Kettering University

Senate, Faculty

5/19/2021: Faculty Senate Unnapproved Meeting Minutes

7/28/2021: Faculty Senate Unapproved Meeting Minutes

7/28/2021: Faculty Senate Approved Meeting Minutes

8/25/2021: ACBSP Unified Standards and Criterion

8/25/2021: MGMT 333 Competency In Professional Management  Draft Syllabus

8/18/2021: MGMT 621, Lean Operations Management Common Professional Component Syllabus

The Quanser Qube is an integrated servomotor lab hardware platform [1], shown in Figure 1. This platform includes not only the direct-drive brushed DC motor, but also two encoders as well as the data acquisition system. One encoder is used to measure the rotation of the DC motor’s shaft itself. It is supplied with two standard items, an inertia disk and an inverted pendulum. The inertia disk is a small aluminum part, which mounts to the equipment using magnets. The inverted pendulum also mounts to the equipment with magnets, with an encoder that plugs into the equipment to provide an additional sensor input to the system. The system includes an amplifier and other necessary components in order to be controlled with either LabVIEW or with MATLAB/Simulink, with the LabVIEW control requiring the National Instruments myRIO device [1]

Peters, Diane

Jones, Aaron Joseph Michael

Kettering University 
Digital Commons @ Kettering University 
Mechanical Engineering Presentations And Conference Materials Mechanical Engineering 
7-2021 
Development of Attachments for the Quanser Qube 
Diane Peters 
Aaron Joseph Michael Jones 
Follow this and additional works at: https://digitalcommons.kettering.edu/mech_eng_conference 
 Part of the Automotive Engineering Commons 
Paper ID #32254
Development of Attachments for the Quanser Qube
Dr. Diane L. Peters, Kettering University
Dr. Peters is an Associate Professor of Mechanical Engineering at Kettering University.
Mr. Aaron-Joseph Michael Jones
c©American Society for Engineering Education, 2021
Development of Attachments for the Quanser Qube 
 
Introduction 
The Quanser Qube is an integrated servomotor lab hardware platform [1], shown in Figure 1. 
This platform includes not only the direct-drive brushed DC motor, but also two encoders as well 
as the data acquisition system. One encoder is used to measure the rotation of the DC motor’s 
shaft itself. It is supplied with two standard items, an inertia disk and an inverted pendulum. The 
inertia disk is a small aluminum part, which mounts to the equipment using magnets. The 
inverted pendulum also mounts to the equipment with magnets, with an encoder that plugs into 
the equipment to provide an additional sensor input to the system. The system includes an 
amplifier and other necessary components in order to be controlled with either LabVIEW or with 
MATLAB/Simulink, with the LabVIEW control requiring the National Instruments myRIO 
device [1]. 
 
 
Figure 1: Quanser Qube 
In this work, a set of new attachments for the Quanser Qube was developed. The attachments 
have been fully designed, and some of them (the inertial disks) have been produced and tested; 
the remaining attachments will be fabricated and tested when the ongoing COVID-19 pandemic 
has abated and full access to the necessary facilities and labs is possible. 
Background 
The value of lab exercises in system dynamics and controls has been widely acknowledged for 
some time, e.g., [2], [3]. At times, equipment is designed and produced exclusively by the lab 
instructors or others at the university, as in [4], [5], [6]. In other cases, an existing, commercially 
available solution is used for modeling or experimental labs, as in [7], [8], [9]. It is also possible 
to have a combined approach, in which an existing solution is used, with some enhancements or 
additional attachments developed, as described in [10]. 
At the authors’ institution, the Quanser Qube is used in the senior-level controls lab, with the 
National Instruments myRIO used as the controller. Students are provided with LabVIEW files 
to run the Qube, which allow them to save their data for later analysis and plotting. The labs 
conducted with the Qube all utilize the inertial disk. The inertial disk is chosen because this 
allows a linear model to be used, which aligns with the knowledge level and course content 
appropriate to this class; the inverted pendulum is occasionally used for demonstration purposes, 
but no labs utilizing this attachment are performed. 
Currently, the labs carried out with the Qube focus on model validation and on control. In the 
area of model validation, students have a purely simulation-based lab where they compare first 
and second order models of the device, to determine under what conditions the motor inductance 
could be neglected, and are given the Qube’s inductance and asked to justify their choice of 
model for it. In the next lab, they utilize the device itself, and compare experimental data to the 
theoretical performance that they expect, then are asked to reflect on the reasons for 
discrepancies. In a later lab, students develop a controller for the Qube. They are given a set of 
specifications that the controller should achieve; these differ from one term to the next, but 
typically involve parameters such as overshoot and settling time. They are then asked to 
calculate values of gains, then apply those gains to the device, determine how well it meets 
specifications, and attempt to use their conceptual knowledge of what different gains accomplish 
to tune the controller appropriately. 
Use of the inertial disk allows a number of labs to be performed, involving system identification, 
model verification, and control, but there are limits to the range of what can be done. The DC 
motor can be approximated as a first order system, due to the low value of the motor’s 
inductance, which leads to a second-order system for the angular position of the disk. 
Furthermore, with only one possible inertia for the disk, all students will have the same values 
for the lab both within the class and across terms. This reduces the ability to easily change labs to 
detect incidents of copying from “crib files”, and does not allow for a model that is validated for 
one set of circumstances (the original disk) to be tested for a different case. Therefore, it was 
decided that additional attachments of varying types would be developed for the Qube. These 
would allow for a greater range of lab exercises, as well as the possibility of using the Qube for 
more open-ended projects. 
Design of New Attachments 
A variety of new attachments were created, ranging from a variety of simple disks of varying 
inertia to more complex devices. The various inertial disks are described first, followed by the 
other types of attachments. 
  
Alternative Disks 
Four disks were designed which varied in size, mass, and inertia. These disks were produced 
using 3D printing, and then had the appropriate magnets glued into recesses in the bottom of the 
disks in order to mount them on the Qube. The disks are summarized in Table 1, with isometric 
views of them shown in Figures 2 – 5. 
Table 1: Inertial Disks 
Disk # Max. Dia. (mm) Max. Height (mm) Mass (g) 
(with magnets) 
Inertia (10-6 kg-m2) 
1 49.60 12.40 15 4.27 
2 49.60 12.40 10 2.34 
3 68.00 6.20 14 8.09 
4 30.00 22.20 9 1.01 
 
 
Figure 2: Disk #1 - Mass Distribution at Outer Edges 
For the first disk, as shown in Figure 1, the mass is distributed to the outer edges, producing a 
relatively higher inertia, compared to the second disk (Figure 2). Both disks have the same outer 
diameter and overall height; however, the mass and inertia are different due to the mass 
distribution. 
 
Figure 3: Disk #2 - Mass Concentrated Nearer the Center 
In contrast to these two, the third disk, shown in Figure 3, is a simple flat disk with a larger 
diameter. 
 
Figure 4: Disk #3 - Constant Thickness Disk 
The fourth and final disk, shown in Figure 4, is also of uniform thickness, but is taller and 
smaller in diameter than the third one. 
 
Figure 5: Disk #4 – Tall Constant Thickness Disk 
Simulations were performed with these disks, in order to determine how much the results of a 
simple lab might vary. It was assumed that a simple square wave, varying from 0 to 2 V, would 
be input to the motor. The results of the simulation are shown in Figure 5. As these results show, 
the steady-state value of the Qube’s speed does not change; however, the time constant clearly 
does experience changes. 
 
Figure 6: Comparison of Simulated Results for Different Disks 
Use of these varying disks could allow a lab to be conducted in which students would be asked 
whether they expect the steady-state value to vary, and if so, why; and then to test their 
hypothesis using the disks. Other labs could be conducted based on the types of analyses 
conducted in [7], with other disks aside from the provided one. Additional labs could be 
designed to look at the impact of uncertainty; the variation from one provided aluminum disk to 
another can be expected to be minimal, compared to the variation seen in the 3D printed plastic 
disks. 
Gear Drive 
The gear drive attachment is a setup that utilizes two Qubes. One is the driving Qube, and the 
other is driven, and is used for sensing purposes. This gear drive is comprised of four gears: a 
0.7-inch diameter gear, a 1.3-inch diameter gear, a 1.9-inch diameter gear, and a 2.5-inch 
diameter gear. It also includes 2 custom made components that are used to attach the 2.5-inch 
gear and the 0.7-inch gear to the Qubes in the same matter as the standard inertia disk. The last 
piece is an assembly plate that secures the other two gears, the 1.3-inch and 1.9-inch, in place 
and keeps the Qubes at a nominal distance for the gear drive to work. This setup is shown in 
Figure 7. 
 
Figure 7: Gear Drive Arrangement 
By creating this more complex system, students could perform a lab in which they examined 
issues that could be seen in a gear drive. A comparison could be performed between simulated 
results and actual results, with the goal of determining what the impact of backlash might be, or 
where there could be unmodeled losses in the system. In addition to its use in modeling 
exercises, a control system could be applied, allowing students to see what the impact of 
backlash is when a system is subject to a controller.  
Drive Belt with Synchronous Pulleys 
Another configuration using two Qubes uses a synchronous belt and pulleys. This arrangement is 
comprised of a 1.63-inch diameter pulley, a 0.88-inch diameter pulley, and a timing belt. Much 
like the gear drive, there are custom made components to connect the components to the Qubes. 
Similarly, the first Qube will drive the system, with the second sensing the results, as shown in 
Figure 8. 
 
Figure 8: Belt & Pulley Arrangement 
This configuration allows similar types of lab exercises to the gear drive system. In addition to 
asking questions about backlash with synchronous belts, students could be asked to evaluate 
whether belt stretch is likely to be a significant concern, and then support their evaluation with 
experimental data.  
Four-Bar Linkage 
The four-bar linkage uses a single Qube, although an arrangement using a second Qube for 
sensing purposes could be designed as well. This setup consists of a 1-inch long bar segment that 
attaches to the rotating part of the Qube, a 2-inch long bar segment, a 1.5-inch long bar segment, 
and a plate that acts like a stationary 1.96-inch long bar segment, as shown in Figure 9. 
 
 
Figure 9: Linkage Arrangement 
The four bar linkage differs from the previous two setups in that it is inherently a nonlinear 
system. This setup could be used in a more complex modeling class, with students validating a 
nonlinear model. It could also be used to investigate issues of linearization; if a linearized model 
is used, students could evaluate for what range of motion they felt it was accurate enough to be 
useful. 
Wind Turbine 
The final attachment designed in this project is an adjustable wind turbine. This device, utilizing 
a single Qube, is comprised of a base plate to attach the turbine to the Qube, a vertical shaft that 
mounts to the Qube’s output shaft, a horizontal shaft with slots for propeller poles, 6 propeller 
poles, and two miter gears. The propeller poles are adjustable, and additional ones could be made 
of different materials and with different geometries. The Qube is used not to drive the device, but 
as a sensing means; its internal encoder reads the rotational speed of the turbine, which could be 
used to compare different configurations and understand how much electricity could be 
generated. This setup is shown in Figure 10. 
 
Figure 10: Wind Turbine Arrangement 
This attachment could be used in labs beyond the Dynamic Systems courses in which the Qubes 
are often used. It could be used for a course in which the focus is on materials, on fluid flow, or 
on energy generation; if a university has a wind tunnel of moderate size, the Qube with this wind 
turbine could be placed in it for a variety of different experiments. 
Summary and Conclusion 
In this paper, designs for a variety of different attachments for the Quanser Qube have been 
presented. The use of such custom designs for a purchased piece of hardware combines the 
advantages of a well-supported commercial platform with the flexibility of a custom-designed 
lab solution. The designs presented could be used for a variety of different types of labs, 
expanding the usefulness of the equipment. In addition, they could be used to spark student 
ideas, in a more open-ended lab in which students could design their own attachments and run 
them with the Qube. By doing so, the effectiveness of lab experiences can be enhanced. 
Acknowledgments 
The authors would like to thank Kettering University’s co-operative education program, for 
funding the on-campus co-op experience for Aaron-Joseph Jones. 
  
References 
[1] https://www.quanser.com/products/qube-servo-2/ accessed 1/15/2021 
[2] J. S. Dalton, D. S. Stutts, and R. L. Montgomery, “Mini-lab projects in the undergraduate 
controls course,” in Proceedings of the ASEE Annual Conference, Nashville, TN, June 
2003. 
[3] Z. Alavi and K. Meehan, “Enhancing a control systems design course by using 
experiential learning model,” in Proceedings of the ASEE Annual Conference, Tampa, 
FL, June 2019. 
[4] M. A. Hopkins and A. M. Kibbe, “Open-source hardware in controls education,” ASEE 
Computers in Education (CoED) Journal, vol. 5 (4), pp. 62 – 70, December 2014. 
[5] P. K. Karra, “A cost-effective laboratory setup for teaching system dynamics and 
controls,” in Proceedings of the ASEE Annual Conference, Salt Lake City, UT, June 
2018. 
[6] R. M. Reck, “Developing an affordable and portable control systems laboratory kit with a 
Raspberry Pi,” Electronics, vol. 5 (3) pp. 36 – 50, 2016. 
[7] R. M. Reck, “Validating DC motor models on the Quanser Qube Servo,” in Proceedings 
of the ASME Dynamic Systems and Control Conference, Atlanta, GA, September 30 – 
October 3, 2018. 
[8] A. Kathpal and A. Singla, “SimMechanicsTM based modeling, simulation and real-time 
control of rotary inverted pendulum,” in 11th International Conference on Intelligent 
Systems and Control (ISCO), January 2017. 
[9] D. L. Peters, R. Stanley, C. J. Hoff, and J. Casci, “Redesign of lab experiences for a 
senior level course in dynamic systems with controls,” in Proceedings of the ASEE 
Annual Conference, Seattle, WA, June 2015. 
[10] D. L. Peters, “Design of a higher order attachment for the Quanser Qube,” in Proceedings 
of the American Control Conference, Boston, MA, July 2016. 


Kettering University

Not a member yet

5/19/2021: Faculty Senate Unnapproved Meeting Minutes

7/28/2021: Faculty Senate Unapproved Meeting Minutes

7/28/2021: Faculty Senate Approved Meeting Minutes

8/25/2021: ACBSP Unified Standards and Criterion

8/25/2021: MGMT 333 Competency In Professional Management Draft Syllabus

8/18/2021: MGMT 621, Lean Operations Management Common Professional Component Syllabus

Development of Attachments for the Quanser Qube

2021 EAC Final Statement of Accreditation

3/24/2021: Faculty Senate Meeting Agenda

5/5/2021: Faculty Senate Meeting Agenda

310

3,854