In the academic year 2020–2021, my wife and I prepped, packed, and shipped “At-Home Lab Kits” as shown in Fig. 3. I assigned “virtual lab partners” facilitated by a GroupMe12 mobile group chat messaging platform to minimize student anxiety in the remote environment. The labs13–18 we developed (see Table I) were performed by a total of 80 students in my four sections of PHY 111, the calculus-based mechanics class at Furman. We began with two orientation labs. The first was an introduction to the Sphero EDU5 app. The second was learning to take data, export the CSV data files to Excel, and then do simple analysis in Excel. We then did four “real physics” labs, as described in the Furman Physics YouTube Channel.6 Curriculum material documents are available in the online appendix.19
Contents of the “At-home Lab Kit” that was sent to the students. Beginning at the top (12 o’clock position) and proceeding clockwise, the kit contents consist of the modified Sphero Mini sensor board, a rice-filled pharmacy prescription bottle, Scotch tape, a 4-m length of string, a 2-m length of elastic, a 3-m tape measure, two restaurant sauce cups, and two cardboard tubes. At the center are the cardboard “paddle” for the centrifuge experiment, and a 6-in. ruler for the slingshot sled experiment.
Contents of the “At-home Lab Kit” that was sent to the students. Beginning at the top (12 o’clock position) and proceeding clockwise, the kit contents consist of the modified Sphero Mini sensor board, a rice-filled pharmacy prescription bottle, Scotch tape, a 4-m length of string, a 2-m length of elastic, a 3-m tape measure, two restaurant sauce cups, and two cardboard tubes. At the center are the cardboard “paddle” for the centrifuge experiment, and a 6-in. ruler for the slingshot sled experiment.
The six labs in mechanics developed in this project.
| Orientation labs | Unboxing the Furman-Modified Sphero Mini Sensor Board13 |
| Intro to Data Capture from Sphero Mini Sensor Board14 | |
| Physics labs | Slingshot Sled Kinematics with Sphero Mini Sensor Board15 |
| Centripetal Acceleration vs. Angular Velocity Using Sphero Mini Sensor Board16 | |
| Solid Cylinder Rolling Down a Ramp Using Sphero Mini Sensor Board17 | |
| Simple Pendulum—Acceleration due to Gravity Using Sphero Mini Sensor Board18 |
| Orientation labs | Unboxing the Furman-Modified Sphero Mini Sensor Board13 |
| Intro to Data Capture from Sphero Mini Sensor Board14 | |
| Physics labs | Slingshot Sled Kinematics with Sphero Mini Sensor Board15 |
| Centripetal Acceleration vs. Angular Velocity Using Sphero Mini Sensor Board16 | |
| Solid Cylinder Rolling Down a Ramp Using Sphero Mini Sensor Board17 | |
| Simple Pendulum—Acceleration due to Gravity Using Sphero Mini Sensor Board18 |
Slingshot sled kinematics lab
In this lab, students measure the time-varying acceleration a(t) of the sled (essentially a heavily damped spring-mass system) and then numerically integrate twice to obtain velocity v(t) and position x(t). Typical data are shown in Fig. 4. When the sled comes to rest, the v(t) curve should go to zero, and the x(t) curve should be flat. Due to the limited temporal resolution of the sensor, the calculated v(t) does not quite go to zero. Still, there is a reasonable agreement of the value of x(t) at the effective “end” of the motion with the measured stopping distance.
Time histories of measured acceleration and Mathematica-calculated velocity and position.
Time histories of measured acceleration and Mathematica-calculated velocity and position.
Centripetal acceleration vs. angular velocity lab
After they construct a cardboard centrifuge paddle (Fig. 5) using the contents of the “at-home” lab kit (Fig. 3), students measure the centripetal acceleration aradial vs. angular velocity ω of an object executing circular motion. Students “wind up” the top and bottom pairs of strings and release the paddle. Typical data are shown in Figs. 6–8. A least-squares quadratic fit of the data aradial = Rω2 yields a value for the inferred rotation radius R = 12.6 cm, which compares very well with the measured distance of 12.1 cm from the center of rotation to the IMU chip.
Completed centrifuge assembly showing the sensor board and two quarters for counterweights.
Completed centrifuge assembly showing the sensor board and two quarters for counterweights.
Graph of radial acceleration (units of g) vs. time collected by the Sphero Mini sensor board for the cardboard centrifuge motion.
Graph of radial acceleration (units of g) vs. time collected by the Sphero Mini sensor board for the cardboard centrifuge motion.
Graph of angular velocity (in degrees per second) vs. time collected by the Sphero Mini sensor board for the cardboard centrifuge motion.
Graph of angular velocity (in degrees per second) vs. time collected by the Sphero Mini sensor board for the cardboard centrifuge motion.
Solid cylinder rolling down a ramp lab
The purpose of this classic lab is to study the work/energy relationship for a solid cylinder rolling without slipping down an inclined plane (Fig. 9). The supplied prescription bottle filled with rice has a bottle cap that mates with the sensor board holder. Students measure the peak value of the angular velocity ω using the “yaw” channel on the sensor board, as shown in Fig. 10.
A solid cylinder of mass m and radius R starts from rest starting from a vertical height h on an inclined plane.
A solid cylinder of mass m and radius R starts from rest starting from a vertical height h on an inclined plane.
Simple pendulum lab—Acceleration due to gravity
This classical simple pendulum lab is the easiest to perform and gives excellent agreement (better than 1% accuracy) with the accepted value of the acceleration due to gravity g. Typical student results are shown in Fig. 11.
Student data for the Simple Pendulum Lab. Editorial note: The label heading in the fourth column should read “Period (sec)” without the obviously incorrect equation T = L/N that the student submitted. It is the author’s intention to present the information as given to him by the student.
Student data for the Simple Pendulum Lab. Editorial note: The label heading in the fourth column should read “Period (sec)” without the obviously incorrect equation T = L/N that the student submitted. It is the author’s intention to present the information as given to him by the student.
Discussion
Note that the Sphero sensor board technical specs (especially the sampling rate) are no match for those of the IOLab,20 PocketLab,21 and Phyphox22 commercial platforms, which are designed specifically for data acquisition. We hope to mitigate this limitation in future applications. On the positive side, two properties of the sensor board provide unique advantages over the three commercial platforms noted. The small size of the sensor board allows it to be mounted easily on everyday objects. One of my students taped the sensor board to the rim of his bicycle wheel. Several students dubbed the sensor board the “Apple AirTag”23 for physics labs! Suppose the sensor is mounted in the original Sphero Mini shell. Students can then explore systems using the vector angular velocity (roll, pitch, and yaw) suggested in “Sphero Mini Sensor Board Rolling Down A Shallow Cone.”24
Student responses on course evaluations were generally positive, as indicated by the selected responses noted in Table II. While the world may never need to return to large-scale remote instruction, I believe this project has applicability to both high school and college first-year physics labs, inquiry-based labs, individual student STEM special projects, and science fairs and robotics club projects.
Selected student responses to end-of-semester course evaluations.
| Generally Positive Responses | Responses Citing Problems or Difficulties |
|---|---|
| “The Sphero labs helped my learning for the most part. They were helpful because they gave us real-life ways to apply the equations we learned. I also liked having the virtual lab partners.” | “… some partners did far less or far more work than the others. Maybe there could be roles that each partner would have to complete included in the assignment to improve this problem.” |
| “I really enjoyed the Sphero labs. I felt like an “at-home” physicist working away in my apartment to prove the known theories of the physical world. I felt like the time needed to perform the labs was very reasonable for the amount of work it required.” | “The Sphero labs definitely helped with my understanding, but it would have been more helpful if an explicit correlation could be made between the lecture material and the lab work. This was done a few times, but not consistently.” |
| “Virtual Lab partnership helped a ton, and the Sphero was fun and made me go out of my way to Google any topics I was confused about how to try and understand a little bit better. The time required was beyond reasonable and caused no stress at all.” | “It may be helpful to use the designated lab time to have groups meet with each other (breakout rooms). This could help with scheduling for virtual lab partners. I know my group struggled with finding times to complete the labs and then meeting virtually to make sure the report was cohesive.” |
| “I thought that the Sphero labs were unique and helped incentivize higher levels of effort because it felt like we were using real pieces of technology. I think using new technology that measures mass quantities of data incentivizes effort.” | “The Sphero labs did not help me understand the concepts. Though they were relatively simple and easy to do; I felt them to be just another assignment rather than a true learning experience.” |
| “I have a pretty positive opinion of the Sphero labs. They did take some effort to build, but the data collection part was easy. I liked working with the Sphero sensor because it automatically collected data, and the program is really easy to use.” | “I think there should be a lab where the students come up with the design.” |
| “I liked the hands-on action and being able to work with the data provided in the graphs. I also liked how once you learned how to use the Sphero, you could manipulate your design to better the results.” |
| Generally Positive Responses | Responses Citing Problems or Difficulties |
|---|---|
| “The Sphero labs helped my learning for the most part. They were helpful because they gave us real-life ways to apply the equations we learned. I also liked having the virtual lab partners.” | “… some partners did far less or far more work than the others. Maybe there could be roles that each partner would have to complete included in the assignment to improve this problem.” |
| “I really enjoyed the Sphero labs. I felt like an “at-home” physicist working away in my apartment to prove the known theories of the physical world. I felt like the time needed to perform the labs was very reasonable for the amount of work it required.” | “The Sphero labs definitely helped with my understanding, but it would have been more helpful if an explicit correlation could be made between the lecture material and the lab work. This was done a few times, but not consistently.” |
| “Virtual Lab partnership helped a ton, and the Sphero was fun and made me go out of my way to Google any topics I was confused about how to try and understand a little bit better. The time required was beyond reasonable and caused no stress at all.” | “It may be helpful to use the designated lab time to have groups meet with each other (breakout rooms). This could help with scheduling for virtual lab partners. I know my group struggled with finding times to complete the labs and then meeting virtually to make sure the report was cohesive.” |
| “I thought that the Sphero labs were unique and helped incentivize higher levels of effort because it felt like we were using real pieces of technology. I think using new technology that measures mass quantities of data incentivizes effort.” | “The Sphero labs did not help me understand the concepts. Though they were relatively simple and easy to do; I felt them to be just another assignment rather than a true learning experience.” |
| “I have a pretty positive opinion of the Sphero labs. They did take some effort to build, but the data collection part was easy. I liked working with the Sphero sensor because it automatically collected data, and the program is really easy to use.” | “I think there should be a lab where the students come up with the design.” |
| “I liked the hands-on action and being able to work with the data provided in the graphs. I also liked how once you learned how to use the Sphero, you could manipulate your design to better the results.” |
Acknowledgments
I gratefully acknowledge the donation of the 44 Sphero Minis from Sphero, and the enthusiasm and tireless support from Sphero employees Colleen Witmuss, Micah Daby, Jeff Wiencrot, Amanda Vaden, Brian Kellner, and Paul Copioli. Many thanks to Furman physics major teaching assistants Madeline Edenton, Maggie Fritts, Maddie Klumb, Ian Maitland, and Lina Zaharias for their help with developing these labs, and to Mr. Wade Shepherd and Ms. Belinda Hilliard for technical and administrative support. I am grateful to my physics department colleagues, Profs. Bill Baker, David Moffett, and Susan D’Amato, for their friendship, encouragement, and support throughout my career at Furman—especially during this project in the throes of the COVID era.
References
John R. Conrad earned a BS in physics from St. Mary’s University in San Antonio, TX, and a PhD in physics from Dartmouth College. After 28 years of teaching and research at the University of Wisconsin-Madison, he retired as Wisconsin Distinguished Professor of Engineering Physics Emeritus. After retirement from UW-Madison, Conrad was an adjunct professor of physics for 15 years at Furman University in Greenville, SC.