MODELING PIROUETTE POSES IN BALLET PERFORMANCES BY USING A

ROTATING PLATFORM WITH A CHANGING MOMENT OF INERTIA

Research Question: What is the effect of changing moment of inertia on the angular momentum
of a rotating platform when the position of the object with respect to the rotation center is
increased under the conditions that force applied on the object is kept constant?

1. Introduction
From the time that I started it in the first grade, ballet has always been a field that I am
passionate about and a major part of my whole life. In ballet, while a lot of technical rules and
traditions are taught during our training, the most important experiences are gained through
trial-and-error method, just like a scientist does his/her investigation in a laboratory. While
dancing, I experienced the principles and limitations of physics through my own body and
unintentionally used those principles in order to perfect my technique and achieve the graceful
and floating image that ballet aims to create. One of the most common movements in ballet is
pirouettes, or “rotation of the body around a vertical axis over one supporting foot on the floor”1.
In this paper, I will investigate the physics behind pirouettes, and while doing that see the
real-life application of concepts such as torque, angular momentum, the moment of inertia, and
how they are related to a field that has been one of the major parts of my life for years.

1.1 Rotational Mechanics

When ballet dancers perform a pirouette,
they begin with positioning their body
with one foot forward, one foot back in
the fourth position, legs bent doing a plié,
and their body balanced in the middle2,
which can be seen in Figure 1. Then, by
pushing off their back foot up the knee of
the supporting leg, which is called retiré,
they create the momentum that gives them
enough speed to turn around themselves.
This preparation is essential due to the
concept of torque in physics. Torque is
defined as “the tendency of a force to
rotate the body to which it is applied”,
around an axis of rotation3. In this case,
the force is the foot pushing off the

1 Laws, Kenneth. Physics and the Art of Dance : Understanding Movement. New York, Oxford University Press, Inc., 2002.

2 Illumin Magazine. “The Science behind the Perfect Pirouette – and How It Has Changed the World of Prosthetics.” vol. 19, no. 2, 2019. Illumin Magazine.

3 Britannica, The Editors of Encyclopaedia. "torque". Encyclopedia Britannica.

1
ground, and the axis of rotation is the body, shown in Figure 2. The equation for the initial torque
is τ = 𝑟 × 𝐹. Once the ballet dancer is spinning, there is no external force acting on them, other
than the friction force. Friction is an example of an external force that produces torque. This is
because when a dancer pushes on the floor one way, the friction between the leg and the floor
creates the push the opposite way. This is modeled in my experiment by using a pendulum and a
rotating platform with two blocks on both ends. The pendulum, as it hits the rotating platform,
creates momentum that lets the rotating platform turn around itself. The setup and methodology
will be discussed later thoroughly.

The principle of the conservation of angular momentum is important while performing

pirouettes. The formula of angular momentum is 𝐿 = 𝑟 × 𝑝 , where 𝑟 is radius and 𝑝 is
momentum. This also can be written as 𝐿 = 𝐼 × ω where L is angular momentum, I is moment
of inertia, and ω is angular velocity. Angular velocity, ω, is “the time rate at which an object
rotates, or revolves, about an axis, or at which the angular displacement between two bodies
changes4.” It can be expressed by the equation ω = 𝑣 ÷ 𝑟 . Moment of inertia is defined as
“quantity expressed by the body resisting angular acceleration which is the sum of the product of
the mass of every particle with its square of a distance from the axis of rotation5.” Moment of

4 Britannica, The Editors of Encyclopaedia. "angular velocity". Encyclopedia Britannica, 4 Dec. 2018.

5 “Moment of inertia.” Merriam-Webster.com Dictionary, Merriam-Webster

2
 2
inertia of a point mass is 𝐼 = 𝑚 × 𝑟 . If there are multiple point masses then the total moment
of inertia equals the sum of them. Moment of inertia is dependent on the mass, radius and shape
of an object6. In this experiment, I will be calculating the moment of inertias of two masses, one
rectangular rotating platform and a cylinder shaft. The equations for these differently shaped
objects are as given below:
2
𝑚𝑟
For solid cylinder: 𝐼 = 2
2 2
For rectangular plane: 𝐼 = (
𝑚 𝑎 +𝑏 )
12
2
For point masses: 𝐼 = 𝑚𝑟
Ballet dancers change the speed of their pirouettes by extending their arms or bringing their arms
2
together. Since 𝐼 = 𝑚𝑟 and ω = 𝑣 ÷ 𝑟 , as ballet dancers extend their arms, increasing the
radius with respect to the center of mass, the moment of inertia increases, however, the angular
velocity decreases. As stated in the law of conservation of angular momentum, angular
momentum is conserved when the radius of the circle is changed.
𝐿 = 𝐼𝑖ω𝑖 = 𝐼𝑓ω𝑓
Therefore, in order to investigate this relationship, I decided to conduct an experiment by
modeling pirouette movement with the help of a simple pendulum made by a tripod stand, rod
and rod clamp, and a rotating platform with two equal masses on it, positioned on both sides. The
positions of the masses with respect to the rotation center (r) are equal and are increased 4 cm in
each trial.

2. Variables
Independent variable: Position of the object with respect to the rotation center.
● Starting from 6.25 cm, increasing the radius 4 cm in each trial.
Dependent variable: Angular momentum
● By measuring the angular velocity and rotational inertia five times for each value
with a data logger.
Controlled variables:
1) Force applied on the object
Method for controlling the variable: Measuring the mass used in the setup and using
the same mass throughout the whole experiment (100 g).
2) Mass of the object
Method for controlling the variable: Using the same system and masses (275 g)
throughout the whole experiment.
3) The angle that the force is applied on the system
Method for controlling the variable: By keeping the position of the tripod stand and the
rod stable and releasing the 100 g mass on the pendulum from the same angle in each trial.

6 Fleishman, Stephen. "Moment of Inertia." Scholastic GO!

3
3. Methodology

3.1 List of materials

● Rotating platform (0.587 kg ±0.001 kg; 0.506 m ±0.005 m width; 0.0545 m ±0.0005 m
length)
● Center shaft (0.188 ±0.001 kg; 0.008415 m radius)
● 275 g masses (x2)
● Data logger
● Digital adaptor (± 0. 001 uncertainty)
● Weighing scale
● The base of the rotating platform, “A” platform
● Tripod stand
● Rod
● Rod clamp
● 70 g mass carrier
● 100 g mass
● iPhone 8 mobile phone
● LoggerPro data-collection software from Vernier Software & Technology

4
 3.2 Experiment procedure
1. Set up all the equipment according to the diagram above.
2. Put the 275 g masses at both sides of the rotating platform, with a distance of 6.25 cm to
the center. Make sure that the screws on top of them are not loose. Measure and note the
radius.
3. Connect the data logger to the rotating platform to measure the angular velocity (rad/s).
4. Put the 100 g mass into the mass carrier and pull it until it touches the rod.
5. Release the mass carrier without applying any forces on it. Pull it back as it touches the
rotating platform.
6. Repeat steps 4-5 four more times.
7. Repeat steps 4-6 four more times by increasing the radius by 4 cm in each repeat.

3.3 Safety and Environmental Issues

● Check if the screws are loose every time the experiment is repeated.
● Hold the mass carrier once it touches the rotating platform.
● Do not change the position of the tripod stand or the angle that the mass carrier is
released.
● Make sure that the rotating platform is perpendicular to the mass carrier and not moving
before the mass carrier hits the rotating platform.

4. Hypothesis
When the radius with respect to the axis of rotation is increased, the angular velocity will
decrease. Increased radius will result in an increasing rotational inertia. If there is no torque
applied, since angular momentum is equal to rotational inertia times angular velocity, as it is
being stated in the Law of Conservation of Angular Momentum, the angular momentum will be
conserved, will stay the same. Thus, while
ballet dancers are increasing or decreasing
their angular velocity by bringing their arms
together or extending their arms, their
angular momentum will be kept constant.

5. Data Collection:
For each level (6.25 cm, 10.25 cm, 14.25
cm, 18.25 cm, 22.25 cm) the experiment has
been repeated five times. In each of these
trials of each level, the data was collected
through PASCO’s Xplorer GLX data logger.
The data logger, as can be seen in Figure 4,
had a lot of values on its screen. For a more
accurate result, the averages of five

5
consecutive values of angular velocity are used for each trial of each level. These consecutive
values are taken where the angular acceleration is at its closest to zero, and the angular velocity
is almost constant. I chose these values in order to eliminate the effect of friction force, thus the
initial torque on the system is ignored.

Table 1 is an example of the process discussed above. This process is repeated for all levels. The
values of angular velocity from 7 to 11 in Figure 4 are the values of Trial 5 of the level 6.25 cm
in Table 1.

Table 1: Raw data table for r= 6.25 cm

Distance to the axis of Angular Velocity (rads-1) (± 0. 001)

rotation (cm)

D1 (cm) D2 (cm) Trial 1 Trial 2 Trial 3 Trial 4 Trial 5

(± 0. 05) (± 0. 05)

2.268 2.276 2.276 2.264 2.286

2.264 2.257 2.268 2.249 2.256

2.265 2.227 2.263 2.229 2.253

6.25 6.25
2.263 2.223 2.262 2.202 2.251

2.253 2.218 2.249 2.198 2.247

Averages= 2.263 2.240 2.264 2.228 2.259

Sample calculation for measuring average values of each trial:

2.268 + 2.264 + 2.265 + 2.263 + 2.253 + 2.263

5
= 2. 263

Then, repeat this step for each trial of all levels. The average values of the raw data for all levels
are given in the raw data table (Table 2) below.

6
Table 2: Raw data table

Distance to the axis of Angular velocity averages (rads-1) (± 0. 001)

rotation (cm)

D1 (cm) D2 (cm) Trial 1 Trial 2 Trial 3 Trial 4 Trial 5

(± 0. 05) (± 0. 05)

6.25 6.25 2.263 2.240 2.264 2.228 2.259

10.25 10.25 1.869 1.885 1.888 1.929 1.945

14.25 14.25 1.675 1.608 1.611 1.659 1.633

18.25 18.25 1.307 1.352 1.340 1.379 1.343

22.25 22.25 0.889 0.920 0.951 0.947 0.917

Graph 1: Angular velocity vs. distance to the axis of rotation

7
Uncertainty Calculations for the Raw Data:

● Uncertainty in the angular velocity measurement: It is taken as the smallest reading of the
data logger used in the experiment (Xplorer GLX) which is 0.001. All of the angular
velocity measurements with the uncertainty (± 0. 001) are given in raw data (Table 2).
● Uncertainty in the measurements of distance to the axis of rotation: The measurements
are made by using a ruler. The uncertainty of the ruler is taken as (± 0. 05) since the
closest marks on the ruler are 0.1 cm apart. The measured values of D1 and D2 are given
with the uncertainty (± 0. 05) in the raw data table (Table 2).

6. Data Processing:

Calculating the uncertainties of average values of angular velocity:

First, deduce the average value of angular velocity where distance to the axis of rotation is 6.25
cm.
2.263 + 2.240 + 2.264 + 2.228 + 2.259
ω𝑎𝑣𝑔 = 5
= 2. 251rads-1

𝑚𝑎𝑥. 𝑣𝑎𝑙𝑢𝑒 − 𝑚𝑖𝑛. 𝑣𝑎𝑙𝑢𝑒

Then, calculate the absolute uncertainty by using the 2
equation.

2.264−2.228
∆ω𝑎𝑣𝑔 = 2
= 0. 018
ω𝑎𝑣𝑔 = 2. 251 ± 0. 018

Calculating the total moment of inertia:

2
𝑚𝑟
For solid cylinder: 𝐼 = 2
2 2
For rectangular plane: 𝐼 = (
𝑚 𝑎 +𝑏 )
12
2
For point masses: 𝐼 = 𝑚𝑟

𝐼𝑠𝑦𝑠𝑡𝑒𝑚 = 2 × 𝐼𝑝𝑜𝑖𝑛𝑡 𝑚𝑎𝑠𝑠 + 𝐼𝑟𝑒𝑐𝑡𝑎𝑛𝑔𝑢𝑙𝑎𝑟 𝑝𝑙𝑎𝑛𝑒 + 𝐼𝑠𝑜𝑙𝑖𝑑 𝑐𝑦𝑙𝑖𝑛𝑑𝑒𝑟

Moment of inertia of the rotating platform (rectangular plane):

2 2
𝐼= (
𝑚 𝑎 +𝑏 )
12
2 2
(
0.587· 0.545 +0.506 ) = 0. 01266
12
Moment of inertia of the center shaft (solid cylinder):
2
𝑚𝑟
𝐼= 2
2
0.188×0.008415 −6
2
= 6. 66 × 10

8
Moment of inertia of the masses (point mass x2):
2
𝐼 = 𝑚𝑟
This value is going to be changing since the r value, distance to the axis of rotation, is changed.
2
2 × 0. 276 × (0. 0625) = 0. 00216 (for r= 6.25 cm)
Repeat this step for every level of distance to the axis of rotation. After calculating the moment
of inertia of the masses for each level, add the values of the rectangular plane, solid cylinder and
the two masses calculated above. The calculated values of the total moment of inertia of the
system (Itotal) are given in Table 3.
−6
0. 00216 + 0. 01266 + 6. 66 × 10 = 0. 1289 kgm2 (for r= 6.25 cm)

Table 3: Processed data table including the average values of angular velocity (ωavg) and
moments of inertia for each level of distance to the axis of rotation.
D1 (cm) (± 0. 05) D2 (cm) (± 0. 05) Itotal (kgm2) ωavg (rads-1)

6.25 6.25 0.1289 2.251 ± 0. 018

10.25 10.25 0.1325 1.903 ± 0. 038

14.25 14.25 0.1380 1.637 ± 0. 034

18.25 18.25 0.1452 1.344 ± 0. 036

22.25 22.25 0.1641 0.9248 ± 0. 031

Since the gradient of the graph of the processed data must give us the angular momentum, the
equation 𝐼 × ω must give the gradient of the graph. Because the gradient of a graph equals to
𝑦−𝑎𝑥𝑖𝑠
𝑥−𝑎𝑥𝑖𝑠
the graph of I vs 1/ω should be used. Table 4 includes the processed values that will be
used to form the graph.

Sample calculation for 1/ω𝑎𝑣𝑔 for r= 6.25 cm:

1
2.251
= 0. 444
Sample calculation for the uncertainty of 1/ω𝑎𝑣𝑔 for r= 6.25 cm:

( 0.018
2.251 )
· 100 · (− 1) =− 0. 7996
0. 444 · ( −0.7996
100 ) =− 0. 0036 ≃− 0. 004

9
Table 4: Processed data table including the values of Itotal and 1/ωavg.
Itotal (kgm2) 1/ωavg (rads-1)

0.1289 0.444 ± 0. 004

0.1325 0.525 ± 0. 01

0.1380 0.61 ± 0. 01

0.1452 0.744 ± 0. 02

0.1641 1.081 ± 0. 04

Graph 2: Moment of inertia vs. 1/angular velocity

7. Conclusion and Evaluation

The aim of this experiment was to prove that angular momentum will remain conserved
as the position of the object with respect to the rotation center is increased, under the condition
that there is no external torque applied on the system. In order to achieve this purpose, a system
including a simple pendulum and a rotating platform is used to model the pirouette movement in
ballet performances. After all of the data collected and analyzed, it is possible to conclude that
the experiment proves the hypothesis. The raw data table, Table 2, shows that the angular
velocity decreases as the distance to the axis of rotation increases, proving that ballet dancers
extend their arms in order to perform pirouettes more slowly and bring their arms closer to their

10
body in order to perform faster pirouettes. Graph 1 also illustrates the inverse relationship
between angular velocity and the distance to the axis of rotation.
When the processed data table, Table 3, is analyzed, it can be observed that as the
distance to the axis of rotation increases, while angular velocity decreases, the moment of inertia
increases. This, without any further calculation, gives us a clue about the result of the
experiment, that the angular momentum might be conserved, since the values have an inverse
relationship. Consequently, Graph 2 proves that claim. The gradient of Graph 2, the moment of
inertia vs. 1/ω graph, must have given us the angular momentum of the system, and as the
hypothesis suggests, this must be a constant value. In Graph 2, the best fit line is a linear fit with
the equation y= mx+b, meaning that the gradient is a constant value. This proves the hypothesis.
Additionally, the graph has a correlation coefficient of 0.9997, therefore the experiment shows a
high precision with small random and systematic errors. This is one of the strengths of this
experiment.
Strengths and Weaknesses:
One of the strengths of this experiment was that I conducted the experiment in our
school’s physics laboratory in the presence of our physics teacher, being able to use professional
equipments and devices which reduced the amount of systematic errors to low levels, especially
the rotating platform system. At first, I thought of conducting the experiment by the help of a
real person performing pirouettes, however it would be almost impossible to control the
controlled variables such as force applied on the subject. Because of that, there would be a high
level of random errors which would affect the precision and accuracy of the experiment. By
conducting the experiment in our laboratory, using professional equipments needed, the amount
of these errors were reduced.
Even though the experiment showed a high accuracy and precision, it was still not
perfect. The possible systematic and random errors that might have affected the results of the
experiment are given below:

Systematic Errors and Improvements:

Friction force on the rotating platform:
The rotating platform used in the experiment is a mechanism that is designed for
experiments with the rotating motion. Because of that, the friction force was negligible, it could
be observed that the platform would not stop rotating without someone holding it. Even though
the friction force was very low on the system, in order to reduce it even more, the system could
be oiled.
Air resistance:
Even though air resistance was negligible and had an insignificant effect on the results,
there was some amount of air resistance acting against the motion of the rotating platform. The
experiment could be performed in an air vacuumed environment in order to reduce that
systematic error.

11
 Random Errors and Improvements:
The angle that the pendulum was released:
In the experiment, the angle that the pendulum was released was a controlled variable.
The pendulum was released from the point where the end of the mass carrier touched the rod
each time. However, since it was not an automatic mechanism, there may be some degree of
random error due to small shifts in the position of the pendulum or the angle that the pendulum
was released. In order to solve this problem, the position of the pendulum should be fixed
properly and the angle should be measured with a protractor each time the experiment is
repeated.
Energy loss due to sound energy:
Another random error was the energy losses due to the sound energy created when the
pendulum hit the rotating platform. During the experiment, I observed that in some trials where
data was collected, the sound created when the two objects collided was louder than others. This
random error could be prevented by using a soundproofing material such as cloth, or conducting
the experiment in an air vacuumed environment.

References:
“angular velocity | Britannica.” Encyclopedia Britannica,
https://www.britannica.com/science/angular-velocity.
Fleishman, Stephen. “Moment of Inertia.” Scholastic GO!,
go.scholastic.com/content/schgo/C/article/019/649/0196490-0.html.
Illumin Magazine. “The Science behind the Perfect Pirouette – and How It Has Changed the
World of Prosthetics.” vol. 19, no. 2, 2019. Illumin Magazine,
https://illumin.usc.edu/the-science-behind-the-perfect-pirouette-and-how-it-has-changed-t
he-world-of-prosthetics/.
Laws, Kenneth. Physics and the Art of Dance : Understanding Movement. New York, Oxford
University Press, Inc., 2002,
http://www.magisterseniusu.com/uploads/1/8/0/0/1800340/kenneth_laws_martha_swope
_francia_russell-physics_and_the_art_of_dance__understanding_movement-oxford_univ
ersity_press_usa__2002_.pdf.
“Moment of inertia Definition & Meaning.” Merriam-Webster,
https://www.merriam-webster.com/dictionary/moment%20of%20inertia.
“torque | Equation, Definition, & Units | Britannica.” Encyclopedia Britannica,
https://www.britannica.com/science/torque.
Tsokos, K. A. Physics for the IB Diploma Coursebook. Cambridge University Press, 2014.

12