CLASSICAL

MECHANICS

Jeremy Tatum
University of Victoria
University of Victoria
Classical Mechanics

Jeremy Tatum
This text is disseminated via the Open Education Resource (OER) LibreTexts Project (https://LibreTexts.org) and like the hundreds
of other texts available within this powerful platform, it is freely available for reading, printing and "consuming." Most, but not all,
pages in the library have licenses that may allow individuals to make changes, save, and print this book. Carefully
consult the applicable license(s) before pursuing such effects.
Instructors can adopt existing LibreTexts texts or Remix them to quickly build course-specific resources to meet the needs of their
students. Unlike traditional textbooks, LibreTexts’ web based origins allow powerful integration of advanced features and new
technologies to support learning.

The LibreTexts mission is to unite students, faculty and scholars in a cooperative effort to develop an easy-to-use online platform
for the construction, customization, and dissemination of OER content to reduce the burdens of unreasonable textbook costs to our
students and society. The LibreTexts project is a multi-institutional collaborative venture to develop the next generation of open-
access texts to improve postsecondary education at all levels of higher learning by developing an Open Access Resource
environment. The project currently consists of 14 independently operating and interconnected libraries that are constantly being
optimized by students, faculty, and outside experts to supplant conventional paper-based books. These free textbook alternatives are
organized within a central environment that is both vertically (from advance to basic level) and horizontally (across different fields)
integrated.
The LibreTexts libraries are Powered by MindTouch® and are supported by the Department of Education Open Textbook Pilot
Project, the UC Davis Office of the Provost, the UC Davis Library, the California State University Affordable Learning Solutions
Program, and Merlot. This material is based upon work supported by the National Science Foundation under Grant No. 1246120,
1525057, and 1413739. Unless otherwise noted, LibreTexts content is licensed by CC BY-NC-SA 3.0.
Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not
necessarily reflect the views of the National Science Foundation nor the US Department of Education.
Have questions or comments? For information about adoptions or adaptions contact info@LibreTexts.org. More information on our
activities can be found via Facebook (https://facebook.com/Libretexts), Twitter (https://twitter.com/libretexts), or our blog
(http://Blog.Libretexts.org).

This text was compiled on 02/09/2023

TABLE OF CONTENTS
Licensing

1: Centers of Mass
1.1: Introduction and Some De nitions
1.2: Plane Triangular Lamina
1.3: Plane Areas
1.4: Plane Curves
1.5: Summary of the Formulas for Plane Laminas and Curves
1.6: The Theorems of Pappus
1.7: Uniform Solid Tetrahedron, Pyramid and Cone
1.8: Hollow Cone
1.9: Hemispheres
1.S: Centers of Mass (Summary)

2: Moments of Inertia
2.1: De nition of Moment of Inertia
2.2: Meaning of Rotational Inertia
2.3: Moments of Inertia of Some Simple Shapes
2.4: Radius of Gyration
2.5: Plane Laminas and Mass Points distributed in a Plane
2.6: Three-dimensional Solid Figures. Spheres, Cylinders, Cones.
2.7: Three-dimensional Hollow Figures. Spheres, Cylinders, Cones
2.8: Torus
2.9: Linear Triatomic Molecule
2.10: Pendulums
2.11: Plane Laminas. Product Moment. Translation of Axes (Parallel Axes Theorem)
2.12: Rotation of Axes
2.13: Momental Ellipse
2.14: Eigenvectors and Eigenvalues
2.15: Solid Body
2.16: Rotation of Axes - Three Dimensions
2.17: Solid Body Rotation and the Inertia Tensor
2.18: Determination of the Principal Axes
2.19: Moment of Inertia with Respect to a Point
2.20: Ellipses and Ellipsoids
2.21: Tetrahedra

3: Systems of Particles
3.1: Introduction to Systems of Particles
3.2: Moment of Force
3.3: Moment of Momentum
3.4: Notation
3.5: Linear Momentum
3.6: Force and Rate of Change of Momentum
3.7: Angular Momentum
3.8: Torque
3.9: Comparison

1 https://phys.libretexts.org/@go/page/19041
 3.10: Kinetic energy
3.11: Torque and Rate of Change of Angular Momentum
3.12: Torque, Angular Momentum and a Moving Point
3.13: The Virial Theorem

4: Rigid Body Rotation

4.1: Introduction to Rigid Body Rotation
4.2: Angular Velocity and Eulerian Angles
4.3: Kinetic Energy of Rigid Body Rotation
4.4: Lagrange's Equations of Motion
4.5: Euler's Equations of Motion
4.6: Force-free Motion of a Rigid Asymmetric Top
4.7: Nonrigid Rotator
4.8: Force-free Motion of a Rigid Symmetric Top
4.9: Centrifugal and Coriolis Forces
4.10: The Top
4.11: Appendix

5: Collisions
5.1: Introduction
5.2: Bouncing Balls
5.3: Head-on Collision of a Moving Sphere with an Initially Stationary Sphere
5.4: Oblique Collisions
5.5: Oblique (Glancing) Elastic Collisions, Alternative Treatment

6: Motion in a Resisting Medium

6.1: Introduction
6.2: Uniformly Accelerated Motion
6.3: Uniformly Accelerated Motion
6.3A: Resistive Force Only
6.3B: Body falling under gravity in a resisting medium, resistive force proportional to the speed
6.3C: Body thrown vertically upwards, initial speed \(v_{0}\)
6.4: Motion in which the Resistance is Proportional to the Square of the Speed
6.4A: Resistive Force Only

7: Projectiles
7.1: No Air Resistance
7.2: Air Resistance Proportional to the Speed
7.3: Air Resistance Proportional to the Square of the Speed

8: Impulsive Forces
8.1: Introduction
8.2: Problem

9: Conservative Forces
9.1: Introduction
9.2: The Time and Energy Equation
9.3: Virtual Work
9.E: Conservative Forces (Exercises)

2 https://phys.libretexts.org/@go/page/19041
10: Rocket Motion
10.1: Introduction
10.2: An Integral
10.3: The Rocket Equation
10.E: Rocket Motion (Exercises)

11: Simple and Damped Oscillatory Motion

11.1: Simple Harmonic Motion
11.2: Mass Attached to an Elastic Spring
11.3: Torsion Pendulum
11.4: Ordinary Homogeneous Second-order Differential Equations
11.5: Damped Oscillatory Motion
11.5i: Light damping- \( \gamma < 2\omega_{0}\)
11.5ii: Heavy damping- \( \gamma > 2\omega_{0}\)
11.5iii: Critical damping- \( \gamma = 2\omega_{0}\)
11.6: Electrical Analogues

12: Forced Oscillations

12.1: More on Differential Equations
12.2: Forced Oscillatory Motion
12.3: Electrical Analogue

13: Lagrangian Mechanics

13.1: Introduction to Lagrangian Mechanics
13.2: Generalized Coordinates and Generalized Forces
13.3: Holonomic Constraints
13.4: The Lagrangian Equations of Motion
13.5: Acceleration Components
13.6: Slithering Soap in Conical Basin
13.7: Slithering Soap in Hemispherical Basin
13.8: More Lagrangian Mechanics Examples
13.9: Hamilton's Variational Principle

14: Hamiltonian Mechanics

14.1: Introduction to Hamiltonian Mechanics
14.2: A Thermodynamics Analogy
14.3: Hamilton's Equations of Motion
14.4: Hamiltonian Mechanics Examples
14.5: Poisson Brackets

15: Special Relativity

15.1: Introduction to Special Relativity
15.2: Preparation
15.3: Preparation
15.4: Speed is Relative - The Fundamental Postulate of Special Relativity
15.5: The Lorentz Transformations
15.6: But This De es Common Sense
15.7: The Lorentz Transformation as a Rotation
15.8: Timelike and Spacelike 4-Vectors

3 https://phys.libretexts.org/@go/page/19041
 15.9: The FitzGerald-Lorentz Contraction
15.10: Time Dilation
15.11: The Twins Paradox
15.12: A, B and C
15.13: Simultaneity
15.14: Order of Events, Causality and the Transmission of Information
15.15: Derivatives
15.16: Addition of Velocities
15.17: Aberration of Light
15.18: Doppler Effect
15.19: The Transverse and Oblique Doppler Effects
15.20: Acceleration
15.21: Mass
15.22: Momentum
15.23: Some Mathematical Results
15.24: Kinetic Energy
15.25: Addition of Kinetic Energies
15.26: Energy and Mass
15.27: Energy and Momentum
15.28: Units
15.29: Force
15.30: The Speed of Light
15.31: Electromagnetism

16: Hydrostatics
16.1: Introduction to Hydrostatics
16.2: Density
16.3: Pressure
16.4: Pressure on a Horizontal Surface. Pressure at Depth
16.5: Pressure on a Vertical Surface
16.6: Centre of Pressure
16.7: Archimedes' Principle
16.8: Some Simple Examples
16.9: Floating Bodies

17: Vibrating Systems

17.1: Introduction
17.2: The Diatomic Molecule
17.3: Two Masses, Two Springs and a Brick Wall
17.4: Double Torsion Pendulum
17.5: Double Pendulum
17.6: Linear Triatomic Molecule
17.7: Two Masses, Three Springs, Two brick Walls
17.8: Transverse Oscillations of Masses on a Taut String
17.9: Vibrating String
17.10: Water
17.11: A General Vibrating System
17.12: A Driven System
17.13: A Damped Driven System

4 https://phys.libretexts.org/@go/page/19041
18: The Catenary
18.1: Introduction
18.2: The Intrinsic Equation to the Catenary
18.3: Equation of the Catenary in Rectangular Coordinates, and Other Simple Relations
18.4: Area of a Catenoid

19: The Cycloid

19.1: Introduction to Cycloids
19.2: Tangent to the Cycloid
19.3: The Intrinsic Equation to the Cycloid
19.4: Variations
19.5: Motion on a Cycloid, Cusps Up
19.6: Motion on a Cycloid, Cusps Down
19.7: The Brachystochrone Property of the Cycloid
19.8: Contracted and Extended Cycloids
19.9: The Cycloidal Pendulum
19.10: Examples of Cycloidal Motion in Physics

20: Miscellaneous
20.1: Introduction
20.2: Surface Tension
20.2.1: Excess Pressure Inside Drops and Bubbles
20.2.2: Angle of Contact
20.2.3: Capillary Rise
20.3: Shear Modulus and Torsion Constant
20.4: Viscosity
20.4.1: Poiseuille's Law
20.4.2: The Couette Viscometer

21: Central Forces and Equivalent Potential

21.1: Introduction to Central Forces
21.2: Motion Under a Central Force
21.3: Inverse Square Attractive Force
21.4: Hooke’s Law
21.5: Inverse Fourth Power Attractive Force
21.6: A General Central Force
21.7: Inverse Cube Attractive Force

22: Dimensions
22.1: Mass, Length and Time
22.2: Table of Dimensions
22.3: Checking Equations
22.4: Deducing Relationships
22.5: Dimensionless Quantities
22.6: Different Fundamental Quantities
22.7: Appendix A
22.8: Appendix B

5 https://phys.libretexts.org/@go/page/19041
Index
Detailed Licensing

6 https://phys.libretexts.org/@go/page/19041
Licensing
A detailed breakdown of this resource's licensing can be found in Back Matter/Detailed Licensing.

1 https://phys.libretexts.org/@go/page/65341
 CHAPTER OVERVIEW
1: Centers of Mass
Topic hierarchy
1.1: Introduction and Some Definitions
1.2: Plane Triangular Lamina
1.3: Plane Areas
1.4: Plane Curves
1.5: Summary of the Formulas for Plane Laminas and Curves
1.6: The Theorems of Pappus
1.7: Uniform Solid Tetrahedron, Pyramid and Cone
1.8: Hollow Cone
1.9: Hemispheres
1.S: Centers of Mass (Summary)

This page titled 1: Centers of Mass is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1
1.1: Introduction and Some Definitions
This chapter deals with the calculation of the positions of the centres of mass of various bodies. We start with a brief explanation of
the meaning of centre of mass, centre of gravity and centroid, and a very few brief sentences on their physical significance. Many
students will have seen the use of calculus in calculating the positions of centres of mass, and we do this for

Plane areas
i for which the equation is given in x − y coordinates;
ii for which the equation is given in polar coordinates.
Plane curves
i for which the equation is given in x − y coordinates;
ii for which the equation is given in polar coordinates.
Three-dimensional figures such as solid and hollow hemispheres and cones.
There are some figures for which interesting geometric derivations can be done without calculus; for example, triangular laminas,
and solid tetrahedra, pyramids and cones. And the theorems of Pappus allows you to find the centres of mass of semicircular
laminas and arcs in your head with no calculus.
First, some definitions.
Consider several point masses in the x − y plane:
m1 at (x 1, y1 )
m2 at ( x 2, y2 )
etc.
The centre of mass is a point (x̄, ȳ ) whose coordinates are defined by
∑ mi xi ∑ mi yi
¯¯
¯ ¯
¯¯
x = y = (1.1.1)
M M

where M is the total mass ∑ m . The sum ∑ m

i i xi is the first moment of mass with respect to the y axis. The sum ∑ m i yi is the
first moment of mass with respect to the x axis.
If the masses are distributed in three-dimensional space, with m at (x 1 1, y1 , z1 ) , etc,. the centre of mass is a point ( x, y, z ) such
¯¯
¯ ¯
¯¯ ¯
¯¯

that
∑ mi xi ∑ mi yi ∑ mi zi
¯¯ ¯¯ ¯¯
x̄ = ȳ = ȳ = (1.1.2)
M M M

In this case, ∑ mi xi , ∑ mi yi , ∑ mi zi are the first moments of mass with respect to the y − z, z − x and x −y planes
respectively.
In either case we can use vector notation and suppose that r , r , r are the position vectors of
1 2 3 m1 , m2 , m3 with respect to the
origin, and the centre of mass is a point whose position vector r̄ is defined by ¯
¯

∑ mi ri
¯
r̄ = (1.1.3)
M

In this case the sum is a vector sum and ∑ m r a vector quantity, is the first moment of mass with respect to the origin. Its scalar
i i

components in the two-dimensional case are the moments with respect to the axes; in the three dimensional case they are the
moments with respect to the planes.
Many early books, and some contemporary ones, use the term "centre of gravity". Strictly the centre of gravity is a point whose
position is defined by the ratio of the first moment of weight to the total weight. This will be identical to the centre of mass
provided that the strength of the gravitational field g (or gravitational acceleration) is the same throughout the space in which the
masses are situated. This is usually the case, though it need not necessarily be so in some contexts.

1.1.1 https://phys.libretexts.org/@go/page/6925
For a plane geometrical figure, the centroid or centre of area, is a point whose position is defined as the ratio of the first moment of
area to the total area. This will be the same as the position of the centre of mass of a plane lamina of the same size and shape
provided that the lamina is of uniform surface density.
Calculating the position of the centre of mass of various figures could be considered as merely a make-work mathematical exercise.
However, the centres of gravity, mass and area have important applications in the study of mechanics.
For example, most students at one time or another have done problems in static equilibrium, such as a ladder leaning against a wall.
They will have dutifully drawn vectors indicating the forces on the ladder at the ground and at the wall, and a vector indicating the
weight of the ladder. They will have drawn this as a single arrow at the centre of gravity of the ladder as if the entire weight of the
ladder could be "considered to act" at the centre of gravity. In what sense can we take this liberty and "consider all the weight as if
it were concentrated at the centre of gravity"? In fact the ladder consists of many point masses (atoms) all along its length. One of
the equilibrium conditions is that there is no net torque on the ladder. The definition of the centre of gravity is such that the sum of
the moments of the weights of all the atoms about the base of the ladder is equal to the total weight times the horizontal distance to
the centre of gravity, and it is in that sense that all the weight "can be considered to act" there. Incidentally, in this example, "centre
of gravity" is the correct term to use. The distinction would be important if the ladder were in a nonuniform gravitational field.
In dynamics, the total linear momentum of a system of particles is equal to the total mass times the velocity of the centre of mass.
This may be "obvious", but it requires formal proof, albeit one that follows very quickly from the definition of the centre of mass.
Likewise the kinetic energy of a rigid body in two dimensions equals M V + I ω where M is the total mass, V the speed of
1

2
2 1

2
2

the centre of mass, I the rotational inertia and ω the angular speed, both around the centre of mass. Again it requires formal proof,
but in any case it furnishes us with another example to show that the calculation of the positions of centres of mass is more than
merely a make-work mathematical exercise and that it has some physical significance.
If a vertical surface is immersed under water (e.g. a dam wall) it can be shown that the total hydrostatic force on the vertical surface
is equal to the area times the pressure at the centroid. This requires proof (readily deduced from the definition of the centroid and
elementary hydrostatic principles), but it is another example of a physical application of knowing the position of the centroid.

This page titled 1.1: Introduction and Some Definitions is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

1.1.2 https://phys.libretexts.org/@go/page/6925
1.2: Plane Triangular Lamina
Definition: A median of a triangle is a line from a vertex to the midpoint of the opposite side.
Theorem I. The three medians of a triangle are concurrent (meet at a single, unique point) at a point that is two-thirds of the
distance from a vertex to the midpoint of the opposite side.
Theorem II. The centre of mass of a uniform triangular lamina (or the centroid of a triangle) is at the meet of the medians.
The proof of I can be done with a nice vector argument (Figure I.1):
Let A , B be the vectors OA , OB . Then A+B is the diagonal of the parallelogram of which OA and OB are two sides, and the
position vector of the point C is (A+B) .
1
1

To get C , we see that

2

2 2 2 1 1
C2 = A + (AM2 ) = A + (M2 − A) = A + ( B − A) = (A + B)
3 3 3 2 3

Thus the points C and C are identical, and the same would be true for the third median, so Theorem I is proved.
1 2

Now consider an elemental slice as in Figure I.2. The centre of mass of the slice is at its mid-point. The same is true of any similar
slices parallel to it. Therefore the centre of mass is on the locus of the mid-points - i.e. on a median. Similarly, it is on each of the
other medians, and Theorem II is proved.

That needed only some vector geometry. We now move on to some calculus.

This page titled 1.2: Plane Triangular Lamina is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1.2.1 https://phys.libretexts.org/@go/page/6926
1.3: Plane Areas
Plane areas in which the equation is given in x − y coordinates

We have a curve y = y(x) (Figure I.3) and we wish to find the position of the centroid of the area under the curve between x = a
and x = b . We consider an elemental slice of width δx at a distance x from the y axis. Its area is yδx, and so the total area is
b

A =∫ ydx (1.3.1)
a

b
The first moment of area of the slice with respect to the y axis is xyδx, and so the first moment of the entire area is ∫ a
.
xydx

Therefore
b b
∫ xydyx ∫ xydyx
a a
¯¯
x̄ = = (1.3.2)
b
A
∫ ydyx
a

labeleq : 1.3.2

For ȳ we notice that the distance of the centroid of the slice from the x axis is
¯
¯ 1

2
y , and therefore the first moment of the area about
the x axis is y. yδx .
1

Therefore
b 2
∫ y dx
a
¯¯
ȳ = (1.3.3)
2A

 Example 1.3.1
Consider a semicircular lamina, x 2
+y
2
=a
2
, see Figure I.4:

a
We are dealing with the parts both above and below the \(x \) axis, so the area of the semicircle is 2∫
0
ydx and the first
a
moment of area is 2 ∫ xydx.
0

You should find x = 4a/(3π) = 0.4244a.

¯¯
¯

Now consider the lamina x 2

+y
2
=a
2
, y > 0 (Figure I.5):

1.3.1 https://phys.libretexts.org/@go/page/6927
The area of the elemental slice this time is yδx (not 2yδx ), and the integration limits are from −a to +a . To find y , use Equation ¯
¯¯

1.3.3, and you should get y = 0.4244a .

Plane areas in which the equation is given in polar coordinates.

We consider an elemental triangular sector (Figure I.6) between θ and θ + δθ . The "height" of the triangle is r and the "base" is
rδθ. The area of the triangle is r δθ.
1 2

Therefore the whole area =

β
1 2
∫ r dθ (1.3.4)
2 α

The horizontal distance of the centroid of the elemental sector from the origin (more correctly, from the "pole" of the polar
coordinate system) is r cos θ . The first moment of area of the sector with respect to the y axis is
2

2 1 2 1 3
r cos θ × r δθ = r cos θδθ
3 2 3

so the first moment of area of the entire figure between θ = α and θ = β is

1 β
3
∫ r cos θdθ
3 α

Therefore
β 3
2∫ r cos θdθ
α
¯¯
x̄ = (1.3.5)
β
2
3∫ r dθ
α

Similarly
β 3
2∫ r sin θdθ
α
¯¯
x̄ = (1.3.6)
β
2
3∫ r dθ
α

 Example 1.3.2
−π +π
Consider the semicircle r = a , θ = 2
to 2

+π/2
2a ∫ cos θdθ +π/2
−π/2 2a 4a
¯¯
¯
x = = ∫ cos θdθ = (1.3.7)
+π/2 3π 3π
3∫ dθ −π/2
−π/2

The reader should now try to find the position of the centroid of a circular sector (slice of pizza!) of angle 2α . The integration
limits will be −α to +α .

1.3.2 https://phys.libretexts.org/@go/page/6927
 When you arrive at a formula (which you should keep in a notebook for future reference), check that it goes to 4α

3π
if α =
π

2
,
and to if α = 0 .
2π

This page titled 1.3: Plane Areas is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1.3.3 https://phys.libretexts.org/@go/page/6927
1.4: Plane Curves
Plane Curves Expressed in x − y coordinates

Figure I.7 shows how an elemental length δs is related to the corresponding increments in x and y :
−−−−−−−−− −−−−−−−−−
2 2
−−−−−−−−
dy dx
2 2
δs = √ δx + δy = √1 + ( ) δx = √ ( ) + 1 dy (1.4.1)
dx dy

Consider a wire of mass per unit length (linear density) λ bent into the shape y = y(x) between x = a and x = b . The mass of an
element ds is λδs, so the total mass is
−−−−−−−−−
b 2
dy
∫ λ ds = ∫ λ√ 1 + ( ) dx (1.4.2)
a
dx

The first moments of mass about the y - and x -axes are respectively
−−−−−−−−−
b 2
dy
∫ λx √ 1 + ( ) dx (1.4.3)
a
dx

and
−−−−−−−−−
b 2
dy
∫ λy √ 1 + ( ) dx (1.4.4)
a
dx

If the wire is uniform and λ is therefore not a function of x or y , λ can come outside the integral signs in Equations 1.4.2 - 1.4.4,
and we hence obtain
−−−−−−−−−
b 2
dy
∫ x√ 1 + ( ) dx
a
dx
¯¯
¯
x = −−−−−−−−− (1.4.5)
b 2
dy
∫ √1 + ( ) dx
a
dx

and
−−−−−−−−−
b 2
dy
∫ y√ 1 + ( ) dx
a dx
¯
¯¯
y = −−−−−−−−− (1.4.6)
b 2
dy
∫ √1 + ( ) dx
a
dx

the denominator in each of these expressions merely being the total length of the wire.

1.4.1 https://phys.libretexts.org/@go/page/6928
  Example 1.4.1

Consider a uniform wire bent into the shape of the semicircle x 2

+y
2
=a
2
, x > 0.
First, it might be noted that one would expect x > 0.4244a (the value for a plane semicircular lamina).
¯¯
¯

The length (i.e. the denominators in Equations 1.4.5 and 1.4.6) is just πa. Since there are, between x and x + δx , two
elemental lengths to account for, one above and one below the x axis, the numerator of Equation 1.4.5 must be
−−−−−−−−−
a 2
dy
2∫ x√ 1 + ( ) dx
0 dx

In this case
−−−−−−
2 2
y = √a − x

and
dy −x
= − −−−− −
dx √ a2 − x2

The first moment of length of the entire semicircle is

−−−−−−−−−−
a 2 a
x x dx
¯¯
x̄ =2∫ x√ 1 + dx = 2a ∫ − −−−− −
2 2
0 a −x 0 √ a2 − x2

2a
From this point the student is left to his or her own devices to solve this integral and derive x = ¯¯
¯
= 0.6366a .
π

Plane Curves Expressed in Polar Coordinates

Figure I.8 shows how an elemental length δs is related to the corresponding increments in r and θ :
−−−−−−−−−− −−−−−−−−−−
2 2
−−−−−−−−−−− dr dθ
2 2 2
δs = √ (δr) + (rδθ) = √( ) +r δθ = √ 1 + (r ) δr. (1.4.7)
dθ dr

The mass of the curve (between θ = a and θ = b ) is

−−−−−−−−−−
β 2
dr
2
∫ λ√ ( ) +r dθ.
α
dθ

The first moments about the y - and x -axes are (recalling that x = r cos θ and y = r sin θ )
−−−−−−−−−−
β 2
dr 2
∫ λr cos θ√ ( ) +r dθ
α
dθ

and

1.4.2 https://phys.libretexts.org/@go/page/6928
 −−−−−−−−−−
β 2
dr
2
∫ λr sin θ√ ( ) +r dθ.
α
dθ

If λ is not a function of r or θ , we obtain

−−−−−−−−−−
β 2
1 dr 2
¯¯
¯
x = ∫ r cos θ√ ( ) +r dθ (1.4.8)
L α
dθ

and
−−−−−−−−−−
β 2
1 dr 2
¯
¯¯
y = ∫ r sin θ√ ( ) +r dθ (1.4.9)
L α
dθ

where L is the length of the wire.

 Example 1.4.2

Again consider the uniform wire of Figure I.8 bent into the shape of a semicircle. The equation in polar coordinates is simply
−π +π
r =a , and the integration limits are θ = to θ = and the length is πa.
2 2

Thus
+π/2
1 1 2a
¯¯ 2
x̄ = ∫ acosθ[0 − a ] 2 dθ = .
πa −π/2
π

The reader should now find the position of the center of mass of a wire bent into the arc of a circle of angle 2α. The expression
2a π
obtained should go to as α goes to , and to a as α goes to zero.
π 2

This page titled 1.4: Plane Curves is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1.4.3 https://phys.libretexts.org/@go/page/6928
1.5: Summary of the Formulas for Plane Laminas and Curves

Uniform Plane Lamina

y = y(x) r = r(θ)

β 3
2∫ r cosθdθ
α
b ¯¯
¯
¯¯
¯ 1 x =
x = ∫ xydx β
A a 3∫ r2 dθ
α
b β 3
¯
¯¯ 1 2 2∫ r sinθdθ
y = ∫ y dx ¯
¯¯ α
2A a y =
β
3∫ r2 dθ
α

Uniform Plane Curve

y = y(x) r = r(θ)

1 1
1 b dy 2 1 β dr 2 2
¯¯
¯ ¯¯
¯
x = ∫ x[1 + ( ) ] 2 x = ∫ rcosθ[( ) +r ] 2
L a dx L α dθ
1 1
1 b dy 2 1 β dr 2 2
¯
¯¯ ¯
¯¯
y = ∫ y[1 + ( ) ] 2 y = ∫ rsinθ[( ) +r ] 2
L a dx L α dθ

This page titled 1.5: Summary of the Formulas for Plane Laminas and Curves is shared under a CC BY-NC 4.0 license and was authored,
remixed, and/or curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit
history is available upon request.

1.5.1 https://phys.libretexts.org/@go/page/6929
1.6: The Theorems of Pappus
(Pappus Alexandrinus, Greek mathematician, approximately 3rd or 4th century AD.)
I. If a plane area is rotated about an axis in its plane, but which does not cross the area, the volume swept out equals the area times
the distance moved by the centroid.
II. If a plane curve is rotated about an axis in its plane, but which does not cross the curve, the area swept out equals the length
times the distance moved by the centroid.
These theorems enable us to work out the volume of a solid of revolution if we know the position of the centroid of a plane area, or
vice versa; or to work out the area of a surface of revolution if we know the position of the centroid of a plane curve or vice versa.
It is not necessary that the plane or the curve be rotated through a full 360o.
We prove the theorems first. We then follow with some examples.

Consider an area A in the zx plane (Figure I.9), and an element δA within the area at a distance x from the z axis. Rotate the area
through an angle ϕ about the z axis. The length of the arc traced by the element dA in moving through an angle ϕ is xϕ , so the
volume swept out by δA is xϕδA. The volume swept out by the entire area is ϕ ∫ xdA . But the definition of the centroid of A is
such that its distance from the z axis is given by xA = ∫ xdA . Therefore the volume swept out by the area is ϕxA. But ϕx is the
¯¯
¯ ¯¯
¯ ¯¯
¯

distance moved by the centroid, so the first theorem of Pappus is proved.

1.6.1 https://phys.libretexts.org/@go/page/6930
Consider a curve of length L in the zx plane (Figure I.10), and an element δs of the curve at a distance x from the z axis. Rotate
the curve through an angle ϕ about the z axis. The length of the arc traced by the element δsin moving through an angle ϕ is xϕ,
so the area swept out by δs is xϕδs. The area swept out by the entire curve is ϕ ∫ xds . But the definition of the centroid is such
that its distance from the z axis is given by x̄L = ∫ xds . Therefore the area swept out by the curve is ϕx̄L. But ϕx̄ is the distance
¯
¯ ¯
¯ ¯
¯

moved by the centroid, so the second theorem of Pappus is proved.

Applications of the Theorems of Pappus

Rotate a plane semicircular figure of area π a through 360o about its diameter. The volume swept out is
1

2
2 4

3
3
πa , and the distance
moved by the centroid is 2πx Therefore by the theorem of Pappus, x =
¯¯
¯ ¯¯
¯
.
4a

(3π)

o
Rotate a plane semicircular arc of length π a through 360 about its diameter. Use a similar argument to show that x = ¯¯
¯ 2a

π
.

Consider a right-angled triangle, height h , base a (Figure I.11). Its centroid is at a distance from the height h . The area of the
a

triangle is . Rotate the triangle through 360o about h . The distance moved by the centroid is
ah

2
. The volume of the cone swept
2πa

3
2

out is ah

2
times 2π

3
, equals πa h

3
.
Now consider a line of length l inclined at an angle α to the y axis (Figure I.12). Its centroid is at a distance l sin α from the y
1

axis. Rotate the line through 360o about the y axis. The distance moved by the centroid is 2π × l sin α = πl sin α . The surface
1

area of the cone swept out is l × π l sin α = π l sin α .2

The centre of a circle of radius b is at a distance a from the y axis. It is rotated through 360o about the y axis to form a torus
(Figure I.13). Use the theorems of Pappus to show that the volume and surface area of the torus are, respectively, 2π ab and 2 2

4 π ab .
2

1.6.2 https://phys.libretexts.org/@go/page/6930
This page titled 1.6: The Theorems of Pappus is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1.6.3 https://phys.libretexts.org/@go/page/6930
1.7: Uniform Solid Tetrahedron, Pyramid and Cone
 Definition

A median of a tetrahedron is a line from a vertex to the centroid of the opposite face.

 Theorem I.

The four medians of a tetrahedron are concurrent at a point 3/4 of the way from a vertex to the centroid of the opposite face.

 Theorem II

The centre of mass of a uniform solid tetrahedron is at the meet of the medians.

Theorem I can be derived by a similar vector geometric argument used for the plane triangle. It is slightly more challenging than
for the plane triangle, and it is left as an exercise for the reader. I draw two diagrams (Figure I.14). One shows the point C that is
1

3/4 of the way from the vertex A to the centroid of the opposite face. The other shows the point C that is 3/4 of the way from the
2

vertex B to the centroid of its opposite face. You should be able to show that
C1 = (A+B+D)/4

In fact this suffices to prove Theorem I, because, from the symmetry between A, B and D, one is bound to arrive at the same
expression for the three-quarter way mark on any of the four medians. But for reassurance you should try to show, from the second
figure, that
\( {\bf \text{C}_{2}={\bf\text{A+B+D}})/{4}\)
The argument for Theorem II is easy, and is similar to the corresponding argument for plane triangles.

Pyramid.
A right pyramid whose base is a regular polygon (for example, a square) can be considered to be made up of several tetrahedra
stuck together. Therefore the centre of mass is 3/4 of the way from the vertex to the mid point of the base.

Cone.
A right circular cone is just a special case of a regular pyramid in which the base is a polygon with an infinite number of
infinitesimal sides. Therefore the centre of mass of a uniform right circular cone is 3/4 of the way from the vertex to the centre of
the base.
We can also find the position of the centre of mass of a solid right circular cone by calculus. We can find its volume by calculus,
too, but we'll suppose that we already know, from the theorem of Pappus, that the volume is × base × height.
1

1.7.1 https://phys.libretexts.org/@go/page/8346
Consider the cone in Figure I.15, generated by rotating the line y = ax

h
(between x = 0 and x = h ) through 360o about the x axis.
2 2

The radius of the elemental slice of thickness dx at x is ax

h
. Its volume is πa x δx
2
.
h

Since the volume of the entire cone is πa h

3
, the mass of the slice is
2 2 2 2
πa x δx πa h 3M x δx
M × 2
÷ = 3
h 3 h

where M is the total mass of the cone. The first moment of mass of the elemental slice with respect to the y axis is 3M x dx
3
.
h

The position of the centre of mass is therefore

3 h 3 3
¯¯
¯
x = ∫ x dx = h
3 0 4
h

This page titled 1.7: Uniform Solid Tetrahedron, Pyramid and Cone is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or
curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is
available upon request.

1.7.2 https://phys.libretexts.org/@go/page/8346
1.8: Hollow Cone
The surface of a hollow cone can be considered to be made up of an infinite number of infinitesimally slender isosceles triangles,
and therefore the centre of mass of a hollow cone (without base) is 2/3 of the way from the vertex to the midpoint of the base.

This page titled 1.8: Hollow Cone is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1.8.1 https://phys.libretexts.org/@go/page/8347
1.9: Hemispheres
Uniform solid hemisphere
Figure I.4 will serve. The argument is exactly the same as for the cone. The volume of the elemental slice is
3

π y δx = π(a − x )δx and the volume of the hemisphere is , so the mass of the slice is
2 2 2 2πa

2 2
3M( a −x )δx
2 2
M × π(a − x )δx ÷ (2πa/3) = 3
2a

where M is the mass of the hemisphere. The first moment of mass of the elemental slice is x times this, so the position of the
centre of mass is
3 a 2 2 3a
¯¯
¯
x = ∫ x(a − x )dx =
2a
3 0 8

Hollow hemispherical shell.

We may note to begin with that we would expect the centre of mass to be further from the base than for a uniform solid
hemisphere.
Again, Figure I.4 will serve. The area of the elemental annulus is 2πaδx (NOT 2πyδx!) and the area of the hemisphere is 2πa
2
.
Therefore the mass of the elemental annulus is
2
M × 2πaδx ÷ (2π a ) = M δx/a

The first moment of mass of the annulus is x times this, so the position of the centre of mass is
a xdx a
¯¯
x̄ =∫ =
0 a 2

This page titled 1.9: Hemispheres is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1.9.1 https://phys.libretexts.org/@go/page/8348
1.S: Centers of Mass (Summary)
SUMMARY
Triangular lamina: 2/3 of way from vertex to midpoint of opposite side
Solid Tetrahedron, Pyramid, Cone: 3/4 of way from vertex to centroid of opposite face.
Hollow cone: 2/3 of way from vertex to midpoint of base.
Semicircular lamina: 4a

3π

(2a sin α)
Lamina in form of a sector of a circle, angle 2α : (3α)

Semicircular wire: 2a

(a sin α)
Wire in form of an arc of a circle, angle 2α : α

Solid hemisphere: 3a

Hollow hemisphere: a

This page titled 1.S: Centers of Mass (Summary) is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1.S.1 https://phys.libretexts.org/@go/page/8349
 CHAPTER OVERVIEW
2: Moments of Inertia
In this chapter we shall consider how to calculate the (second) moment of inertia for different sizes and shapes of body, as well as
certain associated theorems. But the question should be asked: "What is the purpose of calculating the squares of the distances of
lots of particles from an axis, multiplying these squares by the mass of each, and adding them all together?
2.1: Definition of Moment of Inertia
2.2: Meaning of Rotational Inertia
2.3: Moments of Inertia of Some Simple Shapes
2.4: Radius of Gyration
2.5: Plane Laminas and Mass Points distributed in a Plane
2.6: Three-dimensional Solid Figures. Spheres, Cylinders, Cones.
2.7: Three-dimensional Hollow Figures. Spheres, Cylinders, Cones
2.8: Torus
2.9: Linear Triatomic Molecule
2.10: Pendulums
2.11: Plane Laminas. Product Moment. Translation of Axes (Parallel Axes Theorem)
2.12: Rotation of Axes
2.13: Momental Ellipse
2.14: Eigenvectors and Eigenvalues
2.15: Solid Body
2.16: Rotation of Axes - Three Dimensions
2.17: Solid Body Rotation and the Inertia Tensor
2.18: Determination of the Principal Axes
2.19: Moment of Inertia with Respect to a Point
2.20: Ellipses and Ellipsoids
2.21: Tetrahedra

This page titled 2: Moments of Inertia is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1
2.1: Definition of Moment of Inertia
Consider a straight line (the "axis") and a set of point masses m , m , m ... such that the distance of the mass m from the axis is
1 2 3 i

r . The quantity m r is the second moment of the i th mass with respect to (or "about") the axis, and the sum ∑ m r is the
2 2
i i i i i

second moment of mass of all the masses with respect to the axis.
Apart from some subtleties encountered in general relativity, the word "inertia" is synonymous with mass - the inertia of a body is
merely the ratio of an applied force to the resulting acceleration. Thus ∑ m r can also be called the second moment of inertia.
i
2
i

The second moment of inertia is discussed so much in mechanics that it is usually referred to as just "the" moment of inertia.
In this chapter we shall consider how to calculate the (second) moment of inertia for different sizes and shapes of body, as well as
certain associated theorems. But the question should be asked: "What is the purpose of calculating the squares of the distances of
lots of particles from an axis, multiplying these squares by the mass of each, and adding them all together? Is this merely a
pointless make-work exercise in arithmetic? Might one just as well, for all the good it does, calculate the sum ∑ m r ? Does i
2
i

∑ r m have any physical significance?"

2
i i

This page titled 2.1: Definition of Moment of Inertia is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

2.1.1 https://phys.libretexts.org/@go/page/6932
2.2: Meaning of Rotational Inertia
If a force acts of a body, the body will accelerate. The ratio of the applied force to the resulting acceleration is the inertia (or mass)
of the body.
If a torque acts on a body that can rotate freely about some axis, the body will undergo an angular acceleration. The ratio of the
applied torque to the resulting angular acceleration is the rotational inertia of the body. It depends not only on the mass of the
body, but also on how that mass is distributed with respect to the axis.
Consider the system shown in Figure II.1.

A particle of mass m is attached by a light (i.e. zero or negligible mass) arm of length r to a point at O, about which it can freely
rotate. A force F is applied, and the mass consequently undergoes a linear acceleration a = . The angular acceleration is then
F

Θ = F . mr . Also, the torque is τ = F r . The ratio of the applied torque to the angular acceleration is therefore mr . Thus the
¨ 2

rotational inertia is the second moment of inertia. Rotational inertia and (second) moment of inertia are one and the same thing,
except that rotational inertia is a physical concept and moment of inertia is its mathematical representation.

This page titled 2.2: Meaning of Rotational Inertia is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

2.2.1 https://phys.libretexts.org/@go/page/6933
2.3: Moments of Inertia of Some Simple Shapes
A student may well ask: "For how many different shapes of body must I commit to memory the formulas for their moments of
inertia?" I would be tempted to say: "None". However, if any are to be committed to memory, I would suggest that the list to be
memorized should be limited to those few bodies that are likely to be encountered very often (particularly if they can be used to
determine quickly the moments of inertia of other bodies) and for which it is easier to remember the formulas than to derive them.
With that in mind I would recommend learning no more than five. In the following, each body is supposed to be of mass m and
rotational inertia I .

 Formula 1.
A rod of length 2l about an axis through the middle, and at right angles to the rod:
1 2
I = ml (2.3.1)
3

 Formula 2.

A uniform circular disc of radius a about an axis through the center and perpendicular to the plane of the disc:
1
2
I = ma (2.3.2)
2

 Formula 3.
A uniform right-angled triangular lamina about one of its shorter sides - i.e. not the hypotenuse. The other not-hypotenuse side
is of length a :
1 2
I = ma (2.3.3)
6

 Formula 4.

A uniform solid sphere of radius a about an axis through the center.

2
2
I = ma (2.3.4)
5

 Formula 5.
A uniform spherical shell of radius a about an axis through the center.
2 2
I = ma (2.3.5)
3

I shall now derive the first three of these by calculus. The derivations for the spheres will be left until later.
1.Rod, length 2l (Figure II.2)

mδx
The mass of an element δx at a distance x from the middle of the rod is .
2l

and its second moment of inertia is

2.3.1 https://phys.libretexts.org/@go/page/6934
 2
m x δx
.
2l

l l
m 2
m 2
1 2
∫ x dx = ∫ x dx = ml .
2l −l l 0 3

2. Disc, radius a . (Figure II.3)

The area of an elemental annulus, radii r is r + δr is 2πrδr.

The area of the entire disc is πa . 2

Therefore the mass of the annulus is

2πrδrm 2mrδr
= .
2 2
πa a

and its second moment of inertia is

3
2m r δr
.
2
a

The moment of inertia of the entire disc is

a
2m 3
1 2
∫ r dr = ma .
2
a 0
2

3. Right-angled triangular lamina. (Figure II.4)

The equation to the hypotenuse is y = b(1 − x/a) .

ab
The area of the elemental strip is yδx = b(1 − x/a)δx and the area of the entire triangle is .
2

2m(a − x)δx
Therefore the mass of the elemental strip is 2
.
a

2
2m x (a − x)δx
and its second moment of inertia is 2
.
a

2.3.2 https://phys.libretexts.org/@go/page/6934
 2
ma
The second moment of inertia of the entire triangle is the integral of this from x = 0 to x = a , which is .
6

Uniform circular lamina about a diameter.

For the sake of one more bit of integration practice, we shall now use the same argument to show that the moment of inertia of a
2
ma
uniform circular disc about a diameter is . However, we shall see later that it is not necessary to resort to integral calculus to
4
arrive at this result, nor is it necessary to commit the result to memory. In a little while it will become immediately apparent and
2
ma
patently obvious, with no calculation, that the moment of inertia must be . However, for the time being, let us have some
4
more calculus practice. See Figure II.5.

The disc is of radius a , and the area of the elemental strip is 2yδx. But y and x are related through the equation to the circle, which
is y = (a − x ) . Therefore the area of the strip is 2(a − x ) δx . The area of the whole disc is πa , so the mass of the strip
2 2 1/2 2 x 1/2 2

is
2 2 1/2
2(a −x ) δx 2m
2 2 1/2
m× = × (a −x ) δx.
2 2
πa πa

The second moment of inertia about the y -axis is

2m
2 2 2 1/2
× x (a −x ) δx.
2
πa

For the entire disc, we integrate from x = −a to x = +a , or, if you prefer, from x =0 to x =a and then double it. The result
2
ma
should follow. If you need a hint about how to do the integration, let x = acosθ (which it is, anyway), and be sure to get the
4
limits of integration with respect to θ right.
2
ma
The moment of inertia of a uniform semicircular lamina of mass m and radius a about its base, or diameter, is also , since the
4
2
ma
mass distribution with respect to rotation about the diameter is the same. , since the mass distribution with respect to rotation
4
about the diameter is the same.

This page titled 2.3: Moments of Inertia of Some Simple Shapes is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or
curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is
available upon request.

2.3.3 https://phys.libretexts.org/@go/page/6934
2.4: Radius of Gyration
The second moment of inertia of any body can be written in the form mk . Thus, for the rod, the disc (about an axis perpendicular
2

to its plane), the triangle and the disc (about a diameter), k has the values
l
– = 0.866l ,
√3
a
–
= 0.707a ,
√2
a
– = 0.408a , and
√6
a
= 0.500a
2

respectively.
kis called the radius of gyration. If you were to concentrate all the mass of a body at its radius of gyration, its moment of inertia
would remain the same.

This page titled 2.4: Radius of Gyration is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

2.4.1 https://phys.libretexts.org/@go/page/6935
2.5: Plane Laminas and Mass Points distributed in a Plane
In Figure II.6a, the two unbroken lines represent two fixed coordinate axes. I have drawn several point masses m1 , m2 , m3

distributed in a plane.

The x-coordinate of mass m is x . The dashed line is moveable, and it x-coordinate is x, so that the distance of
i i mi this line is
x − x The moment of inertia of the system of masses about the dashed line is
i

2 2 2
 I = m1 (x1 − x ) + m2 (x2 − x ) + m3 (x3 − x ) +. . . . (2.5.1)

Now imagine what happens if the dashed line is moved to the right. The moment of inertia decreases – and decreases - and
decreases. But eventually the line finds itself to the right of m , and then of m , and then of m . After that is by no means obvious
4 5 6

that the moment of inertia is going to continue to decrease. Indeed, by this time it is clear that at some point I is going to go
through a minimum and then start to increase again as more and more of the masses find themselves to the left of the dashed line.
Just where is the dashed line when the moment of inertia is a minimum? I’ll leave you to differentiate Equation 2.5.1 with respect
to x, and hence show that I is least when
m1 x1 + m2 x2 + m3 x3 +. . .
x = (2.5.2)
m1 + m2 + m3 +. . .

That is, the moment of inertia is least when x̄ = x . That is, the moment of inertia is least for an axis passing through the centre of
¯
¯

mass.
In Figure II.6b, the line CC passes through the centre of mass; the moment of inertia is least about this line. The line AA is at a
distance x from CC, and the moment of inertia is greater about AA than about CC. The Parallel Axes Theorem tells us by how
¯¯
¯

much.

Let us measure distances from CC, so that the distance of m from CC is x and the distance of m from AA is x
i i i i
¯¯
+ x̄ .

2.5.1 https://phys.libretexts.org/@go/page/6936
It is clear that I CC = ∑ mi x
2
i

and that
2 2 2
¯¯ ¯¯ ¯¯
IAA = ∑ mi (xi + x̄) = ∑ mi x + 2 x̄ ∑ mi xi + x̄ ∑ mi . (2.5.3)
i

The first term on the right hand side is I . The sum in the second term is the first moment of mass about the centre of mass, and is
CC

zero. The sum in the third term is the total mass. We therefore arrive at the Parallel Axes Theorem.
2
¯¯
¯
IAA = IC C + M x . (2.5.4)

In words, the moment of inertia about an arbitrary axis is equal to the moment of inertia about a parallel axis through the centre of
mass plus the total mass times the square of the distance between the parallel axes. The theorem holds also for masses distributed in
three-dimensional space.
The Perpendicular Axes Theorem, on the other hand, holds only for masses distributed in a plane, or for plane laminas.

Figure II.7 shows some point masses distributed in the xy plane, the z axis being perpendicular to the plane of the paper. The
moments of inertia about the x, y and z axes are denoted respectively by A, B and C . The distance of m from the z axis is
i
1

(x
2
i
2
+y )
i
2 Therefore the moment of inertia of the masses about the z axis is
2 2
C = ∑ mi (x +y ) (2.5.5)
i i

That is to say:

C = A+B (2.5.6)

This is the Perpendicular Axes Theorem. Note again very carefully that, unlike the parallel axes theorem, this theorem applies only
to plane laminas and to point masses distributed in a plane.

Examples of the Use of the Parallel and Perpendicular Axes Theorems.

From Section 2.3 we know the moments of inertia of discs, rods and triangular laminas. We can make use of the parallel and
perpendicular axes theorems to write down the moments of inertia of most of the following examples almost by sight, with no
calculus.
Hoop and discs, radius a .

Rods, length 2l.

2.5.2 https://phys.libretexts.org/@go/page/6936
Rectangular laminas, sides 2a and 2b; a > b .

Square laminas, side 2a.

Triangular laminas.

1 2 2 1 2 2 1 2 2
I = m a (1 + 3cot θ) = m b (3 + tan θ) = m c (3 − 2si n θ)
6 6 6

1 2 2 1 2 2 1 3 2
= m(2 b +c ) = m(3 c − 2a ) = m(a b )
6 6 6

This page titled 2.5: Plane Laminas and Mass Points distributed in a Plane is shared under a CC BY-NC 4.0 license and was authored, remixed,
and/or curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is
available upon request.

2.5.3 https://phys.libretexts.org/@go/page/6936
2.6: Three-dimensional Solid Figures. Spheres, Cylinders, Cones.
Sphere, mass m, radius a .

The volume of an elemental cylinder of radii x , x + δx , height 2y is 4πyxδx = 4π(a

2
−x )
2 1/2
xδx . Its mass is
1

2 2 1 1
4π( a −x ) 2 xδx
m× 4
3
=
3m

a
3
2
× (a
2
−x ) 2 xδx. It' second moment of intertia is =
3m
3
a
× (a
2
−x )
2
2
3
x δx. The second moment of
πa
3

inertia of the entire sphere is

1
3m a 2 2 3 2 2
= ×∫ (a −x ) 2
x δx = ma .
3
a 0 5

The moment of inertia of a uniform solid hemisphere of mass m and radius a about a diameter of its base is also , 2

5
ma
2
, because
the distribution of mass around the axis is the same as for a complete sphere.

 Exercise 2.6.1
A hollow sphere is of mass M , external radius a and internal radius xa. Its rotational inertia is 0.5M a . Show that x is given 2

by the solution of

3 5
1 − 5x + 4x =0

and calculate x to four significant figures.

(Answer = 0.6836.)

Solid cylinder, mass m, radius a , length 2l

The mass of an elemental disc of thickness δx is mδx

2l
. Its moment of inertia about its diameter is 1

4
mδx

2l
2
a =
ma δx

8l
. Its moment of
2 2
2 m( a +4 x )δx
inertia about the dashed axis through the centre of the cylinder is ma δx

8l
+
mδx

2l
2
x =
8l
. The moment of inertia of the
2 2
1 m( a +4 x )δx
entire cylinder about the dashed axis is 2 ∫ 0 8l
= m(
1

4
a
2
+
1

3
l )
2
.
In a similar manner it can be shown that the moment of inertia of a uniform solid triangular prism of mass m, length 2l, cross
section an equilateral triangle of side 2aabout an axis through its centre and perpendicular to its length is m( a + l ) . 1

6
2 1

3
2

2.6.1 https://phys.libretexts.org/@go/page/6937
Solid cone, mass m, height h, base radius a .

The mass of elemental disc of thickness δx is

2 2
πy δx 3my δx
m×
1 2
=
a2 h
.
πa h
3

Its second moment of inertia about the axis of the cone is

2 4
3my δx 3my δx
1

2
×
a h
2
×y
2
=
2a h
2
.
But y and x are related through y = ax

h
, so the moment of inertia of the elemental disk is
2 4
3ma x δx
5
.
2h

The moment of inertia of the entire cone is

2 2
h
3ma
5
∫
0
4
x dx =
3ma

10
.
2h

The following, for a solid cone of mass m, height h , base radius a , are left as an exercise:

This page titled 2.6: Three-dimensional Solid Figures. Spheres, Cylinders, Cones. is shared under a CC BY-NC 4.0 license and was authored,
remixed, and/or curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit
history is available upon request.

2.6.2 https://phys.libretexts.org/@go/page/6937
2.7: Three-dimensional Hollow Figures. Spheres, Cylinders, Cones
Hollowspherical shell, mass m, radius a

The area of the elemental zone is 2π a 2

sin θδθ . Its mass is
2
2πa sin θδθ 1
m× 2
= m sin θδθ
4πa 2

Its moment of inertia is 1

2
m sin θδθ × a
2 2
sin θ =
1

2
ma
2
sin
3
θδθ

The moment of inertia of the entire spherical shell is

π
1 2 3
2 2
ma ∫ sin θδθ  = ma
2 0
3

This result can be used to calculate, by integration, the moment of inertia ma of a solid sphere. Or, if you start with ma for a
2

5
2 2

5
2

solid sphere, you can differentiate to find the result ma for a hollow sphere. Write the moment of inertia for a solid sphere in
2

3
2

terms of its density rather than its mass. Then add a layer da and calculate the increase dI in the moment of inertia. We can also
use the moment of inertia for a hollow sphere ( ma ) to calculate the moment of inertia of a nonuniform solid sphere in which
2

3
2

−−−−−−−
the density varies as ρ = ρ(r) . For example, if ρ = ρ 0 √1 −(
r

a
2
) , see if you can show that the mass of the sphere is 2.467ρ0 a
3

and that its moment of inertia is 1

3
2
ma . A much easier method will be found in Section 19.
Using methods similar to that given for a solid cylinder, it is left as an exercise to show that the moment of inertia of an open
hollow cylinder about an axis perpendicular to its length passing through its centre of mass is m( a + l ) , where a is the radius1

2
2 1

3
2

and 2l is the length.

The moment of inertia of a baseless hollow cone of mass m, base radius a , about the axis of the cone could be found by
integration. However, those who have an understanding of the way in which the moment of inertia depends on the distribution of
mass should readily see, without further ado, that the moment of inertia is ma . (Look at the cone from above; it looks just like a
1

2
2

disc, and indeed it has the same radial mass distribution as a uniform disc.)

This page titled 2.7: Three-dimensional Hollow Figures. Spheres, Cylinders, Cones is shared under a CC BY-NC 4.0 license and was authored,
remixed, and/or curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit
history is available upon request.

2.7.1 https://phys.libretexts.org/@go/page/8357
2.8: Torus
The rotational inertias of solid and hollow toruses (large radius a , small radius b ) are given below for reference and without
derivation. They can be derived by integral calculus, and their derivation is recommended as a challenge to the reader.
Solid torus:

Hollow torus:

This page titled 2.8: Torus is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

2.8.1 https://phys.libretexts.org/@go/page/8358
2.9: Linear Triatomic Molecule
Here is an interesting problem. It should be straightforward to calculate the rotational inertia of the above molecule with respect to
an axis perpendicular to the molecule and passing through the center of mass. In practice it is quite easy to measure the rotational
inertia very precisely from the spacing between the lines in a molecular band in the infrared region of the spectrum.

If you know the three masses (which you do if you know the atoms that make up the molecule) can you calculate the two
interatomic spacings x and y ? That would require determining two unknown quantities, x and y , from a single measurement of the
rotational inertia, I . Evidently that cannot be done; a second measurement is required. Can you suggest what might be done? We
shall answer that shortly. In the meantime, it is an exercise to show that the rotational inertia is given by
2 2
ax + 2hxy + b y + c = 0, (2.9.1)

where

a = m1 (m2 + m3 )/M (2.9.2)

h = m1 m3 /M (2.9.3)

b = m3 (m1 + m2 )/M (2.9.4)

M = m1 + m2 + m3 (2.9.4)

c = −I (2.9.5)

 Example 2.9.1: OCS

Suppose the molecule is the linear molecule OCS, and the three masses are 16, 12 and 32 respectively, and, from infrared
spectroscopy, it is determined that the moment of inertia is 20. (For this hypothetical illustrative example, I am not concerning
myself with units). In that case, equation 2.9.1 becomes
2 2
11.7 3̄x + 17.0 6̄xy + 14.9 3̄y − 20 = 0. (2.9.6)

We need another equation to solve for x and y. What can be done chemically is to prepare an isotopically-substituted molecule
(isotopomer) such as 18OCS, and measure its moment of inertia from its spectrum, making the probably very justified
assumption that the interatomic distances are unaffected by the isotopic substitution. This results in a second equation:
′ 2 ′ ′ 2 ′
a x + 2 h xy + b y + c = 0. (2.9.7)

Let's suppose that the new moment of inertia is I = 21 , and I leave it to the reader to work out the numerical values of a , h
′ ′ ′

and b with the stern caution to retain all the decimal places on your calculator. That is, do not round off the numbers until the
′

very end of the calculation.

You now have two equations, 2.9.1 and 2.9.7, to solve for x and y . These are two simultaneous quadratic equations, and it may be
that some guidance in solving them would be helpful. I have three suggestions.
1. Treat equation 2.9.1 as a quadratic equation in x and solve it for x in terms of y . Then substitute this in equation 2.9.7. I expect
you will very soon become bored with this method and will want to try something a little less tedious.
2. You have two equations of the form S(x, y) = 0, S (x, y) = 0 . There are standard ways of solving these iteratively by an
′

extension of the Newton-Raphson process. This is described, for example, in Section 1.9 of my Celestial Mechanics notes, and
this general method for two or more nonlinear equations should be known by anyone who expects to engage in much numerical
calculation.
For this particular case, the detailed procedure would be as follows. This is an iterative method, and it is first necessary to make a
guess at the solutions for x and y . The guesses need not be particularly good. That done, compute the following six quantities:
2
S = x(ax + 2hy) + b y +c

2.9.1 https://phys.libretexts.org/@go/page/8359
 ′ ′ ′ ′ 2 ′
S = x(a x + 2 h y) + b y +c

Sx = 2(ax + hy)

Sy = 2(hx + by)

′ ′ ′
Sx = 2(a x + h y)

′ ′ ′
Sy = 2(h x + b y)

Here the subscripts denote the partial derivatives. Now if

x (true) = x (guess) + ϵ
and
y (true) = y (guess) + η
the errors ϵ and η can be found from the solution of
Sx ϵ + Sy η + S = 0

and
′ ′ ′
Sx ϵ + Sy η + S =0

If we calculate
1
F = ′ ′
Sy Sx −Sx Sy

The solutions for the errors are

′ ′
ϵ = F (Sy S − Sy S )

′ ′
η = F (Sx S − Sx S)

This will enable a better guess to be made, and the procedure can be repeated until the errors are as small as desired. Generally only
a very few iterations are required. If this is not the case, a programming mistake is indicated.
3. While method 2 can be used for any nonlinear simultaneous equations, in this particular case we have two simultaneous
quadratic equations, and a little familiarity with conic sections provides a rather nice method.
Thus, if S = 0 and S = 0 are equations 2.9.1 and 2.9.7 respectively. Each of these equations represents a conic section, and they
′

intersect at four points. We wish to find the point of intersection that lies in the all-positive quadrant - i.e. with x and y both
positive. Since the two conic sections are very similar, in order to calculate where they intersect it is necessary to calculate with
great accuracy. Therefore, do not round off the numbers until the very end of the calculation. Form the equation c S − cS = 0 . ′ ′

This is also a quadratic equation representing a conic section passing through the four points. Furthermore, it has no constant term,
and it therefore represents the two straight lines that pass through the four points. The equation can be factorized into two linear
terms, αβ − 0 , where α = 0 and β = 0 are the two straight lines. Choose the one with positive slope and solve it with S = 0 or
with S = 0 (or with both, as a check against arithmetic mistakes) to find x and y . In this case, the solutions are
′

x = 0.2529, y = 1.000.

This page titled 2.9: Linear Triatomic Molecule is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

2.9.2 https://phys.libretexts.org/@go/page/8359
2.10: Pendulums
In Section 2.2, we discussed the physical meaning of the rotational inertia as being the ratio of the applied torque to the resulting
angular acceleration. In linear motion, we are familiar with the equation F = ma . The corresponding Equation when dealing with
torques and angular acceleration is τ = I θ¨ . We are also familiar with the equation of motion for a mass vibrating at the end of a
−−
spring of force constant k : m ẍ = −kx . This is simple harmonic motion of period 2π √
m

k
. The mechanics of the torsion
pendulum is similar.
The torsion constant c of a wire is the torque required to twist it through unit angle. If a mass is suspended from a torsion wire, and
the wire is twisted through an angle θ , the restoring torque will be cθ, and the Equation of motion is I θ̈ = −cθ , which is simple
−
−
harmonic motion of period 2π √ . The torsion constant of a wire of circular cross-section, by the way, is proportional to its shear
I

modulus, the fourth power of its radius, and inversely as its length. The derivation of this takes a little trouble, but it can be verified
by dimensional analysis. Thus a thick wire is very much harder to twist than a thin one. A wire of narrow rectangular cross-section,
such as a strip or a ribbon has a relatively small torsion constant.
Now let's look not at a torsion pendulum, but at a pendulum swinging about an axis under gravity.

We suppose the pendulum, of mass m, is swinging about a point O, which is at a distance h from the center of mass C. The
rotational inertia about O is I . The line OC makes an angle θ with the vertical, so that the horizontal distance between O and C is
h sin θ . The torque about O is mgh sin θ , so that the equation of motion is

¨
I θ = −mgh sin θ. (2.10.1)

For small angles (sin θ ≈ θ ), this is

¨
I θ = −mghθ. (2.10.2)

This is simple harmonic motion of period

−−−−−
I
P = 2π √ . (2.10.3)
mgh

We'll look at two examples - a uniform rod, and an arc of a circle.

 Example 2.10.1

First, a uniform rod.

2.10.1 https://phys.libretexts.org/@go/page/8360
The center of mass is C. The rotational inertia about C is ml , so the rotational inertia about O is
1

3
2
I =
1

3
2
ml + mh
2
. If we
substitute this in equation 2.10.3, we find for the period of small oscillations
−−−−−−−
2 2
l + 3h
P = 2π √ . (2.10.4)
3gh

This can be written

− −−−−−− −
−
−− h 2
 l + 3(
l )
l
P = 2π √ ⋅ . (2.10.5)
h
3g ⎷
l

or, if we write P = P

l
and h = h

l
:
2π√
3g

−−−−−−−
2
1 + 3h
P =√ . (2.10.6)
h

The figure shows a graph of P versus h.

Equation 2.10.6 can be written

1
2
P = + 3h. (2.10.7)
h

and, by differentiation of P with respect to h, we find that the period is least when h =
2 1
.
√3

−−
This least period is given by P 2
= √12 ,or P = 1.861.
Equation 2.10.7 can also be written

2.10.2 https://phys.libretexts.org/@go/page/8360
 2 2
 3 h −P h+1 = 0 (2.10.8)

This quadratic Equation shows that there are two positions of the support O that give rise to the same period of small
oscillations. The period is least when the two solutions of Equation 2.10.8 are equal, and by the theory of quadratic Equations,
−−
then, the least period is given by P = √12 as we also deduced by differentiation of Equation 2.10.7, and this occurs when
2

h =
1

√3
.

For periods longer than this, there are two solutions for h. Let h be the smaller of these, and let h be the larger. By the theory
1 2

of quadratic Equations, we have

1 2
 h1 + h2 = P (2.10.9)
3

and
1
 h1 h2 = (2.10.10)
3

Let H = h 2 − h1 be the distance between two points O that give the same period of oscillation.
Then
4
P − 12
2 2 2
 H = (h2 − h1 ) = (h2 + h1 ) − 4 h1 h2 = (2.10.11)
9

If we measure H for a given period P and recall the definition of P we see that this provides a method for determining g .
Although this is a common undergraduate laboratory exercise, the graph shows that the minimum is very shallow and
consequently H and hence g are very difficult to measure with any precision.

 Example 2.10.2

For another example, let us look at a wire bent into the arc of a circle of radius a oscillating in a vertical plane about its mid-
point. In the figure, C is the center of mass.

The rotational inertia about the centerof the circle is ma . By two applications of the parallel axes theorem, we see that the
2

rotational inertia about the point of oscillation is I = ma − m(a − h) + mh = 2mah .Thus, from Equation 2.10.3 we
2 2 2

find
−−
−
2a
P = 2π √ , (2.10.12)
g

and the period is independent of the length of the arc.

This page titled 2.10: Pendulums is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

2.10.3 https://phys.libretexts.org/@go/page/8360
2.11: Plane Laminas. Product Moment. Translation of Axes (Parallel Axes Theorem)
We consider a set of point masses distributed in a plane, or a plane lamina. We have hitherto met three second moments of inertia:
2
A = ∑ mi y , (2.11.1)
i

2
B = ∑ mi x , (2.11.2)
i

2 2
C = ∑ mi (x + y ), (2.11.3)
i i

These are respectively the moments of inertia about the x- and y -axes (assumed to be in the plane of the masses or the lamina) and
the z -axis (normal to the plane). Clearly, C = A + B , which is the perpendicular axes theorem for a plane lamina.
We now introduce another quantity, H , called the product moment of inertia with respect to the x- and y -axes, defined by

H = ∑ mi xi yi (2.11.4)

We'll need sometime to ask ourselves whether this has any particular physical significance, or whether it is merely something to
calculate for the sake of passing the time of day. In the meantime, the reader should recall the parallel axes theorems (Section 2.5)
and, using arguments similar to those given in that section, should derive
¯¯¯¯
H = HC + M x̄ȳ (2.11.5)

It may also be noted that Equation 2.11.4 does not contain any squared terms and therefore the product moment of inertia,
depending on the distribution of masses, is just as likely to be a negative quantity as a positive one.
We shall defer discussing the physical significance, if any, of the product moment until section 2.12. In the meantime let us try to
calculate the product moment for a plane right triangular lamina:

2Mδxδy
The area of the triangle is 1

2
ab and so the mass of the element δxδx is ab
, where M is the mass of the complete triangle. The
2Mxyδxδy
product moment of the element with respect to the sides OA, OB is and so the product moment of the entire triangle is
ab

∫ ∫ xydxdy . We have to consider carefully the limits of integration. We'll integrate first with respect to x ; that is to say we
2M

ab

integrate along the horizontal (y constant) strip from the side OB to the side AB. That is to say we integrate xδx from where x = 0
y
to where x = a(1 − ) . The product moment is therefore
b

2M 1 y
2 2
∫ y a (1 − ) dy.
ab 2 b

We now have to add up all the horizontal strips from the side OA, where y = 0 , to B, where y = b .
Thus
b
Ma y 2

H = ∫ y (1 − ) dy,
b 0 b

which, after some algebra, comes to H = 1

12
M ab .
The coordinates of the centre of mass with respect to the sides OA, OB are a, b, so that, from Equation 2.11.5, we find that the
1

3
1

product moment with respect to axes parallel to OA, OB and passing through the centre of mass is − M ab . 1

36

2.11.1 https://phys.libretexts.org/@go/page/8361
  Exercise 2.11.1

Calculate the product moments of the following eight laminas, each of mass M , with respect to horizontal and vertical axes
through the origin, and with respect to horizontal and vertical axes through the centroid of each. (We have just done the first of
these, above.) The horizontal base of each is of length a , and the height of each is b . You are going to have to take great care
with the signs, and with the limits of integration. If you get an answer right except for the sign, then you have got the answer
wrong.

I make the answers as follows.

This page titled 2.11: Plane Laminas. Product Moment. Translation of Axes (Parallel Axes Theorem) is shared under a CC BY-NC 4.0 license and
was authored, remixed, and/or curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a
detailed edit history is available upon request.

2.11.2 https://phys.libretexts.org/@go/page/8361
2.12: Rotation of Axes
We start by recalling a result from elementary geometry. Consider two sets of axes Oxy and Ox y , the latter being inclined at an
1 1

angle θ to the former. Any point in the plane can be described by the coordinates (x, y) or by (x , y ). 1 1

These coordinates are related by a rotation matrix:

x1 cos θ sin θ x
( ) =( )( ), (2.12.1)
y1 − sin θ cos θ y

The rotation matrix is orthogonal; one of the several properties of an orthogonal matrix is that its reciprocal is its transpose.
x cos θ − sin θ x1
( ) =( )( ). (2.12.2)
y sin θ cos θ y1

Now let us apply this to the moments of inertia of a plane lamina. Let us suppose that the axes are in the plane of the lamina and
that O is the centre of mass of the lamina. A, B and H are the moments of inertia with respect to the axes Oxy, and A , B and H 1 1 1

are the moments of inertia with respect to Ox y . Strictly speaking a lamina implies a continuous distribution of matter in a plane,
1 1

but, since matter, we are told, is composed of discrete atoms, there is little difficulty in justifying treating a lamina as though it we a
distribution of point masses in the plane. In any case the results that follow are valid either for a collection of point masses in a
plane or for a genuine continuous lamina.
We have, by definition:
2
A1 = ∑ m y (2.12.3)
1

2
B1 = ∑ m x (2.12.4)
1

H1 = ∑ m x1 y1 (2.12.5)

Now let us apply Equation 2.12.1 to Equation 2.12.3:

2
A1 = ∑ m(−x sin θ + y cos θ)

2 2 2 2
= sin θ ∑ mx − 2 sin θ cos θ ∑ mxy + cos θ ∑ my .

That is to say (writing the third term first, and the first term last)
2 2
A1 = A cos θ − 2H sin θ cos θ + B sin θ. (2.12.6)

In a similar fashion, we obtain for the other two moments

2 2
B1 = A sin θ + 2H sin θ cos θ + B cos θ. (2.12.7)

and
2 2
H1 = A sin θ cos θ + H sin(cos θ − sin θ) − B sin θ cos θ. (2.12.8)

It is usually more convenient to make use of trigonometric identities to write these as

1 1
A1 = (B + A) − (B − A) cos 2θ − H sin 2θ, (2.12.9)
2 2

1 1
B1 = (B + A) + (B − A) cos 2θ + H sin 2θ, (2.12.10)
2 2

2.12.1 https://phys.libretexts.org/@go/page/8362
 1
H1 = H cos 2θ − (B − A) sin 2θ (2.12.11)
2

These equations enable us to calculate the moments of inertia with respect to the axes Ox 1 y1 if we know the moments with respect
to the axes Oxy. Further, a matter of importance, we see, from Equation 2.12.11, that if
2H
tan 2θ = , (2.12.12)
B−A

the product moment H with respect to the axes Oxy is zero. This gives some physical meaning to the product moment, namely: If
1

we can find some axes (which we can, by means of Equation 2.12.12) with respect to which the product moment is zero, these axes
are called the principal axes of the lamina, and the moments of inertia with respect to the principal axes are called the principal
moments of inertia. I shall use the symbols A and B for the principal moments of inertia, and I shall adopt the convention that
0 0

A ≤B .
0 0

 Example 2.12.1

Consider three point masses at the coordinates given below:

Mass Coordinates

5 (1 , 1)

3 (4 , 2 )

2 (3 , 4)

The moments of inertia are A = 49, B = 71, C = 53 . The coordinates of the centre of mass are (2.3 , 1.9). If we use the
parallel axes theorem, we can find the moments of inertia with respect to axes parallel to the original ones but with origin at the
centre of mass. With respect to these axes we find A = 12.9, B = 18.1, H = +9.3 . The principal axes are therefore inclined
at angles θ to the x -axis given (Equation 2.12.12) by tan 2θ = 3.57669; That is θ = 37°11' and 127° 11'. On using Equation
2.12.9 or 2.12.10 with these two angles, together with the convention that A ≤ B , we obtain for the principal moments of
0 0

inertia A = 5.84 and B = 25.16.

0 0

 Example 2.12.2

Consider the right-angled triangular lamina of Section 11. The moments of inertia with respect to axes passing through the
centre of mass and parallel to the orthogonal sides of the triangle are A = M b , B = M a , H = − M ab . The angles
1

18
2 1

18
2 1

36

that the principal axes make with the a - side are given by tan 2θ = . The interested reader will be able to work out
2
ab

2
b −a

expressions, in terms of M , a, b, for the principal moments.

This page titled 2.12: Rotation of Axes is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

2.12.2 https://phys.libretexts.org/@go/page/8362
2.13: Momental Ellipse
Consider a plane lamina such that its radius of gyration about some axis through the centre of mass is k . Let P be a vector in the
direction of that axis, originating at the centre of mass, given by
2
a
P= ^
r (2.13.1)
k

Here ^r is a unit vector in the direction of interest; k is the radius of gyration, and a is an arbitrary length introduced so that the

dimensions of P are those of length, and the length of the vectorP is inversely proportional to the radius of gyration. The moment
4

of inertia is M k =2 Ma
. That is to say
2
P

4
Ma 2 2
= A cos θ − 2H sin θ cos θ + B sin θ, (2.13.2)
2
P

where A, H and B are the moments with respect to the x- and y -axes. Let (x, y) be the coordinates of the tip of the vector P , so
that x = P cos θ and y = P sin θ . Then
4 2 2
Ma = Ax − 2H xy + By . (2.13.2)

Thus, no matter what the shape of the lamina, however irregular and asymmetric, the tip of the vector P traces out an ellipse,
whose axes are inclined at angles tan (1

2
−1
) to the x - axis.
2H

B−A

This is the momental ellipse, and the axes of the momental ellipse are the principal axes of the lamina.

 Example 2.13.1
Consider a regular n -gon. By symmetry the moment of inertia is the same about any two axes in the plane inclined at 2π/n to
each other. This is possible only if the momental ellipse is a circle. It follows that the moment of inertia of a uniform polygonal
plane lamina is the same about any axis in its plane and passing through its centroid.

 Exercise 2.13.1
Show that the moment of inertia of a uniform plane n - gon of side 2a about any axis in its plane and passing through its
centroid is ma (1 + 3 cot (π/n)) .
1

12
2 2

What is this for a square? For an equilateral triangle?

This page titled 2.13: Momental Ellipse is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

2.13.1 https://phys.libretexts.org/@go/page/8363
2.14: Eigenvectors and Eigenvalues
In Sections 11-13, we have been considering some aspects of the moments of inertia of plane laminas, and we have discussed such
matters as rotation of axes, and such concepts as product moments of inertia, principal axes, principal moments of inertia and the
momental ellipse. We next need to develop the same concepts with respect to three-dimensional solid bodies. In doing so, we shall
need to make use of the algebraic concepts of eigenvectors and eigenvalues. If you are already familiar with such matters, you may
want to skip this section and move on to the next. If the ideas of eigenvalues and eigenvectors are new to you, or if you are a bit
rusty with them, this section may be helpful. I do assume that the reader is at least familiar with the elementary rules of matrix
multiplication.

 Example 2.14.1
0
Consider what happens when you multiply a vector, for example the vector ( ,
)
1

4 −1
by a square matrix, for example the matrix ( ) we obtain:
2 1

4 −1 0 −1
( )( ) =( )
2 1 1 1

The result of the operation is another vector that is in quite a different direction from the original one.
1 3
However, now let us multiply the vector ( ) by the same matrix. The result is ( . The result of the multiplication is
)
1 3

1
merely to multiply the vector by 3 without changing its direction. The vector ( ) is a very special one, and it is called an
1

eigenvector of the matrix, and the multiplier 3 is called the corresponding eigenvalue. "Eigen" is German for "own" in the
1
sense of "my own book". There is one other eigenvector of the matrix; it is the vector ( . Try it; you should that the
)
2

corresponding eigenvalue is 2.

In short, given a square matrix A, if you can find a vector a such that Aa = λ a, where λ is merely a scalar multiplier that does not
change the direction of the vector a, then a is an eigenvector and λ is the corresponding eigenvalue.
In the above, I told you what the two eigenvectors were, and you were able to verify that they were indeed eigenvectors and you
were able to find their eigenvalues by straightforward arithmetic. But, what if I hadn't told you the eigenvectors? How would you
find them?
A11 A12 x1
Let A = ( ) and let x = ( ) be an eigenvector with corresponding eigenvalue λ . Then we must have
A21 A22 x2

A11 A12 x1 λx1

( )( ) =( ).
A21 A22 x2 λx2

That is,
(A11 − λ)x1 + A12 x2 = 0

and
A21 x1 + (A22 − λ)x2 = 0.

These two equations are consistent only if the determinant of the coefficients is zero. That is,
A11 − λ A12
[ ] =0
A21 A22 − λ

This equation is a quadratic equation in λ , known as the characteristic equation, and its two roots, the characteristic or latent roots
are the eigenvalues of the matrix. Once the eigenvalues are found the ratio of x to x is easily found, and hence the eigenvectors.
1 2

Similarly, if A is a 3 × 3 matrix, the characteristic equation is

2.14.1 https://phys.libretexts.org/@go/page/8364
 A11 − λ A12 A13
⎡ ⎤

⎢ A21 A22 − λ A23 ⎥ =0

⎣ ⎦
A31 A32 A33 − λ

This is a cubic equation in λ , the three roots being the eigenvalues. For each eigenvalue, the ratio x : x : x can easily be found
1 2 3

and hence the eigenvectors. The characteristic equation is a cubic equation, and is best solved numerically, not by algebraic
formula. The cubic equation can be written in the form
3 2
λ + a2 λ + a1 λ + a0 = 0,

and the solutions can be checked from the following results from the theory of equations:
λ1 + λ2 + λ3 = −a2 ,

λ2 λ3 + λ3 λ1 + λ1 λ2 = a1 ,

λ1 λ2 λ3 = −a0 .

This page titled 2.14: Eigenvectors and Eigenvalues is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

2.14.2 https://phys.libretexts.org/@go/page/8364
2.15: Solid Body
The moments of inertia of a collection of point masses distributed in three-dimensional space (or of a solid three-dimensional body,
which, after all, is a collection of point masses (atoms)) with respect to axes Oxyz are:
2 2
A = ∑ m(y +z ) F = ∑ myz

2 2
B = ∑ m(z +x ) G = ∑ mzx

2 2
C = ∑ m(x +y ) H = ∑ mxy

Suppose that A, B, C , F , G, H , are the moments and products of inertia with respect to axes whose origin is at the centre of mass.
The parallel axes theorems (which the reader should prove) are as follows: Let P be some point not at the centre of mass, such that
the coordinates of the centre of mass with respect to axes parallel to the axes Oxyz but with origin at P are (x̄, ȳ, z̄).
¯
¯ ¯
¯ ¯
¯

Then the moments and products of inertia with respect to the axes through P are
2 2
¯
¯¯ ¯
¯¯ ¯
¯¯¯
¯
A + M (y +z ) F + M yz

2 2
¯
¯¯ ¯¯
¯ ¯¯¯¯
¯
B + M (z +x ) G + M zx

2 2
¯¯
¯ ¯
¯¯ ¯¯¯¯
¯
C + M (x +y ) H + M yx

where M is the total mass.

Unless stated otherwise, in what follows we shall suppose that the moments and products of inertia under discussion are referred to
a set of axes with the centre of mass as origin.

This page titled 2.15: Solid Body is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

2.15.1 https://phys.libretexts.org/@go/page/8365
2.16: Rotation of Axes - Three Dimensions
Let Oxyz be one set of mutually orthogonal axes, and let Ox y z be another set of axes inclined to the first. The coordinates (
1 1 1

x , y , z ) of a point with respect to the second basis set are related to the coordinates (x, y, z) with respect to the first by
1 1 1

x1 c11 c12 c13 x

⎛ ⎞ ⎛ ⎞⎛ ⎞

⎜ y1 ⎟ = ⎜ c21 c22 c23 ⎟ ⎜ y ⎟ (2.16.1)

⎝ ⎠ ⎝ ⎠⎝ ⎠
z1 c31 c32 c33 z

Here the c are the cosines of the angles between the axes of one basis set with respect to the axes of the other. For example, c
ij 12 is
the cosine of the angle between Ox and Oy , and c is the cosine of the angles between Oy and Oz .
1 23 1

Some readers may know how to express these cosines in terms of complicated expressions involving the Eulerian angles. While
these are important, they are not essential for following the present development, so we shall not make use of the Eulerian angles
just here.
The matrix of direction cosines is orthogonal. Among the several properties of an orthogonal matrix is the fact that its reciprocal
(inverse) is equal to its transpose - i.e. the reciprocal of an orthogonal matrix is found merely my interchanging the rows and
columns. This enables us easily to find (x, y, z) in terms of (x , y , z ) .
1 1 1

A number of other properties of an orthogonal matrix are useful in detecting, locating and even correcting arithmetic mistakes in
computing the elements. These properties are
1. The sum of the squares of the elements in any row or column is unity. This merely expresses the fact that the magnitude of a
unit vector along any of the six axes is indeed unity.
2. The sum of the products of corresponding elements of any two rows or of any two columns is zero. This merely expresses the
fact that the scalar product of any two orthogonal vectors is zero. It will be noted that checking for property 1 will not detect
any mistakes in sign of the elements, whereas checking for property 2 will do so.
3. Every element is equal to ± its own cofactor. This expresses the fact that the cross product of two unit orthogonal vectors is
equal to the third.
4. The determinant of the matrix is ± 1. If the sign is negative, it means that the chiralities (handedness) of the two basis sets of
axes are opposite; i.e. one of them is a right-handed set and the other is a left-handed set. It is usually convenient to choose both
sets as right- handed.
If it is possible to find a set of axes with respect to which the product moments F, G and H are all zero, these axes are called the
principal axes of the body, and the moments of inertia with respect to these axes are the principal moments of inertia, for which we
shall use the notation A , B , C , with the convention A ≤ B ≤ C . We shall see shortly that it is indeed possible, and we shall
0 0 0 0 0 0

show how to do it. A vector whose length is inversely proportional to the radius of gyration traces out in space an ellipsoid, known
as the momental ellipsoid.
In the study of solid body rotation (whether by astronomers studying the rotation of asteroids or by chemists studying the rotation
of molecules) bodies are classified as follows.
1. A ≠ B ≠ C The ellipsoid is a triaxial ellipsoid, and the body is an asymmetric top.
0 0 0

2. A < B = C The ellipsoid is a prolate spheroid and the body is a prolate symmetric top.
0 0 0

3. A = B < C The ellipsoid is an oblate spheroid and the body is an oblate symmetric top.
0 0 0

4. A = B = C The ellipsoid is a sphere and the body is a spherical top.

0 0 0

5. One moment is zero. The ellipsoid is an infinite elliptical cylinder, and the body is a linear top.

 Example 2.16.1

We know from Section 2.5 that the moment of inertia of a plane square lamina of side 2a about an axis through its centroid and
perpendicular to its area is ma , and it will hence be obvious that the moment of inertia of a uniform solid cube of side 2a
2

3
2

about an axis passing through the mid-points of opposite sides is also ma . It will clearly be the same about an axis passing
2

3
2

through the mid-points of any pairs of opposite sides. Therefore the cube is a spherical top and the momental ellipsoid is a
sphere. Therefore the moment of inertia of a uniform solid cube about any axis through its centre (including, for example, a
diagonal) is also ma . 2

3
2

2.16.1 https://phys.libretexts.org/@go/page/8366
  Exercise 2.16.1

What is the ratio of the length to the diameter of a uniform solid cylinder such that it is a spherical top?
Answer:
–
√3/2 = 0.866.

Let us note in passing that

2 2 2 2
A + B + C = 2 ∑ m(x +y + z ) = 2 ∑ mr , (2.16.2)

which is independent of the orientation of the basis axes In other words, regardless of how A, B and C may depend on the
orientation of the axes with respect to the body, the sum A + B + C is invariant under a rotation of axes.
We shall deal with the determination of the principal axes in Section 2.18 - but don't skip Section 2.17.

This page titled 2.16: Rotation of Axes - Three Dimensions is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated
by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

2.16.2 https://phys.libretexts.org/@go/page/8366
2.17: Solid Body Rotation and the Inertia Tensor
It is intended that this chapter should be limited to the calculation of the moments of inertia of bodies of various shapes, and not
with the huge subject of the rotational dynamics of solid bodies, which requires a chapter on its own. In this section I mention
merely for interest two small topics involving the principal axes, and a third topic in a bit more detail as necessary before
proceeding to Section 2.18.
Everyone knows that the relation between translational kinetic energy and linear momentum is E = p /(2m) . Similarly rotational
2

kinetic energy is related to angular momentum L by E = L /(2I ) , where I is the moment of inertia. If an isolated body (such as
2

an asteroid) is rotating about a non-principal axis, it will be subject to internal stresses. If the body is nonrigid this will result in
distortions (strains) which may cause the body to vibrate. If in addition the body is inelastic the vibrations will rapidly die out (if
the damping is greater than critical damping, indeed, the body will not even vibrate). Energy that was originally rotational kinetic
energy will be converted to heat (which will be radiated away.) The body loses rotational kinetic energy. In the absence of external
torques, however, L remains constant. Therefore, while E diminishes, I increases. The body adjusts its rotation until it is rotating
around its axis of maximum moment of inertia, at which time there are no further stresses, and the situation remains stable.
In general the rotational motion of a solid body whose momental ellipse is triaxial is quite complicated and chaotic, with the body
tumbling over and over in apparently random fashion. However, if the body is nonrigid and inelastic (as all real bodies are in
practice), it will eventually end up rotating about its axis of maximum moment of inertia. The time taken for a body, initially
tumbling chaotically over and over, until it reaches its final blissful state of rotation about its axis of maximum moment of inertia,
depends on how fast it is rotating. For most irregular small asteroids the time taken is comparable to or longer than the age of
formation of the solar system, so that it is not surprising to find some asteroids with non-principal axis (NPA) rotation. However, a
few rapidly-rotating NPA asteroids have been discovered, and, for rapid rotators, one would expect PA rotation to have been
reached a long time ago. It is thought that something (such as a collision) must have happened to these rapidly-rotating NPA
asteroids relatively recently in the history of the solar system.
Another interesting topic is that of the stability of a rigid rotator that is rotating about a principal axis, against small perturbations
from its rotational state. Although I do not prove it here (the proof can be done either mathematically, or by a qualitative argument)
rotation about either of the axes of maximum or of minimum moment of inertia is stable, whereas rotation about the intermediate
axis is unstable. The reader can observe this for him- or herself. Find anything that is triaxial - such as a small block of wood
shaped as a rectangular parallelepiped with unequal sides. Identify the axes of greatest, least and intermediate moment of inertia.
Toss the body up in the air at the same time setting it rotating about one or the other of these axes, and you will be able to see for
yourself that the rotation is stable in two cases but unstable in the third.

Inertia Tensor
I now deal with a third topic in rather more detail, namely the relation between angular momentum L and angular velocity ω. The
reader will be familiar from elementary (and two- dimensional) mechanics with the relation L = I ω . What we are going to find in
the three- dimensional solid-body case is that the relation is L = Iω . Here L and ω are, of course, vectors, but they are not
necessarily parallel to each other. They are parallel only if the body is rotating about a principal axis of rotation. The quantity I is a
tensor known as the inertia tensor. Readers will be familiar with the equation F = ma . Here the two vectors are in the same
direction, and m is a scalar quantity that does not change the direction of the vector that it multiplies. A tensor usually (unless its
matrix representation is diagonal) changes the direction as well as the magnitude of the vector that it multiplies. The reader might
like to think of other examples of tensors in physics. There are several. One that comes to mind is the permittivity of an anisotropic
crystal; in the equation D = ϵE and E are not parallel unless they are both directed along one of the crystallographic axes.
If there are no external torques acting on a body, L is constant in both magnitude and direction. The instantaneous angular velocity
vector, however, is not fixed either in space or with respect to the body - unless the body is rotating about a principal axis and the
inertia tensor is diagonal.
So much for a preview and a qualitative description. Now down to work.
I am going to have to assume familiarity with the equation for the components of the cross product of two vectors:
^ + (Az Bx − Az Bz )y
A × B = (Ay Bz − Az By )x ^ + (Ax By − Ay Bz )^
z (2.17.1)

I am also going to assume that the reader knows that the angular momentum of a particle of mass m at position vector r
(components (x, y, z ) ) and moving with velocity v (components ( ẋ, ẏ , ż )) is mr × v . For a collection of particles, (or an

2.17.1 https://phys.libretexts.org/@go/page/8367
extended solid body, which, I'm told, consists of a collection of particles called atoms), the angular momentum is

L = ∑ mr × v (2.17.2)

^ + m(zẋ − x ż )y
= ∑[m(y ż − zẏ )x ^ + m(x ẏ − y ẋ)^
z] (2.17.3)

I also assume that the relation between linear velocity v ( ẋ, ẏ , ż )and angular velocity ω (ωx , ωy , ωz ) is understood to be
v = ω × r , so that , for example ż = ω y − ω x. then
x y

L ^ + (etc.)y
= ∑[m(y(ωx y − ωy x) − z(ωz x − ωx z)x ^ + (etc.)^
z] (2.17.4)

2 2
= (ωx ∑ m y ^ + etc.
− ωy ∑ mxy − ωz ∑ mzx + ωx ∑ m z )x (2.17.5)

= (Aωx − H ωy − Gωz )x
^ + ()y
^ + ()^
z. (2.17.6)

Finally we obtain
Lx A  − H   − G ωx
⎛ ⎞ ⎛ ⎞⎛ ⎞

L = ⎜ Ly ⎟ = ⎜ −H B  − F ⎟ ⎜ ωy ⎟ (2.17.7)

⎝ ⎠ ⎝ ⎠⎝ ⎠
Lz −G  − F C ωz

This is the equation L = Iω referred to above. The inertia tensor is sometimes written in the form
Ixx  Ixy  Ixz
⎛ ⎞

II = ⎜ Ixy  Iyy  Iyz ⎟

⎝ ⎠
Ixz  Iyz  Izz

so that, for example, Ixy = −H . It is a symmetric matrix (but it is not an orthogonal matrix).

This page titled 2.17: Solid Body Rotation and the Inertia Tensor is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or
curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is
available upon request.

2.17.2 https://phys.libretexts.org/@go/page/8367
2.18: Determination of the Principal Axes
We now need to address ourselves to the determination of the principal axes. Unlike the two- dimensional case, we do not have a
nice, simple explicit expression similar to Equation 2.12.12 to calculate the orientations of the principal axes. The determination is
best done through a numerical example.

 Example 2.18.1
Consider four masses whose positions and coordinates are as follows:

M x y z

1 3 1 4

2 1 5 9

3 2 6 5

4 3 5 9

Relative to the first particle, the coordinates are

1 0 0 0

2 -2 4 5

3 -1 5 1

4 0 4 5

From this, it is easily found that the coordinates of the centre of mass relative to the first particle are ( −0.7, 3.9, 3.3), and the
moments of inertia with respect to axes through the first particle are
A = 324

B = 164

C = 182

F = 135

G = −23

H = −31

From the parallel axes theorems we can find the moments of inertia with respect to axes passing through the centre of mass:
A = 63.0

B = 50.2

C = 25.0

F = 6.3

G = 0.1

H = −3.7

The inertia tensor is therefore

63.0 3.7 −0.1
⎛ ⎞

⎜ 3.7 50.2 −6.3 ⎟

⎝ ⎠
−0.1 −6.3 25.0

We understand from what has been written previously that if ω, the instantaneous angular velocity vector, is along any of the
principal axes, then lω will be in the same direction as ω. In other words, if (l, m, n) are the direction cosines of a principal
axis, then
A −H G l l
⎛ ⎞⎛ ⎞ ⎛ ⎞
⎜ −H B −F ⎟ ⎜ m ⎟ = λ ⎜ m ⎟ ,
⎝ ⎠⎝ ⎠ ⎝ ⎠
−G −F C n n

2.18.1 https://phys.libretexts.org/@go/page/8368
where λ is a scalar quantity. In other words, a vector with components l, m, n(direction cosines of a principal axis) is an
eigenvector of the inertia tensor, and λ is the corresponding principal moment of inertia. There will be three eigenvectors (at
right angles to each other) and three corresponding eigenvalues, which we’ll initially call λ , λ , λ , though, as soon as we 1 2 3

know which is the largest and which the smallest, we'll call A , B , C , according to our convention A ≤ B ≤ C .
0 0 0 0 0 0

The Characteristic Equation is

a−λ −H −G
⎡ ⎤

⎢ −H B−λ −F ⎥ = 0.
⎣ ⎦
−G −F C −λ

In this case, this results in the cubic equation

2 3
a0 + a1 λ + a2 λ −λ = 0,

where
a0 = 76226.44

a1 = −5939.21

a2 = 138.20

The three solutions for λ , which we shall call A 0, B0 , C0 in order of increasing size are
A0 = 23.498256

B0 = 50.627521

C0 = 64.074223

and these are the principal moments of inertia. From the theory of equations, we note that the sum of the roots is exactly equal
to a , and we also note that it is equal to A + B + C , consistent with what we wrote in Section 2.16 (Equation 2.16.2). The
2

sum of the diagonal elements of a matrix is known as the trace of the matrix. Mathematically we say that "the trace of a
symmetric matrix is invariant under an orthogonal transformation".
Two other relations from the theory of equations may be used as a check on the correctness of the arithmetic. The product of
the solutions equals a , which is also equal to the determinant of the inertia tensor, and the sum of the products taken two at a
0

time equals −a .1

We have now found the magnitudes of the principal moments of inertia; we have yet to find the direction cosines of the three
principal axes. Let's start with the axis of least moment of inertia, for which the moment of inertia is A = 23.498256. Let the 0

direction cosines of this axis be (l , m , n ). Since this is an eigenvector with eigenvalue 23.498 256 we must have
1 1 1

63.0 3.7 −0.1 l1 l1

⎛ ⎞⎛ ⎞ ⎛ ⎞

⎜ 3.7 50.2 −6.3 ⎟ ⎜ m1 ⎟ = 23.498256 ⎜ m1 ⎟

⎝ ⎠⎝ ⎠ ⎝ ⎠
−0.1 −6.3 25.0 n1 n1

These are three linear equations in l m , n , with no constant term. Because of the lack of a constant term, the theory of
1 1 1

equations tells us that the third equation, if it is consistent with the other two, must be a linear combination of the first two. We
have, in effect, only two independent equations, and we are going to need a third, independent equation if we are to solve for
the three direction cosines. If we let l = l/n and m = m/n , then the first two equations become
′ ′

′ ′
39.501744 l + 3.7 m − 0.1 = 0

′ ′
3.7 l + 26.701744 m − 6.3 = 0.

The solutions are

′
l = −0.019825485
′
m = +0.238686617.

The correctness of the arithmetic can and should be checked by verifying that these solutions also satisfy the third equation.
The additional equation that we need is provided by Pythagoras's theorem, which gives for the relation between three direction
cosines

2.18.2 https://phys.libretexts.org/@go/page/8368
 2 2 2
l +m +n = 1,
1 1 1

or
1
2
n =
1 ′2 ′2
l +m +1

whence

n1 ± 0.972495608.

Thus we have, for the direction cosines of the axis corresponding to the moment of inertia A 0,

l1 = ∓0.019280197

m1 = ±0.232121881

n1 = ±0.972495608

(Check that l 2
1
+m
2
1
2
+n
1
= 1. )
It does not matter which sign you choose - after all, the principal axis goes both ways.
Similar calculations for B yield0

l2 = ±0.280652440

m2 = ∓0.932312706

n2 = ±0.228094774

and for C 0

l3 = ±0.959615796

m3 = ±0.277330987

n3 = ∓0.047170415

For the first two axes, it does not matter whether you choose the upper or the lower sign. For the third axes, however, in order
to ensure that the principal axes form a right-handed set, choose the sign such that the determinant of the matrix of direction
cosines is +1.

We have just seen that, if we know the moments and products of inertia A, B, C , F , G, H with respect to some axes (i.e. if we
know the elements of the inertia tensor) we can find the principal moments of inertia A , B , C by diagonalizing the inertia tensor,
0 0 0

or finding its eigenvalues. If, on the other hand, we know the principal moments of inertia of a system of particles (or of a solid
body, which is a collection of particles), how can we find the moment of inertia I about an axis whose direction cosines with
respect to the principal axes are (l, m, n) ?
First, some geometry.
Let Oxyz be a coordinate system, and let P(x, y, z) be a point whose position vector is
r = xi + yj + zk.

Let L be a straight line passing through the origin, and let the direction cosines of this line be
(l, m, n). A unit vector e directed along L is represented by

e = li + mj + nk

The angle θ between r and e is found from the scalar product r ⋅ e , given by

r cos θ = r ⋅ e.

I.e.
1
2 2 2
(x +y +z ) 2
cos θ = lx + my + nz

The perpendicular distance p from P to L is

1
2 2 2
p = r sin θ = (x +y +z ) 2 sin θ.

2.18.3 https://phys.libretexts.org/@go/page/8368
 1

If we write sin θ = (1 − cos 2

θ) 2 , we soon obtain
2 2 2 2 2
p =x +y +z − (lx + my + nz) .

Noting that l
2
= 1 −m
2 2
−n , m
2
= 1 −n
2
−l , n
2 2
= 1 −l
2
−m ,
2
we find, after further manipulation:
2 2 2 2 2 2 2 2 2 2
p = l (y + z ) + m (z + x ) + n (x + y ) − 2(mnyz + nlzx + lmxy).

Now return to our collection of particles, and let Oxyz be the principal axes of the system. The moment of inertia of the system
with respect to the line L is
2
I = ∑ Mp .

where I have omitted a subscript i on each symbol. Making use of the expression for p and noting that the product moments of the
system with respect to Oxyz are all zero, we obtain
2 2 2
I = l A0 + m B0 + n C0 . (2.18.1)

Also, let A, B, C , F , G, H be the moments and products of inertia with respect to a set of nonprincipal orthogonal axes; then the
moment of inertia about some other axis with direction cosines l, m, n with respect to these nonprincipal axes is
2 2 2
I = l A + m B + n C − 2mnF − 2nlG − 2lmH . (2.18.2)

 Example 2.18.2: Consider a brick

We saw in Section 2.16 that the moment of inertia of a uniform solid cube of mass M and side 2a about a body diagonal is
M a , and we saw how very easy this was. At that time the problem of finding the moment of inertia of a uniform solid
2 2

rectangular parallelepiped of sides 2a, 2b, 2c must have seemed intractable, but by now it is not at all hard.

1 2 2
A0 = M (b +c )
3

1 2 2
B0 = M (c +a )
3

1 2 2
C0 = M (a +b )
3

Thus we have:
a
l = 1

2 2 2
( a +b +c ) 2

b
m =
1
2
( a2 +b +c2 ) 2

c
n = 1

2 2 2
( a +b +c ) 2

We obtain:
2 2 2 2 2 2
2M( b c +c a +a b )
I = 2
2 2
3( a +b +c )

We note:
i. This is dimensionally correct;
ii. It is symmetric in a, b, c;
iii. If a = b = c, it reduces to M a .2

3
2

2.18.4 https://phys.libretexts.org/@go/page/8368
This page titled 2.18: Determination of the Principal Axes is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

2.18.5 https://phys.libretexts.org/@go/page/8368
2.19: Moment of Inertia with Respect to a Point
By “moment of inertia” we have hitherto meant the second moment of mass with respect to an axis. We were easily able to identify
it with the rotational inertia with respect to the axis, namely the ratio of an applied torque to the resulting angular acceleration.
I am now going to define the (second) moment of inertia with respect to a point, which I shall take unless otherwise specified to
mean the origin of coordinates. If we have a collection of mass points m at distances r from the origin, I define
i i

2 2 2 2
ι = ∑ mi r = ∑ mi (x +y +z ) (2.19.1)
i i i i

i i

as the (second) moment of inertia with respect to the origin, also sometimes called the “geometric moment of inertia”. I cannot
relate it in an obvious way to a simple dynamical concept in the same way that I related moment of inertia with respect to an axis to
rotational inertia, but we shall see that it is by no means merely a tedious exercise in arithmetic, and it does have its uses. The
symbol I has probably been used rather a lot in this chapter; so to describe the geometric moment of inertia I am going to use the
symbol ι.
The moment of inertia with respect to the origin is clearly something that does not depend on the orientation of any particular basis
set of orthogonal axes, since it depends only on the distances of the particles from the origin.
If you recall the definitions of A, B and C from Section 2.15, you will easily see that
1
ι = (A + B + C ) (2.19.2)
2

and we already noted (see Equation 2.16.2) that A + B + C is invariant under rotation of axes. In Section 2.18 we expressed it
slightly more generally by saying "the trace of a symmetric matrix is invariant under an orthogonal transformation". By now it
probably seems slightly less mysterious.

The trace of a symmetric matrix is invariant under an orthogonal transformation

Let us now calculate the geometric moment of inertia of a uniform solid sphere of radius a , mass m, density ρ, with respect to the
center of the sphere. It is

2
ι =∫ r dm. (2.19.3)
sphere

The element of mass, dm, here is the mass of a shell of radii r, r + dr; that is 4πρr2dr. Thus
a
4
4 5
ι = 4πρ ∫ r dr = πρa . (2.19.4)
0
5

With m = 4

3
3
πa ρ , this becomes
3 2
ι = ma . (2.19.5)
5

Indeed, for any spherically symmetric distribution of matter, since A = B = C , it will be clear from Equation 2.19.2, that the
moment of inertia with respect to the center is 3/2 times the moment of inertia with respect to an axis through the center. For
example, it is obvious from the definition of moment of inertia with respect to the center that for a hollow spherical shell it is just
M a , and therefore the moment of inertia with respect to an axis through the center is m a . In other words, you can work out
2 2 2

that the moment of inertia of a hollow spherical shell with respect to an axis through its center is ma in your head without any 2

3
2

of the integration that we did in Section 2.7!

By way of illustration, consider three spheres, each of radius a and mass M , but the density between center and surface varies as
2
−−−−−−
2
kr kr kr
ρ = ρ0 (1 − ), ρ = ρ0 (1 − ), ρ = ρ0 √1 −
a a2 a2

for the three spheres.

2.19.1 https://phys.libretexts.org/@go/page/8369
  Example 2.19.1

Calculate for each the moment of inertia about an axis through the center of the sphere. Express the answer in the form
M a × f (k) .
2 2

Solution
The mass of a sphere is
a
2
M = 4π ∫ ρ(r)r dr
0

and so
2 a
2 2
8πa 2
Ma = ∫ ρ(r)r dr
5 5 0

The moment of inertia about the center is

a
4
ι = 4π ∫ ρ(r)r dr.
0

and so the moment of inertia about an axis through the center is

a
8
4
I = ∫ ρ(r)r dr.
3 0

Therefore
a 4
I 5∫ ρ(r)r dr
o
=
2 a
2 2 2
Ma 3a ∫ ρ(r)r dr
5 o

For the first two spheres the integrations are straightforward. I make it
I 12 − 10k
=
2
Ma
2 12 − 9k
5

for the first sphere, and

I 35 − 25k
=
2 2
Ma 35 − 21k
5

for the second sphere. The integrations for the third sphere need a little more patience, but I make the answer
I 5(12α − 3 sin 2α − 3 sin 4α + sin 6α)
=
2 2 2
Ma 18 sin α(4α − sin 4α)
5

−
−
where sin α = √k .

Example 2.19.1 should be enough to convince that the concept of ι is useful – but it is not its only use. We shall meet it again in
Chapter 3 on the dynamics of systems of particles; in particular, it will play a role in what we shall become familiar with as the
virial theorem.

This page titled 2.19: Moment of Inertia with Respect to a Point is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or
curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is
available upon request.

2.19.2 https://phys.libretexts.org/@go/page/8369
2.20: Ellipses and Ellipsoids
Here are some problems concerning ellipses and ellipsoids that might be of interest.
Determine the principal moments of inertia of the following:
1. A uniform plane lamina of mass m in the form of an ellipse of semi axes a and b .
2. A uniform plane ring of mass m in the form of an ellipse of semi axes a and b .
3. A uniform solid triaxial ellipsoid of mass m and semi axes a, b and c .
4. A uniform hollow triaxial ellipsoid of mass m and semi axes a, b and c .
1. By integration, an elliptical lamina is slightly difficult, but by physical insight it is very easy!
The distribution of mass around the minor axis is the same as for a circular lamina of radius a , and therefore the moment B is the
same as for the circular lamina, namely B = ma . Similarly, A = mb , and hence, by the perpendicular axes theorem,
1

4
2 1

4
2

m(a + b ) .
1 2 2
C =
4

I think you will find that the shape of the momental ellipse is the same as the shape of the original elliptical lamina.
2. An elliptical ring (hoop) is remarkably difficult. It cannot be expressed in terms of elementary functions, and it has to be
calculated numerically. It can be expressed in terms of elliptic integrals (no surprise there), but most of us aren’t sure what elliptic
integrals are and they hardly count as elementary functions, and they have to be calculated numerically anyway. We take the ellipse
2
2 y
to be x

a
2
+ 2
= 1, with b ≤ a .
b

Even calculating the circumference of an ellipse isn’t all that easy. The circumference is
1
a dy 2

∮ ds = 4 ∫
0
[1 + (
dx
2
) ]dx , with y = b(1 − x
2
a
) 2
.
After a bit of algebra, this can be written as
−−
2
−−
2 4
a
4a

x
∫
0
√
c −x
2
a −z
2
dx , where c 2
=
a

2 2
.
a −b

At first this looks easy, but I do not think you can do it in terms of elementary functions. No problem, then – just integrate it
numerically. Unfortunately the integrand becomes infinite at the upper limit, so there is still a bit of a problem. However, a change
of variable to x = a sin θ solves that problem. The expression for the circumference becomes simply
2 2 1
π/2
4a ∫
0
[1 − (
a −b

a2
2
) sin θ] 2 dθ ,
which can be integrated numerically without infinity problems at the limits. According to my calculations, the circumference of the
ellipse is ha, where h is a function of b/a as follows:

To find the moment of inertia (or the second moment of length) about the minor axis, we have to multiple the integrand by x , or 2

a sin θ , and integrate. Thus the moment of inertia of the elliptical hoop about its minor axis is c m a , where
2 2 2
1

2.20.1 https://phys.libretexts.org/@go/page/8370
 2 2
π/2 a −b 2 1/2 2
∫ [1−( ) sin θ] sin θdθ
0 2
a
c1 =
2 2
π/2 a −b 2 1/2
∫ [1−( ) sin θ] dθ
0 2
a

The moment of inertia about the major axis is c 2 ma

2
, where
2 2 2
b π/2 a −b
2 1/2 2
∫ [1−( )co s ] sin θdθ
2 0 2
a a
c2 = 2 2
π/2 a −b 2 1/2
∫ [1−( ) sin θ] dθ
0 2
a

These two coefficients of ma are shown below as a function of b/a.

2

The moments of inertia of an elliptical ring of mass m and semi major and semi minor axes a and b are 2
c1 m a about the minor
axis and c ma about the major axis, where c and c are shown as functions of b/a.
2
2
1 2

The moment of inertia about the major axis can also be conveniently expressed in terms of b rather than a . If we write the moment
of inertia about the major axis as c mb , then c as a function of b/a is shown below.
4
2
4

The moment of inertia about an axis perpendicular to the plane of the ellipse and passing through its centre is 2
c3 m a , where, of
course (by the perpendicular axes theorem), c = c + c .
3 1 2

It is also equal to c
1 ma
2
+ c4 m b
2
.
3. For a uniform solid triaxial ellipsoid, the moments of inertia are

2.20.2 https://phys.libretexts.org/@go/page/8370
 1 2 2 1 2 2 1 2 2
A = m(b +c ) B = m(c +a ) C = m(c +a )
5 5 5

The momental ellipsoid is not of the same shape. Its axes are in the ratio

For example, if the axial ratios of the original ellipsoid are 1 : 2 : 3, the axial ratios of the corresponding momental ellipsoid is
−− −−
1 : √
13

10
: √
13

5
= 1 : 1.140 : 1.612 , which is slightly more spherical than the original ellipsoid.

4. Triaxial elliptical shell. We have to think carefully about what a triaxial elliptical shell is. If we imagine the inner surface of the
shell to be an ellipsoid, and the outer surface to be a similar ellipsoid, but with all linear dimensions increased by the same small
fractional increment, then we obtain a figure like this:

In this drawing the linear size of the outer surface is 3 percent larger than that of the inner surface. E. J. Routh correctly shows in
his treatise on rigid bodies that the principal moments of inertia of such a figure are m(b + c ), m(c + a ), m(a + b ) . 1

3
2 2 1

3
2 2 1

3
2 2

But it can be seen that such a figure is not (as presumably a rugger ball is) of uniform thickness. I draw below a shell of uniform
thickness. In such a case the inner and outer surfaces are not exactly similar.

In attempting to calculate the moment of inertia of such a figure I shall restrict myself to the case of a spheroidal shell of uniform
thickness. That is to say, an ellipsoid with two equal axes, represented by the equation, in cylindrical coordinates
2
2
ρ z
+ = 1,
a2 c2

where ρ 2
=x
2
+y
2
. Further, if I put c = χa , the equation to the spheroid can be written
2
2 z 2
ρ + 2
=a ,
χ

If χ < 1 , the spheroid is oblate. If χ > 1 , the spheroid is prolate.

We’ll first need to calculate its surface area, which is
1
c dρ 2
A = 4π ∫ ρ[1 + ( ) ] 2 dz
0 dz

After some algebra, this comes to

2
A = 4π a f (χ),

where
−−−− −
2
1 χ
2 1 + √1 − χ
f (χ) = [
2
ln( ) + 1] for χ ≤ 1
2 √1−χ χ

2.20.3 https://phys.libretexts.org/@go/page/8370
and
−−−−−
1 2 √χ2 − 1
χ
f (χ) = [ sin
−1
( ) + 1]] for χ ≥ 1
√χ2 −1
2 χ

This function is shown below as far as χ = 2 . For χ = 0 , the figure is a disc whose total area
(upper and lower surface) is 2πa , and f
2
=
1

2
. For χ = 1 , the figure is a sphere whose area is 4πa , and f 2
=1 . The function goes
to infinity as χ goes to infinity.

The moment of inertia about the z -axis is

4πm c dρ
3 2 1/2
I = ∫ ρ [1 + ( ) ] dz.
A 0 dz

After some algebra this becomes

2
I = m a g(χ)

2 2 4
(2−χ )(1−χ )−χ ln[(1+√1−χ2 )/χ]
g(χ) =
3/2
for χ ≤ 1
2 2 2 2
4{(1−χ ) +χ (1−χ ) ln[(1+√1−χ )/χ]}

4 √χ2 −1 2
χ χ −2
−1
sin ( )+
χ 2
3/2 χ −1
( χ−1)

g(χ) = 1 − for χ ≥ 1
2 √χ2 −1
χ
−1
4{ sin ( )+1}
χ
√χ2 −1

This function is shown below as far as χ = 2 For χ = 0 , the figure is a disc whose moment of inertia is π a , and f = . For 1

2
2 1

χ = 1 , the figure is a hollow sphere whose moment of inertia is π a , and f = . The function goes to 1 as χ goes to infinity; the
2 2 2

3 3

moment of inertia then approaches that of a hollow cylinder.

2.20.4 https://phys.libretexts.org/@go/page/8370
This page titled 2.20: Ellipses and Ellipsoids is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

2.20.5 https://phys.libretexts.org/@go/page/8370
2.21: Tetrahedra
 Exercise 2.21.1

Show that the moment of inertia about an axis through the centre of mass of a uniform solid regular tetrahedron of mass m and
edge length a is ma1

20
2

 Exercise 2.21.2

Show that the moment of inertia of a methane molecule about an axis through the carbon atom is 8

3
2
ml , where l is the bond
length and m is the mass of a hydrogen atom.

And, in case you are wondering that I haven’t specified the orientation of the axis in either case, the solid regular tetrahedron and
the methane molecule are both spherical tops, and the moment of inertia is the same about any axis through the centre of mass.

This page titled 2.21: Tetrahedra is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

2.21.1 https://phys.libretexts.org/@go/page/8371
 CHAPTER OVERVIEW
3: Systems of Particles
3.1: Introduction to Systems of Particles
3.2: Moment of Force
3.3: Moment of Momentum
3.4: Notation
3.5: Linear Momentum
3.6: Force and Rate of Change of Momentum
3.7: Angular Momentum
3.8: Torque
3.9: Comparison
3.10: Kinetic energy
3.11: Torque and Rate of Change of Angular Momentum
3.12: Torque, Angular Momentum and a Moving Point
3.13: The Virial Theorem

This page titled 3: Systems of Particles is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1
3.1: Introduction to Systems of Particles
By systems of particles I mean such things as a swarm of bees, a star cluster, a cloud of gas, an atom, a brick. A brick is indeed
composed of a system of particles – atoms - which are constrained so that there is very little motion (apart from small amplitude
vibrations) of the particles relative to each other. In a system of particles, there may be very little or no interaction between the
particles (as in a loose association of stars separated from each other by large distances) or there may be (as in the brick) strong
forces between the particles. Most (perhaps all) of the results to be derived in this chapter for a system of particles apply equally to
an apparently solid body such as a brick. Even if scientists are wrong and a brick is not composed of atoms but is a genuine
continuous solid, we can in our imagination suppose the brick to be made up of an infinite number of infinitesimal mass and
volume elements, and the same results will apply.
What sort of properties shall we be discussing? Perhaps the simplest one is this: The total linear momentum of a system of particles
is equal to the total mass times the velocity of the center of mass. This is true, and it may be “obvious” - but it still requires proof. It
2
may be equally “obvious” to some that “the total kinetic energy of a system of particles is equal to M v where M is the total
1

2
¯
¯¯

mass and v̄ is the velocity of the center of mass” - but this one, however “obvious”, is not true!
¯
¯

Before we get round to properties of systems of particles, I want to clarify what I mean by the moment of a vector such as a force or
a momentum. You are already familiar, from Chapters 1 and 2, with the moments of mass, which is a scalar quantity.

This page titled 3.1: Introduction to Systems of Particles is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

3.1.1 https://phys.libretexts.org/@go/page/6939
3.2: Moment of Force
First, let’s look at a familiar two-dimensional situation. In Figure III.1 I draw a force F and a point O. The moment of the force
with respect to O can be defined as
Force times perpendicular distance from O to the line of action of F.

Alternatively, (Figure III.2) the moment can be defined equally well by

Transverse component of force times distance from O to the point of application of the force.

Either way, the magnitude of the moment of the force, also known as the torque, is rF sin θ We can regard it as a vector, τ ,
perpendicular to the plane of the paper:
 τ = r × F (3.2.1)

Now let me ask a question. Is it correct to say the moment of a force with respect to (or “about”) a point or with respect to (or
“about”) an axis?
In the above two-dimensional example, it does not matter, but now let me move on to three dimensions, and I shall try to clarify.
In Figure III.3, I draw a set of rectangular axes, and a force F, whose position vector with respect to the origin is r .

The moment, or torque, of F with respect to the origin is the vector

 τ = r × F (3.2.2)

The x−, y− and z -components of τ are the moments of F with respect to the x−, y− and z-axes. You can easily find the
components of τ by expanding the cross product 3.2.2:
τ =x
^ (y Fz − zFy ) + y
^ (y Fx − x Fz ) + z
^(x Fy − y Fx ) (3.2.3)

where x
^, y z are the unit vectors along the x, y, z axes. In Figure III.4, we are looking down the x-axis, and I have drawn the
^, ^

components F and F , and you can see that, indeed, τ = y F − zF .

y z x z y

3.2.1 https://phys.libretexts.org/@go/page/6940
The dimensions of moment of a force, or torque, are ML2T-2, and the SI units are N m. (It is best to leave the units as N m rather
than to express torque in joules.)

This page titled 3.2: Moment of Force is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

3.2.2 https://phys.libretexts.org/@go/page/6940
3.3: Moment of Momentum
In a similar way, if a particle at position r has linear momentum p = mv , its moment of momentum with respect to the origin is the
vector l defined by
l = r×p (3.3.1)

and its components are the moments of momentum with respect to the axes. Moment of momentum plays a role in rotational
motion analogous to the role played by linear momentum in linear motion, and is also called angular momentum. The dimensions
of angular momentum are M L T . Several choices for expressing angular momentum in SI units are possible; the usual choice is
2 −1

J s (joule seconds).

This page titled 3.3: Moment of Momentum is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum
via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

3.3.1 https://phys.libretexts.org/@go/page/6942
3.4: Notation
In this section I am going to suppose that we n particles scattered through three-dimensional space. We shall be deriving some
general properties and theorems – and, to the extent that a solid body can be considered to be made up of a system of particles,
these properties and theorems will apply equally to a solid body.
In the Figure III.5, I have drawn just two of the particles, (the rest of them are left to your imagination) and the centre of mass C of
the system.

A given particle may have an external force F acting upon it. (It may, of course, have several external forces acting on it, but I
i

mean by F the vector sum of all the external forces acting on the i th particle.) It may also interact with the other particles in the
i

system, and consequently it may have internal forces F acting upon it, where j goes from 1 to n except for i. I define the vector
ij

sum F = ∑ F as the total external force acting upon the system.

i

I am going to establish the following notation for the purposes of this chapter.
Mass of the i th particle = m i

Total mass of the system M = ∑ m i

Position vector of the i th particle referred to a fixed point O: r = x x

^ +y y
^ +z z
^ i i i i

Velocity of the i th particle referred to a fixed point O: r or v (Speed = v )

i i i

Linear momentum of the i th particle referred to a fixed point O: p = m v i i i

Linear momentum of the system: P = ∑ P = ∑ m vi i i

External force on the i th particle: F i

Total external force on the system: F = ∑ F i

Angular momentum (moment of momentum) of the i th particle referred to a fixed point O:

li = ri × pi

Angular momentum of the system: L = ∑ l = ∑ r × p

i i i

Torque on the i th particle referred to a fixed point O: τ = r × F

i i i

Total external torque on the system with respect to the origin:

τ = ∑ τi = ∑ ri × Fi

Kinetic energy of the system: (We are dealing with a system of particles – so we are dealing only with translational kinetic energy
– no rotation or vibration):
1 2
T =∑ mi v
2 i

Position vector of the centre of mass referred to a fixed point O: ¯¯

r̄i
¯
¯^ ¯¯^
= x̄x + ȳ ¯¯^
y + z̄ z

The centre of mass is defined by M r = ∑ m

¯
¯¯
i ri

Velocity of the centre of mass referred to a fixed point O: r = v (Speed = v )

¯
¯¯ ¯¯
¯ ¯
¯¯

For position vectors, unprimed single-subscript symbols will refer to O. Primed single-subscript symbols will refer to C. This will
be clear, I hope, from Figure III.5.

3.4.1 https://phys.libretexts.org/@go/page/6941
Position vector of the i th particle referred to the centre of mass C: r ′
i
= ri − ri
¯
¯¯

Position vector of particle j with respect to particle i: r ij = rj − ri

(Internal) force exerted on particle i by particle j : F

ij

(Internal) force exerted on particle j by particle i: Fji

If the force between two particles is repulsive (e.g. between electrically-charged particles of the same sign), then F and r are in
ji ji

the same direction. But if the force is an attractive force, F and r are in opposite directions.
ji ji

According to Newton’s Third Law of Motion (Lex III), F ij = −Fji

Total angular momentum of system referred to the centre of mass C: L C

Total external torque on system referred to the centre of mass C: τ C

For the velocity of the centre of mass I may use either r or v

˙
¯
¯¯ ¯¯
¯

O is an arbitrary origin of coordinates. C is the centre of mass.

Note that
¯
¯¯ ′
 ri = r + r (3.4.1)
i

and therefore
¯
¯˙
¯ ˙
¯
¯¯ ˙′
 ri = r + r ; (3.4.2)
i

that is to say
¯¯
¯ ′
 vi = v + v (3.4.3)
i

Note also that

′
  ∑ mi r =0 (3.4.4)
i

Note further that

′ ¯¯ ¯¯ ¯¯ ¯¯
∑ mi v = ∑ mi (vi − v̄) = ∑ mi vi − v̄ ∑ mi = M v̄ − v̄M =0 (3.4.5)
i

That is, the total linear momentum with respect to the centre of mass is zero.
Having established our notation, we now move on to some theorems concerning systems of particles. It may be more useful for you
to conjure up a physical picture in your mind what the following theorems mean than to memorize the details of the derivations.

This page titled 3.4: Notation is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

3.4.2 https://phys.libretexts.org/@go/page/6941
3.5: Linear Momentum
 Theorem:

The total momentum of a system of particles equals the total mass times the velocity of the centre of mass.

Thus:
′
¯¯
¯ ¯¯
¯
P = ∑ mi vi = ∑ mi (v + v ) = M v + 0. (3.5.1)
i

This page titled 3.5: Linear Momentum is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

3.5.1 https://phys.libretexts.org/@go/page/6943
3.6: Force and Rate of Change of Momentum
 Theorem:

The rate of change of the total momentum of a system of particles is equal to the sum of the external forces on the system.

Thus, consider a single particle. By Newton’s second law of motion, the rate of change of momentum of the particle is equal to the
sum of the forces acting upon it:
˙
Pi = Fi + ∑ Fij (j ≠ i) (3.6.1)

Now sum over all the particles:

˙
Pi = ∑ Fi + ∑ ∑ Fij (j ≠ i)

i i j

1 1
F+ ∑ ∑ Fij + ∑ ∑ Fij
2 2
i j j i

1
F+ ∑ ∑ Fji + Fij (3.6.2)
2
i j

But, by Newton’s third law of motion, F ji + Fij = 0 , so the theorem is proved.

 Corollary:

If the sum of the external forces on a system is zero, the linear momentum is constant. (Law of Conservation of Linear
Momentum.)

This page titled 3.6: Force and Rate of Change of Momentum is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated
by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

3.6.1 https://phys.libretexts.org/@go/page/6944
3.7: Angular Momentum
Notation:
LC = angular momentum of system with respect to centre of mass C.
L = angular momentum of system relative to some other origin O.
r = position vector of C with respect to O.
¯
¯¯

P = linear momentum of system with respect to O.

(The linear momentum with respect to C is, of course, zero.)

 Theorem:
¯
¯¯
L = LC + r × P (3.7.1)

Thus:
¯
¯¯ ′ ¯¯
¯ ′
L = ∑ ri × pi = ∑ mi (ri × vi ) = ∑ mi (r + r ) × (v + v )
i i

¯
¯¯ ¯¯
¯ ¯
¯¯ ′ ′ ¯¯
¯ ′ ′
= (r × v) ∑ mi + r × ∑ mi v + (∑ mi r ) × v + ∑ r × p
i i i i

¯¯ ¯¯ ¯¯ ¯¯
= M (r̄ × v̄) + r̄ × 0 + 0 × v̄ + LC

therefore
¯¯
L = LC + r̄ ×P

 Example 3.7.1

A hoop of radius a rolling along the ground (Figure III.6):

The angular momentum with respect to C is LC = I Cω where I is the rotational inertia about C. The angular momentum about
C

O is therefore
¯
¯¯ 2 2
I = IC ω + M va = IC ω + M a ω = (IC + M a ) = I ω

where
2
I = IC + M a

is the rotational inertia about O.

This page titled 3.7: Angular Momentum is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum
via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

3.7.1 https://phys.libretexts.org/@go/page/8379
3.8: Torque
Notation:
τC = vector sum of all the torques about C.
τ = vector sum of all the torques about the origin O.
F = vector sum of all the external forces.

 Theorem
¯¯
τ = τC + r̄ ×F

Thus:
′ ¯
¯¯
τ = ∑ ri × Fi = ∑(r + r) × Fi (3.8.1)
i

′ ¯
¯
= ∑ r × Fi + r̄ ∑ Fi (3.8.2)
i

therefore
¯¯
τ = τC + r̄ ×F

This page titled 3.8: Torque is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

3.8.1 https://phys.libretexts.org/@go/page/8380
3.9: Comparison
At this stage I compare some somewhat similar formulas.
¯
¯¯ ¯
¯¯
L = LC + r × P τ = τC + r × F

L = ∑ mi ri × vi τ = ∑ mi ri × v̇i

′ ′ ′ ′

LC = ∑ mi r × v τ = ∑ mi r × v̇i
i i i

P = ∑ mi vi F = ∑ mi v̇i

This page titled 3.9: Comparison is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

3.9.1 https://phys.libretexts.org/@go/page/8381
3.10: Kinetic energy
We remind ourselves that we are discussing particles, and that all kinetic energy is translational kinetic energy.
Notation:
TC = kinetic energy with respect to the centre of mass C.
T = kinetic energy with respect to the origin O.

 Theorem:
1 2
¯¯
T = TC + M v̄ (3.10.1)
2

Thus:
1 2 1 ′ ¯¯
¯ ′ ¯¯
¯
T = ∑ mi v = ∑ mi (v + v) ⋅ (v + v)
2 i 2 i i

=
1

2
∑ mv
′2
i
¯¯
× v̄ ∑ mv +
′
i
1

2
v
−2
∑ mi .

∴ T = TC +
1

2
¯¯
M v̄
2
.

 Corollary:

If v = 0, T
¯¯
¯
= TC . (Think about what this means.)

 Corollary:

Corollary: For a non-rotating rigid body, T = 0, and therefore T

C =
1

2
¯¯
M v̄
2

(Think about what this means.)

This page titled 3.10: Kinetic energy is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

3.10.1 https://phys.libretexts.org/@go/page/8382
3.11: Torque and Rate of Change of Angular Momentum
 Theorem:

The rate of change of the total angular momentum of a system of particles is equal to the sum of the external torques on the
system.

Thus:

L = ∑ ri × pi (3.11.1)

˙
∴ L = ∑ ṙi × ṗi (3.11.2)

But the first term is zero, because ṙ and p are parallel.

i

Also

ṙi = Fi + ∑ Fij (3.11.3)

˙
Li = ∑ ri × (ri + ∑ Fij ) = ∑ ri × Fi + ∑ ri × ∑ Fii
i j i i j

∴ ∑ ri × Fi + ∑ ri × ∑ Fii
i i j

But ∑ i
∑j Fij = 0 by Newton’s third law of motion, and so ∑ i
∑j ri × Fij = 0 .
Also ∑ i
ri × Fi = τ , and so we arrive at
˙
L =τ (3.11.4)

which was to be demonstrated.

 Corollary: Law of Conservation of Angular Momentum

If the sum of the external torques on a system is zero, the angular momentum is constant.

This page titled 3.11: Torque and Rate of Change of Angular Momentum is shared under a CC BY-NC 4.0 license and was authored, remixed,
and/or curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is
available upon request.

3.11.1 https://phys.libretexts.org/@go/page/8383
3.12: Torque, Angular Momentum and a Moving Point
In Figure III.7 I draw the particle m , which is just one of n particles, n − 1 of which I haven’t drawn and are scattered around in
i

3-space. I draw an arbitrary origin O, the centre of mass C of the system, and another point Q, which may (or may not) be moving
with respect to O. The question I am going to ask is: Does the equation L ˙
= τ apply to the point Q? It obviously does if Q is

stationary, just as it applies to O. But what if Q is moving? If it does not apply, just what is the appropriate relation?

The theorem that we shall prove – and interpret - is

˙ ′
LQ = τQ + M r × r̈Q . (3.12.1)
Q

We start:

LQ = ∑(ri − rQ ) × [ mi (vi − vQ )] (3.12.2)

˙
∴ LQ = ∑(ri − rQ ) × mi (v̇i − v̇Q ) + ∑(ṙi − ṙQ ) × mi (vi − vQ ). (3.12.3)

The second term is zero, because ṙ = v

Continue:
˙
L = ∑(ri − rQ ) × mi v̇i − ∑ mi ri × v̇Q + ∑ mi rQ × v̇Q (3.12.4)

Now m i v̇i = Fi , so that the first term is just τQ

Continue:
˙
L = τQ − ∑ mi ri × dotvQ + ∑ Mi rQ × dotvQ

¯¯
= τQ − M r̄ × r̈Q + M rQ × r̈Q

¯
¯¯
= τQ + M (rQ − r) × r̈Q

˙ ′
∴ LQ = τQ + M r × r̈ Q Q. E. D (3.12.5)
Q

Thus in general, L
˙
Q ≠ τQ , but L
˙
Q = τQ under any of the following three circumstances:
i. r = 0 - that is, Q coincides with C.
′
Q

ii. r̈ = 0 - that is, Q is not accelerating.

Q

iii. r̈ and r are parallel, which would happen, for example, if O were a centre of attraction or repulsion and Q were
Q
′
Q

accelerating towards or away from O.

This page titled 3.12: Torque, Angular Momentum and a Moving Point is shared under a CC BY-NC 4.0 license and was authored, remixed,
and/or curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is
available upon request.

3.12.1 https://phys.libretexts.org/@go/page/8384
3.13: The Virial Theorem
First, let me say that I am not sure how this theorem got its name, other than that my Latin dictionary tells me that vis, viris means
force, and its plural form, vires, virium is generally translated as strength. The term was apparently introduced by Rudolph Clausius
of thermodynamics fame. We do not use the word strength in any particular technical sense in classical mechanics, although we do
talk about the tensile strength of a wire, which is the force that it can summon up before it snaps. We use the word energy to mean
the ability to do work; perhaps we could use the word strength to mean the ability to exert a force. But enough of these idle
speculations.
Before proceeding, I define the quantity
2
ι = ∑ mi r (3.13.1)
i

as the second moment of mass of a system of particles with respect to the origin. As discussed in Chapter 2, mass is (apart from
some niceties in general relativity) synonymous with inertia, and the second moment of mass is used so often that it is nearly
always called simply “the” moment of inertia, as though there were only one moment, the second, worth considering. Note
carefully, however, that you are probably much more used to thinking about the moment of inertia with respect to an axis rather
than with respect to a point. This distinction is discussed in Section 2.19. Note also that, since the symbol I tends to be heavily
used in any discussion of moments of inertia, for moment of inertia with respect to a point I am using the symbol ι
I can also write Equation 3.13.1 as

ι = ∑ mi (ri × ri ) (3.13.2)

Differentiate twice with respect to time:

ι̇ = 2 ∑ mi (ri ṙi ), (3.13.3)

and
2
ϊ = 2 ∑ mi (ṙ + ri r̈i ), (3.13.4)
i

or

ϊ = 4T + 2 ∑ ri mi r̈i , (3.13.5)

where T is the kinetic energy of the system of particles. The sums are understood to be over all particles - i.e. i from 1 to n .
mi r̈i is the force on the ith particle. I am now going to suppose that there are no external forces on any of the particles in the
system, but the particles interact with each other with conservative forces, F being the force exerted on particle i by particle j . I
ij

am also going to introduce the notation \( {\bf r} _{ij} = {\bf r} _{i} -{\bf r} _{i}\), which is a vector directed from particle i to
particle j . The relation between these three vectors in shown in Figure III.8.

I have not drawn the force F , but it will be in the opposite direction to r if it is a repulsive force and in the same direction as r
ij ji ji

if it is an attractive force.
The total force on particle i is ∑ j≠i
Fij and this is equal to m i r̈ i Therefore, Equation 3.13.5 becomes

3.13.1 https://phys.libretexts.org/@go/page/8827
 ϊ = 4T + 2 ∑ ri ∑ Fij (3.13.6)

i j≠i

Now it is clear that

∑ ri ∑ Fij = ∑ ∑ rij Fij (3.13.7)

i j≠i i j<i

However, in case, like me, you find double subscripts and summations confusing and you have really no idea what Equation 3.13.7
means, and it is by no means at all clear, I write it out in full in the case where there are five particles. Thus:
∑ ri ∑ Fij = r1 (F12 + F13 + F14 + F15 )
i j≠i

+r2 (F21 + F23 + F24 + F25 )

+r3 (F31 + F33 + F34 + F35 )

+r4 (F41 + F43 + F44 + F45 )

+r5 (F51 + F53 + F54 + F55 )

Now apply Newton’s third law of motion:

∑ ri ∑ Fij = r1 (−F21 − F31 − F41 + F51 )
i j≠i

+r2 (F21 − F32 − F42 − F52 )

+r3 (F31 − F32 − F43 − F53 )

+r4 (F41 − F42 − F43 − F54 )

+r5 (F51 − F52 − F53 + F54 )

Now bear in mind that r 2 − r1 = r21 , and we see that this becomes
∑ ri ∑ j ≠ i Fij = F21 ∗ r21 + F31 ∗ r31 + F41 ∗ r41 + F51 ∗ r51
i

F32 ∗ r32 + F42 ∗ r42 + F52 ∗ r52

+F43 ∗ r43 + F53 ∗ r53

+F54 ∗ r54

and we have arrived at Equation 3.13.7. Equation 3.13.6 then becomes

ϊ = 4T + 2 ∑ ∑ rij Fij (3.13.8)

i j<i

This is the most general form of the virial Equation. It tells us whether the cluster is going to disperse ( ϊ positive) or collapse (
ϊ negative) – though this will evidently depend on the nature of the force law F . ij

Now suppose that the particles attract each other with a force that is inversely proportional to the n th power of their distance apart.
For gravitating particles, of course, n = 2. The force between two particles can then be written in various forms, such as
k k
Fij = −Fij ^
rij = − ^
rij = − rij (3.13.9)
n n+1
r r
ij ij

and the mutual potential energy between two particles is minus the integral of F ij dr , or
k
Uij = − (3.13.10)
(n − 1)rn−1

I now suppose that the forces between the particles are gravitational forces, such that
Gmi mj
Fij = − rij (3.13.11)
3
r
ij

Now return to the term r ij Fij which occurs in Equation 3.13.8:

3.13.2 https://phys.libretexts.org/@go/page/8827
 k k
rij Fij = − rij rij = − = (n − 1)Uij (3.13.12)
n+1 n−1
r r
ij ij

Thus Equation 3.13.8 becomes

ϊ = 4T + 2(n − 1)U , (3.13.13)

where T and U are the kinetic and potential energies of the system. Note that for gravitational interaction (or any attractive) forces,
the quantity U is negative. Equation 3.13.13 is the virial theorem for a system of particles with an r attractive force −2

between them. The system will disperse or collapse according the sign of ϊ For a system of gravitationally-interacting particles,
n = 2 , and so the virial theorem takes the form

ϊ = 4T + 2U (3.13.14)

changing from moment to moment, but always in such a manner that Equation 3.13.13 is satisfied.
In a stable, bound system, by which I mean that, over a long period of time, there is no long-term change in the moment of inertia
of the system, and the system is neither irreversibly dispersing or contracting, that is to say in a system in which the average value
of ϊ over a long period of time is zero (I’ll define “long” soon), the virial theorem for a stable, bound system of r-n particles takes
the form
2⟨t⟩ + (n − 1) < u >= 0, (3.13.15)

and for a stable system of gravitationally-interacting particles,

2⟨t⟩+ < u >= 0, (3.13.16)

Here the angular brackets are understood to mean the average values of the kinetic and potential energies over a long period of
time. By a “long” period we mean, for example, long compared with the time that a particle takes to cross from one side of the
system to the other, or long compared with the time that a particle takes to move in an orbit around the centre of mass of the
system. (In the absence of external forces, of course, the centre of mass does not move, or it moves with a constant velocity.)
2

For example, if a bound cluster of stars occupies a spherical volume of uniform density, the potential energy is (Equation 3GM

5a

5.9.1 of Celestial Mechanics), so the virial theorem (Equation 3.13.16) will enable you to work out the mean kinetic energy and
hence speed of the stars. A globular cluster has roughly spherical symmetry, but it is not of uniform density, being centrally
condensed. If you assume some functional form for the density distribution, this will give a slightly different formula for the
potential energy, and you can then still use the virial theorem to calculate the mean kinetic energy.

 Example 3.13.1

Consider a planet of mass m moving in a circular orbit of radius a around a Sun of mass M , such that m <<M and the Sun
does not move.
The potential energy of the system is U = −GM . m

a
2

The speed of the planet is given by equating to mv

a
GMm

a
2
, from which T = GM
m

(2a)
,

so we easily see in this case that 2T + U =0 .

This page titled 3.13: The Virial Theorem is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum
via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

3.13.3 https://phys.libretexts.org/@go/page/8827
 CHAPTER OVERVIEW
4: Rigid Body Rotation
No real solid body is perfectly rigid. A rotating nonrigid body will be distorted by centrifugal force* or by interactions with other
bodies. Nevertheless most people will allow that in practice some solids are fairly rigid, are rotating at only a modest speed, and
any distortion is small compared with the overall size of the body. No excuses, therefore, are needed or offered for analyzing, to
begin with the rotation of a rigid body.
4.1: Introduction to Rigid Body Rotation
4.2: Angular Velocity and Eulerian Angles
4.3: Kinetic Energy of Rigid Body Rotation
4.4: Lagrange's Equations of Motion
4.5: Euler's Equations of Motion
4.6: Force-free Motion of a Rigid Asymmetric Top
4.7: Nonrigid Rotator
4.8: Force-free Motion of a Rigid Symmetric Top
4.9: Centrifugal and Coriolis Forces
4.10: The Top
4.11: Appendix

This page titled 4: Rigid Body Rotation is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1
4.1: Introduction to Rigid Body Rotation
No real solid body is perfectly rigid. A rotating nonrigid body will be distorted by centrifugal force* or by interactions with other
bodies. Nevertheless most people will allow that in practice some solids are fairly rigid, are rotating at only a modest speed, and
any distortion is small compared with the overall size of the body. No excuses, therefore, are needed or offered for analyzing, to
begin with the rotation of a rigid body.
*I do not in this chapter delve deeply into whether there really is “such thing” as “centrifugal force”. Some would try to persuade
us that there is no such thing. But is there “such thing” as a “gravitational force”? And is one any more or less “real” than the
other? These are deep questions best left to the philosophers. In physics we use the concept of “force” – or indeed any other
concept – according to whether it enables us to supply a description of how physical bodies behave. Many of us would, I think, be
challenged if we were faced with an examination question: “Explain, without using the term centrifugal force, why Earth bulges at
its equator.”
We have already discussed some aspects of solid body rotation in Chapter 2 on Moment of Inertia, and indeed the present Chapter 4
should not be plunged into without a good understanding of what is meant by “moment of inertia”. One of the things that we found
was that, while the comfortable relation L = I ω we are familiar with from elementary physics is adequate for problems in two
dimensions, in three dimensions the relation becomes L = lω , where l is the inertia tensor, whose properties were discussed at
some length in Chapter 2. We also learned in Chapter 2 about the concepts of principal moments of inertia, and we introduced the
notion that, unless a body is rotating about one of its principal axes, the equation L = Iω implies that the angular momentum and
angular velocity vectors are not in the same direction. We shall discuss this in more detail in this chapter.
A full treatment of the rotation of an asymmetric top (whose three principal moments of inertia are unequal and which has as its
momental ellipsoid a triaxial ellipsoid) is very lengthy, since there are so many cases to consider. I shall restrict consideration of the
motion of an asymmetric top to a qualitative argument that shows that rotation about the principal axis of greatest moment of
inertia or about the axis of least moment of inertia is stable, whereas rotation about the intermediate axis is unstable.
I shall treat in more detail the free rotation of a symmetric top (which has two equal principal moments of inertia) and we shall see
how it is that the angular velocity vector precesses while the angular momentum vector (in the absence of external torques) remains
fixed in magnitude and direction.
I shall also discuss the situation in which a symmetric top is subjected to an external torque (in which case L is certainly not fixed),
such as the motion of a top. A similar situation, in which Earth is subject to external torques from the Sun and Moon, causes
Earth’s axis to precess with a period of 26,000 years, and this will be dealt with in a chapter of the notes on Celestial Mechanics.
Before discussing these particular problems, there are a few preparatory topics, namely, angular velocity and Eulerian angles,
kinetic energy, Lagrange’s equations of motion, and Euler’s equations of motion.

This page titled 4.1: Introduction to Rigid Body Rotation is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

4.1.1 https://phys.libretexts.org/@go/page/6946
4.2: Angular Velocity and Eulerian Angles
Let Oxyz be a set of space-fixed axis, and let Ox 0 y0 z0 be the body-fixed principal axes of a rigid body.

The orientation of the body-fixed principal axes Ox y z with respect to the space-fixed axes Oxyz can be described by the three
0 0 0

Euler angles: θ , ϕ , and ψ . These are illustrated in Figure IV.1a. Those who are not familiar with Euler angles or who would like a
reminder can refer to their detailed description in Chapter 3 of my notes on Celestial Mechanics.
We are going to examine the motion of a body that is rotating about a non-principal axis. If the body is freely rotating in space with
no external torques acting upon it, its angular momentum L will be constant in magnitude and direction. The angular velocity
vector ω, however, will not be constant, but will wander with respect to both the space-fixed and body-fixed axes, and we shall be
examining this motion. I am going to call the instantaneous components of ω relative to the body-fixed axes ω , ω , ω , and its
1 2 3

magnitude ω. As the body tumbles over and over, its Euler angles will be changing continuously. We are going to establish a
geometrical relation between the instantaneous rates of change of the Euler angles and the instantaneous components of ω. That is,
we are going to find how ω , ω and ω are related to θ̇ , ϕ̇ and ψ̇ .
1 2 3

I have indicated, in Figure VI.2a, the angular velocities ˙ ˙

θ, ϕ and ˙
ψ as vectors in what I hope will be agreed are the appropriate
directions.

It should be clear that ω is equal to the x -component of ϕ

1 0
˙
plus the x -component of θ˙ and that ω is equal to the y -component
0 2 0

of ϕ˙
plus the y -component of θ˙ and that ω is equal to the z -component of ϕ
0 3 0
˙
plus ψ
˙

Let us look at Figure IV.2b

4.2.1 https://phys.libretexts.org/@go/page/6947
We see that the x0 and y0 components of ˙
θ are ˙
θ cos ψ and ˙
−θ sin ψ respectively. The x0 , y0 and z0 components of ˙
ϕ are,
respectively:
˙
ϕ sin θ sin ψ ,
ϕ̇ sin θ sin ψ , and
˙
ϕ cos θ .
Hence we arrive at
˙ ˙
ω1 = ϕ sin θ sin ψ + θ cos ψ. (4.2.1)

˙ ˙
ω2 = ϕ sin θ cos ψ − θ sin ψ. (4.2.2)

ω3 = ϕ̇ cos θ + ψ̇ (4.2.3)

Template:HypTest

This page titled 4.2: Angular Velocity and Eulerian Angles is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

4.2.2 https://phys.libretexts.org/@go/page/6947
4.3: Kinetic Energy of Rigid Body Rotation
1
Most of us are familiar with the formula Iω
2
for the rotational kinetic energy of a rotating solid body. This formula is adequate
2
for simple situations in which a body is rotating about a principal axis, but is not adequate for a body rotating about a non-principal
axis.
I am going to think of a rotating solid body as a collection of point masses, fixed relative to each other, but all revolving with the
same angular velocity about a common axis – and those who believe in atoms assure me that this is indeed the case. (If you believe
that a solid is a continuum, you can still divide it in your imagination into lots of small mass elements.)

In Figure IV.3, I show just one particle of the rotating body. The position vector of the particle is {\bf r}. The body is rotating at
angular velocity ω. I hope you’ll agree that the linear velocity {\bf v} of the particle is (now think about this carefully) \( v =
\boldsymbol \omega \times \bf{r} \).
The rotational kinetic energy of the solid body is
1 2
1
Trot = ∑ mν = ∑ v ⋅ mv = ∑(ω × r) ⋅ p
2 2

The triple scalar product is the volume of a parallelepiped, which justifies the next step:
1
Trot = ∑ ω ⋅ (r × p)
2

All particles have the same angular velocity, so:

1 1 1
Trot = ω ⋅ ∑(r × p) = ω⋅L = ω ⋅ lω
2 2 2

Thus we arrive at the following expressions for the rotational kinetic energy:
1 1
 Trot = ωL = ωIω (4.3.1)
2 2

If the body is rotating about a nonprincipal axis, the vectors ω and L are not parallel (we shall discuss this in more detail in later
1
sections). If it is rotating about a principal axis, they are parallel, and the expression reduces to the familiar Iω
2

In matrix notation, this can be written

1
 Trot = ω̃Iω (4.3.2)
2

Here I is the inertia tensor, ω is a column vector containing the rectangular components of the angular velocity and ω̃ is its
transpose, namely a row vector.
That is:

A −H −G ωx
⎛ ⎞⎛ ⎞

Trot = (ωx ωy ωz ) ⎜ −H B −F ⎟ ⎜ ωy ⎟ (4.3.3)

⎝ ⎠⎝ ⎠ (4.3.4)
−G −F C ωz

1
2 2 2
= (Aωx + Bωy + C ωz − 2F ωy ωz − 2Gωz ωx − 2H ωx ωy (4.3.1)
2

4.3.1 https://phys.libretexts.org/@go/page/6948
This expression gives the rotational kinetic energy when the components of the inertia tensor and the angular velocity vector are
referred to an arbitrary set of axes. If we refer them to the principal axes, the off-diagonal elements are zero. I am going to call the
principal moments of inertia I , I and I . (I could call them A , B and C , but I shall often use the convention that A < B < C ,
1 2 3

and I do not want to specify at the present which of the three moments is the greatest and which is the greatest, so I’ll call them I , 1

I and I , with I = ∑ m(y + z , etc.). I’ll also call the angular velocity components referred to the principal axes ω , ω , ω .
2 2
2 3 1 1 2 3

Referred, then, to the principal axes, the rotational kinetic energy is

1
2 2 2
 T = (I1 ω + I2 ω + I3 ω ). (4.3.5)
1 2 3
2

I have now dropped the subscript “rot”, because in this chapter I am dealing entirely with rotational motion, and so T can safely be
understood to mean rotational kinetic energy.
We can also now write the kinetic energy in terms of the rates of change of the Eulerian angles, and the expression we obtain will
be useful later when we derive Euler’s equations of motion:
1 2
1 2
˙ ˙ ˙ ˙ ˙ ˙ 2
T = I1 (ϕ sin ψ + θ cos ψ ) + I2 (ϕ sin θcosψ − θ sin ψ ) + I1 (ϕ cos θ + ψ ) (4.3.6)
2 2

You will probably want a concrete example in order to understand this properly,

 Example 4.3.1

Let us imagine that we have a concrete brick of dimensions 10 cm x 15 cm x 20 cm, and of density 4 g cm-3 , and that it is
rotating about a body diameter (the ends of which are fixed) at an angular speed of 6 rad s-1 .

I hope you’ll agree that the mass is 12000 g = 12 kg.

The principal moment of inertia about the vertical axis is
1
I3 = × 12000 × (10
2
+ 7.5 )
2
= 625,000 g cm2 = 0.0625 kg m2.
3

Similarly the other principal moments are

I1 = 0.0500 kg m2 and I = 0.0325 kg m2 .
2

The direction cosines of the vector ω are

15 20 10
, ,
−−−−−−−−−−− − −−−−−−−−−−− − −−−−−−−−−−− −
√ 152 + 202 + 102 √ 152 + 202 + 102 √ 152 + 202 + 102

Therefore ω = 3.34252 rad s-1, ω = 4.45669 rad s-1, ω = 2.22834 rad s-1 .
1 2 3

Hence T = 0.02103 J

This page titled 4.3: Kinetic Energy of Rigid Body Rotation is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated
by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon

4.3.2 https://phys.libretexts.org/@go/page/6948
request.

4.3.3 https://phys.libretexts.org/@go/page/6948
4.4: Lagrange's Equations of Motion
In Section 4.5 I want to derive Euler’s equations of motion, which describe how the angular velocity components of a body change
when a torque acts upon it. In deriving Euler’s equations, I find it convenient to make use of Lagrange’s equations of motion. This
will cause no difficulty to anyone who is already familiar with Lagrangian mechanics. Those who are not familiar with Lagrangian
mechanics may wish just to understand what it is that Euler’s equations are dealing with and may wish to skip over their derivation
at this stage. Later in this series, I hope to add a longer chapter on Lagrangian mechanics, when all will be made clear (maybe). In
the meantime, for those who are not content just to accept Euler’s equations but must also understand their derivation, this section
gives a five-minute course in Lagrangian mechanics.
To begin with, I have to introduce the idea of generalized coordinates and generalized forces.
The geometrical description of a mechanical system at some instant of time can be given by specifying a number of coordinates.
For example, if the system consists of just a single particle, you could specify its rectangular coordinates xyz or its cylindrical
coordinates ρϕz, or its spherical coordinates rθϕ. Certain theorems to be developed will be equally applicable to any of these, so
we can think of generalized coordinates q q q , which could mean any one of the rectangular, cylindrical of spherical set.
1 2 3

In a more complicated system, for example a polyatomic molecule, you might describe the geometry of the molecule at some
instant by a set of interatomic distances plus a set of angles between bonds. A fairly large number of distances and angles may be
necessary. These distances and angles can be called the generalized coordinates. Notice that generalized coordinates need not
always be of dimension L. Some generalized coordinates, for example, may have the dimensions of angle.
[See Appendix of this Chapter for a brief discussion as to whether angle is a dimensioned or a dimensionless quantity.]
While the generalized coordinates at an instant of time describe the geometry of a system at an instant of time, they alone do not
predict the future behaviour of the system.
I now introduce the idea of generalized forces. With each of the generalized coordinates there is associated a generalized force.
With the generalized coordinate q there is associated a corresponding generalized force P . It is defined as follows. If, when the
i i

generalized coordinate q increases by δq , the work done on the system is P δq then P is the generalized force associated with
i i i i i

the generalized coordinate q . For example, in our simple example of a single particle, if one of the generalized coordinates is
i

merely the x-coordinate, the generalized force associated with x is the x-component of the force acting on the particle.
Note, however, that often one of the generalized coordinates might be an angle. In that case the generalized force associated with it
is a torque rather than a force. In other words, a generalized force need not necessarily have the dimensions MLT-2.
Before going on to describe Lagrange’s equations of motion, let us remind ourselves how we solve problems in mechanics using
Newton’s law of motion. We may have a ladder leaning against a smooth wall and smooth floor, or a cylinder rolling down a
wedge, the hypotenuse of which is rough (so that the cylinder does not slip) and the smooth base of which is free to obey Newton’s
third law of motion on a smooth horizontal table, or any of a number of similar problems in mechanics that are visited upon us by
our teachers. The way we solve these problems is as follows. We draw a large diagram using a pencil , ruler and compass. Then we
mark in red all the forces, and we mark in green all the accelerations. If the problem is a two-dimensional problem, we write
F = ma in any two directions; if it is a three-dimensional problem, we write F = ma in any three directions. Usually, this is easy

and straightforward. Sometimes it does not seem to be as easy as it sounds, and we may prefer to solve the problem by Lagrangian
methods.
To do this, as before, we draw a large diagram using a pencil , ruler and compass. But this time we mark in blue all the velocities
(including angular velocities).

Lagrange, in the Introduction to his book La méchanique analytique (modern French

spelling omits the h) pointed out that there were no diagrams at all in his book, since all
of mechanics could be done analytically – hence the title of the book. Not all of us,
however, are as mathematically gifted as Lagrange, and we cannot bypass the step of
drawing a large, neat and clear diagram.
Having drawn in the velocities (including angular velocities), we now calculate the kinetic energy, which in advanced texts is often
given the symbol T , presumably because potential energy is traditionally written U or V . There would be no harm done if you
prefer to write E , E and E for kinetic, potential and total energy. I shall stick to T , U or V , and E .
k p

4.4.1 https://phys.libretexts.org/@go/page/6949
Now, instead of writing F = ma , we write, for each generalized coordinate, the Lagrangian equation (whose proof awaits a later
chapter):
d ∂T ∂T
  ( )− = Pi (4.4.1)
dt ∂q̇ i
∂q̇ i

The only further intellectual effort on our part is to determine what is the generalized force associated with that coordinate. Apart
from that, the procedure goes quite automatically. We shall use it in use in the next section.
That ends our five-minute course on Lagrangian mechanics.

This page titled 4.4: Lagrange's Equations of Motion is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

4.4.2 https://phys.libretexts.org/@go/page/6949
4.5: Euler's Equations of Motion
In our first introduction to classical mechanics, we learn that when an external torque acts on a body its angular momentum
changes (and if no external torques act on a body its angular momentum does not change.) We learn that the rate of change of
angular momentum is equal to the applied torque. In the first simple examples that we typically meet, a symmetrical body is
rotating about an axis of symmetry, and the torque is also applied about this same axis. The angular momentum is just I ω, and so
the statement that torque equals rate of change of angular momentum is merely τ = I ω̇ and that’s all there is to it.
Later, we learn that L = I ω, where l is a tensor, and L and ω are not parallel. There are three principal moments of inertia, and L ,
ω and the applied torque τ each have three components, and the statement “torque equals rate of change of angular momentum”

somehow becomes much less easy.

Euler’s Equations sort this out, and give us a relation between the components of the τ , l and ω.
For Figure IV.5, I have just reproduced, with some small modifications, Figure III.19 from my notes on this Web site on Celestial
Mechanics, where I defined Eulerian angles. Again it is suggested that those who are unfamiliar with Eulerian angles consult
Chapter III of Celestial Mechanics.
In Figure IV.5, Oxyz are space-fixed axes, and Ox y z are the body-fixed principal axes. The axis
0 0 0 Oy0 is behind the plane of
your screen; you will have to look inside your monitor to find it.

I suppose an external torque τ acts on the body, and I have drawn the components τ and τ . Now let’s suppose that the body
1 3

rotates in such a manner that the Eulerian angle ψ were to increase by δψ . I think it will be readily agreed that the work done on
the body is τ δψ. This means, following our definition of generalized force in Section 4.4, that τ is the generalized force
3 3

associated with the generalized coordinate ψ . Having established that, we can now apply the Lagrangian Equation 4.4.1:
d ∂T ∂T
  ( )− = τ3 (4.5.1)
dt ∂ψ̇ ∂ψ

Here the kinetic energy is the expression that we have already established in Equation 4.3.6. In spite of the somewhat fearsome
aspect of Equation 4.3.6, it is quite easy to apply Equation 4.5.1 to it. Thus
∂T
˙
  = I3 (ϕ cosθ + ψ) = I3 ω (4.5.2)
˙
∂ψ

where I have made use of Equation 4.2.3.

Therefore
d ∂T
  ( ) = I3 ω
˙3 (4.5.3)
dt ∂ψ̇

And, if we make use of Equations 4.2.1,2,3, it is easy to obtain

∂T
  ) = I1 ω1 ω2 − I2 ω2 ω2 = ω1 ω2 (I1 − I2 ) (4.5.4)
∂ψ

Thus Equation 4.5.1 becomes:

4.5.1 https://phys.libretexts.org/@go/page/6950
  I3 ω
˙3 − (I1 − I2 )ω1 ω1 = τ3 (4.5.5)

This is one of the Eulerian Equations of motion.

Now, although we saw that τ is the generalized force associated with the coordinate y, it will we equally clear that τ is not the
3 1

generalized force associated with q, nor is τ the generalized force associated with ϕ . However, we do not have to think about what
2

the generalized forces associated with these two coordinates are; it is much easier than that. To obtain the remaining two Eulerian
Equations, all that is necessary is to carry out a cyclic permutation of the subscripts in Equation 4.5.5. Thus the three Eulerian
Equation are:
 I1 ω
˙1 − (I2 − I2 )ω2 ω3 = τ1 , (4.5.6)

 I2 ω
˙2 − (I3 − I1 )ω3 ω1 = τ2 , (4.5.7)

 I3 ω
˙3 − (I1 − I2 )ω1 ω2 = τ3 . (4.5.8)

These take the place of τ = I ω̇ which we are more familiar with in elementary problems in which a body is rotating about a
principal axis and a torque is applied around that principal axis.
If there are no external torques acting on the body, then we have Euler’s Equations of free rotation of a rigid body:
˙1 = (I2 − I3 )ω2 ω3 ,
 I1 ω (4.5.9)

˙2 = (I3 − I1 )ω3 ω1 ,
 I1 ω (4.5.10)

 I3 ω
˙3 = (I1 − I2 )ω1 ω2 . (4.5.11)

 Example 4.5.1

In the above drawing, a rectangular lamina is spinning with constant angular velocity ω between two frictionless bearings. We
are going to apply Euler’s Equations of motion to it. We shall find that the bearings are exerting a torque on the rectangle, and
the rectangle is exerting a torque on the bearings. The angular momentum of the rectangle is not constant – at least it is not
constant in direction. We shall calculate the torque (its magnitude and its direction) and see what is happening to the angular
momentum.
We note that the principal (second) moments of inertia are
1 2 1 2 1 2 2
I1 = mb I2 = ma I3 = m(a +b )
3 3 3

and that the components of angular velocity are

ω1 = ω cos θ ω2 = ω sin θ ω3 = 0.

4.5.2 https://phys.libretexts.org/@go/page/6950
Also, ω̇ and all of its components are zero. We immediately obtain, from Euler’s Equations, that τ1 and τ are zero, and that
2

the torque exerted on the rectangle by the bearings is

1 2 2 2
τ3 = (I2 − I1 )ω1 ω2 = m(a − b )ω sinθ cos θ
3

And since
b b
sin θ = and cos θ = ,
√a2 +b2 √a2 +b2

we obtain
2 2
m( a −b )ab
2
τ3 = 2
ω
3( a2 +b )

Thus τ , the torque exerted on the rectangle by the bearings is directed normal to the plane of the rectangle (out of the plane of
the paper in the instantaneous snapshot above).
The angular momentum is given by L = lω . That is to say:
2
L1 b 0 0 ω cos θ
⎛ ⎞ ⎛ ⎞⎛ ⎞
1 2
⎜ L2 ⎟ = m⎜ 0 a 0 ⎟ ⎜ ω sin θ ⎟
3
⎝ ⎠ ⎝ 2 2 ⎠⎝ ⎠
L3 0 0 a +b 0

2
1 2 1 ab
L1 = m b ω cos θ = m ω
3 3 √a2 +b
2

2
1 2 1 ab
L2 = m b ω sin θ = m ω
3 3 √a2 +b2

L3 = 0

1
L = mabω
3

2
a sinθ
L2 / L1 = 2
= cot θ = tan(90° − θ)
b cosθ

This tells us that L is in the plane of the rectangle, and makes an angle 90° - θ with the x-axis, or q with the y -axis, and it
rotates around the vector τ . τ is perpendicular to the plane of the rectangle, and of course the change in L takes place in that
direction. The torque does no work, and ω and T are constant. The reader might find an analogy in the situation of a planet in
orbit around the Sun in a circular orbit.. The planet experiences a force that is always perpendicular to its velocity. The force
does no work, and the speed and kinetic energy remain constant.

4.5.3 https://phys.libretexts.org/@go/page/6950
 The torque on the plate can be represented as a couple of forces exerted by the bearings on the plate, each of magnitude
2 2
τ3 m( a −b )

2
, or 3
2
ω Forces exerted by the plate on the bearings are, of course, in the opposite direction.
2 √a +b
2
2 2
6( a −b ) 2

 Example 4.5.2

Figure IV.6 shows a disc of mass m, radius a , spinning at a constant angular speed ω about at axle that is inclined at an angle θ
to the normal to the disc. I have drawn three body-fixed principal axes. The x- and y - axes are in the plane of the
disc\boldsymbol; the direction of the x-axis is chosen so that the axle (and hence the vector ω ) is in the zx-plane. The disc is
evidently unbalanced and there must be a torque on it to maintain the motion.
Since ω is constant, all components of ω̇ are zero, so that Euler’s Equations are
τ1 = (I3 − I2 )ω3 ω2 ,

τ2 = (I1 − I3 )ω1 ω3 ,

τ3 = (I2 − I1 )ω2 ω1 ,

Now ω 1 = ω sin θ, ω2 = ω cos θ, I1 =

1

4
2
m a , I2 =
1

4
2
m a , I3 =
1

1
ma
2

Therefore τ 1 = τ3 = 0, andτ2 = −
1

4
2 2
m a ω sinθcosθ = −
1

8
2 2
m a ω sin2θ

(Check, as always, that this expression is dimensionally correct.) Thus the torque acting on the disc is in the negative y -
direction.
Can you reconcile the fact that there is a torque acting on the disc with the fact that is it moving with constant angular velocity?
Yes, most decidedly! What is not constant is the angular momentum L , which is moving around the axle in a cone such that
L = −τ j , where j is the unit vector along the y -axis.
˙
2

This page titled 4.5: Euler's Equations of Motion is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

4.5.4 https://phys.libretexts.org/@go/page/6950
4.6: Force-free Motion of a Rigid Asymmetric Top
By “asymmetric top” I mean a body whose three principal moments of inertia are unequal. While we often think of a “top” as a
symmetric body spinning on a table, in this section the “top” will not necessarily be symmetric, and it will not be in contact with
any table, nor indeed subjected to any external forces or torques.
A complete description of the motion of an asymmetric top is quite complicated, and therefore all that we shall attempt in this
chapter is a qualitative description of certain aspects of the motion. That our description is going to be “qualitative” does not by any
means imply that this section is not going to be replete with equations or that we can give our poor brains a rest.
The first point that we can make is that, provided that no external torques act on the body, its angular momentum L is constant in
magnitude and direction. A second point is that, provided the body is rigid and has no internal degrees of freedom, the rotational
kinetic energy T is constant. I deal briefly with nonrigid bodies in Section 4.7. Although the angular velocity vector ω is by no
means fixed in either magnitude and direction, and the body can tumble over and over, these two conditions impose some
constraints of the magnitude and direction of ω.
We are going to examine these two conditions to see what constraints are imposed on ω. One of the things we shall find is that
rotation of a body about a principal axis of greatest or of least moment of inertia is stable against small displacements, whereas
rotation about the principal axis of intermediate moment of inertia is unstable.
Absence of an external torque means that the angular momentum is constant:
2 2 2 2
L =L +L +L = constant, (4.6.1)
1 2 3

so that, at all times,

2 2 2 2 2 2 2
I ω +I ω +I ω =L (4.6.2)
1 1 2 2 3 3

Thus, for a given L. the angular velocity components always satisfy

2 2 2
ω ω ω
1 2 3
+ + = 1. (4.6.3)
2 2 2
(L/I1 ) (L/I2 ) (L/I3 )

That is to say, the angular velocity vector is constrained such that the tip of the vector ω is always on the surface of an ellipsoid of
semi axes , , ,
L

I1
L

I2
L

I3

In addition to the constancy of angular momentum, the kinetic energy is also constant:
1 1 1
2 2 2
I1 ω + I2 ω + I3 ω =T (4.6.4)
1 2 3
2 2 2

Thus the tip of the angular velocity vector must also be on the surface of the ellipsoid
2 2 2
ω ω ω
1 2 3

−−−−− + −−−−− + −−−−− = 1. (4.6.5)

2 2 2
(√2T /I1) (√2T /I2) (√2T /I3)

−−−−− −−−−− −−−−−

This ellipsoid (which is similar in shape to the momental ellipsoid) has semi axes √2T /I 1,
√2T /I2, √2T /I3.

Thus, how ever the body tumbles over and over, ω is constrained in magnitude and direction so that its tip is on the curve where
these two ellipses intersect.

 Example 4.6.1

Suppose, that we have a rigid body with

I1 = 0.2 kgm2
I2 = 0.3 kgm2
I3 = 0.5 kgm2
and that we set it in motion such that the angular momentum and kinetic energy are L = 4 J s and T = 20 J.
(The angular momentum and kinetic energy will be determined by the magnitude and direction of the initial velocity vector by
which it is set in motion.)

4.6.1 https://phys.libretexts.org/@go/page/6951
The tip of ω is constrained to be on the curve of intersection of the two ellipsoids
2 2 2
ω ω ω
1 2 3
+ + =1 (4.6.6)
2 2 2
20 13.3 8

and
2 2 2
ω ω ω
1 2 3
+ + =1 (4.6.7)
2 2 2
14.14 11.55 8.94

It is not easy (or I do not find it so) to imagine what this curve of intersection looks like in three-dimensional space, but one of
my students, Leif Petersen, prepared the drawing below, and I am grateful to him for permission to reproduce it here. You can
see that the curve of intersection is not a plane curve.
In case it’s of any help, you might want to note that equations 4.6.6 and 4.6.7 can be written
2 2 2
4ω + 9ω + 25 ω = 1600 (4.6.8)
1 2 3

and
2 2 2
2ω + 3ω + 5ω = 400 (4.6.9)
1 2 3

but I’m going to leave the equations in the form 4.6.6 and 4.6.7, and in figure IV.7, I’ll sketch one octant of the two ellipsoidal
surfaces.

4.6.2 https://phys.libretexts.org/@go/page/6951
The continuous blue curve shows an octant of the ellipsoid L = constant, and the dashed black curve shows an octant of the
ellipsoid T = constant. The angular momentum vector can end only on the curve (not drawn) where the two ellipsoids
intersect. Two points on the curve are indicated in Figure IV.7. If, for example, ω is oriented so that ω = 0, the other two 1

components must be ω = 8.16 and ω = 6.32. If it is oriented so that ω = 0, the other two components must be ω = 7.30 and
2 3 3 3

ω = 8.16. If ω = 0, there are no real solutions for ω and ω . This means that, for the given values of L and T , ω cannot be
1 3 1 2 3

zero.
Now I’m going to address myself to the stability of rotation when a symmetric top is initially set to spin about one of its
principal axes, which I’ll take to be the z -axis. We’ll suppose that initially ω = ω = 0, and ω = Ω. In that case the angular
1 2 3

momentum and the kinetic energy are L = I Ω In any subsequent motion, the tip of T = I Ω is restricted to move along the
3 3
2

curve of intersection of the ellipsoids given by equations 4.6.3 and 4.6.5. That is to say, along the curve of intersection of the
ellipsoids
2 2 2
ω ω ω
1 2 3
+ + =1 (4.6.10)
I3 I3 2
( Ω) 2
( Ω) 2 Ω
I1 I2

and
2 2 2
ω ω ω
1 2 3
+ + =1 (4.6.11)
−
− −
− 2
I3
2
I3
2 Ω
(√ Ω) (√ Ω)
I1 I2

For a specific example, I’ll suppose that the moments of inertia are in the ratio 2 : 3 : 5 , and we’ll consider three cases in turn.
Case I. Rotation about the axis of least moment of inertia. That is, we’ll take I 3 =2 ,I
1 =3 ,I 2 =5 . Since I is the smallest
3
−
− −
−
I3 I3 I3 I3 I3 I3
moment of inertia, each of the ratios I1
and I2
are less than 1, and √ I1
>
I1
and√
I2
>
I2
The two ellipsoids are
2 2 2
ω ω ω
1 2 3
+ + =1 (4.6.12)
2 2 2
(0.667Ω) (0.400Ω) Ω

and
2 2 2
ω ω ω
1 2 3
+ + =1 (4.6.13)
2 2 2
(0.816Ω) (0.632Ω) Ω

4.6.3 https://phys.libretexts.org/@go/page/6951
I’ll try and sketch these:

Initially, we suppose, the body was set in motion rotating about the z -axis with angular speed Ω, which determines the values
of L and T , which will remain constant. The tip of the vector ω is constrained to remain on the surface of the ellipsoid L = 0
and on the ellipsoid T = 0, and hence on the intersection of these two surfaces. But these two surface touch only at one point,
namely (ω , ω , ω ) = (0 , 0 , Ω). Thus there the vector ω remains, and the rotation is stable.
1 2 3

Case II. Rotation about the axis of greatest moment of inertia. That is, we’ll take I3 = 5 , I1 = 2 , I2 = 3 . Since I3 is the
−
− −
−
I3 I3 I3 I3 I3 I3
greatest moment of inertia, each of the ratios I1
and I2
are greater than 1, and √ I1
<
I1
and √ I2
<
I2
The two ellipsoids
are
2 2 2
ω ω ω
1 2 3
+ + =1 (4.6.14)
2 2 2
(2.50Ω) (1.67Ω) Ω

and
2 2 2
ω ω ω
1 2 3
+ + =1 (4.6.15)
2 2 2
(1.58Ω) (1.29Ω) Ω

I’ll try and sketch these:

4.6.4 https://phys.libretexts.org/@go/page/6951
Again, and for the same reason as for Case I, we see that this motion is stable.
I3 I3
Case III. Rotation about the intermediate axis. That is, we’ll take I 3 =3 ,I 1 =5 ,I
2 =2 . This time I1
is less than 1 and I2
is
−
− −
−
I3 I3 I3 I3
less than 1, and √ I1
<
I1
and √ I2
<
I2
The two ellipsoids are
2 2 2
ω ω ω
1 2 3
+ + =1 (4.6.16)
2 2
(0.60Ω) (1.50Ω) Ω

and
2 2 2
ω ω ω
1 2 3
+ + =1 (4.6.17)
2 2 2
(0.77Ω) (1.22Ω) Ω

I'll try and sketch these:

4.6.5 https://phys.libretexts.org/@go/page/6951
 Unlike the situation for Cases I and II, in which the two ellipsoids touch at only a single point, the two ellipses for Case III
intersect in the curve shown as a dotted line in figure IV.10. Thus ω is not restricted to lying along the z -axis, but it can move
anywhere along the dotted line. The motion, therefore, is not stable.

You should experiment by throwing a body in the air in such a manner as to let it spin around one of its principal axes. A
rectangular block will do, though the effect is particularly noticeable with something like a table-tennis bat.
Here is another approach to reach the same result. We imagine an asymmetric top spinning about one of its principal axes with
angular velocity \( \boldsymbol\omega \ = \omega \hat{\bf z}). It is then given a small perturbation, so that its angular velocity is
now.
ω = ϵx
^ + ηy
^ + ωz z
^ (4.6.18)

Here the “hatted” quantities are the unit orthogonal vectors; ϵ and η are supposed small compared with ω . Euler’s equations are :
z

I1 ϵ̇ = η ωZ (I2 − I3 ), (4.6.19)

I1 η̇ = ωZ ϵ(I3 − I1 ), (4.6.20)

I1 ω˙Z = ϵη(I1 − I2 ). (4.6.21)

If ϵη << ω˙ Z, then ω is approximately constant. Elimination of η from the first two equations yields
Z

2
(I2 − I3 )(I1 − I3 )ωz
ϵ̈ = −[ ]ϵ (4.6.22)
I1 I2

Elimination of ϵ instead results in a similar equation in η .

If I is either the largest or the smallest of the three moments of inertia, the two parentheses in the denominator have the same sign,
3

so the expression in the brackets is positive. Equation 4.6.22 is then the equation for simple harmonic motion, and the motion is
stable. If, however, I is intermediate between the other two, the two parentheses have opposite sign, and the expression in brackets
3

in negative. In that case ϵ and η increase exponentially, and the motion is unstable.
Mr Neil Honkanen of the University of Victoria conducted an experiment to illustrate the stability of rotation about the three
principal axes. The body in question was a small “brick” of mild steel (density 7.83 g/cm3) of dimensions 3/8 inch × 3/4 inch × 1
1/2 inch, mass 54.1 g. In round figures, this corresponds to principal moments of inertia A = 2 ×10-6 kg m2, B = 7 × 10-6 kg m2,
0 0

C = 8 × 10
-6 kg m2. He suspended it from an electromagnet, which he set in rotation at about 25 revolutions per second, and then
0

let it fall, while photographing it stroboscopically. He did three experiments rotation respectively about the three principle axes.
You can see from the photographs below that the rotation is stable when the rotation is about the axes of greatest or least moment
of inertia, but is unstable when the rotation is about the axis of intermediate moment of inertia.

4.6.6 https://phys.libretexts.org/@go/page/6951
This page titled 4.6: Force-free Motion of a Rigid Asymmetric Top is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or
curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is
available upon request.

4.6.7 https://phys.libretexts.org/@go/page/6951
4.7: Nonrigid Rotator
The rotational kinetic energy of a body rotating about a principal axis is I ω , where I is the moment of inertia about that principal
1

2
3

axis, and the angular momentum is L = Iω. (For rotation about a nonprincipal axis, see Section 4.3.) Thus the rotational kinetic
energy can be written as
2
L
K Erot = . (4.7.1)
2I

When an asymmetric top is rotating about a nonprincipal axes, the body experiences internal stresses, which, if the body is
nonrigid, result in periodic strains which periodically distort the shape of the body. As a result of this, rotational kinetic energy
becomes degraded into heat; the rotational kinetic energy of the body gradually decreases. In the absence of external torques,
however, the angular momentum is constant. Equation 4.7.1 shows that the kinetic energy is least for a given angular momentum
when the moment of inertia is greatest. Thus eventually the body rotates about its principal axis of greatest moment of inertia. After
that, it no longer loses kinetic energy to heat, because, when the body is rotating about a principal axis, it is no longer subject to
internal stresses.
The time taken (the “relaxation time”) for a body to reach its final state of rotation about its principal axis of greatest moment of
inertia depends, among other things, on how fast the body is rotating. A fast rotator will reach its final state relatively soon,
whereas it takes a long time for a slow rotator to reach its final state. Thus it is not surprising to find that, among the asteroids, most
of the fast rotators are principal axis rotators, whereas many slow rotators are also nonprincipal axis rotators. There are, however, a
few fast rotators that are still rotating about a nonprincipal axis. It is assumed that such asteroids may have suffered a collision in
the recent past.

This page titled 4.7: Nonrigid Rotator is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

4.7.1 https://phys.libretexts.org/@go/page/8394
4.8: Force-free Motion of a Rigid Symmetric Top
Notation:
I1, I , I are the principal moments of inertia. I is the unique moment. If it is the largest of the three, the body is an oblate
2 3 3

symmetric top; if it is the smallest, it is a prolate spherical top.

Ox , Oy , Oz are the corresponding body-fixed principal axes.
0 0 0

ω , ω , ω are the components of the angular velocity vector ω with respect to the principal axes.
1 2 3

In the analysis that follows, we are going to have to think about three vectors. There will be the angular momentum vector L ,
which, in the absence of external torques, is fixed in magnitude and in the direction in laboratory space. There will be the direction
of the axis of symmetry, the Oz axis, which is fixed in the body, but not necessarily in space, unless the body happens to be
0

rotating about its axis of symmetry; we’ll denote a unit vector in this direction by z
^ . And there will be the instantaneous angular
0

velocity vector ω which is neither space- nor body-fixed.

What we are going to find is the following. We shall find that ω precesses in the body about the body-fixed symmetry axis in a
cone called the body cone. The angle between ω and ^ z is constant (we’ll be calling this angle α ), and the magnitude ω of ω of is
0

constant. We shall find that the sense of the precession is the same as the sense of the spin if the body is oblate, but opposite if it is
prolate. The direction of the symmetry axis, however, is not fixed in space, but it precesses about the space-fixed angular
momentum vector L in another cone. This cone is narrower than the body cone if the body is oblate, but broader than the body
cone if the body is prolate. The net result of these two precessional motions is that precesses ω in space about the space-fixed
angular momentum vector in a cone called the space cone. For a prolate top, the semi vertical angle of the space cone can be
anything from 0° to 90°; for an oblate top, however, the semi vertical angle of the space cone cannot exceed 19° 28' . That’s quite a
lot to take in in one breath!
We can start with Euler’s equations of motion for force-free rotation of a symmetric top:
I1 ω
˙1 = −ω2 ω3 (I3 − I1 ), (4.8.1)

I1 ω
˙2 = −ω1 ω3 (I3 − I1 ), (4.8.2)

˙3 = 0.
I3 ω (4.8.3)

From the first of these we obtain the result

 ω3 = constant (4.8.4)

For brevity, I am going to let

(I3 − I1 )
= Ω, (4.8.5)
I1

although in a moment Ω will have a physical meaning.

Equations 4.8.1 and 4.8.2 become:
˙1 = Ω ω2
 ω (4.8.6)

and
 ω
˙2 = Ω ω1 (4.8.7)

Eliminate ω from these to obtain

2

2
˙1 = −Ω ω1
 ω (4.8.8)

This is the Equation for simple harmonic motion and its solution is
 ω1 = ω0 cos(Ωt + ϵ) (4.8.9)

in which ω and ϵ, the two constants of integration, whose values depend on the initial conditions in the usual fashion, are the
0

amplitude and initial phase angle. On combining this with Equation 4.8.6, we obtain
 ω2 = ω0 sin(Ωt + ϵ) (4.8.10)

4.8.1 https://phys.libretexts.org/@go/page/8395
From these we see that (ω + ω ) , which is the magnitude of the component of ω in the x y -plane, is constant, equal to ω ;
2
1
2
2
1/2
0 0 0

and since ω is also constant, it follows that (ω + ω + ω )

3 , which is the magnitude of ω, is also constant. The cosine of the
2
1
2
2
2
3
1/2

angle α between z^ and ω is ω /(ω + ω + ω ) , and its sine is ω /(ω + ω + ω ) , so that α is constant. Equations 4.8.9
0 3
2
1
2
2
2
3
1/2
0
2
1
2
2
2
3
1/2

and 4.8.10 tell us, then, that the vector ω is precessing around the symmetry axis at an angular speed Ω. Making use of Equation
4.8.5, we find that

ω3 I1 Ω
cos α = = (4.8.11)
ω (I3 − I1 )ω

If we take the direction of the z axis to be the direction of the component of ω along the symmetry axis, then Ω is in the same
0

direction as z if I > I (that is, if the top is oblate) and it is in the opposite direction if the top is prolate. The situation for oblate
0 3 1

and prolate tops is shown in Figure IV.11.

We have just dealt with how the instantaneous axis of rotation precesses about the body-fixed symmetry axis, describing the body
cone of semi vertical angle α .
Now we are going to consider the precession of the body-fixed symmetry axis about the space-fixed angular momentum vector L . I
am going to make use of the idea of Eulerian angles for expressing the orientation of one three-dimensional set of axes with respect
to another. If you are not already familiar with Eulerian angles or would like a refresher, you can go to to Chapter 3 of Celestial
Mechanics especially Section 3.7.
Recall that we are using Ox y z for body-fixed coordinates, referred to the principal axes. I shall use Oxyz for space-fixed
0 0 0

coordinates, and there is no loss of generality if I choose the Oz axis to coincide with the angular momentum vector L . Let me try
to draw the situation in Figure IV.12a. The axes Oxyz are the space-fixed axes. The axes Ox y z are the body-fixed principal 0 0 0

axes. The angular momentum vector L is directed along the axis Oz . The symmetry axis of the body is directed along the axis Oz . 0

The Eulerian angles of the body-fixed axes relative to the space fixed axes are (ϕ , θ , ψ ).
Recall, with the aid of Figure IV.12b, how these Euler angles are formed:
First, a rotation by ϕ about Oz . Second, a rotation by θ about the dashed line Ox to form an intermediate set of axes ′ ′
Ox y z
′ ′
.
Third, a rotation by ψ about Oz to form the body- fixed principal axes Ox y z .
′
0 0 0

Spend a little time trying to visualize these three sets of axes. Please also convince yourself, from the way the Euler angles were
formed through three rotations, that the vector L is in the y z plane and has no x component. It is also in the y z plane and has
′ ′
′
0 0

no x component.
0

You will then agree that

 Lx′ = 0, Ly ′ = L sin θ Lz ′ = L cos θ. (4.8.12)

Now if L x
′ =0 , then ω is also zero, which means that ω, like L , is in the y
x
′
′
z
′
plane.
We have seen that ω makes an angle α with the symmetry axis Oz , where α is given by Equation 4.8.11.0

4.8.2 https://phys.libretexts.org/@go/page/8395
 Figure 4.8.1 : Paste Caption Here
I’ll now add ω to the drawing to make Figure IV.13. Like L , it is in the y'z' plane and has no x' component. I haven’t marked in the
angle α . I leave it to your imagination. It is the angle between ω and z . You should easily agree that
0

 ωx′ = 0, ωy ′ + ω sin α, ωz ′ = ωcosα. (4.8.13)

From these, together with L y

′ = I1 ωy ′ and Lz
′ = I1 ωz ′ we obtain
 I1 = tan α = I3 tan θ (4.8.14)

For an oblate symmetric top, I 3 > I1 , α > θ.

For a prolate symmetric top, I 3 < I1 , α < θ.

Now ω can be written as the vector sum of the rates of change of the three Euler angles:

ω = θ̇ + ϕ̇ + ψ̇ (4.8.15)

The components of θ̇ and ψ̇ along Oy are each zero, and therefore the component of ω along Oy is equal to the component of ψ̇
′ ′

along Oy' .
˙
  ∴ ω sin α = ϕ sin θ (4.8.16)

4.8.3 https://phys.libretexts.org/@go/page/8395
In summary, then:
1. The instantaneous axis of rotation, which makes an angle α with the symmetry axis, precesses around it at angular speed
I3 − I1
 Ω = ω cos α (4.8.17)
I1

which is in the same sense as ω if the top is oblate and opposite if it is prolate.
2. The symmetry axis makes an angle θ with the space-fixed angular momentum vector L , where
I1
  tan θ = tan α. (4.8.18)
I3

For an oblate top, θ < α . For a prolate top, θ > α .

3. The speed of precession of the symmetry axis about L is
sin α
˙
 ϕ = ω, (4.8.19)
sin θ

or, by elimination of θ between 4.8.18 and 4.8.19,

2 2
I −I
3 1 2 1/2
 ϕ̇ = [1 + cos α] ω. (4.8.20)
1
I
3

The net result of this is that ω preceses about L at a rate ϕ̇ in the space cone, which has a semi-vertical angle α − θ for an oblate
rotator, and θ − α for a prolate rotator. The space cone is fixed in space, while the body cone rolls around it, always in contact, ω
being a mutual generator of both cones. If the rotator is oblate, the space cone is smaller than the body cone and is inside it. If the
rotator is prolate, the body cone is outside the space cone and can be larger or smaller than it.
Write
 c = I3 / I1 (4.8.21)

for the ratio of the principal moments of inertia. Note that for a pencil, c = 0 ; for a sphere, c = 1 ; for a plane disc or any regular
plane lamina, c = 2 . (The last of these follows from the perpendicular axes theorem.) The range of c , then, is from 0 to 2, 0 to 1
being prolate, 1 to 2 being oblate.
Equations 4.8.17 and 4.8.20 can be written
Ω
  = (c − 1) cos α (4.8.22)
ω

and

ϕ̇
2 2 1/2
  = [1 + (c − 1) cos α] (4.8.23)
ω

Figures IV.15 and IV.16 show, for an oblate and a prolate rotator respectively, the instantaneouss rotation vector ω precessing
around the body-fixed symmetry axis at a rate Ω in the body cone of semi vertical angle α ; the symmetry axis precessing about the
space-fixed angular momentum vector L at a rate ϕ̇ in a cone of semi vertical angle θ (which is less than α for an oblate rotator,
and greater than α for a prolate rotator; and consequently the instantaneous rotation vector ω precessing around the space-fixed
angular momentum vector L at a rate ϕ˙ in the space cone of semi vertical angle α − θ (oblate rotator) or θ − α (prolate rotator).

4.8.4 https://phys.libretexts.org/@go/page/8395
One can see from figures IV.15 and 16 that the angle between L and ω is limited for an oblate rotator, but it can be as large as 90°
for a prolate rotator. The angle between L and ω is θ − α (which is negative for an oblate rotator). We have

tan θ − tan α (1 − c) tan α

  tan(θ − α) = = (4.8.24)
2
1 + tan θ tan α c + tan α

1 −c
By calculus this reaches a maximum value of for tan α = √c
2 √c

For a rod or pencil (prolate), in which c = 0 , the angle between L and ω can be as large as 90°. Recalling exactly what are meant
by the vectors L and ω, the reader should try now and imagine in his or her mind’s eye a pencil rotating so that L and ω are at
right angles. The spin vector ω is along the length of the pencil and the angular momentum vector L is at right angles to the length
of the pencil.
For an oblate rotator, the angle between L and ω is limited. The most oblate rotator is a flat disc or any regular flat lamina. The
–
parallel axis theorem shows that for such a body, c = 2 . The greatest angle between L and ω for a disc occurs when tan α = √2α
1
= 54° 44'),and then tan α − θ = –
, α − θ =19° 28'.
√8

I3
In the following figures I illustrate some of these results graphically. The ratio goes from 0 for a pencil through 1 for a sphere to
I1

2 for a disc.

4.8.5 https://phys.libretexts.org/@go/page/8395
4.8.6 https://phys.libretexts.org/@go/page/8395
4.8.7 https://phys.libretexts.org/@go/page/8395
4.8.8 https://phys.libretexts.org/@go/page/8395
4.8.9 https://phys.libretexts.org/@go/page/8395
4.8.10 https://phys.libretexts.org/@go/page/8395
4.8.11 https://phys.libretexts.org/@go/page/8395
 (I3 − I1 )
Our planet Earth is approximately an oblate spheroid, its dynamical ellipticity being about 3.285 × 10−3. It is not
I1

rotating exactly abut its symmetry axis; the angle α between ω and the symmetry axis being about one fifth of an arcsecond, which
is about six metres on the surface. The rotation period is one sidereal day (which is a few minutes shorter than 24 solar hours.)
Equation 4.8.17 tells us that the spin axis precesses about the symmetry axis in a period of about 304 days, all within the area of a
tennis court. The actual motion is a little more complicated than this. The period is closer to 432 days because of the nonrigidity of
Earth, and superimposed on this is an annual component caused by the annual movement of air masses. This precessional motion of
a symmetric body spinning freely about an axis inclined to the symmetry axis gives rise to variations of latitude of amplitude about
a fifth of an arcsecond. It is not to be confused with the 26,000 year period of the precession of the equinoxes, which is caused by
external torques from the Moon and the Sun.

This page titled 4.8: Force-free Motion of a Rigid Symmetric Top is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or
curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is
available upon request.

4.8.12 https://phys.libretexts.org/@go/page/8395
4.9: Centrifugal and Coriolis Forces
We are usually told in elementary books that there is “no such thing” as centrifugal force. When a satellite orbits around Earth, it is
not held in equilibrium between two equal and opposite forces, namely gravity acting towards Earth and centrifugal force acting
outwards. In reality, we are told, the satellite is accelerating (the centripetal acceleration); there is only one force, namely the
gravitational force, which is equal to the mass times the centripetal acceleration.
Yet when we drive round a corner too fast and we feel ourselves flung away from the centre of curvature of our path, the
“centrifugal force” certainly feels real enough, and indeed we can solve problems referred to rotating coordinate systems as if there
“really” were such a thing as “centrifugal force”.
Let’s look at an even simpler example, not even involving rotation. A car is accelerating at a rate a towards the right. See Figure
IV.20 – but forgive my limited artistic abilities. Drawing a motor car is somewhat beyond my skills.

There is a plumb-bob hanging from the roof of the car, but, because of the acceleration of the car, it is not hanging vertically. Some
would say that there are but two forces on the plumb-bob – its weight and the tension in the string – and, as a result of these, the
bob is accelerating towards the right. By application of F = ma it is easily possible to find the tension in the string and the angle
that the string makes with the vertical.
The passenger in the car, however, sees things rather differently (Figure IV.21.)

To the passenger in the car, nothing is accelerating. The plumb-bob is merely in static equilibrium under the action of three forces,
one of which is a force ma towards the left. To the question “But what is the agent that is causing this so-called force?” I counter
with the question “What is the agent that is causing the downward force that you attribute to some mysterious ‘gravity’ ”?
It seems that, when referred to the reference frame of Figure IV.20, there are only two forces, but when referred to the accelerating
reference frame of Figure IV.21, the system can be described perfectly well by postulating the existence of a force ma pulling
towards the left. This is in fact a principle in classical mechanics, known as d’Alembert’s principle, whereby, if one refers the
description of a system to an accelerating reference frame, one can replace an acceleration with a force in the opposite direction. It
results in a perfectly valid description of the behavior of a system, and will accurately predict how the system will behave. So
who’s to say which forces are “real” and which are “fictitious”, and which reference frame is better than another?
The situation is similar with respect to centrifugal force. If you consider a satellite in orbit around Earth, some would say that there
is only one force acting on the satellite, namely the gravitational force towards Earth. The satellite, being in a circular orbit, is
accelerating towards the centre of the circle, and all is as expected - F = ma . The acceleration is the centripetal acceleration (peto
– I desire). An astronaut on board the satellite may have a different point of view. He is at a constant distance from Earth, not

4.9.1 https://phys.libretexts.org/@go/page/8396
accelerating; he is in static equilibrium, and he feels no net force on him whatever – he feels weightless. If he has been taught that
Earth exerts a gravitational force, then this must be balanced by a force away from Earth. This force, which becomes apparent
when referred to a corotating reference frame, is the centrifugal force (fugo – I flee, like a fugitive). It would need a good lawyer to
argue that the invisible gravitational force towards Earth is a real force, while the equally invisible force acting away from Earth is
imaginary. In truth, all forces are “imaginary” – in that they are only devices or concepts used in physics to describe and predict the
behavior of matter.
I mentioned earlier a possible awful examination question: Explain why Earth bulges at the equator, without using the term
“centrifugal force”. Just thank yourself lucky if you are not asked such a question! People who have tried to answer it have come
up with some interesting ideas. I have heard (I do not know whether it is true) that someone once offered a prize of $1000 to
anyone who could prove that Earth is rotating, and that the prize has never been claimed! Some have tried to imagine how you
would determine whether Earth is rotating if it were the only body in the universe. There would be no external reference points
against which one could measure the orientation of Earth. It has been concluded (by some) that even to think of Earth rotating in
the absence of any external reference points is meaningless, so that one certainly could not determine how fast, or even whether
and about what axis, Earth was rotating. Since “rotation” would then be meaningless, there would be no centrifugal force, Earth
would not bulge, nor would the Foucault pendulum rotate, nor would naval shells deviate from their paths, nor would cyclones and
anticyclones exist in the atmosphere.
Centrifugal force comes into existence only when there is an external universe. It is the external universe, then, revolving around
the stationary Earth, that causes centrifugal force and all the other effects that we have mentioned. These are deep waters indeed,
and I do not pursue this aspect further here. We shall merely take the pragmatic view that problems in mechanics can often be
solved by referring motions to a corotating reference frame, and that the behavior of mechanical systems can successfully and
accurately be described and predicted by postulating the “existence” of “inertial” forces such as centrifugal and Coriolis forces,
which make themselves apparent only when referred to a rotating frame. Thus, rather than involving ourselves in difficult questions
about whether such forces are real, we shall take things easy with just a few simple Equations.

Σ = Oxyz is an inertial reference frame (i.e. not accelerating or rotating).

′
Σ = Ox y z
′ ′
is a frame that is rotating about the z -axis with angular velocity ω = ωz^ .
Three questions:
1. If P is a point that is fixed with respect to Σ , what is its velocity v with respect to Σ?
′

2. If P is a point that is moving with velocity v with respect to Σ , what is its velocity v with respect to Σ?
′ ′

3. If P is a point that has an acceleration a with respect to Σ , what is its acceleration a with respect to Σ?
′ ′

1. The answer to the first question is, I think, fairly easy. Just by inspection of Figure XV.22, I hope you will agree that it is

 v = ω × r (4.9.1)

In case this is not clear, try the following argument. At some instant the position vector of P with respect to Σ is r . At a time δt

later its position vector is r + δr , where δr = r sin θωδt and δr is at right angles to r , and is directed along the small circle
whose zenith angle is θ , in the direction of motion of P with respect to Σ. Expressed
alternatively,

4.9.2 https://phys.libretexts.org/@go/page/8396
 ^
δr = r sin θωδtϕ .

Draw the vector δr on the Figure if it helps. Divide

both sides by δt and take the limit as δt → 0 to get ṙ . The magnitude and direction of ṙ are then expressed by the single vector
Equation ṙ = ω × r , or v = ω × r .
The only thing to look out for is this. In two-dimensional problems we are often used to expressing the relation between linear and
angular speed by v = rω , and it doesn't matter which way round we write r and ω. When we are doing a three-dimensional
problem using vector notation, it is important to remember that it is v = ω × r .and not the other way round.
2. If P is moving with velocity v with respect to Σ , then its velocity v with respect to Σ must be
′ ′

′
 v = v + ω × r (4.9.2)

This shows that

d d
 ( )Σ = ( )Σ′ + ω× (4.9.3)
dt dt

What this Equation is intended to convey is that the operation of differentiating with respect to time when referred to the inertial
frame Σ has the same result as differentiating with respect to time when referred to the rotating frame Σ , plus the operation ω×. ′

We shall understand this a little better in the next paragraph.

3. If P is accelerating with respect to Σ , we can apply the operation 4.9.3 to the Equation 4.9.2:
′

d
′ ′
a=( )Σ′ (v + ω × r) + ω × (v + ω × r)
dt

′ ′ ′
= a + ω × v + ω × v + ω × (ω × r).

′ ′
∴ a = a + ω × (ω × r) + 2ω × v (4.9.4)

This, then, anwers the third question we posed. All we have to do now is understand what it means.
To start with, let us return to the case where P is neither moving nor accelerating with respect to Σ . In that case, Equation 4.9.4 is
′

just
 a = ω × (ω × r) (4.9.5)

which we could easily have obtained by applying the operator 4.9.3 to 4.9.1. Let us try and understand what this means. In what
follows, a “hat” ( ^ ) denotes a unit vector.
We have
^
ω × r = ωr sin θϕ

and hence
^
^ × ωr sin θϕ ^ ^
ω × (ω × r) = ω z = rω sin θ bf z × ϕ

and z
^×ϕ ^
is a unit vector directed towards the z -axis. Notice that the point P is moving at angular speed ω in a small circle of
radius r sin θ . The expression a = ω × (ω × r) = rω sin θz ^×ϕ ^
2
, then, is just the familiar centripetal acceleration, of magnitude r
rω sin θ , directed towards the axis of rotation.
2

We could also think of ω × (ω × r) as a triple vector product.

We recall that

A × (B × C) = (A ⋅ C)B − (A ⋅ B)C

so that
2
ω × (ω × r) = (w ⋅ r)ω − ω r.

That is ω × (ω × r) =  bf rω 2
^ − rω ^
cosθz r
2
.

4.9.3 https://phys.libretexts.org/@go/page/8396
This can be illustrated by the vector diagram shown in Figure IV.23. The vectors are drawn in green, in accordance with my
convention of red, blue and green for force, velocity and acceleration respectively.

However, Equation 4.9.4 also tells us that, if a particle is moving with velocity v with respect to Σ , it has an additional
′ ′

acceleration with respect to Σ of 2ω × v , which is at right angles to v and to ω. This is the Coriolis acceleration.
′ ′

The converse of Equation 4.9.4 is

′ prime
a = a + ω × (r × ω) = 2 v ×ω (4.9.6)

′ ′
F = F + mω × (r × ω) + 2m v × ω (4.9.7)

It is worth now spending a few moments thinking about the direction of the Coriolis force
′
2m v × ω.

Earth is spinning on its axis with a period of 24 sidereal hours (23h and 56m of solar time.). The vector ω is directed upwards
through the north pole. Now go to somewhere on Earth at latitude 45° N. Fire a naval shell to the north. To the east. To the south.
To the west. Now go to the equator and repeat the experiment. Go to the north pole. There you can fire only due south. Repeat the
experiment at 45° south, and at the south pole. Each time, think about the direction of the vector 2mv × ω . If your thoughts are to
′

my thoughts, your mind to my mind, you should conclude that the shell veers to the right in the northern hemisphere and to the left
in the southern hemisphere, and that the Coriolis force is zero at the equator. As air rushes out of a high pressure system in the
northern hemisphere, it will swirl clockwise around the pressure centre. As it rushes in to a low pressure system, it will swirl
counterclockwise. The opposite situation will happen in the southern hemisphere.
You can think of the Coriolis force on a naval shell as being a consequence of conservation of angular momentum. Go to 45° N and
point your naval gun to the north. Your shell, while waiting in the breech, is moving around Earth’s axis at a linear speed of ωR sin
45°, where R is the radius of Earth. Now fire the shell to the north. By the time it reaches latitude 50° N, it is being carried around
Earth’s axis in a small circle of radius only R sin 40°. In order for angular momentum to be conserved, its angular speed around the
axis must speed up – it will be deviated towards the east.
Now try another thought experiment (Gedanken Prüfung.). Go to the equator and build a tall tower. Drop a stone from the top of
the tower. Think now about the direction of the vector 2mv × ω . I really mean it – think hard. Or again, think about conservation
′

of angular momentum. The stone drops closer to Earth’s axis of rotation. It must conserve angular momentum. It falls to the east of
the tower (not to the west!).

4.9.4 https://phys.libretexts.org/@go/page/8396
The two forces on the stone are its weight mg and the Coriolis force. Earth’s spin vector ω is to the north. The Coriolis force is at
right angles to the stone’s velocity. If we resolve the stone’s velocity into a vertically down component ẏ and a horizontal east
component ż , the corresponding components of the Coriolis force will be 2mωẏ to the east and 2mωẋ upward. However, I’m
going to assume that ż << ẏ and the only significant Coriolis force is the eastward component 2mωẏ , which I have drawn.
Another way of stating the approximation is to say that the upward component of the Coriolis force is negligible compared with the
weight mg of the stone.
After dropping for a time t, the y-coordinate of the stone is found in the usual way from
1 2
ÿ = g, ẏ = gt, y = gt ,
2

and the x- coordinate is found from

1
2 3
ẍ = 2ωẏ = 2ωgt, ẋ = ωgt , x = ωgt
3

Thus you can find out how far to the east it has fallen after two seconds, or how far to the east it has fallen if the height of the tower
is 100 metres. The Equation to the trajectory would be the t -eliminant, which is
2
2
8ω 3
 x = y (4.9.8)
9g

For Earth, ω = 7.292 × 10−5 rad s−1, and at the equator g = 9.780 m s−2, so that
2
8ω −10 −1
= 4.788 × 10 m .
9g

The path is graphed in Figure IV.25 for a 100-metre tower. The horizontal scale is exaggerated by a factor of about 6000.

I once asked myself the question whether a migrating bird could navigate by using the Coriolis force. After all, if it were flying
north in the northern hemisphere, it would experience a Coriolis force to its right; might this give it navigational information? I

4.9.5 https://phys.libretexts.org/@go/page/8396
published an article on this in The Auk 97, 99 (1980). Let me know what you think!
You may have noticed the similarity between the Equation for the Coriolis force
′
F = 2m v × ω

and the Equation for the Lorentz force on an electric charge moving in a magnetic field:

F = qv × B.

The analogy can be pursued a bit further. If you rotate a coil in an electric field, a current will flow in the coil. That’s
electromagnetic induction, and it is the principle of an electric generator. Sometime early in the twentieth century, the American
physicist Arthur Compton (that’s not Denis Compton, of whom only a few of my readers will have heard, and very few indeed in
North America) successfully tried an interesting experiment. He made some toroidal glass tubes, filled with water coloured with
KMnO4, so that he could see the water, and he rotated these tubes about a horizontal or vertical diameter, and, lo, the water flowed
around the tubes, just as a current flows in a coil when it is rotated in a magnetic field. Imagine a toroidal tube set up in an east-
west vertical plane at the equator. The top part of the tube is slightly further from Earth’s rotation axis than the bottom part, and
consequently the water near the top of tube has more angular momentum per unit mass, around Earth’s axis, than the water near the
bottom. Now rotate the tune through 180° about its east-west horizontal diameter. The high angular momentum fluid moves closer
to Earth’s rotation axis, and the low angular momentum fluid moves further from Earth’s axis. Therefore, in order to conserve
angular momentum, the fluid must flow around the tube. By carrying out a series of such experiments, Compton was able, at least
in principle, to measure the speed of Earth’s rotation, and even his latitude, without looking out of the window, and indeed without
even being aware that there was an external universe out there. You may think that this was a very difficult experiment to do, but
you do it yourself every day. There are three mutually orthogonal semicircular canals inside your ear, and, every time you move
your head, fluid inside these semicircular canals flows in response to the Coriolis force, and this fluid flow is detected by little
nerves, which send a message to your brain to tell you of your movements and to help you to keep your balance. You have a
wonderful brain, which is why understanding physics is so easy.
Going back to the Lorentz force, we recall that a moving charge in a magnetic field experiences a force at right angles to its
velocity. But what is the origin of a magnetic field? Well, a magnetic field exists, for example, in the interior of a solenoid in which
there is a current of moving electrons in the coil windings, and it is these circulating electrons that ultimately cause the Lorentz
force on a charge in the interior of the solenoid – just as it is the galaxies in the universe revolving around our stationary Earth
which are the ultimate cause of the Coriolis force on a particle moving with respect to our Earth. But there I seem to be getting into
deep waters again, so perhaps it is time to move on to something easier.

This page titled 4.9: Centrifugal and Coriolis Forces is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

4.9.6 https://phys.libretexts.org/@go/page/8396
4.10: The Top
We have classified solid bodies technically as symmetric, asymmetric, spherical and linear tops, according to the relative sizes of
their principal moments of inertia. In this section, or at least in the title of this section, I mean “top” in the nontechnical sense of the
child’s toy – that is to say, a symmetric body, pointed at one end, spinning around its axis of symmetry, with the pointed end on the
ground or on a table. Technically, it is a “heavy symmetric top with one point fixed.”

I have drawn it is Figure IV.26, spinning about its symmetry axis, which makes an angle θ with the vertical. The distance between
the centre of mass and the point of contact with the table is l. It has a couple of forces acting on it – its weight and the equal,
opposite reaction of the table. In Figure IV.27, I replace these two forces by a torque, τ , which is of magnitude M gl sin θ .
Note that, since there is an external torque acting on the system, the angular momentum vector is not fixed.

Before getting too involved with numerous Equations, let’s spend a little while describing qualitatively the motion of a top, and
also describing the various coordinate systems and angles we shall be discussing. First, we shall be making use of a set of space-
fixed coordinates. We’ll let the origin O of the coordinates be at the (fixed) point where the tip of the top touches the table. The
axis Oz points vertically up to the zenith. The Ox and Oy axes are in the (horizontal) plane of the table. Their exact orientation is
not very important, but let’s suppose that Ox points due south, and Oy points due east. Oxyz then constitutes a right-handed set.
We’ll also make use of a set of body-fixed axes, which I’ll just refer to for the moment as 1, 2 and 3. The 3-axis is the symmetry
axis of the top. The 1- and 2-axes are perpendicular to this. Their exact positions are not very important, but let’s suppose that the
31-plane passes through a small ink-dot which you have marked on the side of the top, and that the 123 system constitutes a right-
handed set.
We are going to describe the orientation of the top at some instant by means of the three Eulerian angles θ , ϕ and ψ (see Figure
IV.28). The symmetry axis of the top is represented by the heavy arrow, and it is tipped at an angle θ to the z -axis. I’ll refer to a
plane normal to the axis of symmetry as the “equator” of the top, and it is inclined at θ to the xy-plane. The ascending node of the

4.10.1 https://phys.libretexts.org/@go/page/8397
equator on the xy-plane has an azimuth ϕ , and ψ is the angular distance of the 1-axis from the node. The azimuth of the symmetry
axis of the top is ϕ - 90 ° = ϕ + 270°.

Now let me anticipate a bit and describe the motion of the top while it is spinning and subject to the torque described above.
The symmetry axis of the top is going to precess around the z -axis, at a rate that will be described as ϕ̇ . Except under some
conditions (which I shall eventually describe) this precessional motion is secular. That means that ϕ increases all the time – it does
not oscillate to and fro. However, the symmetry axis does not remain at a constant angle with the z -axis. It oscillates, or nods, up
and down between two limits. This motion is called nutation (Latin: nutare, to nod). One of our aims will be to try to find the rate
of nutation ϕ˙ and to find the period and amplitude of the nutation.
It may look as though the top is spinning about its axis of symmetry, but this isn’t quite so. If the angular velocity vector were
exactly along the axis of symmetry, it would stay there, and there would be no precession or nutation, and this cannot be while
there is a torque acting on the top. An exception would be if the top were spinning vertically (θ = 0), when there would be no
torque acting on it. The top can in fact do that, except that, unless the top is spinning quite fast, this situation is unstable, and the
top will tip away from its vertical position at the slightest perturbation. At high spin speeds, however, such motion is stable, and
indeed one of our aims must be to determine the least angular speed about the symmetry axis such motion is stable.
However, as mentioned, unless the top is spinning vertically, the vector ω is not directed along the symmetry axis. We’ll call the
three components of ω along the three body- fixed axes ω , ω and ω , the last of these being the component of ω along the
1 2 3

symmetry axis. One of the things we shall discover when we proceed with the analysis is that ω remains constant throughout the
3

motion. Also, you should be able to distinguish between ω and ψ˙ . These are not the same, because of the motion of the node. In
3

fact you will probably understand that ψ = ω − ϕ̇ cos θ . Indeed, we have already derived the relations between the component of
3

the angular velocity vector and the rate of change of the Eulerian angles – see Equations 4.2.1 ,2 and 3. We shall be making use of
these relations in what follows.
To analyse the motion of the top, I am going to make use of Lagrange’s Equations of motion for a conservative system. If you are
familiar with Lagrange’s Equations, this will be straightforward. If you are not, you might prefer to skip this section until you have
become more familiar with Lagrangian mechanics in Chapter 13. However, I introduced Lagrange’s Equation briefly in Section
4.4, in which Lagrange’s Equation of motion was given as
d ∂T ∂T
  ( ) −( ) = Pj . (4.10.1)
dt ∂q̇ j
∂q̇ j

4.10.2 https://phys.libretexts.org/@go/page/8397
Here T is the kinetic energy of the system. P is the generalized force associated with the generalized coordinate q . If the force is
j j

a conservative force, then P can be expressed as the negative of the derivative of a potential energy function:
j

∂V
 Pj = − ( ) (4.10.2)
∂qj

Thus we have Lagrange’s Equation of motion for a system of conservative forces

d ∂V ∂V ∂V
  ( )− =− (4.10.3)
dt ∂q
˙j ∂qj ∂qj

Thus, in solving problem in Lagrangian dynamics, the first line in our calculation is to write down an expression for the kinetic
energy. The first line begins: “T =. . .”.
In the present problem, the kinetic energy is
1 1
2 2 2
 T = I1 (ω +ω )+ I3 ω (4.10.4)
1 2 3
2 2

Here the subscripts refer to the principal axes, 3 being the symmetry axis. The Eulerian angles θ and ϕ are zenith distance and
azimuth respectively of the symmetry axis with respect to laboratory fixed (space fixed) axes. The Eulerian angle ψ is measured
around the symmetry axis. The components of the angular velocity are related to the rates of change of the Eulerian angles by
previously derived formulas (Equations 4.2.1,2,3), so the θ˙, ϕ˙ and ψ˙ .
1 2 2 1
˙ ˙ 2 ˙ ˙ 2
 T = I1 (θ +ϕ sin θ) + I3 (ψ + ϕ cos θ) (4.10.5)
2 2

The potential energy is

 V = M gl cos θ + constant. (4.10.6)

Having written down the expressions for the kinetic and potential energies in terms of the Eulerian angles, we are now in a position
to apply the Lagrangian Equations of motion 4.10.3 for each of the three coordinates. We’ll start with the coordinate ϕ . The
Lagrangian Equation is
d ∂T ∂T ∂V
  ( )− =− (4.10.7)
dt ˙ ∂ϕ ∂ϕ
∂ϕ

We see that ∂T

∂ϕ
and ∂V

∂ϕ
are each zero, so that d

dt
∂T

∂ϕ
= 0, or ∂T

∂ϕ
= constant. This has the dimensions of angular momentum, so I’ll
call the constant L1. On evaluating the derivative ∂T

∂ϕ
, we obtain for the Lagrangian Equation in ϕ :
2 2
 I1 ϕ̇ si n θ + I3 ϕ̇ cos θ + I3 ψ̇ cos θ = L1 (4.10.8)

I’ll leave the reader to carry out exactly the same procedure with the coordinate ψ . You’ll quickly conclude that ∂T

∂ψ
= constant,
which has the dimensions of angular momentum, so call it L3 , and you will then arrive at the following for the Lagrangian
Equation in ψ :
˙
 I3 (ψ + ϕ cos θ) = L3 (4.10.9)

But the expression in parentheses is equal to ω (see Equation 4.2.3, although we have already used it in Equation 4.10.5), so we
3

obtain the result that ω , the component of the angular velocity about the symmetry axis, is constant during the motion of the top. It
3

would probably be worth the reader’s time at this point to distinguish again carefully in his or her mind the difference between ω 3

and ψ˙ .
Eliminate ψ̇ from Equations 4.10.8 and 4.10.9:
L1 − L3 cosθ
˙
 ϕ = (4.10.10)
2
I1 sin θ

This Equation tells us how the rate of precession varies with θ as the top nods or nutates up and down.
We could also eliminate ϕ˙ from Equations 4.10.8 and 4.10.9:

4.10.3 https://phys.libretexts.org/@go/page/8397
 L3 (L1 − L3 cos θ) cos θ
 ψ̇ = − (4.10.11)
2
I3 I1 sin θ

The Lagrangian Equation in θ is a little more complicated, but we can obtain a third Equation of motion from the constancy of the
total energy:
1 2 2
2
1 2
˙ ˙ ˙ ˙
  I1 (θ +ϕ sin θ) + I3 (ψ + ϕ cos θ) + M gl cos θ = E. (4.10.12)
2 2

We can eliminate ϕ̇ and ψ̇ from this, using Equations 4.10.10 and 4.10.11, to obtain an Equation in θ and the time only. After a
little algebra, we obtain
2
2 L1 − L3 cos θ
˙
 θ = A − B cos θ − ( ) , (4.10.13)
I1 sin θ

where
2
1 L
3
 A = (2E − ) (4.10.14)
I1 I3

and
2M gl
 B = (4.10.15)
I1

The turning points in the θ -motion (i.e. the nutation) occur where ˙
θ =0 . This results (after some algebra! – but quite
straightforward all the same) in a cubic Equation in c = cos θ :
2 3
 a0 = a1 c + a2 c + Bc =0 (4.10.16)

where
2 2 2
L1 2E L L
3 1
 a0 = A − ( ) = − − , (4.10.17)
2
I1 I1 I1 I3 I
1

2L1 L3 2L1 L3 2M gl
 a1 = −B = − (4.10.18)
2 2
I I I1
1 1

and
2 2 2
L3 2E L L3
3
 a2 = −A − ( ) = −[ − −( ) ] (4.10.19)
I1 I1 I1 I3 I1

Now Equation 4.10.16 is a cubic Equation in cos θ and it has either one real root or three real roots, and in the latter case two of
them or all three might be equal. We must also bear in mind that θ is real only if cos θ is in the range −1 to+1. We are trying to find
the nutation limits, so we are hoping that we will find two and only two real values of θ . (If the tip of the top were poised on top of
a point – e.g. if it were poised on top of the Eiffel Tower, rather than on a horizontal table − you could have θ > 90°.)
To try and understand this better, I constructed in my mind a top somewhat similar in shape to the one depicted in Figures IV.25
and 26, about 4 cm diameter, 7 cm high, made of brass. For the particular shape and dimensions that I imagined, it worked out to
have the following parameters, rounded off to two significant Figures:
M=0.53 kg l= 0.044 m I1 = 1.7×10−4 kg m2 I3 = 9.8×10−5 kg m2
I thought I’d spin the top so that ω (which, as we have seen, remains constant throughout the motion) is 250 rad s-1, and I'd start
3

the top ( ϕ˙ = θ˙ = 0 ) at θ = 30°. and then let go. Presumably it would then immediately start to fall, and 30° would then be the
upper bound to the nutation. We want to see how far it will fall before nodding upwards again. With ω = 250 rad s−1 we find, from 3

Equation 4.10.9 that

L3 = 2.45 x 10-2 Js
Also, I am assuming that ϕ˙ = 0 when θ = 30°, and Equation 4.10.10 tells us that

4.10.4 https://phys.libretexts.org/@go/page/8397
L1 = 2.121762 x 10-2 Js
Then with g = 9.8 m s−2, we have, from Equation 4.10.15,
B = 2.688659 x 103 s-2.
My initial conditions are that ϕ˙ and ˙
θ are each zero when θ = 30°, and Equations 4.10.10 and 4.10.13 between them tell us that
A = B cos 30°, so that

A = 2.328447 x 103 s-2.

From Equations 4.12.17, 18 and 19 we now have
a0= -1.324989 x 103 s-2
a1= +3.328586 x 103 s-2
a2= -2.309834 x 103 s-2
and we already have
B = 2.688659 x 103 s-2.
The “sign rule” for polynomial Equations, if you are familiar with it, tells us that there are no negative real roots (i.e. no solutions
with θ > 90°), and indeed if we solve the cubic Equation 4.10.16 we obtain
c = 0.824596, 0.866025, 6.9000406
The second of these corresponds to θ = 30°, which we already knew must be a solution. Indeed we could have divided Equation
4.10.16 by c– cos 30° to obtain a quadratic Equation for the remaining two roots, but it is perhaps best to solve the cubic Equation

as it is, in order to verify that cos 30° is indeed a solution, thus providing a check on the arithmetic. The third solution does not give
us a real θ (we were rather hoping this would happen). The second solution is the lower limit of the nutation, corresponding to θ =
34° 27'.
Generally, however, the top will nutate between two values of θ . Let us call these two values α and β, α being the smaller (more
vertical) of the two. I’ll refer to θ = α as the “upper bound” of the motion, even though α < β , since this corresponds to the more
vertical position of the top. We have looked a little at the motion in θ ; now let’s look at the motion in ϕ , starting with Equation
4.10.10:

L1 − L3 cos θ
˙
 ϕ = (4.10.10.)
2
I1 sin θ

If the initial conditions are such that L > L cos α (and therefore always greater than L
1 3 3 cos θ ) ϕ^ is always positive. The motion
is then something like I try to illustrate in Figure IV.29

This motion corresponds to an initial condition in which you give the top an initial push in the forward direction as indicated by the
L1
little blue arrow. If the initial conditions are such that cos α > > cos β , the sign of ϕ̇ is different at the upper and lower bounds.
L3

Thus is illustrated in Figure IV.30

4.10.5 https://phys.libretexts.org/@go/page/8397
This motion would arise if you were initially to give a little backward push before letting go of the top, as indicated by the little
blue arrow.
If the initial conditions are such that L = L cos α , then θ˙ and ϕ˙ are each zero at the upper bound of the nutation, and this was
1 3

the situation in our numerical example. It corresponds to just letting the top drop when you let it go, without giving it either a
forward or a backward push. This is illustrated in Figure IV.31.

As we discovered while doing our numerical example, the initial conditions ˙ ˙

θ =ϕ =0 when θ =α lead, in this third type of
motion, to
 L1 = L3 cos α (4.10.20)

and
 A = B cos α (4.10.21)

In the case Equation 4.10.13 becomes

2 L3 (cos α − cos θ)
˙ 2
 θ = B(cos α − cos θ) − [ ] (4.10.22)
I1 sin θ

The lower bound to the nutation(i.e. how far the top falls) is found by putting θ = β when θ˙ =0. This gives the following quadratic
Equation for β:
2 2
L L cos α
2 3 3
cos β− cos β + (4.10.23)
2
I B I1 sin θ
1

In our numerical example, this is

2
cos β − 7.725002 cos β + 5.690048 = 0, (4.10.24)

which, naturally, has the same two solutions as we obtained when we solved the cubic Equation, namely 0.824 596 and 6.900 406.
Recalling the definition of B (Equation 4.10.15), we see that Equation 4.10.23 can be written
2M glI1
2
  cos α − cos β = sin β, (4.10.25)
2
L
3

from which we see that the greater L , the smaller the difference between α and β - i.e. the smaller the amplitute of the nutation.
3

4.10.6 https://phys.libretexts.org/@go/page/8397
Equation 4.10.12, with the help of Equations 4.10.10 and 4.10.11, can be written:
2
L 1 2 1
3 ˙ 2
 E − − I1 θ = (L1 csc θ − L3 cot θ) + M glcosθ. (4.10.26)
2I3 2 2I1

The left hand side is the total energy minus the spin and nutation kinetic energies. Thus the right hand side represents the effective
potential energy V (θ) referred to a reference frame that is co-rotating with the precession. The term M gl sin θ needs no
e

explanation. The negative of the derivative of the first term on the right hand side would be the “fictitious” force that “exists” in the
corotating reference frame. The effective potential energy V (θ) is given by e

Ve (θ) 2 I1 M gl cos θ
2
  = [csc θ − (L3 / L1 ) cot θ] + . (4.10.27)
2 2
L /(2 I1 ) L
1 1

I draw Ve (θ) in Figures IV.32 and 33 using the values that we used in our numerical example – that is:
2
 Ve (θ) = 1.32081(csc θ − 1.154701 cot θ) + 0.229536 cos θ joules. (4.10.28)

Figure 32 is plotted up to 90° (although as mentioned earlier one could go further than this if the top were not spinning on a
horizontal table), and Figure 33 is a close look close 2 to the minimum. One can see that if E − L /(2I ) = 0.1979 the effective 3
3 2

potential energy (which cannot go higher than this, and reaches this value only when ) θ˙ = 0 , the nutation limits are between 30°
and 34° 24' . For a given L , for a larger total energy, the nutation limits are correspondingly wider. But for a given total energy, the
3

larger the component L of the angular momentum is, the lower will be the horizontal line and the narrower the nutation limits. If
3

the top loses energy (e.g. because of air resistance, or friction at the point of contact with the table), the E = constant line will
become lower 2 and lower, and the amplitude of the nutation will become less and less. If E − L /(2I ) is equal to the minimum 2
3 3

value of V (θ) there is only one solution for θ , and there is no nutation. For energy less than this, there is no stationary value of θ
e

and the top falls over.

4.10.7 https://phys.libretexts.org/@go/page/8397
We can find the rate of true regular precession quite simply as follows – and this is often done in introductory books.

In Figure IV.34, the vector L represents the angular momentum at some time, and in a time interval δt later the change in the
angular momentum is δL . The angular momentum is changing because of the external torque, which is a horizontal vector of
magnitude M gl sin θ (remind yourself from Figure XV .26 and 27). The rate of change of angular momentum is given by L = τ .
In time δt the tip of the vector L moves through a “distance” τ δt. Denote by Ω precessional angular velocity (the magnitude of
which we have hitherto called ϕ˙ ). The tip of the angular momentum vector is moving in a small circle of radius L sin θ . We
therefore see that τ = ΩL sin θ . Further, τ is perpendicular to both Ω and L . Therefore, in vector notation,
τ = Ω×L (4.10.29)

Note that the magnitude of τ is M gl sin θ and the magnitude of Ω × L is ΩL sin θ, so that the rate of precession is
M gl
 Ω = (4.10.30)
L

and is independent of θ .
One can continue to analyse the motion of a top almost indefinitely, but there are two special cases that are perhaps worth noting
and which I shall describe.
Special Case I. L = L .
1 3

In this case, Equation 4.10.27 becomes

Ve (θ)
2
  = C (cscθ − cotθ) + cos θ (4.10.31)
M gl

where

4.10.8 https://phys.libretexts.org/@go/page/8397
 2
L
1
 C = (4.10.32)
2M glI1

It may be rather unlikely that L = L exactly, but this case is of interest partly because it is exceptional in that V (0) does not go
1 3 e

to infinity; in fact V (0) = M gl whatever the value of C . Try substituting θ = 0 in Equation 4.10.31 and see what you get! The
e

right hand side is indeed 1, but you may have to work a little to get there. The other reason why this case is of interest is that it
makes a useful introduction to case II, which is not impossibly unlikely, namely that L is approximately equal to L , which leads
1 3

to motion of some interest.

Ve (0)
In Figure IV.35, I draw Mgl
for several different C .

From the graphs, it looks as though, if C ≥ 2 , there is one equilibrium position, it is at θ = 0°(i.e. the top is vertical), and the
equilibrium is stable, If C < 2 , there are two equilibrium positions: the vertical position is unstable, and the other equilibrium
position is stable. Thus if the top is spinning fast (large C ) it can spin in the vertical position only (a “sleeping top”), but, as the top
slows down owing to friction and air resistance, the vertical position will become unstable, and the top will fall down to a positive
value of θ .
dVe
These deductions are correct, for dθ
=0 results in
2 4
 2C (1 − cosθ) = sin θ (4.10.33)

One solution is θ = 0 , and a second differentiation will show that this is stable or unstable according to whether C is greater than
or less than 2, although the second differentiation is slightly tedious, and it can be avoided. We can also note that 1 − cos θ is a
common factor of the two sides of Equation 4.10.33, and it can be divided out to yield a cubic Equation in cos θ:
2 3
 2C − 1 − (2C + 1) cos θ − cos θ − cos θ = 0, (4.10.34)

which could be solved to find the second equilibrium point – but that again is slightly tedious. A less tedious way might be to take
the square root of each side of Equation 4.10.33:
−
−− 2
 √2C (1 − cos θ) = 1 − cos θ (4.10.35)

and then divide by the common factor (1 − \cos θ) to obtain

−
−−
cos θ = √2C − 1, (4.10.36)

which gives a real θ only if C ≥2 . Note also,if C =

1

2
, θ = 90° .
Special Case II. L 3 ≈ L1 .
Ve (0)
In other words, L and L are not very different. In Figure IX.36 I draw
1 3
Mgl
for several different C , for L3 = 1.01 L1 .

4.10.9 https://phys.libretexts.org/@go/page/8397
We see that for quite a large range of C greater than 2 the stable equilibrium position is close to vertical. Even though the curve for
C = 2 has a very broad minimum, the actual minimum is a little less than 17°. (I haven’t worked out the exact position – I’ll leave

that to the reader.)

This page titled 4.10: The Top is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

4.10.10 https://phys.libretexts.org/@go/page/8397
4.11: Appendix
In Section 4.4 we raised the question as to whether angle is a dimensioned or a dimensionless quantity, and in Section 4.8 we raised
the question as to whether angle is a vector quantity.
I can present two arguments. One of them will prove incontrovertibly that angle is dimensionless. The other will prove, equally
incontrovertibly, and equally convincingly that angle has dimensions. Angle, as you know, is defined as the ratio of arc length to
radius. It is the ratio of two lengths, and is therefore incontrovertibly dimensionless. Q.E.D. On the other hand, it is necessary to
state the units in which angle is expressed. You cannot merely talk about an angle of 1. You must state whether that is 1 degree or 1
degree. Angle therefore has dimensions. Q.E.D. So – you may take your pick. In many contexts, I like to think of angle as a
dimensioned quantity, having dimensions Θ. That is to say, not a combination of mass, length and time, but having its own
dimensions in its own right. I find I can carry on with dimensional analysis successfully like this.
Now for the question: Is angle a vector?
An angle certainly has both magnitude and a direction associated with it. Thus the direction associated with the angle

is at right angles to the plane of the screen, or the paper.

However, this evidently isn’t enough for it to be a vector in the sense that we know it.
For example, if you turn through an angle a, and then through an angle b, you cannot say that the net resultant of these is to turn
through an angle c , where c = a + b − 2ab cos C .
2 2 2

Thus, although angle has both magnitude and direction, and could be thought of thus far as a vector, angles do not obey the
ordinary triangle law of vector addition. For this reason, angles are sometimes called “pseudo-vectors”.
In fact, as any astronomy student will tell you, the correct relation between the angles is
cos c = cos a cos b + sin a sin b cos C .
If the angles a, b, c (not C ) are very small, then the triangle becomes almost plane. The angles add more and more like the usual
plane triangle rule for vector addition. This is probably obvious when thinking about the geometry, but you can also convince
yourself of it by expanding the sines and cosines (except for cos C ) as series, and, to the second order of small quantities
θ , sin θ ≈ 1) , you’ll find that the equation cos c = cos a cos b + sin a sin b cos C reduces to
1 2
(cos θ ≈ 1 −
2

c = a + b − 2ab cos C . For this reason it is sometimes said that an “infinitesimal rotation” can
2 2 2

be regarded as a true vector. Also for this reason, the time rate of change of an angle, , that is to say an angular velocity, can
dθ

dt

quite safely be treated as a true vector, since the numerator and denominator of the derivatives are both infinitesimals.

4.11.1 https://phys.libretexts.org/@go/page/8883
Thus, although angle has direction associated with it, angle is not a true vector in that angles do not follow the usual rules for vector
addition. However, very small angles do approximately follow the addition rules, so that, in the infinitesimal limit, angles can be
treated as vectors. And hence angular velocity, being a ratio of infinitesimals (dθ and dt ), can correctly be treated as vectors.

This page titled 4.11: Appendix is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

4.11.2 https://phys.libretexts.org/@go/page/8883
 CHAPTER OVERVIEW
5: Collisions
In this chapter on collisions, we shall have occasion to distinguish between elastic and inelastic collisions. An elastic collision is
one in which there is no loss of translational kinetic energy. That is, not only must no translational kinetic energy be degraded into
heat, but none of it may be converted to vibrational or rotational kinetic energy. In laying out the principles involved in collisions
between particles, we need not suppose that the particles actually "bang into" – i.e. touch – each other. For example most of the
principles that we shall be describing apply equally to collisions between balls that "bang into" each other and to phenomena such
as Rutherford scattering, in which an alpha particle is deviated from its path by a gold nucleus without actually "touching" it. Of
course, if you think about it at an atomic level, when two billiard balls collide, the atoms don't actually "touch" each other; they are
repelled from each other by electromagnetic forces, just as the alpha particle and the gold nucleus repelled each other in the
Rutherford-Geiger-Marsden experiment.

Topic hierarchy
5.1: Introduction
5.2: Bouncing Balls
5.3: Head-on Collision of a Moving Sphere with an Initially Stationary Sphere
5.4: Oblique Collisions
5.5: Oblique (Glancing) Elastic Collisions, Alternative Treatment

This page titled 5: Collisions is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1
5.1: Introduction
In this chapter on collisions, we shall have occasion to distinguish between elastic and inelastic collisions. An elastic collision is
one in which there is no loss of translational kinetic energy. That is, not only must no translational kinetic energy be degraded into
heat, but none of it may be converted to vibrational or rotational kinetic energy. It is well known, for example, that if a ball makes a
glancing (i.e. not head-on) elastic collision with another ball of the same mass, initially stationary, then after collision the two balls
will move off at right angles to reach other. But this is so only if the balls are smooth. If they are rough, after collision the balls will
be spinning, so this result – and any other results that assume no loss of translational kinetic energy − will not be valid. When
molecules collide, they may be set into rotational and vibrational motion, and in that case the collision will not be elastic in the
sense in which we are using the term. If two atoms collide, one (or both) may be raised to an excited electronic level. Some of the
translational kinetic energy has then been converted to potential energy. If the excited atom subsequently drops down to a lower
level, that energy is radiated away and lost from the system. Superelastic collisions are also possible. If one atom, before collision,
is in an excited electronic state, on collision it may make a radiationless downwards transition, and the potential energy released is
then converted to translational kinetic energy, so the collision is superelastic. None of this is intended to mean that elastic collisions
are impossible or even rare. In the case of collisions involving macroscopic bodies, such as smooth, hard billiard balls, collisions
may not be 100% elastic, but they may be close to it. In the case of low-energy (low temperature) collisions between atoms, there
need be no excitation to excited levels, in which case the collision will be elastic. Some subatomic particles, in particular leptons
(of which the electron is the best-known example), are believed to have no internal degrees of freedom, and therefore collisions
between them are necessarily elastic.
In laying out the principles involved in collisions between particles, we need not suppose that the particles actually "bang into" –
i.e. touch – each other. For example most of the principles that we shall be describing apply equally to collisions between balls that
"bang into" each other and to phenomena such as Rutherford scattering, in which an alpha particle is deviated from its path by a
gold nucleus without actually "touching" it. Of course, if you think about it at an atomic level, when two billiard balls collide, the
atoms don't actually "touch" each other; they are repelled from each other by electromagnetic forces, just as the alpha particle and
the gold nucleus repelled each other in the Rutherford-Geiger-Marsden experiment.
The theory of collisions is used a great deal, of course, in the study of high-energy collisions between particles in particle physics.
Bear in mind, however, that in "atom-smashing" experiments with modern huge particle accelerators, or even in relatively mild
collisions such as Compton scattering of x-rays, the particles involved are moving at speeds that are not negligible compared with
the speed of light, and therefore relativistic mechanics is needed for a proper analysis. In this chapter, collisions are treated entirely
from a nonrelativistic point of view.

This page titled 5.1: Introduction is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

5.1.1 https://phys.libretexts.org/@go/page/6953
5.2: Bouncing Balls
When a ball is dropped to the ground, one of four things may happen:
1. It may rebound with exactly the same speed as the speed at which it hit the ground. This is an elastic collision.
2. It may come to a complete rest, for example if it were a ball of soft putty. I shall call this a completely inelastic collision.
3. It may bounce back, but with a reduced speed. For want of a better term I shall refer to this as a somewhat inelastic collision.
4. If there happens to be a little heap of gunpowder lying on the table where the ball hits it, it may bounce back with a faster speed
than it had immediately before collision. That would be a superelastic collision.
The ratio
speed after collision

speed before collision

is called the coefficient of restitution, for which I shall use the speed before collision symbol e . The coefficient is 1 for an elastic
collision, less than 1 for an inelastic collision, zero for a completely inelastic collision, and greater than 1 for a superelastic
collision. The ratio of kinetic energy (after) to kinetic energy (before) is evidently, in this situation, e .
2

−−−−−
If a ball falls on to a table from a height h , it will take a time t = √2H lg to fall. If the collision is somewhat inelastic it will
0 0 0

then rise to a height h = e h and it will take a time et to reach height h . Then it will fall again, and bounce again, this time to a
1
2
0 1

lesser height. And, if the coefficient of restitution remains the same, it will continue to do this for an infinite number of bounces.
After a billion bounces, there is still an infinite number of bounces yet to come. The total distance travelled is
2 4 6
h = h0 + 2 h0 (e +e + e +. . . ) (5.2.1)

and the time taken is

2 3
t = t0 + 2 t0 (e + e + e +. . . ). (5.2.2)

These are geometric series, and their sums are

2
1 +e
h = h0 ( ), (5.2.3)
2
1 −e

which is independent of g (i.e. of the planet on which this experiment is performed), and
1 +e
t = t0 ( ) (5.2.4)
1 −e

For example, suppose h = 1 m, e = 0.5, g = 9.8 m s−2, then the ball comes to rest in 1.36 s after having travelled 1.67 m after an
0

infinite number of bounces.

 Discuss
Does the ball ever stop bouncing, given that, after every bounce, there is still an infinite number yet to come; yet after 1.36
seconds it is no longer bouncing...?

This page titled 5.2: Bouncing Balls is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

5.2.1 https://phys.libretexts.org/@go/page/6954
5.3: Head-on Collision of a Moving Sphere with an Initially Stationary Sphere
The coefficient of restitution is
relative speed of recession after collision
e = . (5.3.1)
relative speed of approach before collision

We suppose that the two masses m and m , the initial speed u, and the coefficient of restitution e are known; we wish to find v
1 2 1

and v .
2

We evidently need two equations. Since there are no external forces on the system, the linear momentum of the system is
conserved:

m1 u = m1 v1 + m2 v2 . (5.3.2)

The second equation will be the restitution equation (Equation 5.3.1):

v2 − v1 = eu. (5.3.3)

These two equations can be solved to yield

m1 − m2 e
v1 = ( )u (5.3.4)
m1 + m2

and
m1 (1 + e)
v2 = ( ) u. (5.3.5)
m1 + m2

The relation between the kinetic energy loss and the coefficient of restitution isn't quite as simple as in Section 5.2.

 Exercise 5.3.1

Show that
2 2 2
kinetic energy (after) m1 v + m1 v m1 + m2 e
1 2
= = . (5.3.6)
2
kinetic energy (before) m1 u m1 + m2

If m 2 =∞ (as in Section 5.2), this becomes just e . If e = 1 , it becomes unity, so all is well.
2

If m 1 << m2 (Ping-pong ball collides with cannon ball), v = −u, v = 0 .

1 2

If m 1 =m (Ping-pong ball collides with ping-pong ball), v = 0, v = u .

2 1 2

If m 1 >> m (Cannon ball collides with ping-pong ball), v = u, v = 2u .

2 1 2

 Example 5.3.1
A moving sphere has a head-on elastic collision with an initially stationary sphere. After collision the kinetic energies of the
two spheres are equal. Show that the mass ratio of the two spheres is 0.1716.
Which of the two spheres is the more massive? (I guarantee that your answer to this will be correct.)

This page titled 5.3: Head-on Collision of a Moving Sphere with an Initially Stationary Sphere is shared under a CC BY-NC 4.0 license and was
authored, remixed, and/or curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a
detailed edit history is available upon request.

5.3.1 https://phys.libretexts.org/@go/page/6955
5.4: Oblique Collisions
In Figure V.2 I show two balls just before collision, and just after collision. The horizontal line is the line joining the centres – for
short, the "line of centres". We suppose that we know the velocity (speed and direction) of each ball before collision, and the
coefficient of restitution. The direction of motion is to be described by the angle that the velocity vector makes with the line of
centres. We want to find the velocities (speed and direction) of each ball after collision. That is, we want to find four quantities, and
therefore we need four equations. These equations are as follows.

There are no external forces on the system along the line of centres. Therefore the component of momentum of the system along
the line of centres is conserved:
m1 v1 cos β1 + m2 v2 cos β2 = m1 u1 cos α1 + m2 u2 cos α2 . (5.4.1)

If we assume that the balls are smooth - i.e. that there are no forces perpendicular to the line of centres and the balls are not set into
rotation, then the component of the momentum of each ball separately perpendicular to the line of centres is conserved:
v1 sin β1 = u1 sin α1 (5.4.2)

and
v2 sin β2 = u2 sin α2 . (5.4.3)

The last of the four equations is the restitution equation

elative speed of recession along the line of centres after collision
e = .
relative speed of approach along the line of centres before collision

That is,
v2 cos β2 − v1 cos β1 = e(u1 cos α1 − u2 cos α2 ). (5.4.4)

 Example 5.4.1A
Suppose m =3kg, m = 2kg, u = 40ms−1 u = 15ms−1
1 2 1 2

α1 = 10° , α = 70°, e = 0.8

2

Find v , v , β , β .
1 2 1 2

Solution
v1 = 16.28 m s−1 v = 44.43 m s−1
2

β1 = 25°15' β = 18°30' 2

 Example 5.4.1B
Suppose m = 3kg, m = 3kg, u = 12ms−1 u = 15ms−1
1 2 1 2

α1 = 20° , α = 50°, β = 47°

2 2

Find v , v , β , e .
1 2 1

5.4.1 https://phys.libretexts.org/@go/page/6956
 Solution
v1 = 10.50 m s−1 v = 15.71 m s−1
2

β1 = 23°00' e = 0.6418

 Exercise 5.4.1

If u 2 =0 , and if e = 1 and if m 1 = m2 , show that β = 90° and β = 0°.

1 2

This page titled 5.4: Oblique Collisions is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

5.4.2 https://phys.libretexts.org/@go/page/6956
5.5: Oblique (Glancing) Elastic Collisions, Alternative Treatment
In Figure V.3, unlike Figure V.2, the horizontal line is not intended to represent the line of centers. Rather, it is the direction of the
initial velocity of m , and m is initially at rest. The second mass m is slightly off the line of the velocity of m . I am assuming
1 2 2 1

that the collision is elastic, so that e = 1 . In the "before" part of the Figure, I have indicated, as well as the two masses, the position
and velocity V of the center of mass C . The velocity of C remains constant, because there are no external forces on the system. I
have not drawn C in the "after" part of the Figure, because it would get a little in the way. Think about where it is.

Figure V.3 shows the situation in "laboratory space". (Later, we'll look at the situation referred to a reference frame in which C is at
rest – "center of mass space".) The angle θ is the angle through which m has been scattered (the "scattering angle"). I have
1

indicated in the Figure how it is related to the α and β of Section 5.3. Note that m (initially stationary) scoots off along the line
1 1 2

of centers.
The following two equations express the constancy of linear momentum of the system.
(m1 + m2 )V = m1 u = m1 v1 cos θ + m2 v2 cos α1 . (5.5.1)

I'm going to draw, in Figure V.4, the situation "close-up", so that you can see the geometry more clearly. Note that the distance b is
called the impact parameter. It is the distance by which the two centers would have missed each other had the first particle not been
scattered.
In Figure V.5, I draw the situation in center of mass space, in which the center of mass C is stationary. In this reference frame, I just
have to subtract V from all the velocities. Note that in center of mass space the speeds of the particles are unaltered by the
collision. In center of mass space, m is scattered through an angle θ , and I am going to find a relation between θ , θ and the mass
1
′ ′

m2
ratio .
m1

I shall start with the profound statement that

v1 sin θ
tan θ = (5.5.2)
v1 cos θ

Now v sin θ is the y -component of the final velocity of m in laboratory space. The y -component of the final velocity of m in
1 1 1

center of mass space is u sin θ , and these two are equal, since the y -component of the motion is unaffected by the change of
′ ′

reference frame. Therefore

′ ′
u sin θ
tan θ = . (5.5.3)
v1 cos θ

Therefore

5.5.1 https://phys.libretexts.org/@go/page/6957
 ′ ′
v1 sin θ = u sin θ . (5.5.4)

The x-components of the "before" and "after" velocities of m are related by

1

′ ′
v1 cos θ = u sin θ + V . (5.5.6)

Substitute Equations 5.5.4 and 5.5.6 into Equation 5.5.2 to obtain

′
sin θ
tan θ = (5.5.7)
′ ′
cos θ + V / u

But
′
(m1 + m2 )V = m1 u = m1 (u + V ), (5.5.8)

from which
V m1
= . (5.5.9)
′
u m2

On substituting this into Equation 5.5.7, we obtain the relation we sought:

′
sin θ
tan θ = . (5.5.10)
′
cosθ + m1 / m2

This relation is illustrated in Figure V.6 for several mass ratios.

Let us try to interpret the Figure. For m > m , any scattering angle, forward or backward, is possible, but for m < m ,
2 1 2 1

backward scattering is not possible, and forward scattering is possible only up to a maximum. This is only to be expected. Thus for
an impact parameter of zero or of R + R , and m < m , the scattering angle θ must be zero, and therefore for intermediate
1 2 2 1

impact parameters it must go through a maximum. This would be clearer if we could plot the scattering angle versus the impact
parameter, and indeed that is something that we shall try to do. In the meantime it is easy to show, by differentiation of Equation
5.5.10 (do it!), that the maximum scattering angle is sin μ, where
−1

m2
μ = .
m1

That is, if the scattered particle is very massive compared with the scattering particle, the maximum scattering angle is small – just
to be expected.

v1
I want to do two things now - one, to calculate the scattering angle θ as a function of impact parameter, and two, to calculate as
u
a function of scattering angle. I’m going to start with Equations 5.4.1, 5.4.2 and 5.4.4, except for the following. I’ll assume e = 1
(elastic collision), and u = 0 (m is initially stationary), and β = 0 (since m is initially stationary, it must move along the line
2 2 2 2

of centers after collision). Since I want to try to calculate the scattering angle, I’ll write θ + α for β (see Figure V.4). I’m also
1 1
v1 v2 m2
going to write r , r and μ for the dimensionless ratios
1 2 , and respectively. With those small changes, Equations 5.4.1,
u u m1

5.4.2 and 5.4.4 become

r1 cos(θ + α1 ) + μr2 = cos α1 , (5.5.11)

r1 sin(θ + α1 ) = sin α1 , (5.5.12)

5.5.2 https://phys.libretexts.org/@go/page/6957
 r2 − r1 cos(θ + α1 ) = cos α1 . (5.5.13)

Eliminate r from Equations 5.5.11 and 5.5.13 to obtain

2

r1 cos(θ + α1 ) = M cos α1 , (5.5.14)

where
1 −μ m1 − m2
M = = . (5.5.15)
1 +μ m1 + m2

v1
If we now eliminate α from Equations 5.5.12 and 5.5.14, we obtain the relation between
1 and the scattering angle, which was
u
the second of our two aims above. The elimination is easily done as follows. Expand \sin and cos of θ + α in the two equations, 1

divide both sides of each equation by cos α and eliminate tan α between the two equations. The result is
1 1

2
r (1 + M ) cos θ + M = 0. (5.5.16)
1

We’ll have a look at this equation in a moment, but in the meantime, instead of eliminating α from Equations 5.5.12 and 5.5.14, 1

let’s eliminate r . This will give us a relation between the scattering angle θ and α , and, since α is closely relation to the impact
1 1 1

parameter (see Figure V.4) this will achieve the first of our aims, namely to find the scattering angle as a function of the impact
parameter. If you do the algebra, you should find that the relation between θ and α is 1

a(1 − M )
t = , (5.5.17)
a2 + M

where

t = (5.5.17a,b)

tanθ and a =

tanα1 .

Now let
b
′
b = (5.5.18)
R1 + R2

and from Figure V.5 we see that

′
b = sin α1 . (5.5.19)

On elimination of α from Equations 1 5.5.17 and 5.5.19 , we obtain the required relation between scattering angle θ and
(dimensionless) impact parameter b : ′

−−−− −2
′ t
sμb √ 1 − b
tan θ = . (5.5.20)
2
t
1 − μ + 2μb

This relation is shown in Figure V .7. The values of the mass ratio μ ( = f racm 2 m1 ) are (from the
1 1 1 9 10
lowest up) , , , , 1, , 2, 4, 8 and (dashed) ∞. This Figure is perhaps slightly easier to interpret than Figure V.6. One can
8 4 2 10 9
see that for μ > 1 , any scattering angle is possible, but for μ < 1 , the scattering angle has a maximum possible value, less than 90
°, and the scattering angle is zero for b = 0 or 1. ′

 Exercise 5.5.1

We saw, by differentiation of Equation 5.5.10, that the maximum scattering angle was sin
−1
μ. Now show the same thing by
differentiation of Equation 5.5.20. (This is not so easy, is it?)
Show that the scattering angle is greatest for an impact parameter of
−−−−−
1 −μ
′
b =√ . (5.5.21)
2

5.5.3 https://phys.libretexts.org/@go/page/6957
 Solution
You will notice that, for /( b' = 0/) (head-on collision) the scattering angle changes abruptly from 0 to 180° as the mass ratio
changes from less than 1 to more than 1. No problem there. But if the mass ratio is exactly 1 (not the tiniest bit less or the
tiniest bit more) the scattering angle is apparently 90°. This may cause some puzzlement until it is realized that for a head-on
collision with μ = 1 the first sphere comes to a dead halt.
The case of mu = ∞ (second sphere immovable) is of some interest. It is easy in that case to calculate how the scattering
angle varies with impact parameter for an elastic collision, merely by requiring the scattered sphere to obey the law of
reflection, and without any reference to Equation 5.5.20.

 Exercise 5.5.2

Easy Exercise.
Without any reference to Equation 5.5.20 , show that, if the second ball is immovable, the scattering angle is related to the
impact parameter by
−1 ′
θ = 180° − 2si n b . (5.5.22)

Not-so-easy exercise. Show that, in the limit as μ → ∞ , Equation 5.5.20 approaches Equation 5.5.22.
In any case, the limiting case as the second sphere becomes immovable is shown as a dashed curve in Figure V.7.
m2
Exercise of Intermediate Difficulty. The mass ratio is 0.9, and the scattering angle is 50 . What was the impact parameter?
∘

m1

Answers
b' = 0.07270 or 0.58540.

We have now dealt with the direction of motion of m1 after scattering as a function of impact parameter. We should now look at
v1
the speed of m1 after collision, and this takes us back to Equation 5.5.16 , which gives is the speed (r 1 = ) as a function of
u
scattering angle θ . It is 11 quadratic in r , so, for a given scattering angle there are two possible speeds – which is not surprising,
1

because a given scattering angle can arise from two different impact parameters, as we have just found out. We can conveniently
show the relation between r and θ simply by plotting the equation in polar coordinates. I’ll re-write the equation here for easy
1

reference:
2
r = r1 (1 + M ) cos θ + M = 0. (5.5.16.)
1

1 −μ 1
Here, M = = , but I want to write the equation in terms of the mass fractions
1 +μ 1 +μ

m1 1 m2 μ
q1 = = and q2 = = . (5.5.23a,b)
m1 + m2 1 +μ m1 + m2 1 +μ

If you work at this for a short while, you will find that Equation 5.5.16 becomes
2 2 2
r +q − 2 r1 q1 cos θ = q . (5.5.24)
1 1 2

5.5.4 https://phys.libretexts.org/@go/page/6957
and one is then overcome with an overwhelming desire to draw a triangle:

For a given mass ratio, the locus of r1 (the speed) versus θ (the scattering angle) is such that q1 and q2 are constant – in other
words, it is a circle:

One can imagine that the first particle comes in from the left at speed u and the collision takes place at the asterisk, and, after
collision, it is moving at a speed r times u in a direction θ , the magnitude of its velocity vector being determined by where the
1

vector intersects the circle (in two possible places) given by Equation 5.5.24. The maximum scattering angle corresponds to a
velocity vector that is tangent to the circle. If the asterisk is the pole (origin) of the polar coordinates, the centre of the circle is at a
distance q from the pole, and its radius is q . Figure V.10 shows the circles corresponding to several mass ratios. The Figure
1 2

graphically illustrates the relation between u, v , θ and μ . You can see, for example, that if μ > 1 , scattering through any angle is
1

possible, and the relation between v and θ is unique; but if μ < 1 , only forward scattering is possible, up to a maximum θ , and,
1

for a given θ , there are two solutions for v . 1

This deals with what happens to the sphere m . We can now turn our attention to m . Starting from Equations 5.5.11, 5.5.12 and
1 2

5.5.13, we are going to want to eliminate r and θ - indeed anything that pertains to the sphere m .
1 1

If you refer to Figure V.4 you will see that, after collision, m scoots off at an angle α to the original direction of motion of m .
2 1 1

Therefore I think it is of interest to find a relation between r (\dfrac{v_{2}}{u})and α . If we succeed in doing this, it means that
2 1

we can also find a relation, if we want it, between r and the impact parameter, since b = sin α . It is easy to eliminate r from
2
′
1 1

Equations 5.5.12 and 5.5.13, and then you can get tan(θ + α ) from Equation 5.5.14, and hence get the required relation:
1

2 cos α1
r2 = . (5.5.25)
1 +μ

I’ll draw this relation as a polar graph, r versus α , in Figure V.11. I’ll leave the reader to work out and draw the relation between
2 1

1
r2 and b if he or she wishes. Equation V.11 is the polar equation to a circle of radius
′
.
1 +μ

 Example 5.5.1
m2
Suppose the mass ratio μ = = 0.5 and the scattering angle is θ = 20°. Equation 5.5.16 or Figure V.10 will show that r = 1
m1

0.8696 or 0.3833. Equation 5.5.17 will show that α = 58° .4 or 11° .6 . And Equation 5.5.25 or Figure V.11 will show that r
1 2

= 0.6983 or 1.306. I’ll leave it to the reader to determine which alternative values of r , r and α go together.
1 2 1

5.5.5 https://phys.libretexts.org/@go/page/6957
This page titled 5.5: Oblique (Glancing) Elastic Collisions, Alternative Treatment is shared under a CC BY-NC 4.0 license and was authored,
remixed, and/or curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit
history is available upon request.

5.5.6 https://phys.libretexts.org/@go/page/6957
 CHAPTER OVERVIEW
6: Motion in a Resisting Medium
In studying the motion of a body in a resisting medium, we assume that the resistive force on a body, and hence its deceleration, is
some function of its speed. Such resistive forces are not generally conservative, and kinetic energy is usually dissipated as heat. For
simple theoretical studies one can assume a simple force law, such as the resistive force is proportional to the speed, or to the
square of the speed, or to some function that we can conveniently handle mathematically. For slow, laminar, nonturbulent motion
through a viscous fluid, the resistance is indeed simply proportional to the speed, as can be shown at least by dimensional
arguments. One thinks, for example, of Stokes's Law for the motion of a sphere through a viscous fluid. For faster motion, when
laminar flow breaks up and the flow becomes turbulent, a resistive force that is proportional to the square of the speed may
represent the actual physical situation better.
6.1: Introduction
6.2: Uniformly Accelerated Motion
6.3: Uniformly Accelerated Motion
6.3A: Resistive Force Only
6.3B: Body falling under gravity in a resisting medium, resistive force proportional to the speed
6.3C: Body thrown vertically upwards, initial speed \(v_{0}\)
6.4: Motion in which the Resistance is Proportional to the Square of the Speed
6.4A: Resistive Force Only

This page titled 6: Motion in a Resisting Medium is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1
6.1: Introduction
In studying the motion of a body in a resisting medium, we assume that the resistive force on a body, and hence its deceleration, is
some function of its speed. Such resistive forces are not generally conservative, and kinetic energy is usually dissipated as heat. For
simple theoretical studies one can assume a simple force law, such as the resistive force is proportional to the speed, or to the
square of the speed, or to some function that we can conveniently handle mathematically. For slow, laminar, nonturbulent motion
through a viscous fluid, the resistance is indeed simply proportional to the speed, as can be shown at least by dimensional
arguments. One thinks, for example, of Stokes's Law for the motion of a sphere through a viscous fluid. For faster motion, when
laminar flow breaks up and the flow becomes turbulent, a resistive force that is proportional to the square of the speed may
represent the actual physical situation better.

This page titled 6.1: Introduction is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

6.1.1 https://phys.libretexts.org/@go/page/6960
6.2: Uniformly Accelerated Motion
Before studying motion in a resisting medium, a brief review of uniformly accelerating motion might be in order. That is, motion in
which the resistance is zero. Any formulas that we develop for motion in a resisting medium must go to the formulas for uniformly
accelerated motion as the resistance approaches zero.
One may imagine a situation in which a body starts with speed v and then accelerates at a rate a . One may ask three questions:
0

How fast is it moving after time t ?

How far has it moved in time t ?
How fast is it moving after it has covered a distance x ?
The answers to these questions are well known to any student of physics:
 v = v0 + at, (6.2.1)

1
2
 v = v0 t + at , (6.2.2)
2

2 2
 v =v + 2ax. (6.2.3)
0

Since the acceleration is uniform, there is no need to use calculus to derive these. The first follows immediately from the meaning
of acceleration. Distance travelled is the area under a speed : time graph. Figure VI.1 shows a speed : time graph for constant
acceleration, and Equation 6.2.2 is obvious from a glance at the graph. Equation 6.2.3 can be obtained by elimination of t between
Equations 6.2.1 and 6.2.2. (It can also be deduced from energy considerations, though that is rather putting the cart before the
horse.)

Nevertheless, although calculus is not necessary, it is instructive to see how calculus can be used to analyse uniformly accelerated
motion, since calculus will be necessary in less simple situations. We shall be using calculus to answer the three questions posed
earlier in the section.
For uniformly accelerated motion, the Equation of motion is
 ẍ = a. (6.2.4)

To answer the first question, we write ẍ as dv

dt
, and then the integral (with initial condition x = 0 when t = 0 ) is

 v = v0 + at. (6.2.5)

This is the first time integral.

Next, we write v as dv

dt
and integrate again with respect to time, to get
1 2
 x = v0 t + at . (6.2.6)
2

This is the second time integral.

To obtain the answer to the third question, which will be called the space integral, we must remember to write ẍ as v
dv

dt
.
Thus the Equation of motion (Equation 6.2.4) is

6.2.1 https://phys.libretexts.org/@go/page/6961
 dv
 v = a. (6.2.7)
dx

When this is integrated with respect to x (with initial condition v = v when x = 0 ) we obtain
0

2 2
 v =v + 2ax. (6.2.8)
0

This is the space integral.

Examples.
Here are a few quick examples of problems in uniformly accelerated motion. It is probably a good idea to work in algebra and
obtain algebraic solutions to each problem. That is, even if you are told that the initial speed is 15 ms-1, call it v , or, if you are told
0

that the height is 900 feet, call it h . You will probably find it helpful to sketch graphs either of distance versus time or speed versus
time in most of the problems. One last little hint: Remember that the two solutions of a quadratic Equation are equal if b = 4ac . 2

 Example 6.2.1

A body is dropped from rest. The last third of the distance before it hits the ground is covered in time T. Show that the time
taken for the entire fall to the ground is 5.45T.

 Example 6.2.2

The Lady is 8 metres from the bus stop, when the Bus, starting from rest at the bus stop, starts to move off with an acceleration
of 0.4 m s-2. What is the least speed at which the Lady must run in order to catch the Bus?
Answer: 2.53ms-1.

 Example 6.2.3

A parachutist is descending at a constant speed of 10 feet per second. When she is at a height of 900 feet, her friend, directly
below her, throws an apple up to her. What is the least speed at which he must throw the apple in order for it to reach her? How
long does it take to reach her, what height is she at then, and what is the relative speed of parachutist and apple? Assume g =
32 ft s-2. Neglect air resistance for the apple (but not for the parachutist!)
Answer:230fts-1, 7.5s, 825 ft, 0fts-1.

 Example 6.2.4
A lunar explorer performs the following experiment on the Moon in order to determine the gravitational acceleration g there.
He tosses a lunar rock upwards at an initial speed of 15 m s-1. Eight seconds later he tosses another rock upwards at an initial
speed of 10 m s-1. He observes that the rocks collide 16.32 seconds after the launch of the first rock. Calculate g and also the
height of the collision.
Answer: 1.64ms-2, 26.4m

 Example 6.2.5

Mr A and Mr B are discussing the merits of their cars. Mr A can go from 0 to 50 mph in ten seconds, and Mr B can go from 0
to 60 mph in 20 seconds. Mr B gives Mr A a start of one second. Assuming that each driver first accelerates uniformly to his
maximum speed and thereafter travels at each uniform speed, how long does it take Mr B to catch Mr A, and how far have the
cars travelled by then?
Answer: 41 s, half a mile.

I make the answers as follows. Let me know (jtatum@uvic.ca) if you think I have got any of them wrong.

This page titled 6.2: Uniformly Accelerated Motion is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon

6.2.2 https://phys.libretexts.org/@go/page/6961
request.

6.2.3 https://phys.libretexts.org/@go/page/6961
 SECTION OVERVIEW
6.3: Uniformly Accelerated Motion
If the only force on a body is a resistive force that is proportional to its speed, the equation of motion is

m ẍ = −b ẋ.

One thinks, for example, of Stokes's equation for the laminar motion of a sphere through a viscous fluid, in which the resistive
force is 6πηaυ, where η is the coefficient of dynamic viscosity. If we divide both sides of the equation by the mass m , we obtain

m ẍ = −γ ẋ,

where γ = m
b
is the damping constant. It has dimension T-1 and SI units s-1.

Topic hierarchy

6.3A: Resistive Force Only

6.3B: Body falling under gravity in a resisting medium, resistive force proportional to the speed

6.3C: Body thrown vertically upwards, initial speed \(v_{0}\)

This page titled 6.3: Uniformly Accelerated Motion is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

6.3.1 https://phys.libretexts.org/@go/page/6962
6.3A: Resistive Force Only
It is difficult to imagine a real situation in which the one and only force is a resistive force proportional to the speed. A body falling
through the air won't do, because, in addition to the resistive force, there is the acceleration due to gravity. Perhaps we could
imagine a puck sliding across the ice. The ice would have to be presumed to be completely frictionless, and the only force on the
puck would be the resistance of the air. It is a slightly artificial situation, because we want the puck to be going so fast that the
frictional force is negligible compared with the air resistance, but not so fast that the airflow is turbulent - but we need to start
somewhere. The frictional force is, at least to a very good approximation, not a function of speed, but is constant, and we shall start
by assuming that it is negligible and that the only horizontal force on the puck is air resistance and that the air resistance is
proportional to the speed.
In this case, the Equation of motion is indeed Equation 6.3.2. To obtain the first time integral, we write ẋ as v and the first time
integral is readily found to be
−γt
v = v0 e . (6.3.3)

Here v is the initial speed. This is illustrated in Figure VI.2

0

The speed is reduced to half of the initial speed in a time

ln 2 0.693
t 1 = = . (6.3.4)
2 γ γ

The second time integral is found by writing v in Equation 6.3.3 as dx

dt
. Integration, with initial condition x = 0 when t = 0 , gives
−γt
x = x∞ (1 − e ), (6.3.5)

v0
where x = ∞ γ
. This is illustrated in Figure VI.3. It is seen that the puck travels an eventual distance of x∞ , but only after an
infinite time.

6.3A.1 https://phys.libretexts.org/@go/page/8886
We can obtain the space integral either by eliminating t from between the two time integrals, or by writing the Equation of motion
as
dv
v = −γv. (6.3.6)
dx

With initial condition v = v when x = 0 , this becomes

0

v = v0 − γx, (6.3.7)

which is illustrated in Figure VI.4. The speed drops linearly with distance (but exponentially with time) reaching zero after having
v0
travelled a finite distance x =
∞ γ
in an infinite time.

This analysis has assumed that the only force was the resistive force proportional to the speed. In the case of our imaginary ice
puck, we were assuming that the resistive force was that of the air, the friction being negligible. Of course, as the puck slows down
and the resistive force becomes less, there will come a point when the frictional force is no longer negligible compared with the
ever-decreasing air resistance, so that the above Equations no longer accurately describe the motion. We shall come back to this
point in subsection 3c.

This page titled 6.3A: Resistive Force Only is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum
via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

6.3A.2 https://phys.libretexts.org/@go/page/8886
6.3B: Body falling under gravity in a resisting medium, resistive force proportional to
the speed
We are here probably considering a small sphere falling slowly through a viscous liquid, with laminar flow around the sphere,
rather than a skydiver hurtling through the air. In the latter case, the airflow is likely to be highly turbulent and the resistance
proportional to a higher power of the speed than the first.
We'll use the symbol y for the distance fallen. That is to say, we measure y downwards from the starting point. The equation of
motion is
ÿ = g − γv, (6.3.8)

where g is the gravitational acceleration.

g
The body reaches a constant speed when ÿ becomes zero. This occurs at a speed v^ = γ
, which is called the terminal speed.
To obtain the first time integral, we write the equation of motion as
dv
= γ(v
^ − v) (6.3.9)
dt

or
dv
= γdt. (6.3.10)
^−v
v

dv
= −γdt. (6.3.11)
^
v−v

DON'T! In the middle of an exam, while covering this derivation that you know so well, you can suddenly find yourself in
inextricable difficulties. The thing to note is this. If you look at the left hand side of the equation, you will anticipate that a
logarithm will appear when you integrate it. Keep the denominator positive! Some mathematicians may know the meaning of the
logarithm of a negative number, but most of us ordinary mortals do not - so keep the denominator positive!
With initial condition v = 0 when t = 0 , the first time integral becomes
γt
^(1 − e
v= v ). (6.3.12)

This is illustrated in Figure VI.5.

Students will have seen equations similar to this before in other branches of physics - e.g. growth of charge in a capacitor or growth
of current in an inductor. That is why learning physics becomes easier all the time, because you have seen it all before in quite
different contexts. Perhaps you have already noticed that third-year physics is easier than second-year physics; just think how much
easier fourth-year is going to be! At any rate, v approaches the terminal speed asymptotically, never quite reaching it, but reaching

6.3B.1 https://phys.libretexts.org/@go/page/8887
half of the terminal speed in time ln 2

γ
=
.693

γ
(you have seen that before while studying radioactive decay), and reaching (1 − e −1
)
= 63% of the terminal speed in time 1

γ
.
If the body is thrown downwards, so that its initial speed is not zero but is v = v when t = 0 , you will write the equation of
0

motion either as Equation 6.3.10 or as Equation 6.3.11, depending on whether the initial speed is slower than or faster than the
terminal speed, thus ensuring that the denominator is kept firmly positive. In either case, the result is
−γt
v= v
^ + (v0 − v
^)e (6.3.13)

Figure VI.6 shows v as a function of t for initial conditions v 0 = 0,

1

2
^, v
v ^, 2 v
^ .

Returning to the initial condition v = 0 when t = 0 , we readily find the second time integral to be
^
v −γt
^t −
y = v (1 − e ). (6.3.14)
γ

You should check whether this equation is what is expected for when t =0 and when t approaches infinity. The second time
integral is shown in Figure VI.7.

The space integral is found either by eliminating t between the first and second time integrals, or by writing ÿ as v
dv

dy
in the
equation of motion:
dv
v = γ(v
^ − v), (6.3.15)
dy

6.3B.2 https://phys.libretexts.org/@go/page/8887
whence
^
v v v
y = ln(1 − )− . (6.3.16)
γ ^
v γ

This is illustrated in Figure VI.8. Notice that the equation gives y as a function of v , but only
numerical calculation will give v for a given y .

 Exercise 6.3B. 1

Assume g = 9.8 m s-2 . A particle, starting from rest, is dropped through a medium such that the terminal speed is 9.8 m s-1.
How long will it take to fall through 9.8 m?

Solution
We are asked for t , given y , and we know the equation relating t and y - it is the second time integral, Equation 6.3.14 - so
what could be easier? We have γ = = 1s−1, so Equation 6.3.14 becomes
g

^
v

−t
9.8 = 9.8t − 9.8(1 − e ) (6.3.17)

and suddenly we find that it is not as easy as expected!

The equation can be written
−t
f (t) = t + e − 2 = 0. (6.3.18)

For Newton-Raphson iteration we need

′ −t
f (t) = 1 − e . (6.3.19)

f
and, after some rearrangement, the Newton-Raphson iteration (t → t − ′
) becomes
f

1 −t
t = + 2. (6.3.20)
t
e −1

(It may be noticed that 6.3.20, which derives from the Newton-Raphson process, is merely a rearrangement of Equation
6.3.18.)

Starting with an exceedingly stupid first guess of t = 100 s, the iterations proceed as follows:
t = 100.000 000 000
2.000 000 000

6.3B.3 https://phys.libretexts.org/@go/page/8887
 1.843 482 357
1.841 406 066
1.841 405 661
1.841 405 660 s

 Exercise 6.3B. 2
Assume g = 9.8 m s-2 . A particle, starting from rest, falls through a resisting medium, the damping constant being γ= 1.96 s-1
(i.e. v^ = 5 m s-1 ). How fast is it moving after it has fallen 0.3 m?

Solution
We are asked for v , given y . We want the space integral, Equation 6.3.16. On substituting the data, we obtain

f (v) = 5 ln(1 − 0.2v) + v + 0.588 = 0. (6.3.21)

From this,
′
f (v) = v/(v − 5) (6.3.22)

f
The Newton-Raphson process (t → t − f
′
) , after some algebra, arrives at

u(5 ln(0.2u) + 0.588) u(5 ln u − 7.459189560)

v= +5 = + 5, (6.3.23)
v v

where u = 5 − v .
This time Newton-Raphson does not allow us the luxury of an exceedingly stupid first guess, but we know that the answer
must lie between 0 and 5 m s-1 , so our moderately intelligent first guess can be v =2.5ms-1 .
Newton-Raphson iterations:
v = 2.500 000 000
2.122 264 100
2.051 880 531
2.049 766 247
2.049 764 400 m s-1

Problems
Here are four problems concerning a body falling from rest such that the resistance is proportional to the speed. Assume that v =
9.8 m s−2 . Answers to questions 6.3.3 - 6.3.6 are to be given to a precision of 0.0001 seconds.

 Exercise 6.3B. 3

A particle falls from rest in a medium such that the damping constant is γ = 1.0 s−1. How long will it take to fall through 10 m?

 Exercise 6.3B. 4
It takes t seconds to fall through y metres. Construct a table showing t for 201 values of y going from 0 to 20 metres in steps
of 0.1 metre, assuming that γ = 1.0 s−1.

6.3B.4 https://phys.libretexts.org/@go/page/8887
  Exercise 6.3B. 5

Construct a table showing t for 201 values of y going from 0 to 20 metres in steps of 0.1 metres for γ = 0.0, 0.5, 1.0, 1,5, 2.0 s
. The table is to have six columns. The first column gives the distance fallen to a precision of 0.1 metres. The remaining five
columns will give the time, to a precision of 0.0001 seconds, that the body takes to fall a given distance, to a precision of
0.0001 seconds

 Exercise 6.3B. 6

Draw, by computer, a graph showing t (the dependent variable, plotted vertically) versus y (plotted horizontally) for the five
values of γ in question 3.

These four problems are in order of increasing difficulty. The first is merely an exercise in solving an implicit equation (Equation
6.3.14) numerically, and might serve as an introductory example of how, for example, to solve an equation by Newton-Raphson

iteration (I make the answer 1.8656 s.) The last two, if started from scratch, could well take up an entire afternoon before it is
solved to one's complete satisfaction. It might be observed that the graphs of question 4 could be drawn fairly easily by calculating
y explicitly as a function of t , thus obviating the necessity of Newton-Raphson iteration. No such short cuts, however, can be made

for constructing the table of question 3.

In fact I solved questions 3 and 4 in just a few minutes − but I did not start from scratch. As you progress through your scientific
career, you will become aware that there are certain operations that you encounter time and time again. To do questions 3 and 4, for
example, you need to be able to solve an equation by Newton-Raphson iteration; you need to be able to construct a table of a
function y = f (x ; a), or in this case t = f (y ; γ); and you need to be able to instruct a computer to draw graphs of
tabulated values. I learned long ago that all of these are problems that crop up frequently, and I therefore long ago wrote short
programs (only a few lines of Fortran each) for doing each of them. All I had to do on this occasion was to marry these existing
programs together, tailored to the particular functions needed. Likewise a student will recognize similar problems for which he or
she frequently needs a solution. You should accumulate and keep a set of these small programs for use in the future whenever you
may need them. For example, this is by no means the last time you will need Newton-Raphson iteration to solve an equation. Write
a Newton-Raphson program now and keep it for future occasions!

This page titled 6.3B: Body falling under gravity in a resisting medium, resistive force proportional to the speed is shared under a CC BY-NC 4.0
license and was authored, remixed, and/or curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts
platform; a detailed edit history is available upon request.

6.3B.5 https://phys.libretexts.org/@go/page/8887
6.3C: Body thrown vertically upwards, initial speed v 0 v0
If we measure y upwards from the ground, the equation of motion is
^ + v).
ÿ = −g − γv = −γ(v (6.3.24)

The first time integral is

−γt
^ + (v0 + v
v = −v ^)e (6.3.25)

and this is shown in Figure VI.9.

It reaches a maximum height after time T , when v = 0 (at which time the acceleration is just −g ):
1 v0
t+ ln(1 + ). (6.3.26)
γ v
^

dy
The second time integral (obtained by writing v as dt
in Equation 6.3.25) and the space integral (obtained by writing ÿ as v dv

dy
in
the equation of motion) require some patience, but the results are
^
(v0 + v
−γt
y = (1 − e −v
^t, (6.3.27)
γ

v
^ + v0
v = v0 − γy − v
^ ln( ). (6.3.28)
^+v
v

These are illustrated in Figures VI.10 and VI.11.

6.3C.1 https://phys.libretexts.org/@go/page/8889
This page titled 6.3C: Body thrown vertically upwards, initial speed v is shared under a CC BY-NC 4.0 license and was authored, remixed,
0

and/or curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is
available upon request.

6.3C.2 https://phys.libretexts.org/@go/page/8889
 SECTION OVERVIEW
6.4: Motion in which the Resistance is Proportional to the Square of the Speed
There are not really any new principles; it is just a matter of practice with slightly more difficult integrals. We assume that the
resistive force per unit mass is kẋ . Here, although k plays a somewhat similar role to the γ of Section 3, it is not exactly the same
2

thing as γ , and indeed it is not dimensionally the same as γ . What are the dimensions, and the SI units, of k ?

Topic hierarchy

6.4A: Resistive Force Only

This page titled 6.4: Motion in which the Resistance is Proportional to the Square of the Speed is shared under a CC BY-NC 4.0 license and was
authored, remixed, and/or curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a
detailed edit history is available upon request.

6.4.1 https://phys.libretexts.org/@go/page/6963
6.4A: Resistive Force Only
We'll imagine a puck sliding along a frictionless surface against turbulent air resistance. The Equation of motion is:
2
ẍ = −kv . (6.4A.1)

By this time we assume that the student knows how to obtain the first and second time integrals and the space integral. The actual
integrations may be slightly more difficult, but we leave it to the reader to obtain the results
v0
v=
1 + kv0 t

ln(1 + kv0 t)
x =
k

−kx
v = v0 e .

These are illustrated in Figures VI.12,13,14. Note that, provided that Equation 6.4A.1 accurately describes the entire motion
(which may not be the case in a practical situation), there is no finite limit to x, nor does the speed drop to zero in any finite time.

This page titled 6.4A: Resistive Force Only is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum
via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

6.4A.1 https://phys.libretexts.org/@go/page/8938
 CHAPTER OVERVIEW
7: Projectiles
Topic hierarchy
7.1: No Air Resistance
7.2: Air Resistance Proportional to the Speed
7.3: Air Resistance Proportional to the Square of the Speed

This page titled 7: Projectiles is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1
7.1: No Air Resistance
We suppose that a particle is projected from a point O at the origin of a coordinate system, the y -axis being vertical and the x-axis
directed along the ground. The particle is projected in the xy-plane, with initial speed V at an angle α to the horizon. At any 0

subsequent time in its motion its speed is V and the angle that its motion makes with the horizontal is ψ .
The initial horizontal component if the velocity is V cos α , and, in the absence of air resistance, this horizontal component remains
0

constant throughout the motion. I shall also refer to this constant horizontal component of the velocity as u. I.e. u = V cos α = 0

constant throughout the motion.

The initial vertical component of the velocity is V cos α , but the vertical component of the motion is decelerated at a constant rate
0

g . At a later time during the motion, the vertical component of the velocity is V cos ψ , which I shall also refer to as v . 0

In the following, I write in the left hand column the horizontal component of the equation of motion and the first and second time
integrals; in the right hand column I do the same for the vertical component.

Horizontal Vertical

ẍ = 0 ÿ = −g 7.1.1a, b

ẋ = u = V0 cosα ẋ = u = V0 cosα 7.1.2a, b

1 2
ẋ = u = V0 t cosα ẋ = u = V0 t cosα − gt 7.1.3a, b
2

Equations 7.1.3a, b are the parametric equations to the trajectory. In vector form, these two equations could be written as a single
vector equation:
1
2
r = V0 t + gt (7.1.4)
2

Note the + sign on the right hand side of Equation 7.1.4. The vector g is directed downwards.
The xy-equation to the trajectory is found by eliminating t between Equations 7.1.3a and 7.1.3b to yield:
2
gx
y = x tan α − (7.1.5)
2
2V cos2 α
0

Now, re-write this in the form

2
x − ()x = −()y

2
Add to each side (half the coefficient of x) in order to "complete the square" on the left hand side, and, after some algebra, it will
be found that the equation to the trajectory can be written as:
2
(x − A) = −4a(y − B), (7.1.6)

where
2
V sin α cos α V0 sin 2α
0
A = = (7.1.7)
g 2g

2 2
V sin α
0
B = (7.1.8)
2g

and
2 2
V cos α
0
a = (7.1.9)
2g

Having re-arranged Equation 7.1.5 in the form 7.1.6, we see that the trajectory is a parabola whose vertex is at (A , B). The range
2 2

The greatest range on the horizontal plane is obtained when sin 2α = 1, or α = 45o. The
V sin 2α
on the horizontal plane is 2A, or 0

g
2 2 2
V0 V0 sin α
greatest range on the horizontal plane is therefore g
The maximum height reached is B, or 2g
The distance between vertex

7.1.1 https://phys.libretexts.org/@go/page/6967
 2 2
V0 cos α
and focus is a, or 2g
.The focus is above ground if this is less than the maximum height, and below ground if it is greater than
the maximum height. That is, the focus is above ground if cos 2
α < cos
2
α . That is to say, the focus is above ground if α > 45o and
below ground if α < 45o.
3

2
(1+y′ ) 2

The radius of curvature ρ anywhere along the trajectory can be found using the usual formula ρ =
y′′
. At the top of the
trajectory, y′ = 0 , so that  rho = Alternatively (in case one has forgotten or is unfamiliar with the "usual formula"), we note
1

y′

that the speed at the top of the path is just equal to the (constant) horizontal component of the velocity V cos α . We can then 0

equate the centripetal acceleration V cos to g and hence obtain:

2
0
2 α

2 2
V cos α
0
ρ = . (7.1.10)
g

By subtracting this from our expression for the maximum height of the projectile, we find that the height of the center of curvature
2 2

The center of curvature is above ground if α > 54o

V (1−3 cos α)
above the ground is 0

2g

44'.
The range r on a plane inclined at an angle θ to the horizontal can be found by substituting x = r cos θ and y = rsinθ in the
Equation 7.1.5 to the trajectory. This results, after some algebra, in
2
V
0
r = [sin(2α − θ) − sin θ]. (7.1.11)
2
gcos θ

This is greatest when 2α − θ = 90o; i.e. when the angle of projection bisects the angle between the inclined plane and the vertical.
The maximum range is
2
V
0
r = . (7.1.12)
g(1 + sin θ)

This is the equation, in polar coordinates, of a parabola, and this parabola, when rotated about its vertical axis, describes a
paraboloid, known as the paraboloid of safety. It is the envelope of all possible trajectories with an initial speed V . If a gun is 0

firing shells with initial speed V , or a lawn sprinkler is ejecting water at initial speed V , you are safe as long as you are outside
0 0

the paraboloid of safety. Figure VII.1 shows trajectories for a = 20, 40, 60, 80, 100, 120, 140 and 160 degrees, and, as a dashed line,
the paraboloid of safety. Notice how the range changes with a and that it is greatest for a = 45o.

 Exercise 7.1.1

A gun projects a shell, in the absence of air resistance, at an initial angle α to the horizontal. The speed of projection varies
with angle of projection and is given by
Initial speed = V 0 cos
1

2
α

Show that, in order to achieve the greatest range on the horizontal plane, the shell should be projected at an angle to the
horizontal whose cosine c is given by the solution of the equation
3 2
3c + 2c − 2c − 1 = 0

Find the optimum angle to a precision of one arcminute.

This page titled 7.1: No Air Resistance is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

7.1.2 https://phys.libretexts.org/@go/page/6967
7.2: Air Resistance Proportional to the Speed
As in the previous section, I shall write the x-component of the equation of motion, and of the first and second time integrals, in the
left hand column, and the y -component in the right-hand column. The x-component of the air resistance per unit mass is γẋ and the
y -component is γẏ . Here γ is the damping constant, defined in Section 6.3. The x- and y -components of the initial velocity are,

respectively, V cos α and V sin α . It should be readily seen that the equations of motion and their time integrals are as follows:
0 0

Horizontal Vertical

ẍ = −γ ẋ ÿ = −g − γ ẏ 7.2.1a, b

−γt γt
ẏ = v = V0 sin αe ^(1 − e
−v )
−γt
ẋ = u = V0 cosα. e 7.2.2a, b
where v
^ = g/γ

−γt
x = x∞ (1 − e ) 1
−γt
y = ^)(1 − e
(V0 sin α + v ^t
)− v 7.2.3a, b
V0 c os α
where x∞ = γ
γ

(In case it is not "readily seen", for the horizontal motion refer to Chapter 6, Section 3, especially Equations 6.3.2, 6.3.3 and 6.3.5,
and for the vertical motion refer to Chapter 6, Section 3b, especially Equations 6.3.24, 6.3.25 and 6.3.27.) It will be seen that, as
t → ∞ , u → 0 , v → −v ^ , x → x . The xy-equation to the trajectory is the t -eliminant of Equations 6.2.3a and 6.2.3b. After a
∞

small amount of algebra this is found to be:

^)
x(V0 sin α + v ^
v x
y = + ln(1 − ). (7.2.4)
V0 cos α γ x∞

The range on a horizontal plane is found by setting y = 0, to obtain either

x
x = −Aln(1 − ) (7.2.5)
x∞

or
x
−
x = x∞ (1 − e A ), (7.2.6)

where
v
^V0 cosα V0 cosα
A = , x∞ =
^)
γ(V0 sinα + v γ

and

7.2.1 https://phys.libretexts.org/@go/page/6968
 g
^ =
v .
γ

 Example 7.2.1

Suppose
V0 = 20 ms-1
α = 50
∘

g =9.8 ms
-2

γ =1.96s
-1 (∴ v^ =5 ms−1)
Then A =1.613 870 65 m
and x ∞ = 6.55905724 m.
Try to find the range on the horizontal plane, using either Equation 7.2.5 or 7.2.6, to nine significant figures. Which equation
works best? Newton-Raphson may fail with a stupid first guess - but it should not be difficult to make a fairly intelligent first
guess. I should not tell you, but figure VII.2 was calculated using the data of this example.
I make the answer 6.437 584 2 m.

Here’s a more difficult problem.

 Example 7.2.2

It is well known that, in the absence of air resistance, the maximum range on the horizontal plane is effected by choosing the
initial launch elevation to be α = 45 . What if there is air resistance, with damping constant γ? What, then, should be the angle
∘

−x

of launch to achieve the greatest range on the horizontal plane? Given Equation 7.2.6, x = x ∞ (1 −e A
) , for what value of α
is x greatest?

Solution
Equation 7.2.6, written in full, is
V0 cosα ^)x
−γ(V0 sin α + v
x = [1 − exp ( )] . (7.2.7)
γ ^V0 cos α
v

This can be written

(b sin α + 1)x
x = acosα [1 − exp (− )] , (7.2.8)
a cos α

V0 V0 γV0
where a = γ
and b = ^
=
g
. We have to find for what value of α is x greatest. It seems a simple enough problem, but at
v

the moment I can’t find a good way of solving it. If anyone has a clue, let me know (jtatum@uvic.ca). In the meantime, the
best I can offer is, for our particular numerical example, to calculate the range, x, for several values of α and see where it goes
through a maximum. For our particular numerical example, a = 10.204 081 63 m and b = 4. Here is a graph of range versus
launch angle, for an initial speed of 20 ms−1. A launch angle of about 23 59′ gives a range of about 8.4635 m. For a given γ
∘

and g , the optimum launch angle depends on the launch speed V . Is this intuitively obvious?
0

7.2.2 https://phys.libretexts.org/@go/page/6968
This page titled 7.2: Air Resistance Proportional to the Speed is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated
by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

7.2.3 https://phys.libretexts.org/@go/page/6968
7.3: Air Resistance Proportional to the Square of the Speed
Notation: V is the velocity, V is the speed. The horizontal and vertical components of the velocity are, respectively,
u = ẋ = V cos ψ and v = ẏ = V sinψ . Here ψ is the angle that the instantaneous velocity V makes with the horizontal. The

resistive force per unit mass is kV . The horizontal and vertical components of the resistive force per unit mass are kV cosψ and
2 2

kV sinψ respectively. The launch speed is V and the launch angle (i.e. the initial value of ψ ) is α . Distance traveled from the
2
0

launch point, measured along the trajectory, is s and speed V = ṡ . The Equations of motion are:
Horizontal:
2
ẍ = −kV cos ψ (7.3.1)

Vertical:
2
ẍ = −g − kV sin ψ (7.3.2)

These cannot be integrated as conveniently as in the previous cases, but we can get a simple relation between the horizontal
component u of the speed and the intrinsic coordinate s . Thus, when we make use of ẍ = u̇ , V = ṡ and V cosψ = u , Equation
7.3.1 takes the form

u̇ = −ku ṡ (7.3.3)

Integration, with initial condition u = V 0 cosθ , yields

−ks
u = V0 cos α. e . (7.3.4)

We can also obtain an exact explicit intrinsic Equation to the trajectory by consideration of the normal Equation of motion.
The intrinsic Equation to any curve is a relation between the intrinsic coordinates (s , ψ ). The rate at which the slope angle ψ
dψ
changes as you move along the curve, i.e. , is called the curvature at a point along the curve. If the slope is increasing with s , the
ds

curvature is positive. The reciprocal of the curvature at a point, , is the radius of curvature at the point, denoted here by ρ.
ds

dψ

The normal Equation of motion is the Equation F = ma applied in a direction normal to the curve. The acceleration appropriate
2
dψ
here is the centripetal acceleration V

ρ
or V 2

ds
.

In a direction normal to the motion, the air resistance has no component, and gravity has a component −g cos θ . (It is minus
because the curvature is clearly negative.) The normal Equation of motion is therefore
dψ
2
V = −g cos θ. (7.3.5)
ds

But
−ks
u V0 cos α. e
V = = (7.3.6)
cos ψ cos ψ

Therefore
dψ
2 2 −2ks 3
V cos α. e = −g cos ψ. (7.3.7)
0
ds

Separate the variables, and integrate, with appropriate initial conditions:

ψ s
3
g 2ks
∫ sec ψdψ = − ∫ e ds (7.3.8)
2 2
α V cos α 0
0

From here it is good integration practice to show that the intrinsic Equation is
sec ψ + tan ψ g
2ks
sec ψ tan ψ − sec α tan α + ln( ) = (1 − e ) (7.3.9)
2 2
sec α + sec α kV cos α
0

This Equation is of the form

7.3.1 https://phys.libretexts.org/@go/page/6969
 2ks
sec ψ tan ψ + ln(sec ψ + tan ψ) = A − Be . (7.3.10)

While it would be straightforward now to compute s as a function of ψ and hence to plot a graph of s versus ψ , we really want to
show y as a function of x, and x and y as a function of time. I am indebted to Dario Bruni of Italy for the following analysis.
Let (x 1 , y1 ) be a point on the trajectory. When the projectile moves a short distance Δs , the new coordinates will be (x 2 , x2 ),
where

x2 = x1 + Δs cos ψ1 (7.3.11)

and

y2 = y1 + Δs sin ψ1 , (7.3.12)

provided that Δs is taken to be sufficiently small that the path between the two points is approximately a straight line. The
calculation starts with x = y = 0 and ψ = α . At each stage of the calculation, the new value of ψ can be calculated from Equation
1 1

7.3.10. This can be done easily, for example, by Newton-Raphson iteration, since the derivative of the left hand side of this

equation with respect to ψ is just 2 sec ψ. Thus, with a sufficiently small interval Δs, the shape of the trajectory can be built up
3

point by point.
While this gives us the shape of the trajectory, it tells us nothing about the time. To do this, we can write the Equations of motion,
Equations 7.3.1 and 7.3.2 in the forms
−−−−−−
2 2
ẍ = −kẋ√ ẋ + ẏ (7.3.13)

and
−−−−−−
2 2
ÿ = −g − kẏ √ ẋ + ẏ . (7.3.14)

Let (x1 , y1 ) be a point on the trajectory. After a short time Δt, the new coordinates will be (x 2 , y2 ), where
1 2
x2 = x1 + ẋ1 Δt + ẍ1 (Δt) (7.3.15)
2

and
1 2
y2 = y1 + ẏ 1 Δt + ÿ 1 (Δt) , (7.3.16)
2

provided that Δt is taken to be sufficiently small that the acceleration between the two instants of time is approximately constant.
Also, the new velocity components are given by

ẋ2 = ẋ1 + ẍ1 Δt (7.3.17)

and
ẏ = ẏ + ÿ Δt. (7.3.18)
2 1 1

The calculation starts with

2 2
ẋ = V0 cos α, ẏ = V0 sin α, ẍ = −kV cos α, ÿ = −g − kV sin α
0 0

and after each increment Δt the new coordinates and velocity and acceleration components are calculated. The results of Sr Bruni’s
calculations are shown in Figure VII.3 for
k = 0.0177m-1, V = 90.5ms-1, α = 60 , g = 9.8ms-2
0
∘

7.3.2 https://phys.libretexts.org/@go/page/6969
Plotted with step by step method from intrinsic Equation with Δs= 0.025 m. Horizontal range 79.0 m; maximum height 62.4 m.
Total flight duration 7.1 seconds. The time taken to reach the maximum height is 2.8 seconds, so the descent time is longer than the
ascent time.
An alternative approach has been given by Ambrose Okune, of Uganda. In Okune’s analysis, he obtains explicit expressions for t ,
x and y in terms of the angle ψ . (In Equation 7.3.10 we already have a relation between s and ψ .)

We start with Equation 7.3.1, the horizontal Equation of motion

2
ẍ = −kV cos ψ = −kV V cos ψ. (7.3.19)

−−−−−−
Now ẍ = u̇ , 2
V = √u + v
2
, and V cos ψ = u so that
−−−−−−
2 2
u̇ = −ku √ u +v . (7.3.20)

Similarly, Equation 7.3.2, the vertical Equation of motion, is

2
ÿ = −g − kV sin ψ = −g − kV V sin ψ, (7.3.21)

−−−−−−
and, with ÿ = v̇ ,V 2
= √u + v
2
, and V sin ψ = v ,this becomes
− −−−−−
2 2
v̇ = −g − kv√ u + v . (7.3.22)

Now
v̇ dv v g
= = + − −−−−−. (7.3.23)
u̇ du u 2
ku √ u + v
2

Also v = u tan ψ so that

dv dψ
2
= tan ψ + usec . (7.3.24)
du du

On comparison of Equations 7.3.23 and 7.3.24, we see that

g
3
− −−−−− = u sec ψdψ. (7.3.25)
2 2
ku √ u + v

Upon substitution of v = u tan ψ this becomes

g dψ
3
= sec ψ . (7.3.26)
3
ku du

7.3.3 https://phys.libretexts.org/@go/page/6969
and hence
g −3 3
∫ u du = ∫ sec ψdψ. (7.3.27)
k

Upon integration, we obtain

g g
+ ln(sec ψ + tan ψ) + sec ψ tan ψ = A = + ln(sec α + tan α) + sec α tan α. (7.3.28)
2 2
ku ku
0

From this, we obtain

−
−
g 1
u =√ , (7.3.29)
−−−−−−−−−−−−−−−−−−−−−−−−−− −
k √ A − ln(secψ + tan ψ) − secψ tan ψ

and hence
−
−
g tan ψ
v=√ −−−−−−−−−−−−−−−−−−−−−−−−−− −. (7.3.30)
k √ A − ln(secψ + tanψ) − sec ψ tan ψ

Thus we now have the velocity components explicitly in terms of the angle y. For simplicity, let us write
λ = A − ln(sec ψ + tan ψ) − sec ψ tan ψ. (7.3.31)

Then the Equations for the velocity components are

−
−
g 1
u =√ − (7.3.32)
k √λ

and
−
−
g tan ψ
v=√ − . (7.3.33)
k √λ

In the limit, as u → 0 , ψ → −90 , y → −∞ , the motion approaches a vertical asymptote. As ψ → −90 , λ → −secψtanψ and
∘ ∘

−
−
tanψ g
hence limψ→−90∘ = −1 . Thus the limiting value of the vertical component of the velocity is −√
k
. This agrees precisely
√λ

with what one would expect for a body falling vertically at terminal speed, with resistance proportional to the square of the speed
(see Equation 6.4.5).
We now aim to find an expression relating ψ to t , which we do by noting that
du du
dψ dt dt
= = . (7.3.34)
du du dλ
dt
dψ dλ dψ

The derivative du

dt
can be found from the horizontal Equation of motion ẍ = −kV 2
cosψ , which can be written (because u =V

and ẍ = u̇ ) as u̇ = −ku secψ . Then, making use of Equation 7.3.32, we obtain

2

du g
=− sec ψ (7.3.35)
dt λ

The derivative du

dλ
can be found from Equation 7.3.32 and is
−
−
du 1 g 1
=− √ . (7.3.36)
3
dλ 2 k
λ 2

The derivative dλ

dψ
can be found from Equation 7.3.31 and is

dλ 3
= −2 sec ψ. (7.3.37)
dψ

Thus the relation we seek is

7.3.4 https://phys.libretexts.org/@go/page/6969
 dψ −− − 2
= −√gk √λ cos ψ. (7.3.38)
dt

If the initial motion of the projectile at time zero makes an angle α with the horizontal, then integration of Equation 7.3.38 gives
the following expression for the subsequent time t when the motion makes an angle ψ with the horizontal.
α
1 dψ
t = −− ∫ − . (7.3.39)
2
√gk ψ √λ cos ψ

dψ dψ
Also u = dx

dt
= dx

dψ dt
. With u and dt
given respectively by Equations 7.3.32 and 7.3.38, we obtain

dx 1
=− , (7.3.40)
2
dψ kλ cos ψ

from which we can calculate x as a function of ψ :

α
1 dψ
x = ∫ . (7.3.41)
k ψ
λ cos2 ψ

dy dy dψ dψ
Further, v = dt
= dψ dt
. With v and dt
given respectively by Equations 7.3.33 and 7.3.38, we obtain

dy tan ψ
=− , (7.3.42)
2
dψ kλ cos ψ

from which we can calculate y as a function of ψ :

α
1 tan ψdψ
y = ∫ . (7.3.43)
2
k ψ λ cos ψ

Equations 7.3.39, 7.3.41 and 7.3.43 enable us to calculate t , x and y as a function of ψ , and hence to calculate any one of them in
terms of any of the others. In each case a numerical integration is required, such as by Simpson’s rule or by Gaussian quadrature, or
other integration algorithm, and, as is always the case, sufficient points must be sampled to obtain adequate precision. Numerical
integration of these Equations, using the data of Dario Bruno’s example above, produced the same x : y trajectory as calculated for
Figure VII.3 by Bruno, and the x : t and y : t relations shown in Figure VII.4.

I am greatly indebted to Dario Bruni and to Ambrose Okune for their interesting and instructive contributions to this section – an
inspirational example of international scientific cooperation between, Italy, Uganda and Canada!

This page titled 7.3: Air Resistance Proportional to the Square of the Speed is shared under a CC BY-NC 4.0 license and was authored, remixed,
and/or curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is
available upon request.

7.3.5 https://phys.libretexts.org/@go/page/6969
 CHAPTER OVERVIEW
8: Impulsive Forces
Topic hierarchy
8.1: Introduction
8.2: Problem

This page titled 8: Impulsive Forces is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1
8.1: Introduction
As it goes about its business, a particle may experience many different sorts of forces. In Chapter 7, we looked at the effect of
forces that depend only on the speed of the particle. In a later chapter we shall look at forces that depend only on the position of the
particle. (Such forces will be called conservative forces.) In this chapter we shall look at the effect of forces that vary with time. Of
course, it is quite possible that an unfortunate particle may be buffeted by forces that depend on its speed, on its position, and on
the time - but, as far as this chapter is concerned, we shall be looking at forces that depend only on the time.
Everyone knows that Newton's second law of motion states that when a force acts on a body, the momentum of the body changes,
dp
and the rate of change of momentum is equal to the applied force. That is, F =
dt
. If a force that varies with time, F (t), acts on a
T
body for a time T , the integral of the force over the time, ∫ F (t)dt is called the impulse of the force, and it results in a change of
0

momentum ΔP which is equal to the impulse. I shall use the symbol J to represent impulse, or the time integral of a force. Its SI
units would be N s, and its dimensions MLT-1, which is the same as the dimensions of momentum.
Thus, Newton's second law of motion is

F = ṗ .

When integrated over time, this becomes

J = Δp.

Likewise, in rotational motion, the angular momentum L of a body changes when a torque τ acts on it, such that the rate of change
of angular momentum is equal to the applied torque:
˙
τ = L.

T
If the torque, which may vary with time, acts over a time T , the integral of the torque over the time, ∫ 0
τ dt is the angular impulse,
which I shall denote by the symbol K , and it results in a change of the angular momentum:

K = ΔL.

The SI units of angular impulse are N m s, and the dimensions are ML2T-1, which are the same as those of angular momentum.
For example, suppose that a golf ball is struck by a force varies with time as
3
2
2

|t − t0 | 3

^
F = F [1 − ( ) ]
τ

This may look like a highly-contrived and unlikely function, but in Figure VIII.1 I have drawn it for F^ = 1 , t = 3 , τ 0 =1 and you
will see that it is a reasonably plausible function. The club is in contact with the ball from time t − τ to t + τ .0 0

8.1.1 https://phys.libretexts.org/@go/page/6974
If the ball, of mass m, starts from rest, what will be its speed V immediately after it leaves the club? The answer is that its new
momentum, mV , will equal the impulse (or the time integral) of the above force:
3
2
2
t0 +τ
|t − t0 | 3

^
mV = F ∫ [1 − ( ) ] dt.
t0 −τ
τ

This is very easy to understand; if there is any difficulty it might be in the mechanics of working out this integral. It is good
integration practice, but, if you can't do it after a reasonable effort, and you want a hint, ask me (jtatum@uvic.ca) and I'll see what I
can do. You should get
3π
^ ^
mV = F τ = 0.589 F τ
16

 Example 8.1.1

Here is a very similar example, except that the integration is rather easier. A ball of mass 500 g, initially at rest, is struck with a
force that varies with time as
1

2 2
t−t0
^
F = F [1 − (
τ
) ] ,

where F^ = 4000N, t = 10 ms, t = 3 ms. Draw (accurately, by computer) a graph of

0 F versus time (it doesn't look quite like
Figure VIII.1). How fast is the ball moving immediately after impact?
(I make it 37.7 m s-1.)

This page titled 8.1: Introduction is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

8.1.2 https://phys.libretexts.org/@go/page/6974
8.2: Problem
In this section I offer a set of miscellaneous problems. In a typical problem it is assumed that an impulsive force or torque acts for
only a very short time. By "a very short" time, I mean that the time during which the force or torque acts is very small or is
negligible compared with other times that might be involved in the problem. For example, if a golf club hits a ball, the club is in
contact with the ball for a time that is negligible compared with the time in which the ball is in the air. Or if a pendulum is
subjected to an impulsive torque, the time during which the impulsive torque is applied is negligible compared with the period of
the pendulum.
In many problems, you will be told that a body is subjected to an impulse J . The easiest way to interpret this is to say that the
linear momentum of the body suddenly changes by J . Or you may be told that the body is subjected to an impulsive torque K . The
easiest way to interpret this is to say that the angular momentum of the body suddenly changes by K .
In some of the problems, for example the first one, the body concerned is freely hinged about a fixed point; that is, it can freely
rotate about that point.
Before giving the first problem, here is a little story. One of the most inspiring lectures I remember going to was one given by a
science educator. She complained that a professor, instead of inspiring his students with a love and appreciation of the great and
profound ideas of science and civilization, "infantalized" the class with a tiresome insistence that the class use blue pencils for
velocity vectors, green for acceleration, and red for forces. I recognized immediately that this was a great way of imparting to
students an appreciation of the profound ideas of physics, and I insisted on it with my own students ever since. In some of the
following drawings I have used this colour convention, though I don't know whether your computer will reproduce the colours that
I have used. In any case, I strongly recommend that you use the colour convention so deprecated by the educator if you want to
understand the great ideas of civilization, such as the ideas of impulsive forces.

 Exercise 8.2.1

In Figure VIII.2, a body is free to rotate about a fixed axis O. The centre of mass of the body is at C. The distance OC is h. The
body is struck with a force of impulse J at A, such that OA = x. The mass of the body is m. Its rotational inertia about C is
mk , and its rotational inertia about O is m(k + h ) .
2 2 2

As a result of the blow, the body will rotate with angular speed w and the centre of mass will move forward with linear speed
hw. One of the questions in this problem is to calculate ω
The body will also push with an impulsive force against the axis at O. It is not immediately obvious whether the body will push
upwards against the axis in the same direction as J , or whether the left hand end of the body will swing downwards and the
body will push downwards on the axis. You will probably agree that if A is very close to O, the body will push upwards on the
axis, but if A is near the right hand end, the body will push downward on the axis. In the Figure, I am assuming that the body
pushes upwards on the axis; the axis therefore pushes downwards on the body, with a force of impulse P , and what the Figure
shows is the two impulsive forces that act on the body. The second question to be asked in this problem is to find P in terms of
J and x.

If we are right in our intuitive feeling that P acts upwards or downwards according to the position of A - i.e. on where the
body is struck - there is presumably some position of A such that the reactive impulse of the axis on the body is zero. Indeed
there is, and the position of A that gives rise to zero reactive impulse at A is called the centre of percussion, and a third
question in this problem is to find the position of the centre of percussion. Where on the bat should you hit the baseball if you
want zero impulsive reaction on your wrists? Where should you position a doorstop so as to result in zero reaction on the door
hinges? Never let it be said that theoretical physics does not have important practical applications. The very positioning of a
door-stop depends on a thorough understanding of the principles of classical mechanics.

8.2.1 https://phys.libretexts.org/@go/page/6975
The net upward impulse is J − P , and this results in a change in linear momentum mhω:
J − P = mhω (8.2.1)

The impulsive torque about O is Jx, and this results in a change in angular momentum I ω; that is to say m(k 2 2
+ h )ω :
2 2
Jx = m(k + h )ω. (8.2.2)

These two equations enable us to solve for the two unknowns ω and P . Indeed, Equation 8.2.2 gives us ω immediately, and
elimination of ω between the two equations gives us P :
xh
P = J (1 − ). (8.2.3)
2 2
k +h

If the right hand side of Equation 8.2.3 is positive, then Figure VIII.2 is correct: the axis pushes down on the body, and the
2 2 2 2

body pushes upwards on the axis. That is, P acts downwards if x <
k +h

h
, and upwards if x >
k +h

h
. The position of the
2 2
k +h
centre of percussion is x = h
.
If the body is a uniform rod of length l, O is at one end of the rod, then k = l so that, in this case, x = l. This is where
2 1

12
2 2

you should position a door-stop. It is also where you should hit a baseball with the bat - if the bat is a uniform rod. However, I
admit to not knowing a great deal about baseball bats, and if such a bat is not a uniform rod, but is, for example, thicker and
heavier at the distal than the proximal end, the centre of percussion will be further towards the far end.

 Exercise 8.2.2

A heavy rod, of mass m and length 2l, hangs freely from one end. It is given an impulse J as shown at a point at a distance x
from the upper end. Calculate the maximum angular height through which the rod rises.
We can use Equation 8.2.2 to find the angular speed ω immediately after impact. In this equation, m(k 2
+h )
2
is the rotational
2

inertia of the rod about its end, which is , so that 4ml

ω =
3Jx
2
.
4ml

The kinetic energy immediately after impact is 1

2
⋅
4

3
ml
2
⋅ω
2
and we have to equate this to the subsequent gain in potential
energy mgl(1 − cos θ) .
Thus
2 2 2

cos θ = 1 −
2lω

3g
=1−
2J x
3
.
8gm2 l

To get the rod to swing through 180o, the angular impulse applied must be
−
−
gl
Jx = 4ml√
3
.

8.2.2 https://phys.libretexts.org/@go/page/6975
 Exercise 8.2.3

A uniform rod of mass m and length 2l is freely hinged at one end O. A mass cm (where c is a constant) is attached to the rod
at a distance x from O. An impulse J is applied to the other end of the rod from O. Where should the mass cm be positioned if
the linear speed of the mass cm immediately after the application of the impulse is to be greatest?
The angular impulse about O is 2lJ. The rotational inertia about O is ml + cmx . If w is the angular speed immediately
4

3
2 2

after the blow, the angular momentum is ( ml + cmx )ω . If we equate this to the impulse, we find
4

3
2 2

ω =
6lJ
2 2
.
m(4 l +3cx )

The linear speed of the mass cm is x times this, or 6lJ

2
. By calculus, this is greatest when x = 2l
.
m(4 l +3cx2 ) √3c

 Exercise 8.2.4

A uniform rod is of mass m and length 2l. An impulse J is applied as shown at a distance x from the mid-point of the rod. P is
a point at a distance y from the mid-point of the rod. Does P move forward or backward? Which way does A move?
The first thing we can do is to find the linear speed u of the centre of mass of the rod and the angular speed ω of the rod. We do
this by equating the impulse to the increase in linear momentum and the moment of the impulse (i.e. the angular impulse) to
the increase in angular momentum:
J = mu

and
Jx =
1

3
ml ω
2
.
2
3Jxy
The forward velocity of P is u + yω . That is to say m
J
+ 2
. This is positive if y >−
l

3x
but negative otherwise. For the
ml

point A, y = −l , so that A will move forward if x < l

3
, and it will move backwards otherwise.

 Exercise 8.2.5

8.2.3 https://phys.libretexts.org/@go/page/6975
A spherical planet, mass m, radius a , is struck by an asteroid with an impulse J as shown, the impact parameter being x. P is a
point on the diameter, a distance y from the centre of the planet. Does P move forward or backward? Which way does A
move?
As in the previous problem, we can easily find u and ω:
J = mu

and
Jx =
2

5
ma ω
2
.
2
5xy
The forward velocity of P is u + yω . That is to say J

m
(1 +
2a2
) . This is positive if y > − 2a

5x
but negative otherwise. For the
point A, y = −a , so that A will move forward if x < , and it will move backward otherwise. That is to say A will move
2a

backwards if the blow is more that 70% of the way from A to B.

 Exercise 8.2.6

A hoop, radius a , mass m, moving at speed v , hits a kerb of height h as shown. Will it mount the kerb, or will it fall back?
The only impulse acts at the point A. The moment of the impulse about A is therefore zero, and therefore there is no change in
angular momentum with respect to the point A.
Before impact, the angular momentum with respect to the point A is
mva + mv(a − h) = mv(2a − h) .
Let ω be the angular speed about A after impact. The angular momentum about A is then 2ma 2
ω. These are equal, so that
(2a−h)v
ω =
2a2

If it is to mount the kerb, the new kinetic energy 1

2
2
(2m a )ω
2
must be greater than the potential energy that is to be overcome,
mgh .

Therefore
2a√gh
v>
2a−h
.

8.2.4 https://phys.libretexts.org/@go/page/6975
 Exercise 8.2.7

Three equal particles, A, B and C, each of mass m, are joined by light inextensible strings as shown, the angle BAC being 60o.
A is given a sharp tug of impulse J as shown. Find the initial velocities of the particles and the initial tensions in the strings.
Let the initial tensions in BA and AC be T and T respectively.
′

Let the initial velocity of A be ui + vj .

Then the initial velocity of B is ui
−−
and the initial velocity of C is 1

2
(u − √3v) towards A.
The equations of impulsive motion are:
For B:
mu = T

For C:
1 −− ′
(u − √3v) = T
2

For A:
1 ′
mu = J − T − T
2

1 – ′
mv = √3T
2

The solutions of these equations are:

√3J
u =
7J

15m
,v= 15m
,
T =
7J

15
,T ′
=
2J

15
.

 Exercise 8.2.8

8.2.5 https://phys.libretexts.org/@go/page/6975
Two rods, each of mass m and length 2l, are freely jointed as shown. One of them is given an impulse J as shown. What
happens then is that the end of one rod gives the end of the other an impulsive kick P , and the other gives the one an equal
kick in the opposite direction. The centre of mass of the system moves forward with speed u and the two rods rotate with
angular speeds ω and ω . The problem is to determine P , u, ω and ω .
1 2 1 2

The equations of impulsive motion are:

LH rod, translation:
J − P = m(u + lω1 ) ,
RH rod, translation:
P = m(u + lω2 )

LH rod, rotation:
1 2
Pl = m l ω1
3

RH rod, rotation:
1 2
Pl = m l ω1
3

The solutions to these equations are:

u =
2m
J
,
ω1 = ω2 =
8ml
3J
,

P =
J

8
.

 Exercise 8.2.9

Two rods, each of mass m and length 2l, are freely jointed initially at right angles. They are dropped on to a smooth horizontal
table and strike it with speed V . Find the rate θ˙ at which the rods splay apart immediately after impact.

8.2.6 https://phys.libretexts.org/@go/page/6975
 We consider the dynamics of the right hand rod. On impact, it experiences a vertical impulse J at its lower end, and it
experiences a horizontal impulse P (from the other rod) at its upper end. Immediately after impact, let the components of the
velocity of the centre of mass of the right hand rod be u and v , and the angular speed of the rod is the required quantity θ˙ .
From geometry, we have
x = l sin θ and y = l cos θ
and hence
˙
˙
u = ẋ = l cos θθ =
lθ

√2
and v = −ẏ ˙
= l sin θθ =
lθ

√2
.

The impulsive equations of motion are

P = mu

and
J = m(V − v) .
After that, some algebra results in
√18V
˙
θ =
8l
.

 Exercise 8.2.10

A square plate is spinning about a vertical diameter at angular speed ω . It strikes a fixed obstacle at the corner A, so that it
1

subsequently spins about a vertical axis through A at angular speed ω . Find ω . 2 2

Since the impulse is at A, the moment of the impulse about A is zero, so that angular momentum about A is conserved. The
relevant moments of inertia can be calculated from the information in Chapter 2, and hence we obtain
1

3
2
m a ω1 =
7

3
m a ω2
2
.
ω1
ω2 =
7
.

This page titled 8.2: Problem is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

8.2.7 https://phys.libretexts.org/@go/page/6975
 CHAPTER OVERVIEW
9: Conservative Forces
A conservative force is a force with the property that the work done in moving a particle between two points is independent of the
taken path.

Topic hierarchy
9.1: Introduction
9.2: The Time and Energy Equation
9.3: Virtual Work
9.E: Conservative Forces (Exercises)

This page titled 9: Conservative Forces is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1
9.1: Introduction
In Chapter 7 we dealt with forces on a particle that depend on the speed of the particle. In Chapter 8 we dealt with forces that
depend on the time. In this chapter, we deal with forces that depend only on the position of a particle. Such forces are called
conservative forces. While only conservative forces act, the sum of potential and kinetic energies is conserved.
Conservative forces have a number of properties. One is that the work done by a conservative force (or, what amounts to the same
thing, the line integral of a conservative force) as it moves from one point to another is route-independent. The work done depends
only on the coordinates of the beginning and end points, and not on the path taken to get from one to the other. It follows from this
that the work done by a conservative force, or its line integral, round a closed path is zero. (If you are reminded here of the
properties of a function of state in thermodynamics, all to the good.) Another property of a conservative force is that it can be
derived from a potential energy function. Thus for any conservative force, there exists a scalar function V ( x, y , z ) such that the
force is equal to -grad V , or −∇V . In a one-dimensional situation, a sufficient condition for a force to be conservative is that it is
a function of its position alone. In two- and three-dimensional situations, this is a necessary condition, but it is not a sufficient one.
That a conservative force must be derivable from the gradient of a potential energy function and that its line integral around a
closed path must be zero implies that the curl of a conservative force must be zero, and indeed a zero curl is a necessary and a
sufficient condition for a force to be conservative.
This is all very well, but suppose you are stuck in the middle of an exam and your mind goes blank and you can't think what a line
integral or a grad or a curl are, or you never did understand them in the first place, how can you tell if a force is conservative or
not? Here is a rule of thumb that will almost never fail you: If the force is the tension in a stretched elastic string or spring, or the
thrust in a compressed spring, or if the force is gravity or if it is an electrostatic force, the force is conservative. If it is not one of
these, it is not conservative.

 Example 9.1.1
A man lifts up a basket of groceries from a table. Is the force that he exerts a conservative force?
Solution
No, it is not. The force is not the tension in a string or a spring, nor is it electrostatic. And, although he may be fighting against
gravity, the force that he exerts with his muscles is not a gravitational force. Therefore it is not a conservative force. You see,
he may be accelerating as he moves the basket up, in which case the force that he is exerting is greater than the weight of the
groceries. If he is moving at constant speed, the force he exerts is equal to the weight of the groceries. Thus the force he exerts
depends on whether he is accelerating or not; the force does not depend only on the position.

 Example 9.1.2

But you are not in an exam now, and you have ample time to remind yourself what a curl is. Each of the following two forces
are functions of position only - a necessary condition for them to be conservative. But it is not a sufficient condition. In fact
one of them is conservative and the other isn't. You will have to find out by evaluating the curl of each. The one that has zero
curl is the conservative one. When you have identified it, work out the potential energy function from which it can be derived.
In other words, find V (x, y, z) such that F = −∇V .
i. F = (3x 2 2
z − 3 y z)i − 6xyzj + (x
3 2
− xy )k

ii. F = ax 2 2
yzi − bxy zj + cxyz k
2

When you have identified which of these forces is irrotational (i.e. has zero curl), you can find the potential function by
calculating the work done when the force moves from the origin to (x,y ,z ) along any route you choose. Indeed, you might try
more than one route to convince yourself that the line integral is route-independent.
One could devise many exercises in determining whether various force functions are conservative, and, if so, what the
corresponding potential energy functions are, but I am going to restrict this chapter to just one more topic, namely

This page titled 9.1: Introduction is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

9.1.1 https://phys.libretexts.org/@go/page/6981
9.2: The Time and Energy Equation
Consider a one-dimensional situation in which there is a force F (x) that depends on the one coordinate only and is therefore a
conservative force. If a particle moves under this force, its equation of motion is

m ẍ = F (x)

and we can obtain the space integral in the usual fashion by writing ẍ as v dv

dx
.
Thus
dv
mv = F (x)
dx

Integration yields
1 2
mv =∫ F (x)dx + T0
2

Here mv is called the kinetic energy and the integration constant T can be interpreted as the initial kinetic energy. Thus the gain
1

2
2
0

in kinetic energy is

T − T0 = ∫ F (x)dx (9.2.4)

the right hand side merely being the work done by the force.
Since F is a function of x alone, we can find a V such that F = − . [It is true that we could also find a function V such that
dV

dx

F =+ , but we shall shortly find that the choice of the minus sign gives V a desirable property that we can make use of.] If we
dV

dx

integrate this equation, we find

V = −∫ F (x)dx + V0 (9.2.5)

Here V is the potential energy and V is the initial potential energy. From Equations 9.2.4 and 9.2.5 we obtain
0

V + T = V0 + T0 (9.2.6)

Thus the quantity V + T is conserved under the action of a conservative force. (This would not have been the case if we had
chosen the + sign in our definition of V .) We may call the sum of the two energies E , the total energy, and we have

T = E − V (x) (9.2.7)

or
1 2
mv = E − V (x) (9.2.8)
2

With v = dx

dt
, we obtain, by integrating Equation 9.2.8,
−
−− x
m dx
t = ±√ ∫ −−− −−− −− (9.2.9a)
2 x0 √ E − V (x)

This may at first appear to be a very formal and laborious way of arriving at something very obvious and something we have
known since we first studied physics, but we shall see that it can often be a quite useful equation. You might, by the way, check that
this equation is dimensionally correct.
The choice of the sign in Equation 9.2.9a may require some care, as will be evident in the examples that follow in the next section.
If the particle is moving away from the origin, then its speed is v = , and we choose the positive sign. If the particle is moving
dx

dt

towards from the origin, then its speed is v = − , and we choose the negative sign. However, I believe the following to be true:
dx

dt

If the particle is moving away from the origin, then the initial value of x is smaller than the final value. If the particle is moving
toward the origin, then the initial value of x is larger than the final value. It would seem to be safe, then, always to use the positive

9.2.1 https://phys.libretexts.org/@go/page/6982
sign, but then the lower limit of integration is the smaller value of x (not necessarily the initial value), and the upper limit of
integration is the larger value of x (not necessarily the final value). It may therefore be easier to write the equation in the form
−
−− xla rger
m dx
t = ±√ ∫ (9.2.9b)
−−− −−− −−
2 xsma ller √ E − V (x)

All that this means is that, for a conservative force, the time taken for a “return” journey is just equal to the time taken for the
outbound journey, so one might as well always calculate the time for the outbound journey.
In some classes of problem such as pendulums, or rods falling over, the potential energy can be written as a function of an angle,
and the kinetic energy is rotational kinetic energy written in the form I ω where ω = . In that case, Equation 9.2.9b takes the
1

2
2 dθ

dt

form
−
− θ
I dθ
t = ±√ ∫ . (9.2.10)
−−−− −−−−
2 θ0 √ E − V (θ)

You should check that this, too, is dimensionally correct.

This page titled 9.2: The Time and Energy Equation is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

9.2.2 https://phys.libretexts.org/@go/page/6982
9.3: Virtual Work
We have seen that a mechanical system subject to conservative forces is in equilibrium when the derivatives of the potential energy
with respect to the coordinates are zero. A method of solving such problems, therefore, is to write down an expression for the
potential energy and put the derivatives equal to zero.
A very similar method is to use the principle of virtual work. In this method, we imagine that we act upon the system in such a
manner as to increase one of the coordinates. We imagine, for example, what would happen if we were to stretch one of the springs,
or to increase the angle between two jointed rods, or the angle that the ladder makes as it leans against the wall. We ask ourselves
how much work we have to do on the system in order to increase this coordinate by a small amount. If the system starts from
equilibrium, this work will be very small, and, in the limit of an infinitesimally small displacement, this “virtual work” will be zero.
This method is very little different from setting the derivative of the potential energy to zero. I mention it here, however, because
the concept might be useful in Chapter 13 in describing Hamilton’s variational principle.
Let’s start by doing a simple ladder problem by the method of virtual work. The usual uniform ladder of high school physics, of
length 2l and weight mg, is leaning in limiting static equilibrium against the usual smooth vertical wall and the rough horizontal
floor whose coefficient of limiting static friction is μ . What is the angle θ that the ladder makes with the vertical wall?
I have drawn the four forces on the ladder, namely: its weight mg; the normal reaction of the floor on the ladder, which must also
be mg; the frictional force, which is μmg; and the normal (and only) reaction of the wall on the ladder, which must also be μmg.
There are several ways of doing this, which will be familiar to many readers. The only small reminder that I will give is to point out
that, if you wish to combine the two forces at the foot of the ladder into a single force acting upwards and somewhat to the left, so
that there are then just three forces acting on the ladder, the three forces must act through a single point, which will be above the
middle of the ladder and to the right of the point of contact with the wall. But we are interested now in solving this problem by the
principle of virtual work.
Before starting, I should warn that it is important in using the principle of virtual work to be meticulously careful about signs, and
in that respect I remind readers that in the differential calculus the symbols δ and d in front of a scalar quantity x do not mean “a
small change in” or “an infinitesimal change” in x. Such language is vague. The symbols stand for “a small increase in” and “an
infinitesimal increase in”.

Let us take note of the following distances:

C D = l cos θ

9.3.1 https://phys.libretexts.org/@go/page/6984
and

BE = 2l sin θ.

If we were to increase θ by δθ, keeping the ladder in contact with wall and floor, the increases in these distances would be

δ(C D) = −l sin θdθ.

and

δ(BE) = 2l cos θdθ.

Further, if were to increase θ by δθ, the work done by the force at C would be mg times the decrease of the distance CD, and the
work done by the frictional force at E would be minus μmg times the increase of the distance BE. The other two forces do no
work. Thus the “virtual work” done by the external forces on the ladder is

mg. l sin θδθ − μmg.2l cos θδθ.

On putting the expression for the virtual work to zero, we obtain

tan θ = 2μ.

You should verify that this is the same answer as you get from other methods – the easiest of which is probably to take moments
about E.
There is something about virtual work which reminds me of thermodynamics. The first law of thermodynamics, for example is
ΔU = Δq + Δw , where ΔU is the increase of the internal energy of the system, Δq is the heat added to the system, and Δw is

the work done on the system. Prepositions play an important part in thermodynamics. It is always mandatory to state clearly and
without ambiguity whether work is done by the piston on the gas, or by the gas on the system; or whether heat is gained by the
system or lost from it. Without these prepositions, all discussion is meaningless. Likewise in solving a problem by the principle of
virtual work, it is always essential to say whether you are describing the work done by a force on what part of the system (on the
ladder or on the floor?) and whether you are describing an increase or a decrease of some length or angle.
Let us move now to a slightly more difficult problem, which we’ll try by three different methods – including that of virtual work.
In Figure IX.5, a uniform rod AB of weight M g and length 2a is freely hinged at A. The end B carries a smooth ring of negligible
mass. A light inextensible string of length l has one end attached to a fixed point C at the same level as A and distant 2a from it. It
passes through the ring and carries at its other end a weight M g hanging freely. (The “smooth” ring means that the tension in the
1

10

string is the same on both sides of the ring.) Find the angle CAB when the system is in equilibrium.
I have marked in various angles and lengths, which can easily be determined from the geometry of the system, and I have also
marked the four forces on the rod.

9.3.2 https://phys.libretexts.org/@go/page/6984
Let us first try a very conventional method. We know rather little about the force R of the hinge on the rod (though see below), and
therefore this is a good reason for taking moments about the point A. We immediately obtain
1 1 1
M ga cos θ + M g.2a cos θ = M g.2a cos θ.
10 10 2

Divide by M ga and set cos θ = 2c − 1 , where c = cos θ . After a little algebra, we obtain 12c − c − 6 = 0 and hence we find
2 1

2
2

for the equilibrium condition that θ = 82o 49' or 263o 37'. The latter, by the way, is a physically valid solution – you might want to
sketch it.
Now let’s try the same problem using energy conditions. We’ll take the zero of potential energy when the rod is horizontal – at
which time the small mass is at a distance l below the level AC.
When the angle CAB = θ , the distance of the centre of mass of the rod below AC is a sin θ and the distance of the small mass
below AC is l − 4a sin θ + 2a sin θ so that the potential energy is
1

1 1 2 1
V = −M ga sin θ + M g[l − (l − 4a sin θ + 2a sin θ)] = − M ga(3 sin θ − sin θ)
10 2 3 2

The derivative is
dV 2 1 1
=− M ga(3 cos θ − cos θ),
dθ 3 2 2

and setting this to zero will produce the same results as before. Further differentiation (do it), or a graph of V : θ (do it), will show
that the 82o 49' solution is stable and the 263o 37' solution is unstable.
Now let’s try it by virtual work. We are going to increase θ by δθ and see how much work is done.

9.3.3 https://phys.libretexts.org/@go/page/6984
The distance of the centre of mass of the rod below AC is a sin θ , and if θ increases by δθ, this will increase by a cos θδθ , and the
work done by M g will be M ga cos θδθ .
The distance of the ring below AC is 2a sin θ , and if θ increases by δθ, this will increase by 2a cos θδθ, and the work done by the
downward force will be M g.2a cos θδθ .
1

10

The distance BC is 4a sin θ , and if

1

2
θ increases by δθ , this will increase by 2a cos
1

2
θδθ and the work done by the sloping force
will be MINUS M g.2a cos θδθ .
1

10
1

Thus the virtual work is

1 1 1
M g. a cos θδθ + M g.2a cos θδθ − M g.2a cos θδθ.
10 10 2

If we put this equal to zero, we obtain the same result as before.

This page titled 9.3: Virtual Work is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

9.3.4 https://phys.libretexts.org/@go/page/6984
9.E: Conservative Forces (Exercises)
 Exercise 9.E. 1

How long does it take a particle to fall to the ground from a height h , with a uniform acceleration g ? From elementary methods
−−
you know that the answer is √ 2h

g
, but we are going to use this simple example to illustrate the use of Equation 9.2.9. The use
of the equation to solve such a simple problem might seem a rather tedious way of solving an elementary problem, but the
method is useful in more difficult cases, and the use of the method is best introduced with a simple example.
Let us take the origin to be the ground, and we shall measure distances y upward from the ground. The speed is then given by
dy
v=−
dt
, so we use the negative sign in Equation 9.2.9a. The initial and final values of y are h and 0 respectively.
Alternatively, you can use Equation 9.2.9b, with the positive sign, in which case the loser and upper limits of integration are 0
and h respectively.
The total energy E is the initial potential energy mgh (since the initial kinetic energy is zero), and the potential energy at
height y is V (y) = mgy Equation 9.2.9 therefore takes the form
−
−− 0
m dy
t = −√ ∫ .
−−−−−−−− −
2 h √ mgh − mgy

 Exercise 9.E. 2

How long does it take for a particle, thrown vertically upwards with initial speed v , to reach a height h ? Again, by elementary
0

2
v0 −√v −2gh
0

methods, you will easily find (do it!) that the answer is t = g
, but let’s see if we can do it from Equation 9.2.9.
Let us take the origin to be the ground, and we shall measure distances y upward from the ground. The speed is then given by
dy
v=+
dt
, so we use the positive sign in Equation 9.2.9. The initial and final values of y are 0 and h respectively.
The total energy E is the initial kinetic energy mv (since the initial potential energy is zero), and the potential energy at
1

2
2
0

height y is V (y) = mgy . Equation 9.2.9 therefore takes the form

−
−− h
m dy
t = +√ ∫ −−−−−−−−−−.
2 0 1 2
√ mv − mgy
2 0

This yields the expected answer

−−−−−−−
2
v0 − √ v − 2gh
0

t = .
g

 Exercise 9.E. 3

In this example, we shall have a stone falling from a height that is not negligible compared with the radius of the Earth, so that
the acceleration is not constant. We shall suppose that we drop a stone from a point at a distance r = b from the centre of the
Earth, and we ask how long it will take to reach the surface of the Earth, radius a .
Let us take the origin to be the centre of the Earth, and we shall measure distances r radially outward from the centre. The
dy
speed is then given by v=−
dt
, so we use the negative sign in Equation 9.2.9. The initial and final values of r are b and a

respectively.
The total energy E is the initial potential energy − (since the initial kinetic energy is zero), and the potential energy at
GMm

distance r from the centre is V (r) = − . Equation 9.2.9 therefore takes the form
GMm

−
−− a
m dr
t = −√ ∫ (9.E.1)
−−−−−−−−−−−
2 b GMm GMm
√ −
r b

9.E.1 https://phys.libretexts.org/@go/page/6983
The gravity on the surface of Earth is g 0 =
GM

a
2
so that equation 9.E.1 can be written
−
− −
−
a
1 b dr
t =− √ ∫ −−−−− (9.E.2)
a 2g0 b b
√ −1
r

This can be integrated analytically to give

−
− −
−
a
b b 1
t = √ ∫ (α + sin 2α), (9.E.3)
a 2g0 b 2

where
a
2
cos α = . (9.E.4)
b

Here’s a numerical example: How long would it take for a stone to fall to Earth from an initial height of 240,000 miles? In case
the above method is too difficult, here’s another way to do it - in your head in a few seconds!
240,000 miles is the radius of the Moon's orbit. The stone is falling in a highly elliptical orbit of major axis equal to the
distance to the Moon - i.e. its semi major axis is half that of the Moon. Therefore by Kepler's third law, its period is equal to the
–
Moon's period (which is 28 days) divided by 2√2 which is 2.8. The orbital period of the stone is therefore 10 days. The time
taken to drop to Earth is half of this, or five days. You might want to calculate it from Equations 9.E.3 and 9.E.4 and see if you
get the same answer!

 Exercise 9.E. 4

The answer is that it is four times the time that it takes to rise from the vertical position to an angle α from the vertical. Thus
we shall work out this time from Equation 9.2.10 and multiply by four.

Let us adopt the upper end of the string as our level for zero potential energy. The potential energy V (θ) is then −mgl cos θ .
The total energy E is equal to the potential energy when θ = α ; that is to say, E = −mgl cos α . If we take the initial angle to
be 0 and the final angle to be α , then the upward motion is such that ω = + , and we choose the positive sign for Equation
dθ

dt

9.2.10. The rotational inertia I is ml . Thus Equation 9.2.10 gives

2

−−
α
8l dθ
P = 4t = √ ∫ − −−−−−−−− −. (9.E.5)
g 0 √ cos θ − cos α

This can doubtless be expressed in terms of special functions that most of us are unfamiliar with, so you will probably opt to
evaluate this numerically as a function of α . There is a small difficulty at the upper limit of the integration when the integrand
becomes infinite. Indeed, in all cases in which the system is at rest at either the start or the finish, the denominator of Equation
9.2.9 or 9.2.10 will necessarily be zero and hence the integrand will be infinite. In some cases, as in examples (i) to (iii), the
integral can be done analytically and there is no problem. If, however, as in the present example, the integration has to be done
numerically, there is a potential problem and some ingenuity (often by making a change of variable) will have to be exercised.

9.E.2 https://phys.libretexts.org/@go/page/6983
Using a trigonometric identity, you can write Equation 9.E.5 as
−−
α
4l dθ
P =√ ∫ −−−−−−−−−−−−−−.
g 0 2 1 2 1
√ sin α − sin θ
2 2

This doesn't get rid of the infinity, but now make a change of variable by letting
1 1
sin θ = sin α sin ϕ
2 2

and the difficulty will disappear. In particular, the expression for the period becomes
−
− π

l 2 2 dϕ
P =√ × ∫ .
−−−−−−−−−−−−−−
g π 0 2 1 2
√ 1 − sin α. sin ϕ
2

−
−
Below is a graph of the period, in units of 2π √ versus α .
l

A further question might be asked. For example: What is the amplitude of the pendulum swing such that the period is 10
−
−
√2 α
percent more than the small angle limit of 2π √
l

g
? In other words, solve the equation π
∫
0
dθ

√cos θ−cos α
= 1.1 for α . This
will be quite a challenge.
I make it α = 69º.325146

 Exercise 9.E. 5

A Semicircle, a Ring and a String

9.E.3 https://phys.libretexts.org/@go/page/6983
In this example, we have a smooth semicircular wire in a vertical plane. The radius of the semicircle is a . A ring of mass m at
P can slide smoothly around the ring. An elastic string of natural length 2a is attached to the ends of the wire at A and B and is
threaded through the ring. The force constant of the string is k . The ring is subjected to three conservative forces - gravity and
the tensions in the two parts of the string. The ring is smooth, so there is no nonconservative friction.
By geometry the lengths of the following are
AP: 2a cos θ
BP: 2a sin θ
PM: 2a sin θ cos θ = a sin 2θ
We are going to find the equilibrium position(s) of the ring, and see how long it takes to slide from one position on the
semicircle to another.
Before doing any calculations, let's think about the physics qualitatively. Suppose that it is a very heavy ring and a weak string.
In that case the ring will surely slide down to the bottom of the semicircle and stay there. On the other hand suppose that the
ring is not very heavy but the string is quite strong. In that case we may well imagine that the ring may rest in stable
equilibrium farther up the semicircle; indeed, if the string is very strong, the stable equilibrium position of the ring might be
quite near the top. Of course, by the symmetry of the situation, there will always be an equilibrium position at the bottom of the
semicircle, but, if the string is very strong, this position will be unstable, and, upon the slightest displacement, the ring will
snap up to a higher position. Whether the position at the bottom of the semicircle will be stable or unstable depends on the
mg
relative strengths of the weight of the ring and the tension in the string. Let us then, in anticipation, refer to the ratioka
by the
symbol λ . In fact we shall find that if λ > 0.586, the position at the bottom of the semicircle will be stable, but if λ is less than
this the positions of stable equilibrium will be higher up.
The extension of the string above its natural length is 2a(cos θ + sin θ − 1) and the depth of the ring below AB is
a sin 2θ .Therefore, if we take AB to be the level for zero gravitational potential energy, the potential energy (elastic plus

gravitational)) is
1 2
V (θ = ⌊2a(cos θ + sin θ − 1 ) ⌋ − mga sin 2θ
2

2 2
= ka ⌊2(cos θ + sin θ − 1 ) − λ sin 2θ⌋. (9.E.6)

Figure IX.3 shows this potential energy as a function of θ for several values of λ , including 0.586. We can see that, for λ
greater than this, the position at the bottom of the ring (θ = 45o) is the only equilibrium position and it is stable. For small
values of λ , the bottom, while an equilibrium position, is unstable, and there are two stable positions higher up. To calculate
these equilibrium positions exactly, we need to determine where the derivative of V is zero, and to find whether these positions
are stable or unstable, we need to examine the sign of the second derivative.
I leave it to the reader to work through the first derivative and to show that one condition for the derivative to be zero is for
0
cos θ to equal sin θ ; that is, θ = 45 , which corresponds to the bottom of the semicircle. Another condition for the first
derivative to be zero is slightly more challenging to find, but you should find that the derivative is zero if

9.E.4 https://phys.libretexts.org/@go/page/6983
 –
π √2
sin(θ + ) = .
4 2 −λ

–
This corresponds to a real value of θ only if λ ≤ 2 − √2 = 0.586 . The second derivatives are necessary to determine whether
the equilibria are stable (V a minimum) or unstable (V a maximum) or a glance at figure IX.3 will be easier.
Now let's express the potential energy as a function U of the angle ϕ , so that U (ϕ) = V (θ) . From Figure IX.2, we see that
ϕ = 90 − 2θ . After some algebra and trigonometry, I find that the potential energy as a function of ϕ is given by
∘

U (ϕ) 1 – 1 –
2
= 4 cos ϕ − 4 √2 cos ϕ − λ cos ϕ − λ − 4 + 4 √2.
2
ka 2 2

Now expand this carefully by Taylor's theorem as far as ϕ : 2

U (ϕ) – 2
= 2(λ + √2 − 2)ϕ .
2
ka

What we have done is to approximate the potential energy function U (ϕ) by a parabola for small ϕ . Now a parabolic potential
well is characteristic of simple harmonic motion. For example, in linear simple harmonic motion obeying the equation
−−
m ẍ = −kx the potential energy per unit mass is 1

2
kx
2
and the period is 2π √
m

k
. In rotational simple harmonic motion
−
−
obeying the equation I θ¨ = −cθ the potential energy per unit rotational inertia is 1

2
2
cθ and the period is 2π √ . The rotational
I

inertia here is just ma , so we find that the period of small oscillations is

2

−−−−−−−−−−− −
λa
P = π√ , (9.E.7)
–
(λ + √2 − 2)g

–
provided, of course, that λ > 2 − √2 .
If you want to find how long the ring takes to slide from an initial position ϕ = α to the bottom you can use Equations 9.2.10
and 9.E.6, with E = U (α) . You will find the usual difficulty that the integrand is zero at the start. I haven't actually tried the
problem, because it looks slightly tedious, but I am fairly certain that it can be integrated analytically. If you do it and get an
answer, make sure that, in the limit of small ϕ , you get the same as Equation 9.E.7

9.E.5 https://phys.libretexts.org/@go/page/6983
  Exercise 9.E. 6

A rod of length 2l and mass m has one end freely pivoted on a horizontal floor. The rod is held at an initial angle of 45o to the
vertical and then released. How long does it take for the rod to hit the floor? I'll leave you to work this one out. By the way, if I
had started with the rod vertical, you would find that it takes an infinite time to fall - because the vertical position, although
unstable, is an equilibrium position, and it would never get going unless given an infinitesimal displacement.

This page titled 9.E: Conservative Forces (Exercises) is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

9.E.6 https://phys.libretexts.org/@go/page/6983
 CHAPTER OVERVIEW
10: Rocket Motion
Topic hierarchy
10.1: Introduction
10.2: An Integral
10.3: The Rocket Equation
10.E: Rocket Motion (Exercises)

This page titled 10: Rocket Motion is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1
10.1: Introduction
If you are asked to state Newton's Second Law of Motion, I hope you will not reply: "Force equals mass times acceleration" -
because that is not Newton's Second Law of Motion. Newton's Second Law of Motion is:
The alteration of motion is ever proportional to the motive force impressed; and is made in the direction of the right line in
which that force is impressed.
In short: Force equals Rate of Change of Momentum, or, in symbols, F = ṗ . On differentiating the right hand side, we obtain
F = m v̇ + ṁv . In other words, if the mass is constant then indeed force equals mass times acceleration - but only if the mass

is constant. In a rocket, a very appreciable fraction of the mass of the rocket is fuel, which is burned and ejected at a very high
rate, so that the mass of the rocket is rapidly diminishing during the motion. It is one of the great problems of rocket design
that such a high proportion of the initial mass must be fuel. For this reason, other possible methods of driving spacecraft are
being investigated by many groups. For example, in the ion propulsion system of the Deep Space One spacecraft, electrically
accelerated ions are ejected at high speed from the spacecraft. The force produced and the acceleration are minute, but,
because it can be kept up for a very long time, very high speeds can eventually be reached. "Solar sail" systems similarly rely
on the very tiny force that can be exerted by the solar wind, but this tiny force can be exerted during most of the lifetime of a
spacecraft's flight, and hence again high speeds can be reached.
This chapter, however, concerns just conventional rocket motion. In the next section I consider the motion of a rocket in space
subject only to the one force from the high-speed ejection of burned fuel in the absence of any other forces. At a later date, if I
can find the time and energy, I may add further sections on rocket motion against gravity, which might be uniform or might
fall off with distance from Earth, and we might include air resistance or not. But to begin with, we deal solely with a rocket
isolated in space and subject to no additional forces.

This page titled 10.1: Introduction is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

10.1.1 https://phys.libretexts.org/@go/page/6988
10.2: An Integral
So that we don't get bogged down later with an integral that is going to crop up, see if you can do the following integration:

∫ ln(a − bt)dt

You should get −t − 1

b
(a − bt) ln(a − bt)+ constant.

This page titled 10.2: An Integral is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

10.2.1 https://phys.libretexts.org/@go/page/6989
10.3: The Rocket Equation
Initially at time t = 0, the mass of the rocket, including fuel, is m . 0

-1
We suppose that the rocket is burning fuel at a rate of b kg s so that, at time t , the mass of the rocket-plus-remaining-fuel is
m = m − bt . The rate of increase of mass with time is = −b and is supposed constant with time. (The rate of "increase" is,
dm
0
dt

of course, negative.)
We suppose that the speed of the ejected fuel, relative to the rocket, is V . The thrust of the ejected fuel on the rocket is therefore
V b, or −V . This is equal to the instantaneous mass times acceleration of the rocket:
dm

dt

dv dv
V b =m = (m0 − bt) . (10.3.1)
dt dt

Thus
v t
dt
∫ dv = V b ∫ . (10.3.2)
0 0 m0 − bt

t
(Don't be tempted to write the right hand side as −V b ∫
0
dt

bt−m0
. You are anticipating a logarithm, so keep the denominator
positive. We have met this before in Chapter 6.) On integration, we obtain
m0
v = V ln . (10.3.3)
m0 − bt

The acceleration is
dv V b
= . (10.3.4)
dt m0 − bt

At t = 0, the speed is zero and the acceleration is Vb

m0
.
m0
At time t = , the remaining mass is zero and the speed and acceleration are both infinite. However, this is so only if the initial
b

mass is 100% fuel and nothing else. This is not realistic. If the fraction of the total mass was initially f , the fuel will be completely
f m0
expended after a time b
at which time the speed will be (which is, of course, positive), and the speed will remain constant
thereafter. For example, if f = 99%, the final speed will be 4.6 V .
Equations 10.3.3 and 10.3.4 are shown in Figures X.1 and X.2. In Figure X.1, the speed of the rocket is plotted in units of V , the
m0
ejection speed of the burnt fuel. The time is plotted in units of . The fuel initially comprised 90% of the rocket, so that the rocket
b
m0
runs out of fuel in time 0.9 b
, at which time its speed is 2.3V . In Figure X.2, the acceleration is plotted in units of the initial
acceleration, which is Vb

m0
. When the fuel is exhausted, the acceleration is ten times this.

10.3.1 https://phys.libretexts.org/@go/page/6990
In Equation 10.3.3, v is of course dx

dt
, so the equation can be integrated to obtain the distance:time relation:
m0 bt
x = V [t + ( − t) ln(1 − )]. (10.3.5)
b m0

Elimination of t between Equations 10.3.3 and 10.3.5 gives the relation between speed and distance:
V m0 v −v

x = [1 − (1 + )e V
]. (10.3.6)
b V

f m0
If f is the fraction of the initial mass that is fuel, the fuel supply will be exhausted after a time , at which time its speed will be
b

−V ln(1 − f ) (this is positive, because 1 −f is less than 1), its acceleration will be and it will have travelled a distance
1

(1−f )

V m0 m0i

b
[f + (1 − f ) ln(1 − f )] . If the entire initial mass is fuel, so that f = 1, the fuel will burn for a time b
, at which time its
V m0
speed and acceleration will be infinite, it will have travelled a finite distance and the mass will have been reduced to zero,
b

This remarkable result is not very believable, for two reasons. In the first place it is not very realistic. More importantly, when the
speed becomes comparable to the speed of light, the equations which we have developed for nonrelativistic speeds are no longer
approximately valid, and the correct relativistic equations must be used. The speed cannot then reach the speed of light as long as
the remaining mass is non-zero.
Equations 10.3.5 and 10.3.6 are illustrated in Figures X.3 and X.4, in which , the fraction of the initial mass that is fuel, is 0.9. The
V m0 m0
units for distance, time and speed in these graphs are, respectively, , and V .
b b

10.3.2 https://phys.libretexts.org/@go/page/6990
This page titled 10.3: The Rocket Equation is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum
via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

10.3.3 https://phys.libretexts.org/@go/page/6990
10.E: Rocket Motion (Exercises)
 Exercise 10.E. 1

Derive the integral in section 2.

 Exercise 10.E. 2

Integrate equation 10.3.2 to obtain equation 10.3.3

 Exercise 10.E. 3

Integrate equation 10.3.3 to obtain equation 10.3.5

 Exercise 10.E. 4

Obtain equation 10.3.6

In the following problems, (numbers 5 - 8) assume:

V = 2 km s-1.
m0 = 2000 kg
b = 0.5 kg s-1
f = 90%

 Exercise 10.E. 5

What is the maximum speed, and how long does it take to attain it?

 Exercise 10.E. 6

How long does it take to reach a speed of 3 km s-1?

 Exercise 10.E. 7

How long does it take for the rocket to travel 600 km?

 Exercise 10.E. 8

How fast is it moving when it has travelled 300 km?

This page titled 10.E: Rocket Motion (Exercises) is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

10.E.1 https://phys.libretexts.org/@go/page/6991
 CHAPTER OVERVIEW
11: Simple and Damped Oscillatory Motion
Topic hierarchy
11.1: Simple Harmonic Motion
11.2: Mass Attached to an Elastic Spring
11.3: Torsion Pendulum
11.4: Ordinary Homogeneous Second-order Differential Equations
11.5: Damped Oscillatory Motion
11.5i: Light damping- \( \gamma < 2\omega_{0}\)
11.5ii: Heavy damping- \( \gamma > 2\omega_{0}\)
11.5iii: Critical damping- \( \gamma = 2\omega_{0}\)
11.6: Electrical Analogues

This page titled 11: Simple and Damped Oscillatory Motion is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated
by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

1
11.1: Simple Harmonic Motion
I am assuming that this is by no means the first occasion on which the reader has met simple harmonic motion, and hence in this
section I merely summarize the familiar formulas without spending time on numerous elementary examples
Simple harmonic motion can be defined as follows: It a point P moves in a circle of radius a at constant angular speed ω (and
hence period ) its projection Q on a diameter moves with simple harmonic motion. This is illustrated in Figure XI.1, in which
2π

the velocity and acceleration of P and of Q are shown as coloured arrows. The velocity of P is just aω and its acceleration is the
centripetal acceleration aω . As in Chapter 8 and elsewhere, I use blue arrows for velocity vectors and green for acceleration.
2

P0 is the initial position of P - i.e. the position of P at time t = 0 - and a is the initial phase angle. At time t later, the phase angle is
ωt + α . The projection of P upon a diameter is Q. The displacement of Q from the origin, and its velocity and acceleration, are

y = a sin(ωt + α) (11.1.1)

v = ẏ = aω cos(ωt + α) (11.1.2)

2
ÿ = −aω sin(ωt + α). (11.1.3)

Equations 11.1.2 and 11.1.3 can be obtained immediately either by inspection of Figure XI.1 or by differentiation of Equation
11.1.1. Elimination of the time from Equations 11.1.1 and 11.1.2 and from Equations 11.1.1 and 11.1.3 leads to

1
2 2
v = ẏ = ω(a −y ) 2
(11.1.4)

and
2
ÿ = −ω y (11.1.5)

An alternative definition of simple harmonic motion is to define as simple harmonic motion any motion that obeys the differential
Equation 11.1.5. We then have the problem of solving this differential Equation. We can make no progress with this unless we
remember to write ÿ as v (recall that we did this often in Chapter 6.) Equation 11.1.5 then immediately integrates to
dv

dy

2 2 2 2
v = ω (a −y )

dy
A further integration, with v = dt
, leads to

y = a sin(ωt + α)

provided we remember to use the appropriate initial conditions. Differentiation with respect to time then leads to Equation 11.1.2,
and all the other Equations follow.

11.1.1 https://phys.libretexts.org/@go/page/6995
  Exercise 11.1.1

Important Problem.
Show that y = a sin(ωt + α) can be written
y = A sin ωt + B cos ωt (11.1.6)

−−− −−−−
where A = a cos α and B = a sin α . The converse of these are 2 2
a = √A + B , cos α =
A
, sin α =
B
. It is
√A2 +B2 √A2 +B2

important to note that, if A and B are known, in order to calculate a without ambiguity of quadrant it is entirely necessary to
y
calculate both cos α and sin α . It will not do, for example, to calculate α solely from α = tan ( ) because this will give
−1

two possible solutions for a differing by 180o.

Show also that Equation 11.1.6 can also be written
iωt −iωt
y = Me +Ne , (11.1.7)

where M =
1

2
(B − iA) and N =
1

2
(B + iA) show that the right hand side of Equation 11.1.7 is real.

The four large satellites of Jupiter furnish a beautiful demonstration of simple harmonic motion. Earth is almost in the plane of their
orbits, so we see the motion of satellites projected on a diameter. They move to and fro in simple harmonic motion, each with
different amplitude (radius of the orbit), period (and hence angular speed ω ) and initial phase angle α .

This page titled 11.1: Simple Harmonic Motion is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

11.1.2 https://phys.libretexts.org/@go/page/6995
11.2: Mass Attached to an Elastic Spring
I am thinking of a mass m resting on a smooth horizontal table, rather than hanging downwards, because I want to avoid the
unimportant distraction of the gravitational force (weight) acting on the mass. The mass is attached to one end of a spring of force
constant k , the other end of the spring being fixed, and the motion is restricted to one dimension.
I suppose that the force required to stretch or compress the spring through a distance x is proportional to x and to no higher
powers; that is, the spring obeys Hooke's Law. When the spring is stretched by an amount x there is a tension kx in the spring;
when the spring is compressed by x there is a thrust kx in the spring. The constant k is the force constant of the spring.
When the spring is stretched by an distance x, its acceleration ẍ is given by

m ẍ = −kx. (11.2.1)

This is an Equation of the type 11.1.5, with ω 2

=
k

m
, and the motion is therefore simple harmonic motion of period
−
−−
2π m
P = = 2π √ . (11.2.2)
ω k

At this stage you should ask yourself two things: Does this expression have dimensions T? Physically, would you expect the
oscillations to be slow for a heavy mass and a weak spring? The reader might be interested to know (and this is literally true) that
−−
when I first typed Equation 11.2.2 , I inadvertently typed √
k

m
and I immediately spotted my mistake by automatically asking
−−
myself these two questions. The reader might also like to note that you can deduce that P∞ √
m

k
by the method of dimensions,
although you cannot deduce the proportionality constant 2π. Try it.
Energy Considerations. The work required to stretch (or compress) a Hooke's law spring by x is kx , and this can be described as
1

2
2

the potential energy or the elastic energy stored in the spring. I shall not pause to derive this result here. It is probably already
known by the reader, or s/he can derive it by calculus. Failing that, just consider that, in stretching the spring, the force increases
linearly from 0 to kx, so the average force used over the distance x is kx and so the work done is kx .
1

2
1

2
2

If we assume that, while the mass is oscillating, no mechanical energy is dissipated as heat, the total energy of the system at any
time is the sum of the elastic energy kx stored in the spring and the kinetic energy mv of the mass. (I am assuming that the
1

2
2 1

2
2

mass of the spring is negligible compared with m.)

Thus
1 1
2 2
E = kx + mv (11.2.3)
2 2

and there is a continual exchange of energy between elastic and kinetic. When the spring is fully extended, the kinetic energy is
zero and the total energy is equal to the elastic energy then, ka when the spring is unstretched and uncompressed, the energy is
1

2
2

entirely kinetic; the mass is then moving at its maximum speed aω and the total energy is equal to the kinetic energy then,
m a ω . Any of these expressions is equal to the total energy:
1 2 2
2

1 2
1 2
1 2
1 2 2
E = kx + mv = ka = ma ω (11.2.4)
2 2 2 2

This page titled 11.2: Mass Attached to an Elastic Spring is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

11.2.1 https://phys.libretexts.org/@go/page/6996
11.3: Torsion Pendulum
A torsion pendulum consists of a mass of rotational inertia I hanging by a thin wire from a fixed point. If we assume that the torque
required to twist the wire through an angle θ is proportional to θ and to no higher powers, then the ratio of the torque to the angle is
called the torsion constant c . It depends on the shear modulus of the material of which the wire is made, is inversely proportional to
its length, and, for a wire of circular cross-section, is proportional to the fourth power of its diameter. A thick wire is much harder
to twist than a thin wire. Ribbonlike wires have comparatively small torsion constants. The work required to twist a wire through
an angle θ is cθ .
1

2
2

When a torsion pendulum is oscillating, its Equation of motion is

¨
I θ = −cθ. (11.3.1)

−
−
This is an Equation of the form 11.1.5 and is therefore simple harmonic motion in which ω =√
c

I
. This example, incidentally,
shows that our second definition of simple harmonic motion (i.e. motion that obeys a differential Equation of the form of Equation
11.1.5) is a more general definition than our introductory description as the projection upon a diameter of uniform motion in a
circle. In particular, do not imagine that ω here is the same thing as θ˙ !

 Exercise 11.3.1
Write down the torsional analogues of all the Equations given for linear motion in Sections 11.1 and 11.2.

This page titled 11.3: Torsion Pendulum is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

11.3.1 https://phys.libretexts.org/@go/page/6997
11.4: Ordinary Homogeneous Second-order Differential Equations
This is not a full mathematical course on differential Equations, but it may be useful as a reminder for those who have already
studied differential Equations, and may even be just enough for our purposes for those who have not.
dy
We suppose that and y denotes . An ordinary homogenous second-order differential equation is an Equation of the
′

y = y(x)
dx
form
′′ ′

ay + b y + cy = 0, (11.4.1)

and we have to find a function y(x) which satisfies this. It turns out that it is quite easy to do this, although the nature of the
solutions depends on whether b is less than, equal to or greater than 4ac.
2

A first point to notice is that, if y = f (x) is a solution, so is Af (x) - just try substituting this in the Equation 11.4.1. If y = g(x) is
another solution, the same is true of g - i.e. Bg(x) is also a solution. And you can also easily verify that any linear combination,
such as

y = Af (x) + Bg(x), (11.4.2)

is also a solution.
Now Equation 11.4.1 is a second-order Equation - i.e. the highest derivative is a second derivative - and therefore there can be only
two arbitrary constants of integration in the solution - and we already have two in Equation 11.4.1, and consequently there are no
further solutions. All we have to do, then, is to find two functions that satisfy the differential Equation.
′ ′′

It will not take long to discover that solutions of the form y = e kx

satisfy the Equation, because then y = ky and y 2
=k y , and, if
you substitute these in Equation 11.4.1, you obtain
2
(ak + bk + c)y = 0. (11.4.3)

You can always find two values of k that satisfy this, although these may be complex, which is why the nature of the solutions
depends on whether b is less than or greater than 4ac. Thus the general solution is
2

k1 x k2 x
y = Ae + Be (11.4.4)

where k and k are the solutions of the Equation

1 2

2
ax + bx + c = 0. (11.4.5)

−b
There is one complication, however, if b 2
= 4ac because then the two solutions of Equation 11.4.5 are each equal to ( . The
(2a)

solution of the differential Equation is then

−bx
y = (A + B) exp[ ] (11.4.6)
(2a)

and the two constants can be combined into a single constant C = A+B so that Equation 11.4.6 can be written
−bx
y = C exp[ ]. (11.4.7)
(2a)

This solution has only one independent arbitrary constant, and so an additional solution must be possible. Let us try and see
whether a function of the form
mx
y = xe (11.4.8)

′ ′′

might be a solution of Equation . From Equation

11.4.1 11.4.8 we obtain y = (1 + mx)e
mx
and y = m(2 + mx)e
mx
. On
2
b
substituting these into Equation 11.4.1 , remembering that c = , we obtain for the left hand side of Equation 11.4.1 , after
(4a)

some algebra,

11.4.1 https://phys.libretexts.org/@go/page/6998
 mx
e 2
[(2am + b ) x + 4a(2am + b)]. (11.4.9)
4a

−b
This is identically zero if m = , and hence
(2a)

−bx
y = x exp[ ] (11.4.10)
(2a)

is a solution of Equation 11.4.1 the general solution of Equation 11.4.1, if b 2

= 4ac , is therefore
−bx
y = (C = Dx) exp[ ]. (11.4.11)
(2a)

We shall discover what these solutions actually look like in the next section.

This page titled 11.4: Ordinary Homogeneous Second-order Differential Equations is shared under a CC BY-NC 4.0 license and was authored,
remixed, and/or curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit
history is available upon request.

11.4.2 https://phys.libretexts.org/@go/page/6998
 SECTION OVERVIEW
11.5: Damped Oscillatory Motion
As pointed out in Section 11.2, the equation of motion for a mass m vibrating on the end of a spring of force constant k , in the
absence of any damping, is
m ẍ = −kx. (11.5.1)

Here, I am assuming that the displacement x is a function of time, and a dot denotes d

dt
.
However, in most real situations, there is some damping, or loss of mechanical energy, which is dissipated as heat. In the case of
our example of a mass oscillating on a horizontal table, damping may be caused by friction between the mass and the table. For a
mass hanging vertically from a spring, we might imagine the mass to be immersed in a viscous fluid. These are obvious examples.
Slightly less obvious, it may be that the constant bending and stretching of the spring produces heat, and the motion is damped
from this cause. In any case, in this analysis we shall assume that, in addition to the restoring force kx, there is also a damping
force that is proportional to the speed at which the particle is moving. I shall denote the damping force by bẋ. The equation of
motion is then
m ẍ = −kx − b ẋ. (11.5.2)

If I divide by m and write ω for2

0
k

m
and γ for b

m
, we obtain the equation of motion in its usual form
2
ẍ + γ ẋ + ω x = 0. (11.5.3)
0

Here γ is the damping constant, which we have already met in Chapter 10, and, from Section 11.4, we are ready to solve the
differential Equation 11.5.3. Indeed, we know that the general solution is
k1 t k2 t
x = Ae + Be , (11.5.4)

where k and k are the solutions of the quadratic equation

1 2

2 2
k + γk + ω = 0. (11.5.5)
0

An exception occurs if k 1 = k2 , and we shall deal with that exceptional case in due course (subsection 11.5iii). Otherwise, k and 1

k are given by
2

−−−− −−−− −−−−−−−

1 1 2
1 1
2 2
k1 = − γ +√ γ −ω , k2 = − γ −√ γ −ω . (11.5.6)
0 0
2 4 2 4

In Section 11.5.3, we pointed out that the nature of the solution depends on whether b is less than, equal to or greater than 4ac, or,
2

in the present case, upon whether γ is less than, equal to or greater than. These cases are referred to, respectively, as lightly
damped, critically damped and heavily damped. We shall start by considering light damping.

Topic hierarchy

11.5i: Light damping- \( \gamma < 2\omega_{0}\)

11.5ii: Heavy damping- \( \gamma > 2\omega_{0}\)

11.5iii: Critical damping- \( \gamma = 2\omega_{0}\)

This page titled 11.5: Damped Oscillatory Motion is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

11.5.1 https://phys.libretexts.org/@go/page/6999
11.5i: Light damping- γ<2 ω 0 γ<2ω0
Since γ < 2ω , we have to write Equations 11.5.6 as
0

−−−−−− −− −−−−−− −−
1 1 2
1 1 2
2 2
k1 = − γ + i√ ω − γ , k2 = − γ − i√ ω − γ . (11.5.7)
0 0
2 4 2 4

Further, I shall write

−−−−−−− −
′ 1
2 2
ω = √w − γ . (11.5.8)
0
4

Equation 11.5.4 is therefore

1 ′ 1 ′ 1 ′ ′
− γt+iω t − γt−iω t − γt +iω t −iω t
x = Ae 2 + Be 2 =e 2 (Ae + Be ). (11.5.9)

′ ′

If x is to be real, A and B must be complex. Also, since e −iω t

= (e
iω t
) * must equal A *, where the asterisk denotes the complex
conjugate.
Let A = (a − ib) and B =
1

2
1

2
(a + ib) where a and b are real. Then the reader should be able to show that Equation 11.5.9 can
be written as
1
′ ′
− γt
x =e 2 (a cos ω t + b sin ω t). (11.5.10)

−−−−−−
And if C 2 2
= √a + b , sin α =
a
2
, cos α =
b
2
, the equation can be written
√a2 +b √a2 +b

1
′
− γt
x = Ce 2 sin(ω t + α). (11.5.11)

Equations 11.5.9, 11.5.10 or 11.5.11 are three equivalent ways of writing the solution. Each has two arbitrary integration constants
(A, B), (a, b) or (C , α), whose values depend on the initial conditions - i.e. on the values of x and ẋ when t = 0 . Equation

11.5.11 shows that the motion is a sinusoidal oscillation of period a little less than ω , with an exponentially decreasing amplitude. 0

To find C and α in terms of the initial conditions, differentiate Equation 11.5.11 with respect to the time in order to obtain an
equation showing the speed as a function of the time:

−
1
γt
′ ′ 1 ′

ẋ = C e 2 [ ω cos(ω t + α) − γ sin(ω t + α)]. (11.5.12)

2

By putting t = 0 in Equations 11.5.11 and 11.5.12 we obtain

x0 = C sin α (11.5.13)

and
′ 1
(ẋ)0 = C (ω cos α − γ sin α). (11.5.14)
2

From these we easily obtain

1 (ẋ)0 γ
cot α = [ + ] (11.5.15)
′
ω ẋ0 2

and
C = x0 csc α. (11.5.16)

The quadrant of α can be determined from the signs of cot α and csc α , C always being positive.
1

Note that the amplitude of the motion falls off with time as e −
2
γt
, but the mechanical energy, which is proportional to the square of
the amplitude, falls off as e .−γt

Figures XI.2 and XI.3 are drawn for C = 1, α = 0, γ = 1 . Figure XI.2 has ω 0 = 25γ and hence x = e −
2
t
, and
sin 0.24.9949995t
1

figure XI.3 has ω 0 = 4γ and hence x = e −

2
t
sin 3.968626967t .

11.5i.1 https://phys.libretexts.org/@go/page/8954
 Exercise 11.5i. 1

Draw displacement : time graphs for an oscillator with m = 0.02 kg, k = 0.08 N m-1, g = 1.5 s-1, t = 0 to 15 s, for the following
initial conditions:
i. x = 0, (ẋ) = 4 ms-1
0 0

ii. (ẋ) = 0, x = 3 m
0 0

iii. (ẋ) = −2 ms-1, x = 2

0

Although the motion of a damped oscillator is not strictly "periodic", in that the motion does not repeat itself exactly, we could
define a "period" P = as the interval between two consecutive ascending zeroes. Extrema do not occur exactly halfway
2π
′
ω

between consecutive zeroes, and the reader should have no difficulty in showing, by differentiation of Equation 11.5.11 , that
′

extrema occur at times given by . However, provided that the damping is not very large, consecutive
′
2ω
tan(ω t + α) =
γ

extrema occur approximately at intervals of P

2
. The ratio of consecutive maximum displacements is, then,
1
− γt
| xn | e 2

= . (11.5.17)
1 1
| xn+1 | − γ(t+ P)
e 2 2

From this, we find that the logarithmic decrement is

| xn | Pγ
ln( ) = , (11.5.18)
| xn+1 | 4

from which the damping constant can be determined.

11.5i.2 https://phys.libretexts.org/@go/page/8954
This page titled 11.5i: Light damping- γ < 2ω is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
0

Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

11.5i.3 https://phys.libretexts.org/@go/page/8954
11.5ii: Heavy damping- γ>2 ω 0 γ>2ω0
The motion is given by Equations 11.5.4 and 11.5.6 where, this time, k and k are each real and negative. For convenience, I am
1 2

going to write λ = −k and λ = −k . λ are λ both real and positive, with λ > λ given by
1 1 2 2 1 2 2 1

−−−−−−−− − −−−−−−−− −
1 1 2
1 1 2
2 2
λ1 = γ − √( γ) − ω , λ2 = γ + √( γ) − ω (11.5.19)
0 0
2 2 2 2

The general solution for the displacement as a function of time is

−λ1 t −λ2 t
x = Ae + Be . (11.5.20)

The speed is given by

−λ1 t −λ2 t
ẋ = −Aλ1 e − Bλ2 e . (11.5.21)

The constants A and B depend on the initial conditions. Thus:

x0 = A + B (11.5.22)

and

(ẋ)0 = −(Aλ1 + Bλ2 ). (11.5.23)

From these, we obtain

(ẋ)0 + λ2 x0 (ẋ)0 + λ1 x0
A = , B = −[ ]. (11.5.24)
λ2 − λ1 λ2 − λ1

 Example 11.5ii. 1

x0 ≠ 0, (ẋ)0 = 0.

x0
−λ1 t −λ2 t
x = (λ2 e − λ1 e ). (11.5.25)
λ2 − λ1

Figure XI.4 shows x : t for x = 1 m, λ = 1 s-1, λ = 2 s-1.

0 1 2

The displacement will fall to half of its initial value at a time given by putting x

x0
=
1

2
in Equation 11.5.25. This will in
general require a numerical solution. In our example, however, the equation reduces to = 2e − e and if we let 1

2
−t −2t

u =e , this becomes u − 2u + = 0 . The two solutions of this are u = 1.707107 or 0.292893. The first of these gives a
−t 2 1

negative t, so we want the second solution, which corresponds to t = 1.228 seconds.

The velocity as a function of time is given by
λ1 λ2 x0
−λ1 t λ2 t
ẋ = − (e −e ). (11.5.26a)
λ2 − λ1

This is always negative. In figure XI.5, is shown the speed, which is |ẋ| as a function of time, for our numerical example.
Those who enjoy differentiating can show that the maximum speed is reached in a time −ẋ and that the maximum speed is
λ λ
2 1
λ1 λ2 x0 λ1 λ1

λ2 −λ1
[(
λ2
) λ −λ
2 1 −(
λ2
) λ −λ
2 1 ] . (Are these dimensionally correct?) In our example, the maximum speed, reached at
t = ln 2 = 0.6931 seconds, is 0.5 m s-1.

11.5ii.1 https://phys.libretexts.org/@go/page/8955
 Example 11.5ii. 2

x0 = 0, (ẋ)0 = V (> 0) .
In this case it is easy to show that
V −λ1 t −λ2 t
x = (e −e ). (11.5.26b)
λ2 − λ1

λ
2
λ
1 ln( )
λ1
It is left as an exercise to show that reaches a maximum value of V
when . Figure XI.6 illustrates
λ
1
x ( ) λ −λ
2 1 t =
λ2 λ2 λ2 −λ1

Equation 11.5.26a for λ = 1 s-1, λ = 2 s-1, V = 5 m s-1. The maximum displacement of 1.25 m is reached when
1 2

t = ln 2 = 0.6831 s. It is also left as an exercise to show that equation 11.5.26a can be written
1
− λt
2V e 2
1 2 2
x = sinh( γ − ω ). (11.5.27)
0
λ2 − λ1 4

11.5ii.2 https://phys.libretexts.org/@go/page/8955
 Example 11.5ii. 3

x0 ≠ 0, (ẋ)0 = −V .
This is the really exciting example, because the suspense-filled question is whether the particle will shoot past the origin at
some finite time and then fall back to the origin; or whether it will merely tamely fall down asymptotically to the origin
without ever crossing it. The tension will be almost unbearable as we find out. In fact, I cannot wait; I am going to plot x
versus t in figure XI.7 for λ = 1 s-1, λ = 2 s -1, x = 1 m, and three different values of V , namely 1, 2 and 3 m s-1.
1 2 0

We see that if V = 3 m s-1 the particle overshoots the origin after about 0.7 seconds. If V = 1 m s-1, it does not look as though it
will ever reach the origin. And if V = 2 m s-1, I'm not sure. Let's see what we can do. We can find out when it crosses the origin
by putting x= 0 in Equation 11.5.20, where A and B are found from Equations 11.5.24 with (ẋ) = −V . This gives, for the
0

time when it crosses the origin,

1 V − λ1 x0
t = ln( ). (11.5.28)
λ2 − λ1 V − λ2 x0

11.5ii.3 https://phys.libretexts.org/@go/page/8955
 Since λ > λ , this implies that the particle will overshoot the origin if V
2 1 > λ2 x0 , and this in turn implies that, for a given V ,
it will overshoot only if
2
V 2
+ω
x
2 0
0

γ < . (11.5.29)
V

x0

For our example, λ x = 2 m s-1, so that it just fails to overshoot the origin if V = 2 m s-1. For V = 3 m s-1, it crosses the origin
2 0

at t = ln 2 = 0.6931 s. In order to find out how far past the origin it goes, and when, we can do this just as in
I make it that it reaches its maximum negative displacement of -0.125 m at t = ln 4 = 1.386 s.

This page titled 11.5ii: Heavy damping- γ > 2ω is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
0

Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

11.5ii.4 https://phys.libretexts.org/@go/page/8955
11.5iii: Critical damping- γ=2 ω 0 γ=2ω0
Before embarking on this section, you might just want to refresh your memory of differential equations as described in Section
11.4.
γ
In this case, λ and λ are each equal to
1 2
2
. As discussed in Section 4, the general solution is of the form
1 1
− γt − γt
x = Ce 2 + Dte 2 , (11.5.30)

which can also be written in the form

1
− γt
x = Ce 2 (1 + at). (11.5.31)

Either way, there are two arbitrary constants, which can be determined by the initial values of the displacement and speed. It is easy
to show that
(ẋ)0 1
C = x0 and a = + γ. (11.5iii.1)
x0 2

The particle will not go through zero unless

1
V (= −(ẋ)0 ) > γ x0 . (11.5.33)
2

I'll leave it to the reader to draw a graph of Equation 11.5.19.

Ideally the hydraulic door closer that you see near the tops of doors in public buildings should be critically damped. This will cause
the door to close fastest without slamming. And we have already used the physics of impulsive forces in Problem 2.1 of Chapter 8
to work out where to place a door stop for minimum reaction on the hinges. Truly an understanding of physics is of enormous
importance in achieving the task of closing a door!
A more subtle example is in the design of a moving-coil ammeter. In this instrument, the electric current is passed through a coil
between the poles of a magnet, and the coil then swings around against the restoring force of a little spiral spring. The coil is wound
on a light aluminium frame called a former, and, as the coil (and hence the former) moves in the magnetic field, a little current is
induced in the former, and this damps the motion of the coil. In order that the coil and the pointer should move to the equilibrium
position in the fastest possible time without oscillating, the system should be critically damped - which means that the rotational
inertia and the electrical resistance of the little aluminium former has to be carefully designed to achieve this.

This page titled 11.5iii: Critical damping- γ = 2ω is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
0

Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

11.5iii.1 https://phys.libretexts.org/@go/page/8956
11.6: Electrical Analogues
A charged capacitor of capacitance C is connected in series with a switch and an inductor of inductance L. The switch is closed,
and charge flows out of the capacitor and hence a current flows through the inductor. Thus while the electric field in the capacitor
diminishes, the magnetic field in the inductor grows, and a back electromotive force (EMF) is induced in the inductor. Let Q be the
charge in the capacitor at some time. The current I flowing from the positive plate is equal to −Q
˙
. The potential difference across
Q
the capacitor is C
and the back EMF across the inductor is ˙ ¨
LI = −LQ . The potential drop around the whole circuit is zero, so
Q
that C
¨
= −LQ . The charge on the capacitor is therefore governed by the differential equation
Q
¨
Q =− , (11.6.1)
LC

which is simple harmonic motion with ω 0 =

1

√LC
. You should verify that this has dimensions T-1.

If there is a resistor of resistance R in the circuit, while a current flows through the resistor there is a potential drop ˙
RI = −RQ

across it, and the differential equation governing the charge on the capacitor is then
¨ ˙
LC Q + RC Q + Q = 0. (11.6.2)

This is damped oscillatory motion, the condition for critical damping being
4L
2
R = .
C

In fact, it is not necessary actually to have a physical resistor in the circuit. Even if the capacitor and inductor were connected by
superconducting wires of zero resistance, while the charge in the circuit is slopping around between the capacitor and the inductor,
it will be radiating electromagnetic energy into space and hence losing energy. The effect is just as if a resistance were in the
circuit.
If a battery of EMF E were in the circuit, the differential equation for Q would be
¨ ˙
LC Q + RC Q + Q = EC . (11.6.3)

This is not quite an equation of the form 11.4.1, and I shan't spend time on it here. However, if we are interested in the current as a
function of time, we just differentiate Equation 11.6.3 with respect to time:
¨ ˙
LC I + RC I + I = 0. (11.6.4)

This page titled 11.6: Electrical Analogues is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum
via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

11.6.1 https://phys.libretexts.org/@go/page/7000
 CHAPTER OVERVIEW
12: Forced Oscillations
Topic hierarchy
12.1: More on Differential Equations
12.2: Forced Oscillatory Motion
12.3: Electrical Analogue

This page titled 12: Forced Oscillations is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1
12.1: More on Differential Equations
In Section 11.4 we argued that the most general solution of the differential equation
′′ ′
ay + b y + cy = 0 (11.4.1)

is of the form

y = Af (x) + Bg(x). (11.4.2)

In this chapter we shall be looking at equations of the form

′′ ′
ay + b y + cy = h(x). (12.1.1)

If you look back at the arguments that led to the conclusion that Equation 11.4.2 is the most general solution of Equation 11.4.1,
you will be able to conclude that 11.4.2 is still a solution of Equation 11.4.1, but it is not the only solution. There is another
function that is a solution, so that the most general solution to equation 12.1.1 is of the form

y = Af (x) + Bg(x) + H (x). (12.1.2)

The solution H (x) is called the particular integral, while the part Af (x) + Bg(x) is the complementary function. I shall be
dealing in this chapter mainly with the particular integral, though we shall not entirely forget the complementary function.
This is a book on classical mechanics rather than on differential equations, so I am not going into how to obtain the particular
integral H (x) for a given h(x). There are several ways of doing it; for those who know what they are and are in practice with
them, Laplace transforms are among the more attractive methods. Some readers will already know how to do it. They will
doubtless want to go back to Equation 11.6.3 in the previous chapter and try their hand at finding the particular integral for that.
Those who do not may be happy and content to take my word for the particular integral in the sections that follow, or perhaps at
least to differentiate it to verify that it is indeed a solution.

This page titled 12.1: More on Differential Equations is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

12.1.1 https://phys.libretexts.org/@go/page/7002
12.2: Forced Oscillatory Motion
We are thinking of a mass m attached to a spring of force constant k and subject to a damping force bẋ , but also subject to a
periodic sinusoidal force F^ cos ωt . The equation of motion is
^
m ẍ + b ẋ + kx = F cos ωt, (12.2.1)

or, if we divide by m:
2 ^
ẍ + γ ẋ + ω = f cos ωt. (12.2.2)
0

^
Here γ = , ω = , f^ =
b

m
2
0 m
k F

m
. ω is the forcing angular frequency and ω is the natural frequency of mass and spring in the
0

absence of damping. One part of the general solution of Equation 12.2.2 is the complementary function, which we have dealt with
at length in Chapter 11. In this section I shall be interested in the particular integral. I shall not derive it here (those who are familiar
with differential equations will be able to do so), but you should at least verify by differentiation and substitution that the following
is a solution, and it is indeed the particular integral:
^
f 2
2
x = [(ω − ω ) cos ωt + γω sin ωt] . (12.2.3)
2 2 2 2 2 0
(ω −ω ) +γ ω
0

This can also be written

^
f
x = cos(ωt − α), (12.2.4)
1
2 2 2 2 2
[(ω −ω ) +γ ω ] 2
0

where
2 2
ω −ω γω γω
0
cos α = , cos α = , tan α = . (12.2.5)
1 1
2 2
2 2 2 2 2 2 2 2 2 2 ω −ω
[(ω −ω ) +γ ω ] 2 [(ω −ω ) +γ ω ] 2 0
0 0

The response frequency is the same as the forcing frequency, but there is a phase lag between x and F . Figure XII.1 shows α as a
γ
function of Ω = ω
for several different values of Γ = .
ω0 ω0

The particular integral can also be written

x =x
^ cos(ωt − α), (12.2.6)

where the displacement amplitude x

^ varies with forcing frequency ω as

^
f
^ =
x . (12.2.7)
1
2 2 2 2 2
[(ω −ω ) +γ ω ] 2
0

If we now introduce dimensionless quantities

ω γ x
^
^
Ω = , Γ = , X = , (12.2.8)
^
ω0 ω0 f

2
ω
0

Equations 12.2.5 and 12.2.7 become

ΓΩ
tan α = (12.2.9)
1 − Ω2

and
1
^
X = . (12.2.10)
1

[(1 − Ω2 )2 + Γ2 Ω2 ] 2

12.2.1 https://phys.libretexts.org/@go/page/7003
 ^
The phase lag α and the displacement amplitude X
^
(=
x

f
^
) are shown as a function of forcing frequency for various values of the
( )
2
ω
0

damping constant in figures XII.1 and 2.

A common misunderstanding is that the displacement amplitude is greatest when the forcing frequency is equal to the undamped
frequency ω . That this is far from the case is immediately obvious from a glance at figure XII.2. We can find the frequency that
0

results in the greatest displacement amplitude by maximizing Equation 12.2.7 or 12.2.10. This is most easily achieved by
minimizing the square of the denominator. Let D be the square of the denominator of equation 12.2.10, and let W = Ω and 2

–
G = Γ . Then D = (1 − W ) + GW , which is greatest for W = 1 − G or, provided γ < √2ω
2 2 1
0
2

−−−−−− −−
1
2 2
ω = √ω − γ . (12.2.11)
0
2

This is less not only than ω , but also less than ω . For the frequency given by equation 12.2.11, the displacement amplitude will be
′

^
f
^max =
x . (12.2.12)
1
2 1
γ(ω − γ2) 2
0 4

The locus of the maxima in figure XII.2 is found by eliminating γ from equation 12.2.7 and ^
dx

dω
=0 which gives

12.2.2 https://phys.libretexts.org/@go/page/7003
 ^
f
x
^max = −−−−−−. (12.2.13)
4
√ω − ω4
0

The solution given by Equations 12.2.7 and 12.2.8 is the particular integral. As pointed out in Section 1 of this chapter, the
complete solution is the sum of the particular integral and the complementary function, the latter being the unforced solutions of
Chapter 11. The particular integral represents the steady state solution, whereas the complementary function, which dies out with
time, is a transient solution. When a mechanical oscillation is started, or when an alternating current electric circuit is first switched
on, the solution is the sum of transient and steady state parts, the former more or less rapidly dying away. Often when an electric
fuse blows, the overload is caused by the large, but temporary, amplitude of the transient part of the solution.
Equations 12.2.7 and 12.2.8 give the displacement of the system as a function of time. Differentiation with respect to time gives
the velocity as a function of time. Thus:
^ sin(ωt − α),
ẋ = v = −v (12.2.14)

where
^
f ω
^ =
v (12.2.15)
1
2 2 2 2 2
[(ω −ω ) +γ ω ] 2
0

is the velocity amplitude. In dimensionless units, this can be written

Ω
^ =
V , (12.2.16)
1
2 2 2 2
[(1 − Ω ) +Γ Ω ] 2

where
^
v
^
V = . (12.2.17)
^
f

ω0

This is illustrated in figure XII.3.

^
f
It is left to the reader to show that the velocity amplitude is greatest and equal to γ
when ω = ω .
0

We have now found the phase lag and the displacement and velocity amplitudes as a function of forcing frequency, but I must now
try the reader's patience one step further for the most important part of the analysis, which really must not be skipped. Damping of
oscillatory motion implies that some of the mechanical energy (which, in an undamped system, alternates between kinetic and
potential energy) is lost - or, rather, that it is dissipated as heat. This happens if the damping is caused by the oscillator being
immersed in a viscous fluid, or if it is caused by the repeated expansion and compression of a spring, or, in an electric circuit, by

12.2.3 https://phys.libretexts.org/@go/page/7003
the dissipation of heat in the resistive part of the circuit. We aim now to find the rate at which the mechanical energy is dissipated
as heat.
We return to the equation of motion, Equation 12.2.1:
^
m ẍ + b ẋ + kx = F cos ωt

and multiply each side by ẋ:

2 ^
m ẋẍ + b ẋ + kx ẋ = ẋF cos ωt. (12.2.18)

Introduce the total mechanical energy:

1 1
2 2
E = m ẋ + kx . (12.2.19)
2 2

The instantaneous rate of change of E is mẋẍ + kx ẋ while the instantaneous rate at which F does work is ẋF cos ωt. The
difference (see Equation 12.2.18), bẋ , is therefore the rate at which work is being dissipated as heat, which, of course, is zero if
2

b = 0.

The average of bẋ over a complete period is

2

2
^ 2
¯¯¯¯¯¯
¯ bf ω ¯
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
¯
2 2
bẋ = sin (ωt − α) (12.2.20)
2 2 2 2
(ω −ω )+γ ω
0

¯
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
¯
where the bars denote the average value over a period. But 2
sin (ωt − α) =
1

2
, so the average rate at which work is being
dissipated as heat, for which I shall use the symbol Q
˙
is
2
^ 2
bf ω
˙
Q = (12.2.21)
2 2 2 2 2
2[(ω −ω ) +γ ω ]
0

The reader should check that the right hand side has the dimensions of rate of dissipation of energy and hence the SI unit of watts.
∗ ˙
Q
In dimensionless units, in which Q
˙
=
^
2
this can be written
bf
( )
2ω
0

2
∗ Ω
˙
Q = . (12.2.22)
2 2 2 2
(1 − Ω ) +Γ Ω

This is illustrated in figure XII.4. The reader can easily prove that the rate at which work is dissipated as heat is greatest when the
forcing frequency is equal to ω .
0

12.2.4 https://phys.libretexts.org/@go/page/7003
Summary
Phase lag: Equation 12.2.5
2 2
ω −ω γω γω
cos α =
0

1
, cos α = 1
, tan α = 2
ω −ω2
.
2 2 0
2 2 2 2 2 2 2 2
[( ω −ω ) +γ ω ] 2 [( ω −ω ) +γ ω ] 2
0 0

Displacement amplitude: Equation 12.2.7

^
f
^ =
x 1
2 2
2 2 2
[( ω −ω ) +γ ω ] 2
0

Velocity amplitude: Equation 12.2.15

^
f ω
v
^ =
1
2 2
[( ω −ω2 ) +γ 2 ω2 ] 2
0

Rate of dissipation of work as heat: Equation 12.2.21

2
^ 2
mγf ω
˙
Q =
2 2
2
2 2
.
2[( ω −ω ) +γ ω ]
0

In terms of dimensionless variables,

Phase lag: Equation 12.2.9
ΓΩ
tan α =
2
1−Ω

Displacement amplitude: Equation 12.2.10

^ 1
X = 1
2 2 2 2
[(1−Ω ) +Γ Ω ] 2

Velocity amplitude: 12.2.16

^ Ω
V = 1
2 2 2 2
[(1−Ω ) +Γ Ω ] 2

Rate of dissipation of work as heat: 12.2.22

∗ 2
˙ Ω
Q = 2
2 2 2
(1−Ω ) +Γ Ω

This page titled 12.2: Forced Oscillatory Motion is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

12.2.5 https://phys.libretexts.org/@go/page/7003
12.3: Electrical Analogue
Suppose that an alternating potential difference E = E ^ sin ωt is applied across an LCR circuit. We refer to Equation 11.6.3, and we

see that the equation that governs the charge on the capacitor is
Q
LQ̈ + RQ̇ + ^ sin ωt.
=E (12.3.1)
C

We can differentiate both sides with respect to time, and divide by L, and hence see that the current is given by
^
R 1 Eω
¨ ˙
I + I + I = cos ωt. (12.3.2)
L LC L

We can compare this directly with Equation 12.2.2, so that we have

^
R 1 Eω
2 ^
γ = , ω = − ,
−− f = . (12.3.3)
0
L √LC L

Then, by comparison with Equation 12.2.5, we see that I will lag behind E by α , where
Rω

L
R
tan α = = . (12.3.4)
1 2 1
−ω − Lω
LC Cω

This is just what we obtain from the more familiar complex number approach to alternating current circuits.

This page titled 12.3: Electrical Analogue is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum
via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

12.3.1 https://phys.libretexts.org/@go/page/7004
 CHAPTER OVERVIEW
13: Lagrangian Mechanics
Sometimes it is not all that easy to find the equations of motion and there is an alternative approach known as lagrangian
mechanics which enables us to find the equations of motion when the newtonian method is proving difficult. In lagrangian
mechanics we start, as usual, by drawing a large, clear diagram of the system, using a ruler and a compass. But, rather than drawing
the forces and accelerations with red and green arrows, we draw the velocity vectors (including angular velocities) with blue
arrows, and, from these we write down the kinetic energy of the system. If the forces are conservative forces (gravity, springs and
stretched strings), we write down also the potential energy. That done, the next step is to write down the lagrangian equations of
motion for each coordinate. These equations involve the kinetic and potential energies, and are a little bit more involved than
F = ma, though they do arrive at the same results.

13.1: Introduction to Lagrangian Mechanics

13.2: Generalized Coordinates and Generalized Forces
13.3: Holonomic Constraints
13.4: The Lagrangian Equations of Motion
13.5: Acceleration Components
13.6: Slithering Soap in Conical Basin
13.7: Slithering Soap in Hemispherical Basin
13.8: More Lagrangian Mechanics Examples
13.9: Hamilton's Variational Principle

This page titled 13: Lagrangian Mechanics is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum
via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1
13.1: Introduction to Lagrangian Mechanics
The usual way of using newtonian mechanics to solve a problem in dynamics is first of all to draw a large, clear diagram of the
system, using a ruler and a compass. Then mark in the forces on the various parts of the system with red arrows and the
accelerations of the various parts with green arrows. Then apply the equation F = ma in two different directions if it is a two-
dimensional problem or in three directions if it is a three-dimensional problem, or τ = I θ¨ if torques are involved. More correctly, if
a mass or a moment of inertia is not constant, the equations are F = ṗ and τ = L̇ . In any case, we arrive at one or more equations
of motion, which are differential equations which we integrate with respect to space or time to find the desired solution. Most of us
will have done many, many problems of that sort.
Sometimes it is not all that easy to find the equations of motion as described above. There is an alternative approach known as
lagrangian mechanics which enables us to find the equations of motion when the newtonian method is proving difficult. In
lagrangian mechanics we start, as usual, by drawing a large, clear diagram of the system, using a ruler and a compass. But, rather
than drawing the forces and accelerations with red and green arrows, we draw the velocity vectors (including angular velocities)
with blue arrows, and, from these we write down the kinetic energy of the system. If the forces are conservative forces (gravity,
springs and stretched strings), we write down also the potential energy. That done, the next step is to write down the lagrangian
equations of motion for each coordinate. These equations involve the kinetic and potential energies, and are a little bit more
involved than F = ma , though they do arrive at the same results.
I shall derive the lagrangian equations of motion, and while I am doing so, you will think that the going is very heavy, and you will
be discouraged. At the end of the derivation you will see that the lagrangian equations of motion are indeed rather more involved
than F = ma , and you will begin to despair – but do not do so! In a very short time after that you will be able to solve difficult
problems in mechanics that you would not be able to start using the familiar newtonian methods, and the speed at which you do so
will be limited solely by the speed at which you can write. Indeed, you scarcely have to stop and think. You know straight away
what you have to do. Draw the diagram. Mark the velocity vectors. Write down expressions for the kinetic and potential energies,
and apply the lagrangian equations. It is automatic, fast, and enjoyable.
Incidentally, when Lagrange first published his great work La méchanique analytique (the modern French spelling would be
mécanique), he pointed out with some pride in his introduction that there were no drawings or diagrams in the book – because all of
mechanics could be done analytically – i.e. with algebra and calculus. Not all of us, however, are as gifted as Lagrange, and we
cannot omit the first and very important step of drawing a large and clear diagram with ruler and compass and marking all the
velocity vectors.

This page titled 13.1: Introduction to Lagrangian Mechanics is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated
by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

13.1.1 https://phys.libretexts.org/@go/page/7009
13.2: Generalized Coordinates and Generalized Forces
In two-dimensions the positions of a point can be specified either by its rectangular coordinates (x, y) or by its polar coordinates.
There are other possibilities such as confocal conical coordinates that might be less familiar. In three dimensions there are the
options of rectangular coordinates (x, y, z), or cylindrical coordinates ρ, ϕ, z or spherical coordinates r, ω, ϕ – or again there may
be others that may be of use for specialized purposes (inclined coordinates in crystallography, for example, come to mind). The
state of a molecule might be described by a number of parameters, such as the bond lengths and the angles between the bonds, and
these may be varying periodically with time as the molecule vibrates and twists, and these bonds lengths and bond angles constitute
a set of coordinates which describe the molecule. We are not going to think about any particular sort of coordinate system or set of
coordinates. Rather, we are going to think about generalized coordinates, which may be lengths or angles or various combinations
of them. We shall call these coordinates (q , q , q , . . .). If we are thinking of a single particle in three-dimensional space, there will
1 2 3

be three of them, which could be rectangular, or cylindrical, or spherical. If there were N particles, we would need 3N coordinates
to describe the system – unless there were some constraints on the system.
With each generalized coordinate q is associated a generalized force P , which is defined as follows. If the work required to
j j

increase the coordinate q by δq is P δq , then P is the generalized force associated with the coordinate q .
j j j j j j

A generalized force need not always be dimensionally equivalent to a force. For example, if a generalized coordinate is an
angle, the corresponding generalized force will be a torque.

One of the things that we shall want to do is to identify the generalized force associated with a given generalized coordinate.

This page titled 13.2: Generalized Coordinates and Generalized Forces is shared under a CC BY-NC 4.0 license and was authored, remixed,
and/or curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is
available upon request.

13.2.1 https://phys.libretexts.org/@go/page/7010
13.3: Holonomic Constraints
The complete description of a system of N unconstrained particles requires 3N coordinates. You can think of the state of the
system at any time as being represented by a single point in 3N -dimensional space. If the system consists of molecules in a gas, or
a cluster of stars, or a swarm of bees, the coordinates will be continually changing, and the point that describes the system will be
moving, perhaps completely unconstrained, in its 3N -dimensional space.
However, in many systems, the particles may not be free to wander anywhere at will; they may be subject to various constraints. A
constraint that can be described by an equation relating the coordinates (and perhaps also the time) is called a holonomic constraint,
and the equation that describes the constraint is a holonomic equation. If a system of N particles is subject to k holonomic
constraints, the point in 3N -dimensional space that describes the system at any time is not free to move anywhere in 3N -
dimensional space, but it is constrained to move over a surface of dimension 3N − k . In effect only 3N − k coordinates are
needed to describe the system, given that the coordinates are connected by k holonomic equations.
Incidentally, I looked up the word “holonomic” in The Oxford English Dictionary and it said that the word was from the Greek ő
λος , meaning “whole” or “entire” and ν ὸ μ − ος , meaning “law”. It also said “applied to a constrained system in which the

equations defining the constraints are integrable or already free of differentials, so that each equation effectively reduces the
number of coordinates by one; also applied to the constraints themselves.”
As an example, consider a bar of wet soap slithering around in a hemispherical basin of radius a . You can describe its position in
the basin by means of the usual two spherical angles (θ, ϕ) ; the motion is otherwise constrained by its remaining in contact with
the basin; that is to say it is subject to the holonomic constraint r = a . Thus instead of needing three coordinates to describe the
position of a totally unconstrained particle, we need only two coordinates.
Or again, consider the double pendulum shown in Figure XIII.1, and suppose that the pendulum is constrained to swing only in the
plane of the paper – or of the screen of your computer monitor.

Two unconstrained particles would require six coordinates to specify their positions but this system is subject to four holonomic
constraints. The holonomic equations z = 0 and z = 0 constrain the particles to be moving in a plane, and, if the strings are kept
1 2

taut, we have the additional holonomic constraints x + y = l and (x − x ) + (y − y ) = l . Thus only two coordinates
2
1
2
1
2
1 2 1
2
2 1
2 2
2

are needed to describe the system, and they could conveniently be the angles that the two strings make with the vertical.

This page titled 13.3: Holonomic Constraints is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

13.3.1 https://phys.libretexts.org/@go/page/7011
13.4: The Lagrangian Equations of Motion
This section might be tough – but do not be put off by it. I promise that, after we have got over this section, things will be easy. But
in this section I do not like all these summations and subscripts any more than you do.
Suppose that we have a system of N particles, and that the force on the ith particle (i = 1 to N ) is F . If the ith particle undergoes i

a displacement δr , the total work done on the system is ∑ F ⋅ ∂ r . The position vector r of a particle can be written as a
i i i i

function of its generalized coordinates; and a change in r can be expressed in terms of the changes in the generalized coordinates.
Thus the total work done on the system is
∂ri
∑ Fi ⋅ ∑ δqj (13.4.1)
∂qj
i j

which can be written

∂ri
∑ ∑ Fi ⋅ δqj . (13.4.2)
∂qj
j i

But by definition of the generalized force, the work done on the system is also

∑ Pj ⋅ δqj . (13.4.3)

Thus the generalized force P associated with generalized coordinate q is given by

j j

∂ri
Pj = ∑ Fi ⋅ . (13.4.4)
∂qj
i

Now F i = mi r̈ i , so that
∂ri
Pj = ∑ mi r̈ i ⋅ . (13.4.5)
∂qj
i

Also
d ∂ri ∂ri d ∂ri
( ṙi ⋅ ) = r̈i ⋅ + ṙi ⋅ ( ). (13.4.6)
dt ∂qj ∂qj dt ∂qj

∂ri
Substitute for r̈ i ⋅ from Equation 13.4.6 into Equation 13.4.5 to obtain
∂qj

d ∂ri d ∂ri
Pj = ∑ mi [ ( ṙi ⋅ ) − ṙi ⋅ ( )] . (13.4.7)
dt ∂qj dt ∂qj
i

Now
∂ri ∂ṙi
= (13.4.8)
∂qj ∂q̇ j

and
d ∂ri ∂ṙi
( ) = . (13.4.9)
dt ∂qj ∂qj

Therefore
d ∂ṙi ∂ṙi
Pj = ∑ mi [ ( ṙi ⋅ ) − ṙi ⋅ ( )] (13.4.10)
dt ∂q̇ j
∂qj
i

You may not be immediately comfortable with the assertions in Equations 13.4.8 and 13.4.9 so I’ll interrupt the flow briefly
here with an example to try to justify these assertions and to understand what they mean.

13.4.1 https://phys.libretexts.org/@go/page/7012
 Consider the relation between the coordinate x and the spherical coordinates r, θ, ϕ:
x = r sin θ cos ϕ (A1)

In this example, x would correspond to one of the components of r , and r, θ, ϕ are the q
i 1, q2 , q3 .

From Equation A1, we easily derive

∂x
= sin θ cos ϕ (A2.1)
∂r

∂x
= r cos θ cos ϕ (A2.2)
∂θ

∂x
= −r sin θ sin ϕ (A2.3)
∂ϕ

and differentiating Equation A1 with respect to time, we obtain

˙ ˙
ẋ = ṙ sin θ cos ϕ + r cos θθ cos ϕ − r sin θ sin ϕϕ (A3)

And from this we see that

∂ẋ
= sin θ cos ϕ (A4.1)
∂ ṙ

∂ẋ
= r cos θ cos ϕ (A4.2)
˙
∂θ

∂ẋ
= −r sin θ sin ϕ (A4.3)
˙
∂ϕ

Thus the first assertion is justified in this example, and I think you’ll see that it will always be true no matter what the
functional dependence of r on the q .
i j

For the second assertion, consider

∂x
= sin θ cos ϕ (A5)
∂r

and hence
d ∂x
˙ ˙
= cos θθ cos ϕ − sin θ sin ϕϕ . (A6)
dt ∂r

From Equation A3 we find that

∂ẋ
= cos θθ̇ cos ϕ − sin θ sin ϕϕ̇ , (A7)
∂r

and the second assertion is justified. Again, I think you’ll see that it will always be true no matter what the functional
dependence of r on the q .
i j

The kinetic energy T is

1 1
2
T =∑ mi ṙ =∑ mi ṙi ⋅ ṙi (13.4.11)
i
2 2
i i

Therefore
∂T ∂ṙi
= ∑ mi ṙi ⋅ (13.4.12)
∂qj ∂qj
i

and

13.4.2 https://phys.libretexts.org/@go/page/7012
 ∂T ∂ṙi
= ∑ mi ṙi ⋅ . (13.4.13)
∂q̇ ∂q̇
j i j

On substituting these in Equation 13.4.10 we obtain

d ∂T ∂T
Pj = − . (13.4.14)
dt ∂q̇ j
∂qj

This is one form of Lagrange’s equation of motion, and it often helps us to answer the question posed in the last sentence of Section
13.2 – namely to determine the generalized force associated with a given generalized coordinate.

Conservative Forces
If the various forces in a particular problem are conservative (gravity, springs and stretched strings, including valence bonds in a
molecule) then the generalized force can be obtained by the negative of the gradient of a potential energy function – i.e.
∂V
Pj = − . In that case, Lagrange’s equation takes the form
∂qj

d ∂T ∂T ∂V
− =− . (13.4.15)
dt ∂q̇ j
∂qj ∂qj

In my experience, this is the most useful and most often encountered version of Lagrange’s equation.
The quantity L = T − V is known as the lagrangian for the system, and Lagrange’s equation can then be written
d ∂L ∂L
− = 0. (13.4.16)
dt ∂q̇ j
∂qj

This form of the equation is seen more often in theoretical discussions than in the practical solution of problems. It does enable us
∂L
to see one important result. If, for one of the generalized coordinates, =0 (this could happen if neither T nor V depends on q j
∂qj

∂T ∂V
– but of course it could also happen if and were nonzero but equal and opposite in sign), then that generalized coordinate
∂qj ∂qj

is called an ignorable coordinate – presumably because one can ignore it in setting up the lagrangian. However, it does not really
∂L
mean that it should be ignored altogether, because it immediately reveals a constant of the motion. In particular, if = 0, then
∂qj

∂L ∂L
is constant. It will be seen that if q has the dimensions of length,
j has the dimensions of linear momentum. And if q is i
∂q̇ j
∂q̇ j

∂L ∂L
an angle, has the dimensions of angular momentum. The derivative is usually given the symbol pj and is called the
∂q̇ j
∂q̇ j

generalized momentum conjugate to the generalized coordinate qj . If q is an “ignorable coordinate”, then

j pj is a constant of the
motion.
In each of Equations 13.4.14, 13.4.15 and 13.4.16 one of the qs has a dot over it. You can see which one it is by thinking about the
dimensions of the various terms. Dot has dimension T-1.
So, we have now derived Lagrange’s equation of motion. It was a hard struggle, and in the end we obtained three versions of an
equation which at present look quite useless. But from this point, things become easier and we rapidly see how to use the equations
and find that they are indeed very useful.

This page titled 13.4: The Lagrangian Equations of Motion is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

13.4.3 https://phys.libretexts.org/@go/page/7012
13.5: Acceleration Components
In Section 3.4 of the Celestial Mechanics “book”, I derived the radial and transverse components of velocity and acceleration in
two-dimensional coordinates. The radial and transverse velocity components are fairly obvious and scarcely need derivation; they
are just ρ̇ and ρϕ˙. For the acceleration components I reproduce here an extract from that chapter:
2
“The radial and transverse components of acceleration are therefore (ρ̈ − ρϕ̇ ) and (ρϕ̈ + 2ρ̇ ϕ̇ ) respectively.”
I also derived the radial, meridional and azimuthal components of velocity and acceleration in three-dimensional spherical
coordinates. Again the velocity components are rather obvious; they are ṙ , rθ˙ and r sin θϕ̇ while for the acceleration components I
reproduce here the relevant extract from that chapter.

“On gathering together the coefficients of ^

r , θ , ϕ we find that the components of acceleration are:
^ ^

2 2
Radial: r̈ − rθ̇ 2
− r sin θϕ̇
2
Meridional: rθ̈ + 2 ṙ θ̇ − r sin θ cos θϕ̇
Azimuthal: 2 ṙ ϕ˙ sin θ + 2rθ˙ϕ˙ cos θ + r sin θϕ¨. "
You might like to look back at these derivations now. However, I am now going to derive them by a different method, using
Lagrange’s equation of motion. You can decide for yourself which you prefer.

We’ll start in two dimensions. Let R and S be the radial and transverse components of a force acting on a particle. (“Radial” means
in the direction of increasing ρ; “transverse” means in the direction of increasing ϕ .) If the radial coordinate were to increase by δρ,
the work done by the force would be just Rδρ. Thus the generalized force associated with the coordinate ρ is just P = R . If the ρ

azimuthal angle were to increase by δϕ, the work done by the force would be Sρδϕ. Thus the generalized force associated with the
coordinate ϕ is P = Sρ . Now we do not have to think about how to start; in Lagrangian mechanics, the first line is always “T =
ϕ

...”, and I hope you’ll agree that

1 2
2 2 ˙
T = m(ρ̇ + ρ ϕ ). (13.5.1)
2

If you now apply Equation 13.4.12 in turn to the coordinates ρ and ϕ , you obtain
2
˙ ¨ ˙
Pρ = m(ρ̈ − ρϕ ) and Pϕ = mρ(ρϕ + 2 ρ̇ ϕ ), (13.5.2a,b)

and so
2
R = m(ρ̈ − ρϕ̇ ) and S = m(ρϕ̈ + 2 ρ̇ ϕ̇ ). (13.5.3a,b)

2
Therefore the radial and transverse components of the acceleration are (ρ̈ − ρϕ̇ ) and (ρϕ̈ + 2ρ̇ ϕ̇ ) respectively.
We can do exactly the same thing to find the acceleration components in three-dimensional spherical coordinates. Let R , S and F

be the radial, meridional and azimuthal (i.e. in direction of increasing r, θ and ϕ ) components of a force on a particle.

13.5.1 https://phys.libretexts.org/@go/page/7013
 If r increases by δr, the work on the particle done is Rδr.
If θ increases by δθ, the work done on the particle is Srδθ .
If ϕ increases by δϕ, the work done on the particle is F r sin θδϕ.
Therefore P r = R, Pθ = Sr and P
ϕ = F r sin θ .
Start:
1 2 2
2 2 2 2
T = m(ṙ + r θ̇ +r sin θϕ̇ ) (13.5.4)
2

If you now apply Equation 13.4.12 in turn to the coordinates r, θ and ϕ , you obtain
2 2
˙ 2 2 ˙
Pr = m(r̈ − r θ −r sin θϕ ), (13.5.5)

2
2 ¨ ˙ 2 ˙
Pθ = m(r θ + 2r ṙ θ − r sin θ cos θϕ ) (13.5.6)

and
2 2 2 ˙˙ ˙ 2
Pϕ = m(r sin θϕ̈ + 2 r θ ϕ sin θ cos θ + 2r ṙ ϕ sin θ). (13.5.7)

Therefore
2 2
˙ ˙
R = m(r̈ − r θ − r sin θϕ ), (13.5.8)

2
¨ ˙ ˙
S = m(r θ + 2 ṙ θ − r sin θ cos θϕ ) (13.5.9)

and
¨ ˙˙ ˙
F = m(r sin θϕ + 2r θ ϕ cos θ + 2 ṙ ϕ sin θ). (13.5.10)

Thus the acceleration components are

2 2
Radial: r̈ − rθ˙ 2
− r sin
˙
θϕ
2
Meridional: rθ¨ − 2 ṙ θ˙ − r sin θ cos θϕ˙
Azimuthal: 2 ṙ ϕ̇ sin θ − 2rθ̇ ϕ̇ cos θ + r sin θϕ̈ .
Be sure to check the dimensions. Since dot has dimension T-1, and these expressions must have the dimensions of acceleration,
there must be an r and two dots in each term.

This page titled 13.5: Acceleration Components is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

13.5.2 https://phys.libretexts.org/@go/page/7013
13.6: Slithering Soap in Conical Basin
We imagine a slippery (no friction) bar of soap slithering around in a conical basin. An isolated bar of soap in intergalactic space
would require three coordinates to specify its position at any time, but, if it is subject to the holonomic constraint that it is to be in
contact at all times with a conical basin, its position at any time can be specified with just two coordinates. I shall, first of all,
analyse the problem with a newtonian approach, and then, for comparison, I shall analyse it using lagrangian methods. Either way,
we start with a large diagram. In the newtonian approach we mark in the forces in red and the accelerations in green. See Figure
XIII.2. The semi vertical angle of the cone is α .

The two coordinates that we need are r, the distance from the vertex, and the azimuthal angle ϕ , which I’ll ask you to imagine,
measured around the vertical axis from some arbitrary origin. The two forces are the weight mg and the normal reaction R of the
basin on the soap. The accelerations are r̈ and the centripetal acceleration as the soap moves at angular speed ϕ˙ in a circle of radius
2
r sin α is r sin αϕ˙ .
We can write the newtonian equation of motion in various directions:
2
Horizontal: R cos α = m(r sin α ϕ˙ − r̈ sin α)

i.e.
2
R = m tan α(rϕ̇ − r̈ ). (13.6.1)

Vertical:

R sin α − mg = m r̈ cos α. (13.6.2)

Perpendicular to surface:
2
˙
R − mg sin α = mr sin α cos α ϕ . (13.6.3)

Parallel to surface:
2
2 ˙
g cos α = r sin αϕ − r̈ . (13.6.4)

Only two of these are independent, and we can choose to use whichever two we want to at our convenience. There are, however,
three quantities that we may wish to determine, namely the two coordinates r and ϕ , and the normal reaction R . Thus we need
another equation. We note that, since there are no azimuthal forces, the angular momentum per unit mass, which is r sin α ϕ̇ , is
2 2

conserved, and therefore r ϕ˙ is constant and equal to its initial value, which I’ll call l Ω. That is, we start off at a distance l from
2 2

the vertex with an initial angular speed Ω. Thus we have as our third independent equation

13.6.1 https://phys.libretexts.org/@go/page/7014
 2 2
r ϕ̇ = l Ω (13.6.5)

This last equation shows that ϕ˙ → ∞ as r → 0 .

One possible type of motion is circular motion at constant height (put r̈ = 0 ). From Equations 13.6.1 and 13.6.2 it is easily found
that the condition for this is that
2 g
rϕ̇ = . (13.6.6)
sin α tan α

In other words, if the particle is projected initially horizontally (ṙ =0 ) at r = l and ϕ˙ = Ω , it will describe a horizontal circle (for
ever) if
g 1

Ω =( ) 2 = ΩC , say. (13.6.7)
l sin α tan α

If the initial speed is less than this, the particle will describe an elliptical orbit with a minimum r < l ; if the initial speed is greater
than this, the particle will describe an elliptical orbit with a maximum r > l .
Now let’s do the same problem in a lagrangian formulation. This time we draw the same diagram, but we mark in the velocity
components in blue. See Figure XIII.3. We are dealing with conservative forces, so we are going to use Equation 13.4.13, the most
useful form of Lagrange’s equation.

We need not spend time wondering what to do next. The first and second things we always have to do are to find the kinetic energy
T and the potential energy V , in order that we can use Equation 13.4.13.

1 2 2 2
2
˙
T = m(ṙ +r sin αϕ ) (13.6.8)
2

and
V = mgr cos α + constant. (13.6.9)

Now go to Equation 13.4.13, with qi = r , and work out all the derivatives, and you should get, when you apply the lagrangian
equation to the coordinate r:
2
2 ˙
r̈ − r sin αϕ = −g cos α. (13.6.10)

Now do the same thing with the coordinate ϕ . You see immediately that ∂T

∂ϕ
and ∂V

∂ϕ
are both zero. Therefore dt
d ∂T

˙
is zero and
∂ϕ

therefore ∂T

˙
is constant. That is, mr
2 2
sin α ϕ̇ is constant and so r 2
ϕ̇ is constant and equal to its initial value l 2
. Thus the second
Ω
∂ϕ

13.6.2 https://phys.libretexts.org/@go/page/7014
lagrangian equation is
2 2
r ϕ̇ = l Ω. (13.6.11)

Since the lagrangian is independent of ϕ , ϕ is called, in this connection, an “ignorable coordinate” – and the momentum associated
with it, namely mr ϕ̇ is constant.
2

Now it is true that we arrived at both of these equations also by the newtonian method, and you may not feel we have gained much.
But this is a simple, introductory example, and we shall soon appreciate the power of the lagrangian method,
Having got these two equations, whether by newtonian or lagrangian methods, let’s explore them further. For example, let’s
eliminate ϕ̇ between them and hence get a single equation in r:
4 2 2
l Ω sin α
r̈ − = −g cos α. (13.6.12)
r3

We know enough by now (see Chapter 6) to write r̈ as v dv

dr
, where v = ṙ and if we let the constants l 4
Ω
2 2
sin α and g cos α equal
A and B respectively, Equation 13.6.12 becomes

dν A
ν = − B. (13.6.13)
dr r3

(It may just be useful to note that the dimensions of A and B are L4T-2 and LT-2 respectively. This will enable us to keep track of
dimensional analysis as we go.)
If we start the soap moving horizontally (v = 0 ) when r = l , this integrates, with these initial conditions, to
1 1
2
ν = A( − ) + 2B(l − r). (13.6.14)
2 2
l r

Again, so that we can see what we are doing, let A

2
+ 2Bl = C (note that [C ] = L2T-2), and Equation 13.6.14 becomes
l

2
A
ν =C − − 2Br. (13.6.15)
2
r

This gives ν (= ṙ ) as a function of r. The particle reaches is maximum or minimum height when ν =0 ; that is where
3 2
2Br − Cr + A = 0. (13.6.16)

One solution of this is obviously r = l . Of the other two solutions, one is positive (which we want) and the other is negative (which
we do not want).
If we go back to the original meanings of A , B and C , and write x = r

l
equation (16) becomes, after a little tidying up
2 2
3
lΩ sin α tan α 2
lΩ sin α tan α
x −( + 1)x + = 0. (13.6.17)
2g 2g

1
g
Recall from Equation 13.6.7 that Ω c =(
l sin α tan α
) 2 , and the equation becomes
2 2
Ω Ω
3 2
x −( + 1)x + , (13.6.18)
2 2
2Ωc 2Ωc

or, with a = Ω
2
,
2Ω
c

3 2
x − (a + 1)x + a = 0. (13.6.19)

This factorizes to
2
(x − 1)(x − ax − a) = 0. (13.6.20)

The solution we are interested in is

1 −−−−−−−
x = (a + √ a(a + 4) ). (13.6.21)
2

13.6.3 https://phys.libretexts.org/@go/page/7014
This page titled 13.6: Slithering Soap in Conical Basin is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

13.6.4 https://phys.libretexts.org/@go/page/7014
13.7: Slithering Soap in Hemispherical Basin
Suppose that the basin is of radius a and the soap is subject to the holonomic constraint r = a - i.e. that it remains in contact with
the basin at all times. Note also that this is just the same constraint of a pendulum free to swing in three-dimensional space except
that it is subject to the holonomic constraint that the string be taut at all times. Thus any conclusions that we reach about our soap
will also be valid for a pendulum.
We’ll start with the newtonian approach, and I’ll draw in red the two forces on the soap, namely its weight and the normal reaction
of the basin on the soap. Figure XIII.4

We’ll make use of the expressions for the radial, meridional and azimuthal accelerations from Section 13.5 and we’ll write down
the equations of motion in these directions:
Radial:
2 2
2
mg cos θ − R = m(r̈ − r θ̇ − r sin θϕ̇ ). (13.7.1)

Meridional:
2
¨ ˙ ˙
−mg sin θ = m(r θ − 2 ṙ θ − r sin θ cos θϕ ), (13.7.2)

Azimuthal:

0 = m(2 ṙ ϕ̇ sin θ + 2r ṙ ϕ̇ cos θ + r sin θϕ̈ ) (13.7.3)

We also have the constraint that r = a and hence that ṙ = r̈ = 0 , after which these equations become
2 2
˙ 2 ˙
mg sin θ − R = −ma(θ + sin θϕ ), (13.7.4)

2
¨ ˙
−g sin θ = a(θ − sin θ cos θϕ ), (13.7.5)

0 = sin θϕ̈ + 2 θ̇ ϕ̇ cos θ. (13.7.6)

These, then, are the newtonian equations of motion. If you still prefer the newtonian method to the lagrangian method, and you
wish to integrate these and find expressions θ , ϕ and R separately, by all means go ahead and do so – but I’m now going to try the
lagrangian approach.
Although Lagrange himself would not have drawn a diagram, we shall not omit that step – but instead of marking in the forces,
we’ll mark in the velocity components, and then we’ll immediately write down expressions for the kinetic and potential energies.
Indeed the first line of a lagrangian calculation is always “T = ...”.

13.7.1 https://phys.libretexts.org/@go/page/8444
 1 2
2
2
2
˙ ˙
T = m a (θ + sin θϕ ) (13.7.7)
2

V = −mga cos θ + constant (13.7.8)

Now apply Equation 13.4.13 in turn to the coordinates θ and ϕ .

θ :
2
aθ̈ − a sin θ cos θϕ̇ = −g sin θ. (13.7.9)

ϕ :
As for the conical basin, we see that ∂T

∂ϕ
and ∂V

∂ϕ
are both zero (ϕ is an "ignorable coordinate") and therefore ∂T

˙
is constant and
∂ϕ

equal to its initial value. If the initial values of ϕ̇ and θ are Ω and α respectively, then
2 ˙ 2
sin θϕ = sin αΩ. (13.7.10)

This is merely stating that angular momentum is conserved.

We can easily eliminate ϕ˙ from Equations 13.4.9 and 13.4.10 to obtain
g sin θ
¨ 2
θ = k cot θ csc θ − , (13.7.11)
a

where
4 2
k = sin αΩ . (13.7.12)

˙
Write θ¨ as θ˙ dθ

dθ
in the usual way and integrate to obtain the first space integral:
2 2g
˙ 2 2
θ = (cos θ − cos α) − k(csc θ − csc α). (13.7.13)
a

The upper and lower bounds for θ occur when θ˙ = 0

Example. Suppose that the initial value of θ is α = 45o and that we start by pushing the soap horizontally (θ˙ = 0 ) at an initial
angular speed Ω = 3 rad s-1, so that k = 2.25 rad2 s-2. Suppose that the radius of the basin is a = 1.96 m and that g = 9.8 m s-2. You
can then put θ˙ = 0 in Equation 13.7.13 and solve it for θ . One solution, of course, is θ = α . We could find the other solution by
Newton-Raphson iteration, or by putting csc θ = and solving it as a cubic equation in cos θ. Alternatively, try this:
2 1

(1−cos2 θ)

Let ak

2g
= n, cos θ = x, cos α = c,

so that csc 2
θ =
1

(1−x )
2
and csc 2
α =
1

(1−c )
2
.

The equation then becomes a quadratic equation in x, with solution

13.7.2 https://phys.libretexts.org/@go/page/8444
 2 2
−n+√n −4(1−c )[c(n+c)−1]

That is, x = 2(1−c2 )

I’ll leave you to re-write this in terms of what these quantities originally meant.

This page titled 13.7: Slithering Soap in Hemispherical Basin is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated
by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

13.7.3 https://phys.libretexts.org/@go/page/8444
13.8: More Lagrangian Mechanics Examples
 Example 13.8.1

The upper pulley is fixed in position. Both pulleys rotate freely

without friction about their axles. Both pulleys are “light” in the sense
that their rotational inertias are small and their rotation contributes
negligibly to the kinetic energy of the system. The rims of the pulleys
are rough, and the ropes do not slip on the pulleys. The gravitational
acceleration is g.
The mass M moves upwards at a rate ẋ with respect to the upper,
fixed, pulley, and the smaller pulley moves downwards at the same
rate. The mass m moves upwards at a rate ẏ with respect to the
1

small pulley, and consequently its speed in laboratory space is ẋ − ẏ .

The speed of the mass m is therefore ẋ + ẏ in laboratory space. The
2

object is to find ẍ and ÿ in terms of g.

The kinetic energy is

1 1 1
2 2 2
T = M ẋ + m1 (ẋ − ẏ ) + m2 (ẋ + ẏ ) . (13.8.1)
2 2 2

The potential energy is

V = g[M x − m1 (x − y) − m2 (x + y)] + constant. (13.8.2)

Apply Lagrange’s equation (13.4.13) in turn to the coordinates x and y :

x :
M ẍ + m1 (ẍ + ÿ ) + m2 (ẍ + ÿ ) = −g(M − m1 − m2 ). (13.8.3)

y :
−m1 (ẍ + ÿ ) + m2 (ẍ + ÿ ) = −g(m1 − m2 ) (13.8.4)

These two equations can be solved at one’s leisure for ẍ and ÿ .

 Example 13.8.2

A torus of mass M and radius a rolls without slipping on a horizontal plane. A pearl of mass m slides smoothly around inside
the torus. Describe the motion.

13.8.1 https://phys.libretexts.org/@go/page/8445
I have marked in the several velocity vectors. The torus is rolling at angular speed ϕ̇ Consequently the linear speed of the
centre of mass of the hoop is aϕ˙ and the pearl also shares this velocity. In addition, the pearl is sliding relative to the torus at an
angular speed θ˙ and consequently has a component to its velocity of aθ˙ tangential to the torus. We are now ready to start.
The kinetic energy of the torus is the sum of its translational and rotational kinetic energies:
2 2
1 ˙ 1 2 ˙ 2 ˙
M (aϕ ) + (M a )ϕ = Ma ϕ
2 2

The kinetic energy of the pearl is

2 2
1 2 ˙ ˙ ˙˙
m a (θ +ϕ − 2 θ ϕ cos θ)
2

Therefore

2 ˙
2 1 2
2 2
˙ ˙ ˙˙
T = Ma ϕ + m a (θ +ϕ − 2 θ ϕ cos θ). (13.8.5)
2

The potential energy is

V = constant − mga cos θ. (13.8.6)

The lagrangian equation in θ becomes

¨
a(θ − ϕ̈ cos θ) + g sin θ = 0. (13.8.7)

The lagrangian equation in ϕ becomes

2
¨ ¨ ˙
(2M + m)ϕ = m(θ cos θ − θ sin θ) (13.8.8)

These, then, are two differential equations in the two variables. The lagrangian part of the analysis is over; we now have to see
if we can do anything with these equations.
It is easy to eliminate ϕ̈ and hence get a single differential equation in θ :.
2
2 ¨ ˙
(2M + m sin θ)aθ + ma sin θ cos θθ + (2M + m)g sin θ = 0. (13.8.9)

If you are good at differential equations, you might be able to do something with this, and get θ as a function of the time. In the
meantime, I think I can get the “first space integral” (see Chapter 6) – i.e. θ˙ as a function of θ . Thus, the total energy is
constant:
2 1 2 2
2 2 ˙ ˙
M a ϕ̇ + m a (θ + ϕ̇ − 2 θ ϕ̇ cos θ) − mga cos θ = E. (13.8.10)
2

13.8.2 https://phys.libretexts.org/@go/page/8445
 2
Equation 13.8.8 can easily be integrated once with respect to time, since θ¨ cos θ − θ˙ sin θ = (θ˙ cos θ) as would have been
d

dt

apparent during the derivation of Equation 13.8.8. With the condition that the kinetic energy was initially zero, integration of
Equation 13.8.8 gives
˙ ˙
(2M + m)ϕ = m θ cos θ. (13.8.11)

Now we can easily eliminate ϕ˙ between Equations 13.8.10 and 13.8.11, to obtain a single equation relating θ˙ and θ :
2
˙ 2
b θ (1 + c sin θ) − d cos θ − 1 = 0, (13.8.12)

where
2
M ma m mga
b = , c = , d = = − sec α. (13.8.13)
(2M + m)E 2M E

 Example 13.8.3

As in example ii, we have a torus of radius a and mass M , and a pearl of mass m which can slide freely and without friction
around the torus. This time, however, the torus is not rolling along the table, but is spinning about a vertical axis at an angular
speed ϕ˙ . The pearl has a velocity component aθ˙ because it is sliding around the torus, and a component a sin θϕ˙ because the
torus is spinning. The resultant speed is the orthogonal sum of these. The kinetic energy of the system is the sum of the
translational kinetic energy of the pearl and the rotational kinetic energy of the torus:
1 2
2
2
2 1 1 2
2
˙ ˙ ˙
T = m a (θ + sin θϕ ) + ( M a )ϕ . (13.8.14)
2 2 2

If we refer potential energy to the centre of the torus:

V = mga cos θ. (13.8.15)

The lagrangian equations with respect to the two variables are:

θ :
2
¨
a(θ − sin θ cos θϕ̇ ) − g sin θ = 0. (13.8.16)

ϕ :

13.8.3 https://phys.libretexts.org/@go/page/8445
 2
1
˙ ˙
m sin θϕ + M ϕ = constant. (13.8.17)
2

The constant is equal to whatever the initial value of the left hand side was. E.g., maybe the initial values of θ and ϕ˙ were α
and ω. This finishes the lagrangian part of the analysis. The rest is up to you. For example, it would be easy to eliminate ϕ˙
˙
between these two equations to obtain a differential equation between θ and the time. If you then write ¨
θ as ˙
θ
dθ

dθ
in the usual
way, I think it wouldn’t be too difficult to obtain the first space integral and hence get ˙
θ as a function of θ . I haven’t tried it,
but I’m sure it’ll work.

 Example 13.8.4

Figure XIII.10 shows a pendulum. The mass at the end is m. It is at the end not of the usual inflexible string, but of an elastic
spring obeying Hooke’s law, of force constant k . The spring is sufficiently stiff at right angles to its length that it remains
straight during the motion, and all the motion is restricted to a plane. The unstretched natural length of the spring is l, and, as
shown, its extension is r. The spring itself is “light” in the sense that it does not contribute to the kinetic or potential energies.
(You can give the spring a finite mass if you want to make the problem more difficult.) The kinetic and potential energies are
1 2
2
2 ˙
T = m(ṙ + (l + r) θ ) (13.8.18)
2

and
1
2
V = constant − mg(l + r) cos θ + kr . (13.8.19)
2

Apply Lagrange’s equation in turn to r and to θ and see where it leads you.

 Example 13.8.5

Another example suitable for lagrangian methods is given as problem number 11 in Appendix A of these notes.
Lagrangian methods are particularly applicable to vibrating systems, and examples of these will be discussed in Chapter 17.
These chapters are being written in more or less random order as the spirit moves me, rather than in logical order, so that
vibrating systems appear after the unlikely sequence of relativity and hydrostatics.

This page titled 13.8: More Lagrangian Mechanics Examples is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated
by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

13.8.4 https://phys.libretexts.org/@go/page/8445
13.9: Hamilton's Variational Principle
Hamilton’s variational principle in dynamics is slightly reminiscent of the principle of virtual work in statics, discussed in Section
9.4 of Chapter 9. When using the principle of virtual work in statics we imagine starting from an equilibrium position, and then
increasing one of the coordinates infinitesimally. We calculate the virtual work done and set it to zero. I am slightly reminded of
this when discussing Hamilton’s principle in dynamics
Imagine some mechanical system – some contraption including in its construction various wheels, jointed rods, springs, elastic
strings, pendulums, inclined planes, hemispherical bowls, and ladders leaning against smooth vertical walls and smooth horizontal
floors. It may require N generalized coordinates to describe its configuration at any time. Its configuration could be described by
the position of a point in N -dimensional space. Or perhaps it is subject to k holonomic constraints – in which case the point that
describes its configuration in N -dimensional space is not free to move anywhere in that space, but is constrained to slither around
on a surface of dimension N − k .
The system is not static, but it is evolving. It is changing from some initial state at time t to some final state at time t . The
1 2

generalized coordinates that describe it are changing with time – and the point in N -space is slithering round on its surface of
dimension N − k . One can imagine that at any instant of time one can calculate its kinetic energy T and its potential energy V ,
and hence its lagrangian L = T − V . You can multiply L at some moment by a small time interval δt and then add up all of these
products between t and t to form the integral
1 2

t2

∫ L dt.
t1

This quantity – of dimension ML2T-1 and SI unit J s – is sometimes called the “action”. There are many different ways in which we
can imagine the system to evolve from its initial state to its final state – and there are many different routes that we can imagine
might be taken by our point in N -space as its moves from its initial position to its final position, as long as it moves over its surface
of dimension N − k . But, although we can imagine many such routes, the manner in which the system will actually evolve, and
the route that the point will actually take is determined by Hamilton’s principle; and the route, according to this principle, is such
t2
that the integral ∫ Ldt is a minimum, or a maximum, or an inflection point, when compared with other imaginable routes. Stated
t1
t2 t2
otherwise, let us suppose that we calculate ∫ Ldt over the actual route taken and then calculate the variation in ∫ Ldt if the
t1 t1

system were to move over a slightly different adjacent path. Then (and here is the analogy with the principle of virtual work in a
statics problem) this variation
t2

δ∫ L dt
t1

t2
from what ∫ t1
Ldt would have been over the actual route is zero. And this is Hamilton’s variational principle.
The next questions will surely be: Can I use this principle for solving problems in mechanics? Can I prove this bald assertion? Let
me try to use the principle to solve two simple and familiar problems, and then move on to a more general problem.

 Example 13.9.1

Imagine that we have a particle than can move in one dimension (i.e. one coordinate – for example its height y above a table -
suffices to describe its position), and that when its coordinate is y its potential energy is
V = mgy. (13.9.1)

Its kinetic energy is, of course,

1 2
T = m ẏ . (13.9.2)
2

We are going to use the variational principle to find the equation of motion – i.e we are going to find an expression for its
acceleration. I imagine at the moment you have no idea what its acceleration could possibly be – but do not worry, for we
know that the lagrangian is
1 2
L = m ẏ − mgy, (13.9.3)
2

13.9.1 https://phys.libretexts.org/@go/page/8446
 and we’ll make short work of it with Hamilton’s variational principle and soon find the acceleration. According to this
principle, y must vary with t in such a manner that
t2
1 2
mδ ∫ ( ẏ − gy)dt = 0. (13.9.4)
t1
2

Let us vary ẏ by δẏ and y by δy see how the integral varies.

The integral is then
t2

m∫ (ẏ δẏ − gδy)dt, (13.9.5)

t1

which I’ll call I 1 − I2 .

dy
Now ẏ =
dt
and if y varies by δy, the resulting variation in ẏ will be δẏ =
d

dt
δy , or δẏ dt = dδy .
Therefore
t2

I1 = m ∫ ẏ dδy. (13.9.6)
t1

(If unconvinced of this, consider ∫ e t

cos tdt = ∫ e
t d

dt
sin tdt = ∫ e d sin t
t
.)
By integration by parts:
t2
t2
I1 = [m ẏ δy ] −m ∫ δydẏ . (13.9.7)
t1
t1

The first term is zero because the variation is zero at the beginning and end points. In the second term, dẏ = ÿ dt and therefore
t2

I1 = −m ∫ ÿ δydt (13.9.8)
t1

t2 t2

δ∫ Ldt = −m ∫ (ÿ + g)δydt, (13.9.9)

t1 t1

and, for this to be zero, we must have

ÿ = −g. (13.9.10)

This is the equation of motion that we sought. You would never have guessed this, would you?

Now let’s do another one-dimensional problem.

 Example 13.9.2
Only one coordinate, x, describes the particle’s position, and, when its coordinate is x we’ll suppose that its potential energy is
m ω x and its kinetic energy is, of course, T = m ẋ . The equation of motion, or the way in which the acceleration
1 2 2 1 2
V =
2 2

varies with position, must be such as to satisfy

t2
1 2 2 2
mδ ∫ (ẋ − ω x )dt = 0. (13.9.11)
2 t1

If we vary ẋ by δẋ and x by δx the variation in the integral will be

t2
2
m∫ (ẋδẋ − ω xδx)dt = I1 − I2 . (13.9.12)
t1

t2
By precisely the same argument as before, the first integral is found to be −m ∫ t1
ẍδx dt

Therefore

13.9.2 https://phys.libretexts.org/@go/page/8446
 t2 t2 t2
2
δ∫ L dt = −m ∫ ẍδx dt − m ω ∫ xδx dt, (13.9.13)
t1 t1 t1

and, for this to be zero, we must have

2
ẍ = −ω x. (13.9.14)

These two examples must have given the impression that we are doing something very difficult in order to derive something that is
immediately obvious – but the examples were just intended to show the direction of a more general argument we are about to
make.
This time, we’ll consider a very general system, in which we write the lagrangian as a function of the (several) generalized
coordinates and their time rates of change - i.e. L = L(q , q˙ ) - without specifying any particular form of the function – and we’ll
i i

carry out the same sort of argument to derive a very general equation of motion.
We have
t2 t2 t2
∂L ∂L
δ∫ Ldt = ∫ δLdt = ∫ ∑( δqi + δq
˙i ) dt = 0. (13.9.15)
t1 t1 t1
∂qi ∂q
˙i
i

As before, δq˙i =
d

dt
δqi so that
t2 t2 t2 t2 t2
∂L ∂L d ∂L ∂L d ∂L
∫ δq
˙i dt = ∫ δqi dt = ∫ dδqi = [ δqi ] −∫ δqi dt (13.9.16)
t1
∂q
˙i t1
∂q
˙i dt t1
∂q
˙i ∂q
˙i t1
dt ∂q
˙i
t1

t2 t2
∂L d ∂L
δ∫ Ldt = ∫ ∑( − ) δqi dt = 0. (13.9.17)
t1 t1
∂qi dt ∂q
˙i
i

Thus we arrive at the general equation of motion

∂L d ∂L
− ( ) = 0. (13.9.18)
∂qi dt ∂q
˙i

Thus we have derived Lagrange’s equation of motion from Hamilton’s variational principle, and this is indeed the way it is often
derived. However, in this chapter, I derived Lagrange’s equation quite independently, and hence I would regard this derivation not
so much as a proof of Lagrange’s equation, but as a vindication of the correctness of Hamilton’s variational principle.

This page titled 13.9: Hamilton's Variational Principle is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

13.9.3 https://phys.libretexts.org/@go/page/8446
 CHAPTER OVERVIEW
14: Hamiltonian Mechanics
Hamiltonian mechanics can be used to describe simple systems such as a bouncing ball, a pendulum or an oscillating spring in
which energy changes from kinetic to potential and back again over time, its strength is shown in more complex dynamic systems,
such as planetary orbits in celestial mechanics. The more degrees of freedom the system has, the more complicated its time
evolution.
14.1: Introduction to Hamiltonian Mechanics
14.2: A Thermodynamics Analogy
14.3: Hamilton's Equations of Motion
14.4: Hamiltonian Mechanics Examples
14.5: Poisson Brackets

This page titled 14: Hamiltonian Mechanics is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum
via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1
14.1: Introduction to Hamiltonian Mechanics
The hamiltonian equations of motion are of deep theoretical interest. Having established that, I am bound to say that I have not
been able to think of a problem in classical mechanics that I can solve more easily by hamiltonian methods than by newtonian or
lagrangian methods. That is not to say that real problems cannot be solved by hamiltonian methods. What I have been looking for is
a problem which I can solve easily by hamiltonian methods but which is more difficult to solve by other methods. So far, I have not
found one. Having said that, doubt not that hamiltonian mechanics is of deep theoretical significance.
Having expressed that mild degree of cynicism, let it be admitted that Hamilton theory – or more particularly its extension the
Hamilton-Jacobi equations - does have applications in celestial mechanics, and of course hamiltonian operators play a major part in
quantum mechanics, although it is doubtful whether Sir William would have recognized his authorship in that connection.

This page titled 14.1: Introduction to Hamiltonian Mechanics is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated
by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

14.1.1 https://phys.libretexts.org/@go/page/7016
14.2: A Thermodynamics Analogy
Readers may have noticed from time to time – particularly in Chapter 9 - that I have perceived some connection between parts of
classical mechanics and thermodynamics. I perceive such an analogy in developing hamiltonian dynamics. Those who are familiar
with thermodynamics may also recognize the analogy. Those who are not can skip this section without seriously prejudicing their
understanding of subsequent sections.
Please do not misunderstand: The hamiltonian in mechanics is not at all the same thing as enthalpy in thermodynamics, even
though we use the same symbol, H . Yet there are similarities in the way we can introduce these concepts.
In thermodynamics we can describe the state of the system by its internal energy, defined in such a way that when heat is supplied
to a system and the system does external work, the increase in internal energy of the system is equal to the heat supplied to the
system minus the work done by the system:

dU = T dS − P dV . (14.2.1)

From this point of view we are describing the state of the system by specifying its internal energy as a function of the entropy and
the volume:

U = U (S, V ) (14.2.2)

so that
∂U ∂U
dU = ( ) dS + ( ) dV , (14.2.3)
∂S ∂V
V S

from which we see that

∂U
T =( )V (14.2.4)
∂S

and
∂U
−P = ( ) (14.2.5)
∂V
S

However, it is sometimes convenient to change the basis of the description of the state of a system from S and V to S and P by
defining a quantity called the enthalpy H defined by
H = U +PV . (14.2.6)

In that case, if the state of the system changes, then

dH = dU + P dV + V dP (14.2.7)

= T dS − P dV + P dV + V dP . (14.2.8)

I.e.
dH = T dS + V dP . (14.2.9)

Thus we see that, if heat is added to a system held at constant volume, the increase in the internal energy is equal to the heat added;
whereas if heat is added to a system held at constant pressure, the increase in the enthalpy is equal to the heat added.
From this point of view we are describing the state of the system by specifying its enthalpy as a function of the entropy and the
pressure:
H = H (S, P ) (14.2.10)

so that
∂H ∂H
dH = ( ) dS + ( ) dP , (14.2.11)
∂S ∂P
P S

from which we see that

14.2.1 https://phys.libretexts.org/@go/page/7017
 ∂H
T =( )
∂S
P

and
∂H
V =( ) . (14.2.12)
∂P
S

None of this has anything to do with hamiltonian dynamics, so let’s move on.

This page titled 14.2: A Thermodynamics Analogy is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

14.2.2 https://phys.libretexts.org/@go/page/7017
14.3: Hamilton's Equations of Motion
In classical mechanics we can describe the state of a system by specifying its Lagrangian as a function of the coordinates and their
time rates of change:
L = L(qi , q̇ ) (14.3.1)

If the coordinates and the velocities increase, the corresponding increment in the Lagrangian is
∂L ∂L
dL = ∑ dqi + ∑ ˙i .
dq (14.3.2)
∂qi ˙i
∂q
i i

 Definition: generalized momenta

The generalized momentum pi associated with the generalized coordinate qi is defined as
∂L
pi = . (14.3.3)
∂q
˙i

[You have seen this before, in Section 13.4. Remember “ignorable coordinate”?]

It follows from the Lagrangian equation of motion (Equation 13.4.14)

d ∂L ∂L
=
dt ∂q
˙i ∂qi

that
∂L
ṗ = . (14.3.4)
i
∂qi

Thus

dL = ∑ ṗ dqi + ∑ pi d q̇ . (14.3.5)
i i

i i

(I am deliberately numbering this Equation 14.3.5, to maintain an analogy between this section and Section 14.2.)
However, it is sometimes convenient to change the basis of the description of the state of a system from q and q˙ to q and i i i p
˙i by
defining a quantity called the hamiltonian H defined by

˙i − L.
H = ∑ pi q (14.3.6)

 Definition: hamiltonian
In that case, if the state of the system changes, then

dH = ∑ pi dq
˙i + ∑ q
˙i dpi − dL

i i

˙i + ∑ q
= ∑ pi dq ˙i dpi − ∑ p
˙i dqi − ∑ pi dq
˙i

i i i i

That is

dH = ∑ q
˙i dpi − ∑ p
˙i dqi . (14.3.7)

i i

We are regarding the hamiltonian as a function of the generalized coordinates and generalized momenta:

H = H (qi , pi ) (14.3.8)

so that

14.3.1 https://phys.libretexts.org/@go/page/7018
 ∂H ∂H
dH = ∑ dqi + ∑ dpi , (14.3.9)
∂qi ∂pi
i i

from which we see that

∂H
−ṗ i = (14.3.10)
∂qi

and
∂H
q̇ i
= (14.3.11)
∂pi

In summary, then, Equations 14.3.3, 14.3.4, 14.3.10 and 14.3.11:

∂L
pi = (14.3.12)
∂q
˙i

∂L
p
˙i = (14.3.13)
∂qi

∂H
˙i =
−p (14.3.14)
∂qi

∂H
˙i =
q (14.3.15)
∂pi

which I personally find impossible to commit accurately to memory (although note that there is one dot in each equation) except
when using them frequently, may be regarded as Hamilton’s equations of motion. I’ll refer to these equations as A, B, C and D.
Note that, in Equation 14.3.13, if the Lagrangian is independent of the coordinate q the coordinate q is referred to as an
i i

“ignorable coordinate”. I suppose it is called “ignorable” because you can ignore it when calculating the lagrangian, but in fact a
so-called “ignorable” coordinate is usually a very interesting coordinate indeed, because it means (look at the second equation) that
the corresponding generalized momentum is conserved.
1 1
Now the kinetic energy of a system is given by T = ∑ pi q
i
˙i (for example, ), and the hamiltonian (Equation
mν ν 14.3.6 ) is
2 2
defined as H = ∑ p q˙ − L . For a conservative system, L = T − V , and hence, for a conservative system, H = T + V . If you
i i i

are asked in an examination to explain what is meant by the hamiltonian, by all means say it is T + V . That’s fine for a
conservative system, and you’ll probably get half marks. That’s 50% - a D grade, and you’ve passed. If you want an A+, however, I
recommend Equation 14.3.6.

This page titled 14.3: Hamilton's Equations of Motion is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

14.3.2 https://phys.libretexts.org/@go/page/7018
14.4: Hamiltonian Mechanics Examples
I’ll do two examples by hamiltonian methods – the simple harmonic oscillator and the soap slithering in a conical basin. Both are
conservative systems, and we can write the hamiltonian as T + V , but we need to remember that we are regarding the hamiltonian
2
p
as a function of the generalized coordinates and momenta. Thus we shall generally write translational kinetic energy as (2m)
rather
2

than as 1

2
mν
2
, and rotational kinetic energy as (2I)
L
rather than as 1

2
Iω
2

Simple harmonic oscillator

The potential energy is 1

2
2
kx , so the hamiltonian is
2
p 1 2
H = + kx .
2m 2

p
From equation D, we find that ẋ = , from which, by differentiation with respect to the time, ṗ
m
= m ẍ . And from equation C, we
find that ṗ = −kx . Hence we obtain the equation of motion mẍ = −kx .

Conical basin
We refer to Section 13.6:
1 2 2 2
2
˙
T = m(ṙ +r sin αϕ )
2

V = mgr cos α

1 2
2
2 2 ˙
L = m(ṙ +r sin α ϕ ) − mgr cos α
2

1 2
2 2 2
L = m(ṙ +r sin α ϕ̇ ) + mgr cos α
2

But, in the hamiltonian formulation, we have to write the hamiltonian in terms of the generalized momenta, and we need to know
what they are. We can get them from the lagrangian and equation A applied to each coordinate in turn. Thus
∂L
Pr = = m ṙ (14.4.1)
∂ ṙ

and
∂L
2 2
Pϕ = = mr sin α ϕ̇ . (14.4.2)
˙
∂ϕ

Thus the hamiltonian is

2
Pr
2 p
ϕ
H = + + mgr cos α. (14.4.3)
2
2m 2m r 2
sin α

Now we can obtain the equations of motion by applying equation D in turn to r and ϕ and then equation C in turn to r and ϕ :
∂H pr
ṙ = = , (14.4.4)
∂pr m

∂H pϕ
˙
ϕ = = , (14.4.5)
2
∂pϕ mr 2
sin α

2
p
∂H ϕ
ṗ r = − = − mg cos α, (14.4.6)
3 2
∂r mr sin α

∂H
p
˙ϕ = = 0. (14.4.7)
∂ϕ

Equations 14.4.2 and 14.4.7 tell us that mr 2

sin
2 ˙
αϕ is constant and therefore that

14.4.1 https://phys.libretexts.org/@go/page/7019
 2
r ϕ̇ is constant, = h, say. (14.4.8)

This is one of the equations that we arrived at from the lagrangian formulation, and it expresses constancy of angular momentum.
By differentiation of Equation 14.4.1 with respect to time, we see that the left hand side of Equation 14.4.6 is mr̈ . On the right
hand side of Equation 14.4.6, we have p , which is constant and equal to mh sin α. Equation 14.4.6 therefore becomes
ϕ
2

2 2
h sin α
r̈ = − g cos α, (14.4.9)
3
r

which we also derived from the lagrangian formulation.

This page titled 14.4: Hamiltonian Mechanics Examples is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

14.4.2 https://phys.libretexts.org/@go/page/7019
14.5: Poisson Brackets
Let f and g be functions of the generalized coordinates and momenta. Think first of all of one coordinate, say q , and its conjugate i

∂f ∂g ∂f ∂g
momentum p (defined, you may remember, as
i ). I now ask the question: Is
∂L

∂q
˙
the same thing as
∂qi ∂pi
? ∂pi ∂qi
i

After thinking about it you will probably say something like: Well, I dare say that you might be able to find two functions such that
that is so, but I do not see why it should be so for any two arbitrary functions. If that is what you thought, you thought right. Pairs
of functions such that these two expressions are equal are of special significance. And pairs of functions such that these two
expressions are not equal are also of special significance
The Poisson bracket of two functions of the coordinates and momenta is defined as
∂f ∂g ∂f ∂g
[f , g] = ∑( − ) (14.5.1)
∂qi ∂pi ∂pi ∂qi
i

(Poisson brackets are sometimes written as braces - i.e. {}. I’m not sure whether braces {} or brackets [] are the commoner. I have
chosen brackets here, so that I do not have to call them Poisson braces.)
Poisson brackets have important applications in celestial mechanics and in quantum mechanics. In celestial mechanics, they are
used in the developments of Lagrange’s planetary equations, which are used to calculate the perturbations of the elements of the
planetary orbits under small deviations from ideal two-body point-source orbits. See, for example, Chapter 14 of the Celestial
Mechanics set of these notes. Readers who have had an introductory course in quantum mechanics may have come across the
commutator of two operators, and will (or should!) understand the significance of two operators that commute. (It means that a
function can be found that is simultaneously an eigenfunction of both operators.) You may not have thought of the commutator as
being a Poisson bracket, but you soon will.
Let’s suppose (because it does not make any essential difference) that there is just a single generalized coordinate and its conjugate
generalized momentum, so that the Poisson bracket is just
∂f ∂g ∂f ∂g
[f , g] = − . (14.5.2)
∂q ∂p ∂p ∂q

 Example 14.5.1
Now let’s suppose that f is just q, the coordinate, and that g is the Hamiltonian, H , which is defined, you will recall, as
p q̇ − L , and is a function of the coordinate and the momentum. What, then is the Poisson bracket [q, H ]?

Solution
∂q ∂H ∂q ∂H
[q, H ] = − . (14.5.3)
∂q ∂p ∂p ∂q

∂q
The coordinate and the momentum are independent variables, so that ∂p
is zero, so the second term on the right hand side of
∂q
Equation 14.5.3 is zero. In the first term on the right hand side, ∂q
is of course 1, and ∂H

∂p
, by Hamilton’s equations of motion,
is q̇ . Thus, the answer is

[q, H ] = q̇ . (14.5.4)

In a similar vein, you will find (DO IT!!) that

[p, H ] = ṗ (14.5.5)

Thus neither the generalized coordinate nor the generalized momentum commutes with the Hamiltonian.
Now go a little further, and suppose that there are more than one coordinate and more than one momentum. Two will do, so
that
∂f ∂g ∂f ∂g ∂f ∂g ∂f ∂g
[f , g] = − + − (14.5.6)
∂q1 ∂p1 ∂p1 ∂q1 ∂q2 ∂p2 ∂p2 ∂q2

14.5.1 https://phys.libretexts.org/@go/page/7020
  Exercise 14.5.1

Can you show that:

[ p1 , p2 ] = [ q1 , q2 ] = [ p1 , q2 ] = [ q1 , p2 ] = 0; [ q1 , p1 ] = 1.? (14.5.7)

I shan’t go any further than that here, because it would take us too far into quantum mechanics. However, those readers who have
done some introductory quantum mechanics may recall that there are various pairs of operators that do or do not commute, and
may now begin to appreciate the relation between the Poisson brackets of certain pairs of observable quantities and the commutator
of the operators representing these quantities. For example, consider the last of these. It shows that a coordinate such as x does not
commute with its corresponding momentum p . There is nothing more certain that this. So certain is it that it ought to be called
x

Heisenberg’s Certainty Principle. But for some reason people often seem to present quantum mechanics as something uncertain or
mysterious, whereas in reality there is nothing uncertain or mysterious about it at all.

This page titled 14.5: Poisson Brackets is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

14.5.2 https://phys.libretexts.org/@go/page/7020
 CHAPTER OVERVIEW
15: Special Relativity
The phrase “special” relativity deals with the transformations between reference frames that are moving with respect to each other
at constant relative velocities. Reference frames that are accelerating or rotating or moving in any manner other than at constant
speed in a straight line are included as part of general relativity and are not considered in this chapter.
15.1: Introduction to Special Relativity
15.2: Preparation
15.3: Preparation
15.4: Speed is Relative - The Fundamental Postulate of Special Relativity
15.5: The Lorentz Transformations
15.6: But This Defies Common Sense
15.7: The Lorentz Transformation as a Rotation
15.8: Timelike and Spacelike 4-Vectors
15.9: The FitzGerald-Lorentz Contraction
15.10: Time Dilation
15.11: The Twins Paradox
15.12: A, B and C
15.13: Simultaneity
15.14: Order of Events, Causality and the Transmission of Information
15.15: Derivatives
15.16: Addition of Velocities
15.17: Aberration of Light
15.18: Doppler Effect
15.19: The Transverse and Oblique Doppler Effects
15.20: Acceleration
15.21: Mass
15.22: Momentum
15.23: Some Mathematical Results
15.24: Kinetic Energy
15.25: Addition of Kinetic Energies
15.26: Energy and Mass
15.27: Energy and Momentum
15.28: Units
15.29: Force
15.30: The Speed of Light
15.31: Electromagnetism

This page titled 15: Special Relativity is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1
15.1: Introduction to Special Relativity
Why a chapter on relativity in a book on “classical mechanics”? A first excuse might be that the phrase “classical mechanics” is
used by different authors to mean different things. To some, it means “pre-relativity”; to others it means “pre-quantum mechanics”.
For the purposes of this chapter, then, I mean the latter, so that special relativity may fairly be included in “classical” mechanics. A
second excuse is that, apart from one brief foray into an electromagnetic problem, this chapter deals only with mechanical,
kinematic and dynamical problems, and therefore deals with only a rather restricted part of relativity that can be dealt with
conveniently in a single chapter of classical mechanics rather than in a separate book. This is in fact a quite substantial restriction,
because electromagnetic theory plays a major role in special relativity. It was in fact difficulties with electromagnetic theory that
led Einstein to the special theory of relativity. Indeed, Einstein’s theory of relativity was introduced to the world in a paper with the
title Zur Elektrodynamik bewegter Körper (On the Electrodynamics of Moving Bodies), Annalen der Physik, 17, 891 (1905).
The phrase “special” relativity deals with the transformations between reference frames that are moving with respect to each other
at constant relative velocities. Reference frames that are accelerating or rotating or moving in any manner other than at constant
speed in a straight line are included as part of general relativity and are not considered in this chapter.

This page titled 15.1: Introduction to Special Relativity is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

15.1.1 https://phys.libretexts.org/@go/page/7023
15.2: Preparation
The speed of light is, by definition, exactly 2.997 924 58 % 108 m s-1, and is the same relative to all observers. This seemingly
simple sentence invites several comments.
First: Note that I have used the word “speed”. Some writers use the word “velocity” as if it were merely a more impressive and
scientific-sounding synonym for “speed”. I trust that all readers of these notes know the difference and will use the word “speed”
when they mean “speed”, and the word “velocity” when they mean “velocity – surely not an unreasonable demand. To say that the
“velocity” of light is the same for all observers means that the direction of travel of light is the same relative to all observers. This
is doubtless not at all what a writer who uses the word “velocity” intends to convey – but it is the literal (and of course quite
erroneous) meaning of the assertion.
Second: How can we possibly define the speed of light to have a certain exact value? Surely the speed of light is what we find it to
be, and we are not free to define its value. But in fact we are allowed to do this, and the explanation, briefly, is as follows.
Over the course of history, the metre has been defined in several different ways. At one time it was a specified fraction of the
circumference of Earth. Later, it was the distance between two scratches on a bar of platinum-iridium alloy held in Paris. Later still
it was a specified number of wavelengths of a particular line in the spectrum of mercury, or cadmium, or argon or krypton. In our
present state of technology it is far easier to measure and reproduce precise standards of frequency than it is to measure and
reproduce standards of length. Because of that, the current SI (Système International) unit of time is the SI second, which is based
on the frequency of a particular transition in the spectrum of caesium, and from there, the metre is defined as the distance travelled
by light in vacuo in a defined fraction of an SI second, the speed of light being assigned the exact value quoted above.
Detailed discussion of the exact definitions of the units of time, distance and speed is part of the subject of metrology. That is an
important and interesting subject, but it is only marginally relevant to the topic of relativity, and consequently, having quoted the
exact value of the speed of light, we leave further discussion of metrology here.
Third: How can the speed of light be the same relative to all observers? This assertion is absolutely central to the theory of special
relativity, and it may be regarded as its fundamental and most important principle. We shall discuss it further in the remainder of
the chapter.

This page titled 15.2: Preparation is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.2.1 https://phys.libretexts.org/@go/page/7025
15.3: Preparation
The ratio of the speed ν of a body (or a particle, or a reference frame) is often given the symbol β:
ν
β = . (15.3.1)
c

For reasons that will become apparent (I hope!) later, the range of β is usually restricted to between 0 and 1. In our study of special
relativity, we shall find that we have to make frequent use of a number of functions of β. The most common of these are
1
2 −
γ = (1 − β ) 2
, (15.3.2)

−−−−−−−
(1 + β)
k =√ , (15.3.3)
(1 − β)

z = k − 1, (15.3.4)

1 (1 + β)
−1
ϕ = ln[ ] = tanh β = ln k. (15.3.5)
2 (1 − β)

−1 −1
θ = cos γ = sin (iβγ). (15.3.6)

In Figures XV.1-3 I draw γ, k and ϕ as functions of β. The functions γ and k go from 1 to ∞ as b goes from 0 to 1; z, K and ϕ go
from 0 to ∞. The function θ is imaginary.

 Redundancy

Many – one might even say most – problems in special relativity (including examination and homework questions!) amount,
when stripped of their verbiage, to the following:
“Given one of the quantities β, γ, k, z, K, ϕ, θ, calculate one of the others.”
Thus I would suggest that, even before you have any idea what these quantities mean, you might write a program for your
computer (or programmable calculator) such that, when you enter any one of the real quantities, the computer will instantly
return all six of them. This will save you, on future occasions, from having to remember the exact formulas or having to bother
with tedious arithmetic, so that you can concentrate your mind on understanding the relativity.

15.3.1 https://phys.libretexts.org/@go/page/8466
Just for future reference, I tabulate here the relations between these various quantities. This has involved some algebra and
typesetting; I do not think there are any mistakes, but I hope some reader might check through them all carefully and will let me
know (jtatum@uvic.ca) if he or she finds any.
−−−−− 2
z(z+2) √K(K+2) 2ϕ

β = √1 −
γ
1
2
=
k −1
2
=
2
=
K+1
= tanh ϕ or e

e2ϕ +1
−1
= −i tan θ
k +1 (z+1 ) +1

2
2
(z+1 ) +1
γ =
1

2
=
k +1

2k
=
2(z+1)
= K + 1 = cosh ϕ or 1

2
(e
ϕ
+e
−ϕ
) = cos θ
√1−β

−−−
1+β −−−−− −−−−−−−− ϕ −iθ
2
k =√ = γ + √γ − 1 = z + 1 = K + 1 + √K(K + 2) = e = e
1−β

−−−
1+β −−−−− −−−−−−−− ϕ −iθ
2
z =√ − 1 = γ − 1 + √γ − 1 = k − 1 = K + √K(K + 2) = e − 1 = e −1
1−β

2 ϕ 2
2
1 (k−1) z ( e −1 )
K = −1 = γ −1 = = = = cos θ − 1
ϕ
2 2k 2(z+1) 2e
√1−β

15.3.2 https://phys.libretexts.org/@go/page/8466
 1+β −−−−− −−−−−−−−
ϕ = tanh
−1
β or 1

2
ln(
1−β
−1
) = cosh γ orln(γ + √γ
2
− 1 ) = ln k = ln(z + 1) = ln(K + 1 + √K(K + 2) ) = −iθ

i 1+β −−−−− −−−−−−−−

2
θ = ln( ) = i ln(γ + √γ − 1 ) = i ln k = i ln(z + 1) = i ln[K + 1 + √K(K + 2) ] = iϕ
2 1−β

This page titled 15.3: Preparation is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.3.3 https://phys.libretexts.org/@go/page/8466
15.4: Speed is Relative - The Fundamental Postulate of Special Relativity
You are sitting in a railway carriage (or a railroad car, if you prefer the term). The windows and curtains are closed and you cannot
see outside. You are asked to measure the constant speed of the carriage along its tracks. You try a number of experiments. You
measure the period of a simple pendulum. You slide a puck and roll a ball down an inclined plane. You throw a ball vertically up in
the air and catch it as it comes down. You throw it up at an angle and you watch it describe a graceful parabola. You cause billiard
balls to collide on the billiards table thoughtfully provided in your carriage. You experiment with a torsion pendulum. You stand a
pencil on its end and you watch it as it falls to a horizontal position.
All your careful work is to no avail. None of them tells you what speed you are moving at, or even if you are moving at all. After
exhausting all mechanical experiments you can think of, you are led to the conclusion:

It is impossible to determine the speed of motion of a uniformly-moving reference frame

by means of any mechanical experiment performed within that frame.
Frustrated, you open a curtain on one side of the carriage. You look out and you see that there is another train on the line next to
you. It appears to be moving backwards. Or are you moving forwards? Or are you both moving in the same direction but at
different speeds? You still can’t tell.
You move to the other side of the carriage and open the curtain there. This time you see the station platform, and the station
platform is moving backwards. Or are you moving forwards? (Those of you who have not done much travel by train may not
appreciate just how very strong the impression can be that the platform is moving.) What does it mean, anyway, to say that it is you
that is moving rather than the platform?
The following story is not true, but it ought to be. (It is an “apocryphal” story.) Einstein was travelling by train across Canada.
Halfway across the Prairies he leant across and tapped on the knee of his fellow passenger and asked: “Excuse me, mein Herr, bitte,
but does Regina stop at this train?”
You are about to conclude that it is not possible by any means, whether by experiment or by observation, to determine the speed of
your reference frame, or even whether it is moving or stationary.
But not so hasty! I am about to invent a speedometer, which I intend to patent and to use to make myself rich. I am going to use my
invention to measure the speed of our train – without even looking out of the window!
We shall set up two long parallel glass rods in the middle of the corridor, parallel to the railway lines and to the velocity of the train.
We shall suspend the rods horizontally, side by side from a common support, and we shall rub each of them with a silken
handkerchief, so that each of them bears an electrostatic charge of λC m . They will repel each other with an electrostatic force
−1

per unit length of

2
λ
−1
Fe = Nm , (15.4.1)
4π ϵ0 r

where r is their distance apart, and consequently they will hang out of the vertical – see figure XV.4.

15.4.1 https://phys.libretexts.org/@go/page/7026
Now see what happens when the train moves forward at speed ν . Each rod, bearing a charge λ per unit length, is now moving
forward at speed ν , and therefore each rod constitutes an electric current λν A . Therefore, by Ampère’s law, in addition to the
Coulomb repulsion, they will experience a magnetic attraction per unit length equal to
2 2
μ0 λ ν −1
Fm = Nm (15.4.2)
4πr

The net repulsive force per unit length is now

2
λ
2
(1 − μ0 ϵ0 ν ). (15.4.3)
2
4πϵ0 r

This is a little less than it was when the train was stationary, so the angle between the suspending strings is a little less, as shown in
figure XV.5. It might be noted that the force between the strings is reduced to zero (and the angle also becomes zero) when the train
is travelling at a speed 1
. We remember from electromagnetic theory that the permeability of free space is μ = 4π % 10-7 H
0
√μ0 ϵ0

m-1 and that the permittivity ϵ is 8.8542 %

0 10-12F m-1;
consequently the force and the angle drop to zero and the strings hang
vertically, when the train is moving at a speed of 2.998 % 108 m s-1.

15.4.2 https://phys.libretexts.org/@go/page/7026
To complete my invention, I am now going to attach a protractor to the instrument, but instead of marking the protractor in degrees,
I am going to calibrate it in miles per hour, and my speedometer is now ready for use (figure XV.6).

You now have a choice. Either:

i. You can choose to believe that the speedometer will work and you can accompany me to the patent office to see if they will
grant a patent for this invention, which will measure the speed of a train without reference to any external reference frame. If
you choose to believe this, there is no need for you to read the remainder of the chapter on special relativity.
or
ii. You can say that it defies common sense to believe that it is possible to determine whether a given reference frame is moving or
stationary, let alone to determine its speed. Common sense dictates that
It is impossible to determine the speed of motion of a uniformly-moving reference frame by any means whatever, whether
by a mechanical or electrical or indeed any experiment performed entirely or partially within that frame, or even by
reference to another frame.
Your common sense, then, leads you – as it should – to the fundamental principle of special relativity. Whereas some people protest
that relativity “defies common sense”, in fact relativity is common sense, and its predictions (such as your prediction that my
speedometer will not work) are exactly what common sense would lead you to expect.

This page titled 15.4: Speed is Relative - The Fundamental Postulate of Special Relativity is shared under a CC BY-NC 4.0 license and was
authored, remixed, and/or curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a

15.4.3 https://phys.libretexts.org/@go/page/7026
detailed edit history is available upon request.

15.4.4 https://phys.libretexts.org/@go/page/7026
15.5: The Lorentz Transformations
For the remainder of this chapter I am taking, as a fundamental postulate, that

It is impossible to determine the speed of motion of a uniformly-moving reference frame

by any means whatever, whether by a mechanical or electrical or indeed any experiment
performed entirely or partially within that frame, or even by reference to another frame
and consequently I am choosing to believe that my speedometer will not work. If it is impossible by any electrical experiment to
determine our speed, we must assume that all the electromagnetic equations that we know, not just the ones that we have quoted,
but indeed Maxwell’s equations, which embrace all electromagnetic phenomena, are the same in all uniformly-moving reference
frames.
One of the many predictions of Maxwell’s equations is that electromagnetic radiation (which includes light) travels at a speed
1
c = (15.5.1)
−−−−
√μ0 ϵ0

Presumably neither the permeability nor the permittivity of space changes merely because we believe that we are travelling through
space – indeed it would defy common sense to suppose that they would. Consequently, our acceptance of the fundamental principle
of special relativity is equivalent to accepting as a fundamental postulate that the speed of light in vacuo is the same for all
observers in uniform relative motion. We shall take anything other than this to be an outrage against common sense – though
acceptance of the principle will require a careful examination of our ideas concerning the relations between time and space.
Let us imagine two reference frames, ∑ and ∑ . ∑ is moving to the right (positive x-direction) at speed ν relative to ∑. (For
′ ′

brevity, I shall from time to time refer to S as the “stationary” frame, in the hope that this liberty will not lead to misunderstanding.)
At time t = t = 0 the two frames coincide, and at that instant someone strikes a match at the common origin of the two frames. At
′

a later time, which I shall call t if referred to the frame ∑, and t if referred to ∑ , the light from the match forms a spherical
′ ′

wavefront travelling radially outward at speed c from the origin O of ∑, and the equation to this wavefront, when referred to the
frame ∑, is
2 2 2 2 2
x +y +z −c t = 0. (15.5.2)

′
Referred to ∑ , it also travels outward at speed c from the origin O of S , and the equation to this wavefront, when referred to the
′ ′

frame ∑ , is
′

′2 ′2 ′2 2 ′2
x +y +z −c t = 0. (15.5.3)

Most readers will accept, I think, that y = y and z = z . Some formal algebra may be needed for a rigorous proof, but that would
′ ′

distract from our main purpose of finding a transformation between the primed and unprimed coordinates such that
′2 2 ′2 2 2 2
x −c t =x −c t . (15.5.4)

It is easy to show that the “Galilean” transformation ′

x = x − ct , t =t
′
does not satisfy this equality, so we shall have to try
harder.
Let us seek linear transformations of the form
′
x = Ax + Bt, (15.5.5)

′
t = C x + Dt, (15.5.6)

which satisfy Equation 15.5.4.

We have
′
x Ax + Bt
= , (15.5.7)
′
t C x + Dt

and, by inversion,

15.5.1 https://phys.libretexts.org/@go/page/7027
 ′ ′
x Dx − C t
= . (15.5.8)
′ ′
t Ax − C t

′
Consider the motion of O relative to ∑ and to ∑ . We have
′ x

t
=ν and x ′
=0 .
B
ν =− . (15.5.9)
A

Consider the motion of O relative to ∑ and to ∑. We have

′ x
′
t
= −ν and x = 0 .

B
−ν = (15.5.10)
D

From these we find that D = A and B = −Aν , so we arrive at

′
x = A(x − ν t) (15.5.11)

and
′
t = C x + At. (15.5.12)

On substitution of Equations 15.5.11 and 15.5.12 into Equation 15.5.4, we obtain

2 2 2 2 2 2 2
A (x − ν t) − c (C x + At) =x −c t . (15.5.13)

Equate powers of t to obtain

2

1
A = = γ. (15.5.14)
−−−−
2
1−ν
√ 2
c

Equate powers of xt to obtain

νγ
C =− . (15.5.15)
c

Equating powers of x produces no new information.

2

We have now determined A, B, C and D, and we can substitute them into Equations 15.5.5 and 15.5.6, and hence we arrive at
′
x = γ(x − ν t) (15.5.16)

and

′
t − νx
t =γ( ). (15.5.17)
2
c

These, together with y = y and z = z , constitute the Lorentz transformations, which, by suitable choice of axes, guarantee the
′ ′

invariance of the speed of light in all reference frames moving at constant velocities relative to one another.
To express x and t in terms of x and t , you may, if you are good at algebra, solve Equations 15.5.16 and 15.5.17 simultaneously
′ ′

for x and t , or, if instead, you have good physical insight, you will merely reverse the sign of ν and interchange the primed and
′ ′

unprimed quantities. Either way, you should obtain

′ ′
x = γ(x + ν t ) (15.5.18)

and
′ ′
t + νx
t =γ( ) (15.5.19)
2
c

This page titled 15.5: The Lorentz Transformations is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.5.2 https://phys.libretexts.org/@go/page/7027
15.6: But This Defies Common Sense
At this stage one may hear the protest: “But this defies common sense!”. One may hear it again as we encounter several predictions
of the invariance of the speed of light and of the Lorentz transformations. But, if you have read this far, it is too late to make such
protest. You have already, at the end of Section 15.4, made your choice, and you then decided that it defies common sense to
suppose that one can somehow determine the speed of a reference frame by some experiment or observation. You rejected that
notion, and it was the application of common sense, not its abandonment, that led us into the Lorentz transformations and the
invariance of the speed of light.
There may be other occasions when we are tempted to protest “But this defies common sense!”, and it is therefore always salutary
to recall this. For example, we shall later learn that if a train is moving at speed V relative to the station platform, and a passenger
is walking towards the front of the train at a speed ν relative to the train, then, relative to the platform, he is moving at a speed just
a little bit less that V + ν . When we protest, we are often presented with an “explanation” along the following lines:
In every day life, trains do not move at speeds comparable to the speed of light, nor do walking passengers. Therefore, we do not
notice that the combined speed is a little bit less than V + ν . After all, if V = 60 mph and ν = 4 mph, the combined speed is 0.999
999 999 999 999 5 % 64 mph. The formula V + ν is just an approximation, we are told, and we have the erroneous impression that
the combined speed is exactly V + ν only because we are accustomed, in daily life, to experiencing speeds that are small compared
with the speed of light.
This explanation somehow does not seem to be satisfactory – and nor should it, for it is not a correct explanation. It seems to be an
explanation invented for the benefit of the nonscientific layman – but nothing is ever made easy to understand by giving an
incorrect explanation under the pretence of “simplifying” something. It is not correct merely to say that the Galilean
transformations are just an “approximation” to the “real” transformations.
The problem is that it is exceedingly difficult – perhaps impossible – to describe exactly what is meant by “distance” and “time
interval”. It is almost as difficult as describing colours to a blind person, or even describing your sensation of the colour red to
another seeing person. We have no guarantee that every person’s perception of colour is the same. The best that can be done to
describe what we mean by distance and time interval is to define how distances and times transform between reference frames. The
Lorentz transformations, which we have adopted in order to make it meaningless to discuss the absolute velocity of a reference
frame, amount to a useful working definition of the meanings of space and time. Once we have adopted this definition, “common
sense” no longer comes into the matter. There is no longer a mystery which our minds cannot quite grasp; from this point on it
merely becomes a matter of algebra as to how a measurement of length or of time interval, or of speed, or of mass, as appropriately
defined, transforms when referred to one reference or to another. There is no impossible feat of imagination to be done.

This page titled 15.6: But This Defies Common Sense is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

15.6.1 https://phys.libretexts.org/@go/page/7028
15.7: The Lorentz Transformation as a Rotation
The Lorentz transformation can be written
′
x γ 0 0 iβγ x1
⎛ 1 ⎞ ⎛ ⎞⎛ ⎞
′
⎜x ⎟ ⎜ 0 1 0 0 ⎟ ⎜ x2 ⎟
⎜ 2 ⎟ = (15.7.1)
⎜ ′ ⎟ ⎜ ⎟⎜ ⎟
⎜ x3 ⎟ ⎜ 0 0 1 0 ⎟⎜x ⎟
3

⎝ ′ ⎠ ⎝ ⎠⎝ ⎠
x −iβγ 0 0 γ x4
4

where x = x , x = y , x = z and x = −ict , and similarly for primed quantities. Please do not just take my word for this;
1 2 3 4

multiply the matrices, and verify that this Equation does indeed represent the Lorentz transformation. You could, if you wish, also
write this for short:
′
x = λx. (15.7.2)

Another way of writing the Lorentz transformation is

′
x γ 0 0 βγ x1
⎛ 1 ⎞ ⎛ ⎞⎛ ⎞
′
⎜x ⎟ ⎜ 0 1 0 0 ⎟ ⎜ x2 ⎟
⎜ 2 ⎟ =
⎜ ′ ⎟ ⎜ ⎟⎜ ⎟, (15.7.3)
⎜ x3 ⎟ ⎜ 0 0 1 0 ⎟ ⎜ x3 ⎟

⎝ ′ ⎠ ⎝ ⎠⎝ ⎠
x βγ 0 0 γ x0
0

where x 1 = x,  x2 = y,  x3 = z and x

0 = ct , and similarly for primed quantities.
Some people prefer one version; others prefer the other. In any case, a set of four quantities that transforms like this is called a 4-
vector. Those who dislike version 15.7.1 dislike it because of the introduction of imaginary quantities. Those who like version
15.7.1 point out that the expression

−−−−−−−−−−−−−−−−−−−−−−−−−−−−−
2 2 2 2
√ (Δx1 ) + (Δx2 ) + (Δx3 ) + (Δx4 )

(the “interval” between two events) is invariant in four-space – that is, it has the same value in all uniformly-moving reference
1

frames, just as the distance between two points in three-space, [(Δx ) + (Δy ) 2 2
+ (Δz) ]
2
2
, is independent of the position or
orientation of any reference frame. In version 15.7.3, the invariant interval is
−−−−−−−−−−−−−−−−−−−−−−−−−−−−−
2 2 2 2
√ (Δx1 ) + (Δx2 ) + (Δx3 ) + (Δx0 ) .

Those who prefer version 15.7.1 dislike the minus sign in the expression for the interval. Those who prefer version 15.7.3 dislike
the imaginary quantities of version 15.7.1.
For the time being, I am going to omit y and z , so that I can concentrate my attention on the relations between x and t . Thus I am
going to write 15.7.1 as
′ ′
x x γ iβγ x
1
( ) =( ) =( )( ) (15.7.4)
′ ′
x ict −iβγ γ ict
4

and Equation 15.7.3 as

′
x γ βγ x
( ) =( )( ) (15.7.5)
′
ct βγ γ ct

Readers may notice how closely Equation 15.7.4 resembles the Equation for the transformation of coordinates between two
reference frames that are inclined to each other at an angle. (See Celestial Mechanics Section 3.6.) Indeed, if we let cos θ = γ and
sin θ = iβγ , Equation 15.7.4 becomes

′
x cos θ sin θ x
( ) =( )( ) (15.7.6)
′
ict − sin θ cos θ ct

The matrices in Equations 15.7.1, 15.7.4 and 15.7.6 are orthogonal matrices and they satisfy each of the criteria for orthogonality
described, for example, in Celestial Mechanics Section 3.7. We can obtain the converse relations (i.e. we can express x and t in

15.7.1 https://phys.libretexts.org/@go/page/8460
terms of x and t ) by interchanging the primed and unprimed quantities and either reversing the sign of
′ ′
β or of θ or by
interchanging the rows and columns of the matrix.
There is a difficulty in making the analogy between the Lorentz transformation as expressed by Equation 15.7.4 and rotation of
axes as expressed by Equation 15.7.6 in that, since γ > 1 , θ is an imaginary angle. (At this point you may want to reach for your
ancient, brittle, yellowed notes on complex numbers and hyperbolic functions.) Thus
−−−−−
θ = cos
−1
γ and for γ >1 , this means that θ = i cosh γ = i ln(γ + √γ − 1 ) .
−1
And 2

−−−−−− −
θ = sin
−1
(iβγ) = i sinh
−1
(βγ) = i ln(βγ + √β 2 γ 2 + 1 ) . Either of these expressions reduces to θ = i ln[γ(1 + β)] . Perhaps a
yet more convenient way of expressing this is

−1
1 1 +β
θ = i tanh β = i ln( ). (15.7.7)
2 1 −β

For example, if β = 0.8, θ = 1.0986i, which might be written (not necessarily particularly usefully) as i62o 57'.
At this stage, you are probably thinking that you much prefer the version of Equation 15.7.5 , in which all quantities are real, and
1

the expression for the interval between two events is [(Δx ) + (Δx ) + (Δx ) − (Δx
1
2
2
2
3
2 2
0) ] 2 . The minus sign in the expression
is a small price to pay for the realness of all quantities. Equation 15.7.5 can be written
′
x cosh ϕ sinh ϕ x
( ) =( )( ) (15.7.8)
′
ct sinh ϕ cosh ϕ ct

where cosh ϕ = γ,   sinh ϕ = βγ,   tanh ϕ = β . On the face of it, this looks much simpler.
No messing around with imaginary angles. Yet this formulation is not without its own set of difficulties. For example, neither the
matrix of Equation 15.7.5 nor the matrix of Equation 15.7.8 is orthogonal. You cannot invert the Equation to find x and t in terms
of x and t merely by interchanging the primed and unprimed symbols and interchanging the rows and columns. The converse of
′ ′

Equation 15.7.8 is in fact

′
x cosh ϕ − sinh ϕ x
( ) =( )( ) (15.7.9)
′
ct − sinh ϕ cosh ϕ ct

which can also (understandably!) be written

′
x cosh(−ϕ) sinh(−ϕ) x
( ) =( )( ) (15.7.10)
′
ct sinh(−ϕ) cosh(−ϕ) ct

which demands as much skill in handling hyperbolic functions as the other formulation did in handling complex numbers. A
further problem is that the formulation 15.7.5 does not allow the analogy between the Lorenz transformation and the rotation of
axes. You take your choice.
It may be noticed that the determinants of the matrices of Equations 15.7.5 and 15.7.8 are each unity, and it may therefore be
thought that each matrix is orthogonal and that its reciprocal is its transpose. But this is not the case, for the condition that the
determinant is inity is not a sufficient condition for a matrix to be orthogonal. The necessary tests are summarized in Celestial
Mechanics, Section 3.7, and it will be found that several of the conditions are not satisfied.

This page titled 15.7: The Lorentz Transformation as a Rotation is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or
curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is
available upon request.

15.7.2 https://phys.libretexts.org/@go/page/8460
15.8: Timelike and Spacelike 4-Vectors
I am going to refer some events to a coordinate system whose origin is here and now and which is moving at the same velocity as
you happen to be moving. In other words, you are sitting at the origin of the coordinate system, and you are stationary with respect
to it. Let us suppose that an event A occurs at the following coordinates referred to this reference frame, in which the distances
x ,  y ,  z
1 1 1 are expressed in light-years (lyr) the time t is expressed in years (yr).
1

x1   =  2 y1   =  3 z1   =  7 t1   =   − 1

A “light-year” is a unit of distance used when describing astronomical distances to the layperson, and it is also useful in describing
some aspects of relativity theory. It is the distance travelled by light in a year, and is approximately 9.46 % 1015 m or 0.307 parsec
−−
(pc). Event A, then, occurred a year ago at a distance of √62  =  7.87 lyr, when referred to this reference frame. Note that, if
referred to a reference frame that coincides with this one at t = 0 , but is moving with respect to it, all four coordinates might be
− −−−−−−−− −
different, and the distance √x + y + z and the time of occurrence would be different, but, according to the way in which we
2 2 2

−−−−−−−−−−−−−− −
have defined space and time by the Lorentz transformation, the quantity √x + y + z − c t would be the same.
2 2 2 2 2

Imagine now a second event, B, which occurs at the following coordinates:

x2   =  5 y2   =  8 z2   =  10 t2   =   + 2

−−−
That is to say, when referred to the same reference frame, it will occur in two years’ time at a distance of √189 = 13.75 lyr.
The 4-vector s = B - A connects these two events, and the magnitude s of s is the interval between the two events. Note that the
−−−−−−−−−−−−−−−−−−−−−−− −
distance between the two events, when referred to our reference frame, is √(5 − 2) + (8 − 3) + (10 − 7) = 6.56 lyr. The
2 2 2

−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−− −
interval between the two events is √(5 − 2) + (8 − 3) + (10 − 7) − (2 + 1)
2 2 2
= 5.83 lyr, and this is independent of the
2

velocity of the reference frame. That is, if we “rotate” the reference frame, it obviously makes no difference to the interval between
the two events, which is invariant.
As another example, consider two events A and B whose coordinates are
x1   =  2 y1   =  5 z1   =  3 t1   =   − 2

x2   =  3 y2   =  7 z2   =  4 t2   =   + 6

with distances, as before, expressed in lyr, and times in yr. Calculate the interval between these two events – i.e. the magnitude of
the 4-vector connecting them. If you carry out this calculation, you will find that s = -58, so that the interval s is imaginary and
2

equal to 7.62i.
So we see that some pairs of events are connected by a 4-vector whose magnitude is real, and other pairs are connected by a 4-
vector whose magnitude is imaginary. There are differences in character between real and imaginary intervals, but, in order to strip
away distractions, I am going to consider events for which y = z = 0 . We can now concentrate on the essentials without
being distracted by unimportant details.
Let us therefore consider two events A and B whose coordinates are
x1   =  2 lyr t
1 = −2 yr
x2   =  3 lyr t
2 = +6 yr
These events and the 4-vector connecting them are shown in Figure XV.7.Event A happened two years ago (referred to our
reference frame); event B will occur (also referred to our reference frame) in six years’ time. The square of the interval between the
two events (which is invariant) is -63 lyr2, and the interval is imaginary. If someone wanted to experience both events, he would
have to travel only 1 lyr (referred to our reference frame), and he could take his time, for he would have eight years (referred to our
reference frame) in which to make the journey to get to event B in time. He couldn’t totally dawdle, however; he would have to
travel at a speed of at least times the speed of light, but that’s not extremely fast for anyone well versed in relativity.
1

15.8.1 https://phys.libretexts.org/@go/page/8461
Let’s look at it another way. Let’s suppose that event A is the cause of event B. This means that some agent must be capable of
conveying some information from A to B at a speed at least equal to times the speed of light. That may present some technical
1

problems, but it presents no problems to our imagination.

You’ll notice that, in this case, the interval between the two events – i.e. the magnitude of the 4-vector connecting them - is
imaginary. A 4-vector whose magnitude is imaginary is called a timelike 4-vector.There is quite a long time between events A and
B, but not much distance.
Now consider two events A and B whose coordinates are
x1   =  2 lyr t
1 = −1 yr
x2   =  7 lyr t
2 = +3 yr
The square of the magnitude of the interval between these two events is +9 lyr2, and the interval is real. A 4-vector whose
magnitude is real is called a spacelike 4-vector. It is shown in Figure XV.8.

Perhaps I could now ask how fast you would have to travel if you wanted to experience both events. They are quite a long way
apart, and you haven’t much time to get from one to the other. Or, if event A is the cause of event B, how fast would an
information-carrying agent have to move to convey the necessary information from A in order to instigate event B? Maybe you
have already worked it out, but I’m not going to ask the question, because in a later section we’ll find that two events A and B
cannot be mutually causally connected if the interval between them is real. Note that I have said “mutually”; this means that A
cannot cause B, and B cannot cause A. A and B must be quite independent events; there simply is too much space in the interval
between them for one to be the cause of the other. It does not mean that the two events cannot have a common cause. Thus, Figure
XV.9 shows two events A and B with a spacelike interval between them (very steep) and a third event C such the intervals CA and
CB (very shallow) are timelike. C could easily be the cause of both A and B; that is, A and B could have a common cause.But there
can be no mutual causal connection between A and B. (It might be noted parenthetically that Charles Dickens temporarily nodded
when he chose the title of his novel Our Mutual Friend. He really meant our common friend. C was a friend common to A and to
B. A and B were friends mutually to each other.)

15.8.2 https://phys.libretexts.org/@go/page/8461
  Exercise 15.8.1

The distance of the Sun from Earth is 1.496 % 1011 m. The speed of light is 2.998 % 108 m s-1.
How long does it take for a photon to reach Earth from the Sun?
Event A: A photon leaves the Sun on its way to Earth. Event B: The photon arrives at Earth.
What is the interval (i.e. s in 4-space) between these two events?

This page titled 15.8: Timelike and Spacelike 4-Vectors is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

15.8.3 https://phys.libretexts.org/@go/page/8461
15.9: The FitzGerald-Lorentz Contraction
This is sometimes described in words something like the following:
If a measuring-rod is moving with respect to a “stationary” observer, it “appears” to be shorter than it “really” is.
This is not a very precise statement, and the words that I have placed in inverted commas call for some clarification.
We have seen that, while the interval between two events is invariant between reference frames, the distance between two points
(and hence the length of a rod) depends on the coordinate frame to which the points are referred. Let us now define what we mean
by the length of a rod. Figure XV.10 shows a reference frame, and a rod lying parallel to the x-axis. For the moment I am not
specifying whether the rod is moving with respect to the reference frame, or whether it is stationary.

Let us suppose that the x-coordinate of the left-hand end of the rod is x , and that, at the same time referred to this reference frame,
1

the x-coordinate of the right-hand end is x . The length l of the rod is defined as l =  x − x . That could scarcely be a simpler
2 2 1

statement – but note the little phrase “at the same time referred to this reference frame”. That simple phrase is important.
Now let’s look at the FitzGerald-Lorentz contraction. See Figure XV.11.

′
The are two reference frames, ∑ and sum . The frame ∑ is moving to the right with respect to ∑ with speed ν . A rod is at rest
′

with respect to the frame ∑ , and is therefore moving to the right with respect to ∑ at speed ν .
′

In my younger days I often used to travel by train, and I still like to think of railway trains whenever I discuss relativity. Modern
students usually like to think of spacecraft, presumably because they are more accustomed to this mode of travel. In the very early
days of railways, it was customary for the stationmaster to wear top hat and tails. Those days are long gone, but, when thinking
about the FitzGerald-Lorentz contraction, I like to think of ∑ as being a railway station in which there resides a stationmaster in
′
top hat and tails, while ∑ is a railway train.
′
The length of the rod, referred to the frame ∑ , is l = x′ ′
2
−x
′
1
, in what I hope is obvious notation, and of course these two
coordinates are determined at the same time referred to ∑ . ′

The length of the rod referred to a frame in which it is at rest is called its proper length. Thus l is the proper length of the rod.
′

Now it should be noted that, according to the way in which we have defined distance and time by means of the Lorentz
transformation, although x and x are measured simultaneously with respect to ∑ , these two events (the determination of the
′
2
′
1
′

coordinates of the two ends of the rod) are not simultaneous when referred to the frame ∑ (a point to which we shall return in a
later section dealing with simultaneity). The length of the rod referred to the frame ∑ is given by l = x − x , where these two
2 1

15.9.1 https://phys.libretexts.org/@go/page/8462
 ′
x
coordinates are to be determined at the same time when referred to . Now Equation 15.5.16 tells us that
∑ x2 =
γ
2
+ νt and
′
x1
x1 =
γ
. (Readers should note this derivation very carefully, for it is easy to go wrong. In particular, be very clear what is
+ νt

meant in these two equations by the symbol t . It is the single instant of time, referred to ∑, when the coordinates of the two ends
are determined simultaneously with respect to ∑.) From these we reach the result:
′
l
l = . (15.9.1)
γ

This is the FitzGerald-Lorentz contraction.

It is sometimes described thus: A railway train of proper length 100 yards is moving past a railway station at 95% of the speed of
light (γ = 3.2026.) To the stationmaster the train “appears” to be of length 31.22 yards; or the stationmaster “thinks” the length of
the train is 31.22 yards; or, “according to” the stationmaster the length of the train is 31.22 yards. This gives a false impression, as
though the stationmaster is under some sort of misapprehension concerning the length of the train, or as if he is labouring under
some sort of illusion, and it introduces some sort of unnecessary “mystery” into what is nothing more than simple algebra. In fact
what the stationmaster “thinks” or “asserts” is entirely irrelevant. Two correct statements are: 1. The length of the train, referred to
a reference frame in which it is at rest – i.e. the proper length of the train – is 100 yards. 2. The length of the train when referred to
a frame with respect to which it is moving at a speed of 0.95c is 31.22 yards. And that is all there is to it. Any phrase such as “this
observer thinks that” or “according to this observer” should always be interpreted in this manner. It is not a matter of what an
observer “thinks”. It is a matter of which frame a measurement is referred to. Nothing more, nothing less.

It is possible to describe the Lorentz-FitzGerald contraction by interpreting the Lorentz transformations as a rotation in 4-space.
′
Whether it is helpful to do so only you can decide. Thus Figure XV.12 shows ∑ and ∑ related by a rotation in the manner
described in Section 15.7. The thick continuous line shows a rod oriented so that its two ends are drawn at the same time with
respect to ∑ . Its length is, referred to sum , l , and this is its proper length. The thick dotted line shows the two ends at the same
′ ′ ′

time with respect to ∑. Its length referred to ∑ is l = l

. And, since cos θ = γ , which is greater than 1, this means that, in spite
cos θ

of appearances in the Figure, l < l . The Figure is deceptive because, as discussed in Section 15.7, θ is imaginary. As I say, only
′

you can decide whether this way of looking at the contraction is helpful or merely confusing. It is, however, at least worth looking
at, because I shall be using this concept of rotation in a forthcoming section on simultaneity and order of events. Illustrating the
Lorentz transformations as a rotation like this is called a Minkowski diagram.

This page titled 15.9: The FitzGerald-Lorentz Contraction is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

15.9.2 https://phys.libretexts.org/@go/page/8462
15.10: Time Dilation
We imagine the same railway train ∑ and the same railway station ∑ as in the previous section except that, rather than measuring
′

a length referred to the two reference frames, we measure the time interval between two events. We’ll suppose that a passenger in
the railway train ∑ claps his hands twice. These are two events which, when referred to the reference frame ∑ , take place at the
′ ′

′
same place when referred to this reference frame. Let the instants of time when the two events occur, referred to ∑ , be t and t . ′
1
′
2

The time interval T is defined as t − t . But the Lorentz transformation is

′ ′
2
′
1

′
νx
′
t = γ(t + )
2
c

and so the time interval when referred to ∑ is

′
T = γT . (15.10.1)

This is the dilation of time. The situation is illustrated by a Minkowski diagram in Figure XV.13. While it is clear from the figure
that T = T cos θ and therefore that T = γT it is not so clear from the figure that this means that T is greater than T – because
′ ′ ′

cos θ > 1 and θ is imaginary.

Thus, let us suppose that a passenger on the train holds a 1-metre measuring rod (its length in the direction of motion of the train)
and he claps his hands at an interval of one second apart. Let’s suppose that the train is moving at 98% of the speed of light (γ =
5.025). In that case the stationmaster thinks that the length of the rod is only 19.9 cm and that the time interval between the claps is
5.025 seconds.
I deliberately did not word that last sentence very well. It is not a matter of what the stationmaster or anyone else “thinks” or
“asserts”. It is not a matter that the stationmaster is somehow deceived into erroneously believing that the rod is 19.9 cm long and
the claps 5.025 seconds apart, whereas they are “really” 1 metre long and 1 second apart. It is a matter of how length and time are
defined (by subtracting two space coordinates determined at the same time, or two time coordinates at the same place) and how
space-time coordinates are defined by means of the Lorentz transformations. The length is 19.9 cm, and the time interval is 5.025
seconds when referred to the frame ∑. It is true that the proper length and the proper time interval are the length and the time
interval referred to a frame in which the rod and the clapper are at rest. In that sense one could loosely say that they are “really” 1
metre long and 1 second apart. But the Lorentz contraction and the time dilation are not determined by what the stationmaster or
anyone else “thinks”.
Another way of looking at it is this. The interval s between two events is clearly independent of the orientation any reference
frames, and is the same when referred to two reference frames that may be inclined to each other. But the components of the vector
joining two events, or their projections on to the time axis or a space axis are not at all expected to be equal.
By the way, in Section 15.3 I urged you to write a computer or calculator programme for the instant conversion between the several
factors commonly encountered in relativity. I still urge it. As soon as I typed that the train was travelling at 98% of the speed of
light, I was instantly able to generate γ. You need to be able to do that, too.

This page titled 15.10: Time Dilation is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.10.1 https://phys.libretexts.org/@go/page/8463
15.11: The Twins Paradox
During the late 1950s and early 1960s there was great controversy over a problem known as the “Twins Paradox”. The controversy
was not confined to within scientific circles, but was argued, by scientists and others, in the newspapers, magazines and many
serious journals. It goes something like this:
There are two 20-year-old twins, Albert and Betty. Albert is a sedentary type who likes nothing better than to stay at home tending
the family vineyards. His twin sister Betty is a more adventurous type, and has trained to become an astronaut. On their twentieth
birthday, Betty waves a cheery au revoir to her brother and takes off on what she intends to be a brief spaceflight, at which she
travels at 99.98 % of the speed of light (γ = 50). After six months by her calendar she turns back and on her 21st birthday she
arrives back home to greet her brother, only to find that he is now old and sere and has laboured, by his calendar for 50 years and is
now an aged man of 71 years. If we accept what we have derived in the previous section about the dilation of time, there would
seem to be no particular problem with that. It has even been argued that travel between the stars may not be an impossibility.
Whereas to an Earthbound observer it may take many decades for a spacecraft to travel to a star and back, for the astronauts on
board much less time has elapsed.
And yet a paradox was pointed out. According to the principles of the relativity of motion, it was argued, one could refer
everything to Betty’s reference frame, and from that point of view one could regard Betty as being the stationary twin and Albert as
the one who travelled off into the distance and returned later. Thus, it could be argued, it would be Albert who had aged only one
year, while Betty would have aged fifty years. Thus we have a paradox, which is a problem which apparently gives rise to opposite
conclusions depending on how it is argued. And the only way that the paradox could be resolved was to suppose that both twins
were the same age when they were re-united.
A second argument in favour of this interpretation that the twins were the same age when re-united points out that dilation of time
arises because two events that may occur in the same place when referred to one reference frame do not occur in the same place
when referred to another. But in this case, the two events (Betty’s departure and re-arrival) occur at the same place when referred to
both reference frames.
The argument over this point raged quite furiously for some years, and a particularly plausible tool that was used was something
referred to as the "k -calculus” – an argument that is, however, fatally flawed because the “rules” of the k -calculus inherently
incorporate the desired conclusion. Two of the principal leaders of the very public scientific debate were Professors Fred Hoyle and
Herbert Dingle, and this inspired the following letter to a weekly magazine, The Listener, in 1960:
Sir:
The ears of a Hoyle may tingle;
The blood of a Hoyle may boil
When Hoyle pours hot oil upon Dingle,
And Dingle cold water on Hoyle.
But the dust of the wrangle will settle.
Old stars will look down on new soil.
The pot will lie down with the kettle,
And Dingle will mingle with Hoyle.
So what are you, the reader, expected to believe? Let us say this: If you are a student who has examinations to pass, or if you are an
untenured professor who has to hold on to a job, be in no doubt whatever: The original conclusion is the canonically-accepted
correct conclusion, namely that Albert has aged 50 years while his astronaut sister has aged but one. This is now firmly accepted
truth. Indeed it has even been claimed that it has been “proved” experimentally by a scientist who took a clock on commercial
airline flights around the world, and compared it on his return with a stay-at-home clock. For myself I have neither examinations to
pass nor, alas, a job to hold on to, so I am not bound to believe one thing or the other, and I elect to hold my peace.
I do say this, however – that what anyone “believes” is not an essential point. It is not a matter of what Albert or Betty or Hoyle or
Dingle or your professor or your employer “believes”. The real question is this: What is it that is predicted by the special theory of
relativity? From this point of view it does not matter whether the theory of relativity is “true” or not, or whether it represents a
correct description of the real physical world. Starting from the basic precepts of relativity, whether “true” or not, it must be only a
matter of algebra (and simple algebra at that) to decide what is predicted by relativity.

15.11.1 https://phys.libretexts.org/@go/page/8464
A difficulty with this is that it is not, strictly speaking, a problem in special relativity, for special relativity deals with
transformations between reference frames that are in uniform motion relative to one another. It is pointed out that Albert and Betty
are not in uniform motion relative to one another, since one or the other of them has to change the direction of motion – i.e. has to
accelerate. It could still be argued that, since motion is relative, one can regard either Albert or Betty as the one who accelerates –
but the response to this is that only uniform motion is relative. Thus there is no symmetry between Albert and Betty. Betty either
accelerates or experiences a gravitational field (depending on whether her experience is referred to Albert’s or her own reference
frame). And, since there is no symmetry, there is no paradox. This argument, however, admits that the age difference between
Albert and Betty on Betty’s return is not an effect of special relativity, but of general relativity, and is an effect caused by the
acceleration (or gravitational field) experienced by Betty.
If this is so, there are some severe difficulties is describing the effect under general relativity. For example, whether the general
theory allows for an instantaneous change in direction by Betty (and infinite deceleration), or whether the final result depends on
how she decelerates – at what rate and for how long – must be determined by those who would tackle this problem. Further, the
alleged age difference is supposed to depend upon the time during which Betty has been travelling and the length of her journey –
yet the portion of her journey during which she is accelerating or decelerating can be made arbitrarily short compared with the time
during which she is travelling at constant speed.
If the effect were to occur solely during the time when she was accelerating or decelerating, then the total length and duration of the
constant speed part of her journey should not affect the age difference at all.
Since this chapter deals only with special relativity, and this is evidently not a problem restricted to special relativity, I leave the
problem, as originally stated, here, without resolution, for readers to argue over as they will

This page titled 15.11: The Twins Paradox is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum
via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.11.2 https://phys.libretexts.org/@go/page/8464
15.12: A, B and C
A, B and C were three characters in the Canadian humorist Stephen Leacock’s essay on The Human Element in Mathematics. “A ,
B and C are employed to dig a ditch. A can dig as much in one hour as B can dig in two...”

We can ask A , B and C to come to our aid in a modified version of the twins’ problem, for we can arrange all three of them to be
moving with constant velocities relative to each other. It goes like this (figure XV.14):

The scenario is probably obvious from the figure. There are three events:
1. B passes A
2. B meets C
3. C meets A
At event 1, B and A synchronize their watches so that each reads zero. At event 2, C sets his watch so that it reads the same as
B s . At event 3, C and A compare watches. I shall leave the reader to cogitate over this. The only thing I shall point out is that this
′

problem differs from the problem described as the Twins Paradox in two ways. In the first place, unlike in the Twins Paradox, all
three characters, A , B and C are moving at constant velocities with respect to each other. Also, the first and third events occur at
the same place relative to A but at different places referred to B or to C . In the twin paradox problem, the two events occur at the
same place relative to both frames.

This page titled 15.12: A, B and C is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.12.1 https://phys.libretexts.org/@go/page/8465
15.13: Simultaneity
If the time interval referred to one reference frame can be different when referred to another reference frame (and since time
interval is merely one component of a four-vector, the magnitude of the component surely depends on the orientation in four space
of the four axes) this raises the possibility that there might be a time interval of zero relative to one frame (i.e. two events are
simultaneous) but are not simultaneous relative to another. This is indeed the case, provided that the two events do not occur in the
same place as well as at the same time. Look at Figure XV.15.

I have drawn two reference frames at an (imaginary) angle θ to each other. Think of ∑ as the railway station and of ∑ as the ′

railway train, and that the speed of the railway train is c tan θ (You may have to go back to Section 15.3 or 15.7 to recall the
′
relation of θ to the speed.) The thick line represents the interval between two events that are simultaneous when referred to ∑ , but
are separated in space (one occurs near the front of the train; the other occurs near the rear). (Note also in this text that I am using
the phrase “time interval” to denote the time-component of the “interval”. For two simultaneous events, the time interval is zero,
and the interval is then merely the distance between the two events.)
While the thick line has zero component along the ict
′
axis, its component along the ict axis is ′
l sin θ . That is,
ic(t − t ) = l sin θ = l × iβγ .
′ ′
2 1

Hence:
′
βγl
t2 − t1 = . (15.13.1)
c

For example, if the events took place simultaneously 100,000 km apart in the train (it is a long train) and if the train were travelling
at 95% of the speed of light (γ = 3.203; it is a fast train), the two events would be separated when referred to the railway station by
1.01 seconds. The event near the rear of the train occurred first.

This page titled 15.13: Simultaneity is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.13.1 https://phys.libretexts.org/@go/page/9045
15.14: Order of Events, Causality and the Transmission of Information
Maybe it is even possible that if one event precedes another in one reference frame, in another reference frame the other precedes
the one. In other words, the order of occurrence of events may be different in two frames. This indeed can be the case, and
Minkowski diagrams (Figure XV.16) can help us to see why and in what circumstances.

In part (a), of the two events 1 and 2, 1 occurs before 2 in either ∑ or ∑ . (from this point on I shall use a short phrase such as “in
′

∑” rather than the more cumbersome “when referred to the reference frame ∑”. But in part (b), event 1 occurs before event 2 in

∑, but after event 2 in ∑ . One can see that there is reversal of order of events if the slope of the line joining to two events is less
′

than the angle θ . The angle θ , it may be recalled, is an imaginary angle such than tan θ = iβ = , where ν is the relative speed of
iν

the two frames. In Figure XV.17, for simplicity I am going to suppose that event 1 occurs at the origin of both frames, and that
event 2 occurs at coordinates (ν  t ,  ict ) in S. The condition for no reversal of events is then evidently
ict iν
≥ tan θ = iβ = ;
νt c

or
ν ≤c (15.14.1)

This means, in effect, that neither mass nor energy can be transmitted faster than the speed of light. That is not quite the same thing
as saying that “nothing” can be transmitted faster than the speed of light. For example a Moiré pattern formed by two combs with
slightly different tooth spacings can move faster than light if one of the combs is moved relative to the other; but then I suppose it
has to be admitted that in that case “nothing” is actually being transmitted – and certainly nothing that can transmit information or
that can cause an event. An almost identical example would be the modulation envelope of the sum of two waves of slightly
different frequencies. A well-known example from wave mechanics is that of the wave representation of a moving particle. The
wave group (which is the integral of a continuous distribution of wavelengths whose extent is governed by Heisenberg’s principle)
moves with the particle at a sub-luminal speed, but there is nothing to prevent the wavelets within the group moving through the
group at any speed. These wavelets may start at the beginning of the group and rapidly move through the group and extinguish
themselves at the end. No “information” is transmitted from A to B at a speed any faster than the particle itself is moving.

This page titled 15.14: Order of Events, Causality and the Transmission of Information is shared under a CC BY-NC 4.0 license and was
authored, remixed, and/or curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a
detailed edit history is available upon request.

15.14.1 https://phys.libretexts.org/@go/page/8467
15.15: Derivatives
We’ll pause here and establish a few derivatives just for reference and in case we need them later.
We recall that the Lorentz relations are
′ ′
x = γ(x + ν t ) (15.15.1)

and
′
′
βx
t = γ (t + ) (15.15.2)
c

From these we immediately find that

∂x ∂x ∂t βγ ∂t
( ) = γ; ( ) = γν ; ( ) = ; ( ) = γ. (15.15.3a,b,c,d)
′ ′ ′ ′
∂x ′ ∂t ′ ∂x ′ c ∂t ′
t x t x

We shall need these in future sections.

 Caution
It is not impossible to make a mistake with some of these derivatives if one allows one’s attention to wander. For example, one
′

might suppose that, since = γ then “obviously”

∂x

∂x′
= - and indeed this is correct if t is being held constant. However,
∂x

∂x
1

γ
′

we have to be sure that this is really what we want. The difficulty is likely to arise if, when writing a partial derivative, we
neglect to specify what variables are being held constant, and no great harm would be done by insisting that these always be
specified when writing a partial derivative. If you want the inverses rather than the reciprocals of Equations 15.15.3a,b,c,dthe
rule, as ever, is: Interchange the primed and unprimed symbols and change the sign of ν or β. For example, the reciprocal of
′ ′

(
∂x
′
∂x
)
′
is ( ∂x

∂x
)
′
, while its inverse is ( ∂x

∂x
) . For completeness, and reference, then, I write down all the possibilities:
t t t

′ ′ ′ ′
∂x 1 ∂t 1 ∂x c ∂t 1
( ) = ; ( ) = ; ( ) = ; ( ) = . (5.15.3e,f,g,h)
∂x ′ γ ∂x γν ∂t ′ βγ ∂t γ
t x′ t x′

′ ′ ′ ′
∂x ∂x ∂t βγ ∂t
( ) = γ; ( ) = −γν ; ( ) =− ; ( ) = γ. (15.15.3i,j,k,l)
∂x ∂t ∂x c ∂t
t x t x

∂x 1 ∂t 1 ∂x c ∂t 1
( ) = ; ( ) =− ; ( ) =− ; ( ) = . (15.15.3m,n,o,p)
′ ′ ′ ′
∂x γ ∂x γν ∂t βγ ∂t γ
t x t x

Now let’s suppose that ψ = ψ(x, t) where x and t are in turn functions (Equations 15.15.1 and 15.15.2) of x and t . Then ′ ′

∂ψ ∂x ∂ψ ∂t ∂ψ ∂ψ βγ ∂ψ
= + =γ + (15.15.3)
′ ′ ′
∂x ∂x ∂t ∂x ∂t ∂x c ∂t

and
∂ψ ∂x ∂ψ ∂t ∂ψ ∂ψ ∂ψ
= + = γν +γ . (15.15.4)
′ ′ ′
∂t ∂t ∂x ∂t ∂t ∂x ∂t

The reader will doubtless notice that I have here ignored my own advice and I have not indicated which variables are to be held
constant. It would be worth spending a moment here thinking about this.
We can write Equations 15.15.3 and 15.15.4 as equivalent operators:
∂ ∂ β ∂
=γ( + ) (15.15.5)
′
∂x ∂x c ∂t

and
∂ ∂ ∂
= γ (ν + ). (15.15.6)
′
∂t ∂x ∂t

15.15.1 https://phys.libretexts.org/@go/page/8468
We can also, if we wish, find the second derivatives. Thus
2 2 2 2 2
∂ ψ 2
∂ 2β ∂ β ∂
=γ ( + + ). (15.15.7)
′2 2 2 2
∂x ∂x c ∂x∂t c ∂t

In a similar manner we obtain

2 2 2 2
∂ 2
∂ 2
∂ β ∂
=γ (ν + (1 + β ) + ) (15.15.8)
′ ′
∂x ∂t ∂x2 ∂x∂t c ∂t2

and
2 2 2 2
∂ 2 2
∂ ∂ ∂
=γ (ν + 2ν + ). (15.15.9)
∂t′2 ∂x2 ∂x∂t ∂t2

The inverses of all of these relations are to be found by interchanging the primed and unprimed coordinates and changing
the signs of ν and β.

This page titled 15.15: Derivatives is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.15.2 https://phys.libretexts.org/@go/page/8468
15.16: Addition of Velocities
A railway train trundles towards the east at speed ν , and a passenger strolls towards the front at speed ν . What is the speed of the
1 2

passenger relative to the railway station? We might at first be tempted to reply: “Why, ν + ν of course.” In this section we shall 1 2

show that the answer as predicted from the Lorentz transformations is a little less than this, and we shall develop a formula for
calculating it. We have already discussed (in Section 15.6) our answer to the objection that this defies common sense. We pointed
out there that the answer (to the perfectly reasonable objection) that “at the speeds we are accustomed to we would hardly notice
the difference” is not a satisfactory response. The reason that the resultant speed is a little less than ν + ν results from the way in 1 2

which we have defined the Lorentz transformations between references frames and the way in which distances and time intervals
are defined with reference to reference frames in uniform relative motion

Figure XV.17 shows two references frames, Σ and Σ , the latter moving at speed ν with respect to the former. A particle is moving
′

with velocity u in Σ , with components u and u . (“ in Σ ” = “referred to the reference frame Σ ”.)
′ ′ ′
x
′
′
y
′
′ ′

What is the velocity of the particle in Σ?

Let us start with the x-component.
We have:
∂x ∂x ∂x ∂x
′ ′ ′
( ) dx + ( ) dt ( ) u +( )
′ ′ ′ ′
dx ∂x ′ ∂t ′ ∂x ′ ∂t ′
t x t x
u = = = (15.16.1)
dt ∂t ∂t ∂t ∂t
′ ′ ′
( ) dx + ( ) dt ( ) u +( )
∂x′ ′ ∂t′ ∂x′ ′ ∂t′
t x′ t x′

ν
We take the derivatives from Equations 15.15.3a,b,c,d, and, writing for β, we obtain
c

′
ux + ν
ux = . (15.16.2)
ν
′
1 +u ′
x 2
c

The inverse is obtained by interchanging the primed and unprimed symbols and reversing the sign of ν .
The y -component is found in an exactly similar manner, and I leave its derivation to the reader. The result is
′
u +ν
uy = (15.16.3)
ν
′
1 +u
2
c

Special cases:
I. If u
′
x
′
′
=u and u
′
y
′ =0 , then
′
u +ν
ux = (15.16.4a)
ν
′
1 +u
2
c

uy = 0 (15.16.4b)

15.16.1 https://phys.libretexts.org/@go/page/8469
II. If u ′
x′
=0 and u ′
y′
=u
′
then

ux = ν (15.16.5a)

′
u
uy = (15.16.5b)
γ

′
ν u
Equation 15.16.4a as written is not easy to commit to memory, though it is rather easier if we write β1 = ,  β2 = and
c c
ux
β = . Then the equation becomes
c

β1 + β2
β = (15.16.4)
1 + β1 β2

In Figure XV.18, a train Σ is trundling with speed β (times the speed of light) towards the right, and a passenger is strolling
′
1

towards the front at speed β . The speed β of the passenger relative to the station Σ is then given by Equation 15.16.4. In Figure
2

XV.19, two trains, one moving at speed β and the other moving at speed β , are moving towards each other. (If you prefer to think
1 2

of protons rather than trains, that is fine.) Again, the relative speed b of one train relative to the other is given by Equation 15.16.4.

 Example 15.16.1

A train trundles to the right at 90% of the speed of light relative to Σ, and a passenger strolls to the right at 15% of the speed of
light relative to Σ . The speed of the passenger relative to Σ is 92.5% of the speed of light.
′

The relation between β , β and β is shown graphically in Figure XV.20.

1 2

15.16.2 https://phys.libretexts.org/@go/page/8469
 β1
If I use the notation to mean “combining β with β ”, I can write Equation 15.16.4 as
1 2
β2

β1 + β2
β1 ⊕ β2 = (15.16.5)
1 + β1 β2

β1 + β2
You may notice the similarity of Equation 15.16.4β = to the hyperbolic function identity
1 + β1 β1

tanh ϕ1 + tanh ϕ2
tanh(ϕ1 + ϕ2 ) = (15.16.6)
1 + tanh ϕ1 tanh ϕ2

Thus I can represent the speed of an object by giving the value of ϕ , where
β = tanh ϕ (15.16.7)

or

−1
1 1 +β
ϕ = tanh β = ln( ) (15.16.8)
2 1 −β

The factor ϕ combines simply as

ϕ2
= ϕ1 + ϕ2 (15.16.9)
ϕ2

If you did what I suggested in Section 15.3 and programmed your calculator or computer to convert instantly from one
relativity factor to another, you now have a quick way of adding speeds.

 Example 15.16.2

A train trundles to the right at 90% of the speed of light (ϕ = 1.47222) relative to S, and a passenger strolls to the right at 15%
1

of the speed of light (ϕ = 0.15114) relative to Σ . The speed of the passenger relative to Σ is ϕ = 1.62336, or 92.5% of the
2
′

speed of light.

 Example 15.16.3

15.16.3 https://phys.libretexts.org/@go/page/8469
 (Sorry – there is no Figure XV.21.)
An ocean liner Σ sails serenely eastwards at a speed β = 0.9c (g = 2.29416) relative to the ocean Σ. A passenger ambles
′
1 1

athwartships at a speed β = 0.5c relative to the ship. What is the velocity of the passenger relative to the ocean?
2

The northerly component of her velocity is given by Equation 15.16.5b, and is 0.21794c . Her easterly component is just 0.9c .
Her velocity relative to the ocean is therefore 0.92601c in a direction 13o 37' north of east.

 Exercise 15.16.1

Show that, if the speed of the ocean liner is β and the athwartships speed of the passenger is β , the resultant speed β of the
1 2

passenger relative to the ocean is given by

2 2 2 2 2
β =β +β −β β (15.16.10)
1 2 1 2

and that her velocity makes and angle α with the velocity of the ship given by
−−−−−−
2
β
1
tan α = β2 √ 1 − . (15.16.11)
β1

 Example 15.16.4

A railway train Σ of proper length L = 100 yards thunders past a railway station Σ at such a speed that the stationmaster
′
0

thinks its length is only 40 yards. (Correction: It is not a matter of what he “thinks”. What I should have said is that the length
of the train, referred to a reference frame Σ in which the stationmaster is at rest, is 40 yards.) A dachshund waddles along the
corridor towards the front of the train. (A dachshund, or badger hound, is a cylindrical dog whose proper length is normally
several times its diameter.) The proper length l of the dachshund is 24 inches, but to a seated passenger, it appears to be... no,
0

sorry, I mean that its length, referred to the reference frame Σ , is 15 inches. What is the length of the dachshund referred to the
′

reference frame Σ in which the stationmaster is at rest?

We are told, in effect, that the speed of the train relative to the station is given by γ = 2.5, and that the speed of the dachshund
1

relative to the train is given by γ = 1.6. So how do these two gammas combine to make the factor γ for the dachshund relative
2

to the station?
There are several ways in which you could do this problem. One is to develop a general algebraic method of combining two
gamma factors. Thus:

 Exercise 15.16.2
Show that two gamma factors combine according to
−−−−−−−−−−−−−
2 2
γ1 ⊕ γ2 = γ1 γ2 + √ (γ − 1)(γ − 1) . (15.16.12)
1 2

I’ll leave you to try that. The other way is to take advantage of the programme you wrote when you read Section 15.3, by
which you can instantaneously convert one relativity factor to another. Thus you instantly convert the gammas to phis.
Thus γ 1 = 2.5  ⇒  ϕ1 = 1.56680

and γ1 = 1.6  ⇒  ϕ1 = 1.04697

15.16.4 https://phys.libretexts.org/@go/page/8469
 ∴ ϕ = 2.61377  ⇒  γ = 6.86182

Is this what Equation 15.16.12 gets?

24
Therefore, referred to the railway station, the length of the dachshund is = 3.5 inches.
γ

This page titled 15.16: Addition of Velocities is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.16.5 https://phys.libretexts.org/@go/page/8469
15.17: Aberration of Light
The direction of Earth’s velocity on any particular date is called the Apex of the Earth’s Way. In part (a) of Figure XV.23 I show
Earth moving towards the apex at speed ν , and light coming from a star at speed c from an angle χ from the apex.

The x- and y - components of the velocity of light are respectively −c cos χ and −c sin χ . Relative to Earth (part (b)), the x - and′

y - components are, by Equations 15.16.2 and 15.16.3 (or rather their inverses)
′

c cos χ + ν
−
ν
1 +( ) cos χ
c

and
c sin χ
− .
ν
γ(1 + ( ) cos χ)
c

You can verify that the orthogonal sum of these two components is c , as it should be according to our fundamental assumption that
the speed of light is the same referred to all reference frames in uniform relative motion.
The apparent direction of the star is therefore given by

′
sin χ
sin χ = − (15.17.1)
ν
γ(1 + ( ) cos χ)
c

It is left as an exercise to show that, for small ν

c
, this becomes

′
ν sin χ
χ−χ = . (15.17.2)
c

with ν = 29.8 km s-1, is about 20".5. More details about aberration of light, including the derivation of Equation 15.17.2, can be
ν

found in Celestial Mechanics, Section 11.3.

This page titled 15.17: Aberration of Light is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum
via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.17.1 https://phys.libretexts.org/@go/page/8470
15.18: Doppler Effect
It is well known that the formula for the Doppler effect in sound is different according to whether it is the source or the observer
that is in motion. An answer to the question “Why should this be?” to the effect that “Oh, that’s just the way the algebra works out”
is obviously unsatisfactory, so I shall try to explain why, physically, there is a difference. Then, when you have thoroughly
understood that observer in motion is an entirely different situation from source in motion, and the formulas must be different, we
shall look at the Doppler effect in light, and we’ll return to square one when we find that the formulas for source in motion and
observer in motion are the same!
This section on the Doppler effect will probably be rather longer than it need be, just because some aspects interested me – but if
you find it too long, just skip the parts that aren’t of special interest to you. These will quite likely include the parts on the ballistic
Doppler effect.
First, we’ll deal with the Doppler effect in sound. All speeds are supposed to be very small compared with the speed of light, so
that we need not trouble ourselves with Lorentz transformations. First, let’s deal with observer in motion (Figure XV. 24).
When the source is at rest, it emits concentric equally-spaced spherical wavefronts at some frequency. When an observer moves
towards the source, he will pass these wavefronts at a higher frequency that the frequency at which they were emitted, and that is
the cause of the Doppler effect with a stationary source and moving observer.

Now, we’ll look at the source-in-motion situation. (Figure XV.25).

Here we see that the wavefronts are not equally spaced, but are compressed ahead of the motion of the source, and for that reason
they will pass a stationary observer at a higher frequency than the frequency at which they were emitted. Thus the nature of the
effect is a little different according to whether it is the source or the observer that is in motion, and thus one would not expect
identical equations to describe the two situations.
We shall move on shortly to discuss the effect quantitatively and develop the relevant equations. I shall assume that the reader is
familiar with the usual relation connecting wavelength, frequency and speed of a wave. Nevertheless I shall write down the relation
in large print, three times, just to make sure:

SPEED = FREQUENCY % WAVELENGTH

FREQUENCY = SPEED ÷ WAVELENGTH

15.18.1 https://phys.libretexts.org/@go/page/8471
WAVELENGTH = SPEED ÷ FREQUENCY
I am going to start with the Doppler effect in sound, where the speed of the signal is constant with respect to the medium than
transmits the sound – usually air. I shall give the necessary formulas for source and observer each in motion. If you want the
formulas for one or the other stationary, you just put one of the speeds equal to zero. The speeds of the source S and of the observer
O relative to the air will be denoted respectively by ν and ν and the speed of sound in air will be denoted by c . The situation is
1 2

shown in Figure XV.26.

The relevant formulas are shown below:

The way we work this table is just to follow the arrows. Starting at the top left, we suppose that the source emits a signal of
(c−v1 )
frequency ν . The speed of the signal relative to the source is c − v , and so the wavelength is
0 1 . The wavelength is the same
ν0

for the observer (we are supposing that all speeds are very much less than the speed of light, so the Lorentz factor is effectively 1.)
The speed of sound relative to the observer is c − v , and so the frequency heard by the observer is the last (upper right) entry of
2

the table.
Two special cases:
a. Observer in motion and approaching a stationary source at speed v . v 1 =0 and v 2 = −v . In that case the frequency heard by the
observer is
v
ν = ν0 (1 + ). (15.18.1)
c

b. Source in motion and approaching a stationary observer at speed v . v 1 =v and v 2 =0 . In that case the frequency heard by the
observer is
ν0 v v
2
ν = ≈ ν0 (1 + ( )+( ) +. . . ). (15.18.2)
v
(1 − ) c c
c

We might now consider reflection. Thus, suppose you approach a brick wall at speed v while whistling a note of frequency ν . 0

What will be the frequency of the echo that you hear? Let’s make the question a little more general. A source S , emitting a whistle
of frequency ν , approaches a brick wall M at speed v . A separate observer O approaches the wall (from the same side) at speed
0 1

v . And, for good measure, let’s have the brick wall moving at speed v . (The reader may notice at this point that theoretical
2 3

physics is rather easier than experimental physics.) The situation is shown in Figure XV.27.

15.18.2 https://phys.libretexts.org/@go/page/8471
We construct a table similar to the previous one.

At all times, the speed relative to the air is c .

The answer to our initial question, in which the source and the observer were one and the same, and the mirror (wall) was
stationary is found by putting v = v = v and v = 0 in the last (top right) formula in the table. This results in
1 2 3

c +v v v 2
v 3
ν = ν0 ( ) ≈ ν0 (1 + 2( ) + 2( ) + 2( ) . . . ). (15.18.3)
c −v c c c

So much for the Doppler effect in sound. Before moving on to light, I want to look at what I shall call the Doppler effect in
ballistics, or “cops and robbers”. An impatient reader may safely skip this discussion of ballistic Doppler effect. A police (“cop”)
car is chasing a stolen car driven by robbers. The cop car is the “source” and the robber’s car (or, rather the car that they have
stolen, for it is not theirs) are the “observers”. The cop car (“source”) is travelling at speed v and the robbers (“observer”) is
1

travelling at speed v . The cops are firing bullets (the “signal”) towards the robbers. (No one gets hurt in this thought experiment,
2

which is all make-believe.) The bullets leave the muzzle of the revolver at speed c (that is the speed of the bullets, and is nothing to
do with light) relative to the revolver, and hence they travel (relative to the lamp-posts at the side of the road) at speed c + v and 1

relative to the robbers at speed c + v − v . The cops fire bullets at frequency ν , and our task is to find the frequency with which
1 2 0

the bullets are “received” by the robbers. The distance between the bullets is the “wavelength”.
This may not be a very important exercise, but it is not entirely pointless, for fairness dictates that, when we are considering (even
if only to discard) possible plausible mechanisms for the propagation of light, we might consider, at least briefly, the so-called
“ballistic” theory of light propagation, in which the speed of light through space is equal to the speed at which it leaves the source
plus the speed of the source. Some readers may be aware of the Michelson-Morley experiment. That experiment demonstrated that
light was not propagated at a speed that was constant with respect to some all-pervading “luminiferous aether” – but it must be
noted that it did nothing to prove or disprove the “ballistic” theory of light propagation, since it did not measure the speed of light
from moving sources. In the intervening years, some attempts have indeed been made to measure the speed of light from moving
sources, though their interpretation has not been free from ambiguity.
I now construct a table showing the “frequency”, “speed” and “wavelength” for ballistic propagation in exactly the same way as I
did for sound.

15.18.3 https://phys.libretexts.org/@go/page/8471
In order not to spend longer on “ballistic” propagation than is warranted by its importance, I’ll just let the reader spend as much or
as little time pondering over this table as he or she wishes. Just one small point might be noted, namely that the formulas for
“observer in motion” and “source in motion” are the same.
For completeness rather than for any important application, I shall also construct here the table for “reflection”. A source of bullets
is approaching a mirror at speed v . An observer is also approaching the mirror, from the same side, at speed v . And the mirror is
1 2

moving at speed v , and reflection is elastic (the coefficient of restitution is 1.) You are free to put as many of these speeds equal to
3

zero as you wish.

The entries for “speed” give the speed relative to the source or mirror or observer. The speed relative to stationary lampposts at the
side of the road is c + v before reflection and c + v − 2v after reflection.
1 1 3

We now move on to the only aspect of the Doppler effect that is really relevant to this chapter, namely the Doppler effect in light.
In the previous two situations I have been able to assume that all speeds were negligible compared with the speed of light, and we
have not had to concern ourselves with relativistic effects. Here, however, the signal is light and is propagated at the speed of light,
and this speed is the same whether referred to the reference frame in which the source is stationary or the observer is stationary.
Further, the Doppler effect is noticeable only if source or observer are moving at speeds comparable to that of light. We shall see
that the difference between the frequency of a signal relative to an observer and the frequency relative to the source is the result of
two effects, which, while they may be treated separately, are both operative and in that sense inseparable. These two effects are the
Doppler effect proper, which is a result of the changing distance between source and observer, and the relativistic dilation of time.
I am going to use the symbol T to denote the time interval between passage of consecutive crests of an electromagnetic wave. I’ll
call this the period. This is merely the reciprocal of the frequency ν . I am going to start by considering a situation in which a source
and an observer a receding from each other at a speed v . I have drawn this in Figure XV.27, which is referred to a frame in which
the observer is at rest. The speed of light is c .

Let us suppose that S emits an electromagnetic wave of period T 0 =

1

ν0
referred to the frame in which S is at rest. We are going to
have to think about four distinct periods or frequencies:
1. The time interval between the emission of consecutive crests by S referred to the reference frame in which S is at rest. This is
the period T and the frequency ν that we have just mentioned.
0 0

2. The time interval between the emission of consecutive crests by S referred to the reference frame in which O is at rest. By the
relativistic formula for the dilation of time this is
T0
γ T0 or . (15.18.4)
−−−−−
2
v
√1 − 2
c

3. The time interval between the reception of consecutive crests by O as a result of the increasing distance between O and S (the
“true” Doppler effect, as distinct from time dilation) referred to the reference frame in which S is at rest. This is

15.18.4 https://phys.libretexts.org/@go/page/8471
 v
T0 (1 + ). (15.18.5)
c

4. The time interval between the reception of consecutive crests by O as a result of the increasing distance between O and S (the
“true” Doppler effect, as distinct from time dilation) referred to the reference frame in which O is at rest. This is
v
γ times T0 (1 + ). (15.18.6)
c

This, of course, is what O “observes”, and, when you do the trivial algebra, you find that this is
−−−−−
v
1+
c
T = T0 √ , (15.18.7)
v
1−
c

or, in terms of frequency,

−−−−−
v
1−
c
ν = ν0 √ . (15.18.8)
v
1+
c

If source and observer approach each other at speed v , the result is

−−−−−
v
1+
c
ν = ν0 √ . (15.18.9)
v
1−
c

−−−−
v
1+
The factor √ c

v
is often denoted by the symbol k , and indeed that was the symbol I used in Section 15.3 (see Equation 15.3.3).
1−
c

 Exercise 15.18.1

Expand Equation 15.18.9 by the binomial theorem as far as ( v

c
2
) and compare the result with Equations 15.18.1 and 15.18.2.
I make it
v 1 v 2
ν = ν0 (1 + ( )+ ( ) . . . ). (15.18.10)
c 2 c

 Exercise 15.18.2

An observer O sends an electromagnetic signal of frequency ν at speed c to a mirror that is receding at speed v . When the
0

reflected signal arrives back at the observer, what is its frequency (to first order in )? Is it ν (1 − ) or is it ν (1 − ) ?
v

c
0
v

c
0
2v

I can think offhand of two applications of this. If you examine the solar Fraunhofer spectrum reflected of the equatorial limb of
a rotating planet, and you observe the fractional change in the frequency of a spectrum line, will this tell you or ,
Δν

ν0
v

c
2v

where v is the equatorial speed of the planet’s surface? And if a policeman directs a radar beam at your car, does the frequency
of the returning beam tell him the speed of your car, or twice its speed? You could try arguing this case in court – or, better,
stick to the speed limit so there is no need to do so. The answer, by the way, is ν (1 − ) . 0
2v

Redshift. When a galaxy is moving away from us, a spectrum line of laboratory wavelength λ will appear to have a frequency 0

λ−λ0
for the observer of λ = kλ . The fractional increase in wavelength
0 is generally given the symbol z , which is evidently
λ0

λ−λ0
equal to k − 1 . (Only to first order in β is it approximately equal to β. It is important to note that the definition of z is λ0
,
and not v

c
.
A note on terminology: If a source is receding from the observer the light is observed to be shifted towards longer wavelengths,
and if it is approaching the observer the light is shifted towards shorter wavelengths. Traditionally a shift to longer wavelengths

15.18.5 https://phys.libretexts.org/@go/page/8471
 is called a “redshift”, and a shift towards shorter wavelengths is called a “blueshift”. Note, however, that if an infrared source is
approaching an observer, its light is shifted towards the red, and if an ultraviolet source is receding from an observer, its light is
shifted towards the blue! Nevertheless, I shall continue in this chapter to refer to shifts to longer and shorter wavelengths as
redshifts and blueshifts respectively.

 Example 15.18.1

A red galaxy R of wavelength 680.0 nm and a green galaxy G of wavelength 520.0 nm are on opposite sides of an observer X,
both receding from him/her. To the observer, the wavelength of the red galaxy appears to be 820.0 nm, and the wavelength of
the green galaxy appears to be 640.0 nm. What is the wavelength of the green galaxy as seen from the red galaxy?

Solution
We are told that k for the red galaxy is 82/68 = 1.20588, or z = 0.20588, and that k for the green galaxy k is 64/52 = 1.23077,
or z = 0.23077. Because of the preparation we did in Section 15.3, we can instantly convert these to ϕ . Thus for the red galaxy
ϕ = 0.187212 and for the green galaxy ϕ = 0.207639. The sum of these is 0.394851. We can instantly convert this to

k = 1.48416 or z = 0.48416.. Thus, as seen from R, the wavelength of G is 771.8 nm.

Alternatively.
Show that the factor k combines as
k1 ⊕ k2 = k1 k2 (15.18.11)

and verify that 82

68
×
64

52
= 1.48416 . Show also that the redshift factor z combines as

z1 ⊕ z2 = z1 z2 + z1 + z2 . (15.18.12)

This page titled 15.18: Doppler Effect is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.18.6 https://phys.libretexts.org/@go/page/8471
15.19: The Transverse and Oblique Doppler Effects
I pointed out in Section 15.18 that the observed Doppler effect, when the transmitted signal is electromagnetic radiation and
observer or source or both are travelling at speed comparable to that of light, is a combination of two effects – the “true” Doppler
effect, caused by the changing distance between source and observer, and the effect of time dilation. This raises the following
questions.

Transverse Doppler effect

If a source of light is moving at right angles to (transverse to) the line joining observer to source, will the observer see a change in
frequency or wavelength, even though the distance between observer and source at that instant is not changing? The answer is yes,
certainly, and the effect is sometimes called the “transverse Doppler effect”, although it is the effect of relativistic time dilation
rather than of a true Doppler effect.
Thus let us suppose that a source is moving transverse to the line of sight at a speed described by its parameter β or γ, and that the
period of the radiation referred to the reference frame in which the source is at rest is T and the frequency is ν . The time interval
0 0

between emission of consecutive wavecrests when referred to the frame in which the observer is at rest is longer by the gamma-
factor, and the frequency is correspondingly less. That is, the frequency, referred to the observer’s reference frame, is
ν0 −−−−−
2
ν = = ν0 √ 1 − β (15.19.1)
γ

The light from the source is therefore seen by the observer to be redshifted, even though there is no radial velocity component.

Oblique Doppler Effect

This raises a further question. Suppose a source is moving almost but not quite at right angles to the line of sight, so that it has a
large transverse velocity component, and a small velocity component towards the observer. In that case, its “redshift” resulting
from the time dilation might be appreciable, while its “blueshift” resulting from “true” Doppler effect (the decreasing distance
between source and observer) is still very small. Therefore, even though the distance between source and observer is slightly
decreasing, there is a net redshift of the spectrum. This is in fact correct, and is the “oblique Doppler effect”.
In Figure XV.30, a source S is moving at speed β times the speed of light in a direction that makes an angle θ with the line of sight.
It is emitting a signal of frequency ν in S. (I am here using the frame “in S” as earlier in the chapter to mean “referred to a
0

reference frame in which S is at rest.) The signal arrives at the observer O at a slightly greater frequency as a result of the
decreasing distance of S from O, and at a slightly lesser frequency as a result of the time dilation, the two effects opposing each
other.

The frequency of the received signal at O, in O, is

−−−− −
ν0 √ 1 − β 2
ν = . (15.19.2)
1 − β cos θ

For a given angle θ the redshift is zero for a speed of

15.19.1 https://phys.libretexts.org/@go/page/8472
 2 cos θ
β = (15.19.3)
2
1 + cos θ

or, for a given speed, the direction of motion resulting in a zero redshift is given by
2
β cos θ − 2 cos θ + β = 0. (15.19.4)

This relation is shown in Figure XV.30. (Although Equation 15.19.4 is quadratic in cos θ there is only one real solution θ for β
between 0 and 1. Prove this assertion.) It might be noted that if the speed of the source is 99.99% of the speed of light the observer
will see a redshift unless the direction of motion of S is no further than 9o 36' from the line from S to O. That is worth repeating: S
is moving very close to the speed of light in a direction that is close to being directly towards the observer; the observer will see a
redshift.
Equation 15.19.2, which gives ν as a function of θ for a given β, will readily be recognized at the equation of an ellipse of
eccentricity β, semi minor axis ν and semi major axis γν . This relation is shown in Figure XV.31 for several β. The curves are
0 0

red where there is a redshift and blue where there is a blueshift. There is no redshift or blueshift for β = 0 , and the ellipse for that
case is a circle and is drawn in black.

An alternative and perhaps more useful way of looking at Equation 15.19.2 is to regard it as an equation that gives β as a function
of θ for a given Doppler ratio . For example, if the Doppler ratio of a galaxy is observed to be 0.75, the velocity vector of the
ν

ν0

galaxy could be any arrow starting at the black dot and ending on the curve marked 0.75. The curves are ellipses with semi major
axis equal to 1

ν
and semi minor axis
2
1
.ν 2

√1−( ) (1−( ) )
ν ν
0 0

15.19.2 https://phys.libretexts.org/@go/page/8472
This page titled 15.19: The Transverse and Oblique Doppler Effects is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or
curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is
available upon request.

15.19.3 https://phys.libretexts.org/@go/page/8472
15.20: Acceleration
Figure XV.33 shows two references frames, Σ and Σ , the latter moving at speed v with respect to the former.
′

A particle is moving with acceleration a in Σ . (“ in Σ ” = “referred to the reference frame Σ ”.) The velocity is not necessarily,
′ ′ ′ ′

of course, in the same direction as the acceleration, and we’ll suppose that its velocity in Σ is u . The acceleration and velocity ′ ′

components in Σ are a , a , u , u .
′ ′
x′
′
y′
′
x′
′
y′

What is the acceleration of the particle in Σ? We shall start with the x-component.
The x-component of its acceleration in Σ is given by
dux
ax = , (15.20.1)
dt

where
′
ux + v
ux = ′
(15.20.2)
ux v
1+ 2
c

and
′
′
vx
t = γ (t + ) (15.5.19)
c2

Equations 15.16.2 and 15.5.19 give us

′
dux dux
′
dux = dux = ′
(15.20.3)
′ ux v
dux 2 2
γ (1 + 2
)
c

and
∂t ∂t γv
′ ′ ′ ′
dt  = dt + dx = γdt   +   dx (15.20.4)
′ ′
∂t ∂x c2

On substitution of these into Equation 15.20.1 and a very little algebra, we obtain
′
a
ax = (15.20.5)
′
ux v
3 3
γ (1 + )
c2

The y -component of its acceleration in Σ is given by

duy
ay = , (15.20.6)
dt

We have already worked out the denominator dt (Equation 15.20.4). We know that
′
u ′
y
uy = ′
(15.16.3)
u ′
v
x
γ(1 + 2
)
c

from which

15.20.1 https://phys.libretexts.org/@go/page/8473
 ⎛ ′ ⎞
∂uy ∂uy vu
1 y′ 1
duy = +
′
∂u = ⎜− du
′
+
′
du ⎟. (15.20.7)
′ ′ y
′
⎜ 2 x
′
vu
′ y
′
⎟
∂u ′ ∂u ′ γ vu
′
′ x
′
x y
⎝ c 2
(1 +
x
) 1+ 2 ⎠
c2 c

Divide Equation 15.20.7 by Equation 15.20.4 to obtain

⎛ ′ ⎞
vu
1 y′ 1
⎜− ′ ′ ⎟.
ay = a + a (15.20.8)
⎜ x
′
′ y
′
⎟
γ2 vu
′
′
2 vu
x
′
2 x
1+
⎝ c (1 + ) 2 ⎠
c2 c

This page titled 15.20: Acceleration is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.20.2 https://phys.libretexts.org/@go/page/8473
15.21: Mass
It is well known that “in relativity” the mass of an object increases as its speed increases. This may be well known, but I am not
certain that it is a very precise statement of the true situation. Or at least it is no more precise than to say that the length of a rod
decreases as its speed increases. The length of a rod when referred to a frame in which it is at rest is called its proper length l , and
0

the mass of a body when referred to a frame in which it is at rest is called its rest mass m , and both of these things are invariant.
0

The length of a rod when referred to a reference frame that is moving with respect to it (i.e., in Minkowski language, its component
along an inclined axis) and the mass of a body referred to a frame that is moving with respect to it may indeed be different from the
proper length of the rod or the rest mass of the body.
In order to derive the FitzGerald-Lorentz contraction, we had to think about what we mean by “length” and how to measure it.
Likewise, in order to derive the “relativistic increase of mass” (which may be a misnomer) we have to think about what we mean
by mass and how to measure it.
The fundamental unit of mass used at present in science is the International Prototype Kilogram, a platinum-iridium alloy, held in
Sèvres, Paris, France. In order to determine the mass, or inertia, of another body, we need to carry out an experiment to compare its
reluctance to accelerate when a force is applied to it with the reluctance of the standard kilogram when the same force is applied.
We might, for example, attach the body to a spring, stretch the spring, let go, and see how fast the body accelerates. Then we carry
out the same experiment with the International Prototype Kilogram. Or we might apply an impulse (∫ I dt - see Chapter 8) to the
body and to the Kilogram, and measure the speed immediately after applying the impulse. This might be done, for example, but
striking the body and the Kilogram with a golf club, or, for a more controlled experiment, one could press each body up against a
compressed spring, release the spring, and measure the resulting speed imparted to the body and to the Kilogram. (It is probable
that the International Prototype Kilogram is kept under some sort of guard, and its curators may not altogether appreciate such
experiments, so perhaps these experiments had better remain Thought Experiments.) Yet another method would be to cause the
body and the Kilogram to collide with each other, and to assume that the collision is elastic (no internal degrees of freedom) and
that momentum (defined as the product of mass and velocity) are conserved.
All of these experiments measure the reluctance to accelerate under a force; in other words the inertia or the inertial mass or just
the mass of the body.
Another possible experiment to determine the mass of the body would be to place it and the Kilogram at a measured distance from
another mass (such as the Earth) and measure the gravitational force (weight) of each. One has an uneasy feeling that this sort of
measurement is somehow a little different from the others, in that it isn’t a measure of inertia. Some indeed would differentiate
between the inertial mass and the gravitational mass of a body, although the two are in fact observed to be strictly proportional to
one another. Some would not find the proportionality between inertial and gravitational mass particularly remarkable; to others, the
proportionality is a surprising fact of the profoundest significance.
In this chapter we do not deal with general relativity or with gravity, so we shall think of mass in terms of its inertia. I am going to
measure the ratio of two masses (one of which might be the International Prototype Kilogram) by allowing them to collide, and
their masses are to be defined by assuming that the momentum of the system is conserved in all uniformly moving reference
frames.
Figure XV.34 shows two references frames, Σ and Σ , the latter moving to the right at speed v relative to the former. Two bodies,
′

of identical masses in Σ (i.e. referred to the frame Σ ), are moving at speed u in Σ , one of them to the right, the other to the left.
′ ′ ′ ′

Their mutual centre of mass is stationary in Σ .

′

Now let us refer the situation to the frame Σ (see figure XV.35).

15.21.1 https://phys.libretexts.org/@go/page/8474
The total momentum of the system in Σ is m u + m u . But the centre of mass (which is stationary in Σ ) is moving to the right
1 1 2 2
′

in Σ with speed v . Therefore the momentum is also (m + m )v . If they stick together upon collision, we are left with a single
1 2

particle of mass m + m moving at speed v , and, because there are no external forces, the momentum is conserved. In any case,
1 2

whether the collision is elastic or not, we have

m1 u1 + m2 u2 = (m1 + m2 ). (15.21.1)

But
′
u +v
u1 = (15.21.2a)
u′ v
1+ 2
c

and
′
−u + v
u2 = ′
(15.21.2b)
u v
1− 2
c

Our aim is to try to find a relation between the masses and speeds referred to Σ. Therefore we must eliminate v and u from ′

Equations 15.21.1, 15.21.2a and 15.21.2b. This can be a bit fiddly, but the algebra is straightforward, and I leave it to the reader to
show that the result is
−−−−−−
 u
2

 1− 2
m1  c2
= (15.21.2)
2
m2 u
⎷ 1− 1

2
c

This tells us that the mass m of a body referred to Σ is proportional to 1

2
, where u is its speed referred to Σ . If we call the
u
√1−
2
c

proportionality constant m , then

0

m0
m = (15.21.3)
−−−−−
2
u
√1 − 2
c

If u = 0 , then m = m , and m is called the rest mass, and it is the mass when referred to a frame in which the body is at rest. The
0 0

mass m is generally called the relativistic mass, and it is the mass when referred to a frame in which the speed of the body is u.
Equation 15.21.3 gives the mass referred to Σ assuming that the mass is at rest in Σ . But what if the mass is not at rest in Σ ?
′ ′

15.21.2 https://phys.libretexts.org/@go/page/8474
In figure XV.36 we see a mass m moving with velocity u in Σ . Referred to Σ its mass will be m, where
′ ′ ′

−−−−− −
 u
′2

m  1− 2
c
= . (15.21.4)
′ 2
m ⎷ 1− u
c2

Its velocity u will be in a different direction (referred to Σ) from the direction of u in Σ , and the speed will be given by
′ ′

2 2 2
u = ux + uy (15.21.5)

where ux and u are given by Equations 15.16.2 and 15.16.3. Substitute Equations 15.21.5, 15.16.2 and 15.16.3 into Equation
y

15.21.4 . The objective is to replace u entirely by primed quantities. The algebra is slightly boring, but it is worth persisting. You
will find that u appears when you use Equation 15.16.3. Replace that by u − u . Also write
′2
y
′
′2
for γ . After a little while
′2
x
′
1

v2
2

(1− )
2
c

you should arrive at

′
m vu
x′
= γ (1 + ). (15.21.6)
′ 2
m c

The transformation for mass between the two frames depends only on the x component of its velocity in Σ . It would have made
′ ′

no difference, other than to increase the tedium of the algebra, if I had added +u to the right hand side of Equation 15.21.5.
2
z

The inverse of Equation 15.21.6 is found in the usual way by interchanging the primed and unprimed quantities and changing the
sign of v :
′
m vux
= γ (1 − ). (15.21.7)
2
m c

Example.

 Example 15.21.1

Let’s return to the problem of the dachshund that we met in Section 15.16. A railway train Σ is trundling along at a speed
′ v

c
=
′
u
0.9 (γ = 2.294). The dachshund is waddling towards the front of the train at a speed . In the reference frame of the train Σ
′
x ′
c

the mass of the dog is m = 8 kg. In the reference frame of the railway station, the mass of the dog is given by Equation
′

15.21.6 and is 31.6 kg. (Its length is also much compressed, so it is very dense when referred to Σ and is disc-shaped.)

I leave it to the reader to show that the rest mass of the dog is 4.8 kg.

This page titled 15.21: Mass is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.21.3 https://phys.libretexts.org/@go/page/8474
15.22: Momentum
The linear momentum p of a body, referred to a frame Σ, is defined as
p = mu. (15.22.1)

Here m and u are its mass and velocity referred to Σ. Note that m is not the rest mass.

 Example 15.22.1
The rest mass of a proton is 1.67 % 10-27 kg. What is its momentum referred to a frame in which it is moving at 99% of the
speed of light?
Answer = 3.51 % 10-18 kg m s-1.

This page titled 15.22: Momentum is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.22.1 https://phys.libretexts.org/@go/page/8475
15.23: Some Mathematical Results
Before proceeding with the next section, I just want to establish few mathematical results, so that we do not get bogged down in
heavy algebra later on when we should be concentrating on understanding physics.
First, if
2
u
γ = (1 − ), (15.23.1)
2
c

Then, by trivial differentiation,

3
dγ γ u
= . (15.23.2)
2
du c

3
γ u u̇
γ̇ = . (15.23.3)
2
c

From this, we quickly find that

γuu̇ 2 2
=c −u . (15.23.4)
γ̇

Now for a small result concerning a scalar (dot) product.

Let A be a vector such that A * A = A . 2

Then
d

dt
2 ˙
(A ) = 2AA and d

dt
˙
(A ⋅ A) = 2A ⋅ A

˙ ˙
A ⋅ A = AA (15.23.5)

We can now safely proceed to the next section.

This page titled 15.23: Some Mathematical Results is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.23.1 https://phys.libretexts.org/@go/page/8476
15.24: Kinetic Energy
If a force F acts on a particle moving with velocity u , the rate of doing work – i.e. the rate of increase of kinetic energy T is
˙
T =F⋅u . But F = ṗ where p = mu = γ m u . 0

(A point about notation may be in order here. I have been using the symbol v and v for the velocity and speed of a frame Σ ′

relative to a frame Σ, and my choice of axes without significant loss of generality has been such that v has been directed parallel to
the x-axis. I have been using the symbol u for the velocity (speed = u) of a particle relative to the frame Σ. Usually the symbol γ
1 1

2
− 2
−

has meant (1 − , but here I am using it to mean (1 − . I hope that this does not cause too much confusion and that
2 2
v u
) )
c2 c2

the context will make it clear. I toyed with the idea of using a different symbol, but I thought that this might make matters worse.
Just be on your guard, anyway.)
We have, then

F = m0 ( γ̇ u + γ u̇) (15.24.1)

and therefore
˙ = m (γ̇u2 + γ u̇ ⋅ u).
T (15.24.2)
0

Making use of Equations 15.23.5 and 15.23.6 we obtain

˙ = γ̇m c2
T (15.24.3)
0

Integrate with respect to time, with the condition that when γ = 1, T = 0, and we obtain the following expression for the kinetic
energy:
2
T = (γ − 1)m0 c . (15.24.4)

Exercise. Expand γ by the binomial theorem as far as u

2
c
, and show that, to this order, T =
1

2
2
mu .
I here introduce the dimensionless symbol
T
K = = γ −1 (15.24.5)
2
m0 c

to mean the kinetic energy in units of m 0c

2
. The second half of this was already given as Equation 15.3.5.

This page titled 15.24: Kinetic Energy is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.24.1 https://phys.libretexts.org/@go/page/8477
15.25: Addition of Kinetic Energies
I want now to consider two particles moving at nonrelativistic speeds – by which I mean that the kinetic energy is given to a
sufficient approximation by the expression mu and so that parallel velocities add linearly.
1

2
2

Consider the particles in figure XV.37, in which the velocities are shown relative to laboratory space.

Referred to laboratory space, the kinetic energy is 1

2
2
m1 u
1
+
1

2
m2 u
2
2
. However, the centre of mass is moving to the right with
( m1 u1 +m2 u2 )
speed V =
( m1 +m2 )
, and, referred to centre of mass space, the kinetic energy is 1

2
m1 (u1 − V )
2
+
1

2
m(u2 + V )
2
. On the
other hand, if we refer the situation to a frame in which m is at rest, the kinetic energy is
1
1

2
m2 (u1 + u2 )
2
, and, if we refer the
situation to a frame in which m is at rest, the kinetic energy is m (u + u ) .
2
1

2
1 1 2
2

All we are saying is that the kinetic energy depends on the frame to which speed are referred – and this is not something that crops
up only for relativistic speeds.
Let us put some numbers in. Let us suppose, for example that
m1 = 3 kg u 1 =4 m s-1
m2 = 2 kg u 3 =4 m s-1
so that
V = 1.2 m s-1.
In that case, the kinetic energy
referred to laboratory space is 33 J,
referred to centre of mass space is 29.4 J,
referred to m is 49 J,
1

referred to m is 73.5 J.
2

In this case the kinetic energy is least when referred to centre of mass space, and is greatest when referred to the lesser mass.
Exercise. Is this always so, whatever the values of m1, m2 , u1 and u2?
It may be worthwhile to look at the special case in which the two masses are equal (m) and the two speeds(whether in laboratory or
centre of mass space) are equal (u).
In that case the kinetic energy in laboratory or centre of mass space is mu2, while referred to either of the masses it is 2mu2.
We shall now look at the same problem for particles travelling at relativistic speeds, and we shall see that the kinetic energy
referred to a frame in which one of the particles is at rest is very much greater than (not merely twice) the energy referred to a
centre of mass frame.
If two particles are moving towards each other with “speeds” given by g1 and g2 in centre of mass space, the g of one relative to the
other is given by equation 15.16.14, and, since K = g - 1, it follows that if the two particles have kinetic energies K1 and K2 in
centre of mass space (in units of the m0c2 of each), then the kinetic energy of one relative to the other is
−−−−−−−−−−−−−−−−−−
K = K1 ⊕ K2 = K1 + K2 + K1 K2 + √ K1 K2 (K1 + 2)(K2 + 2) . (15.25.1)

If two identical particles, each of kinetic energy K times m 1 0c

2
, approach each other, the kinetic energy of one relative to the other
is
K = 2 K1 (K1 + 2). (15.25.2)

For nonrelativistic speeds as K 1 → 0 , this tends to K = 4K , as expected.

1

15.25.1 https://phys.libretexts.org/@go/page/8478
Let us suppose that two protons are approaching each other at 99% of the speed of light in centre of mass space (K = 6.08881). 1

Referred to a frame in which one proton is at rest, the kinetic energy of the other will be K = 98.5025, the relative speeds being
0.99995 times the speed of light. Thus K = 16K rather than merely 4K as in the nonrelativistic calculation. For more energetic
1 1

particles, the ratio K

K1
is even more. These calculations are greatly facilitated if you wrote, as suggested in Section 15.3, a program
that instantly connects all the relativity factors given there.

 Exercise 15.25.1

Two protons approach each other, each having a kinetic energy of 500 GeV in laboratory or centre of mass space. (Since the
two rest masses are equal, these TWO spaces are identical.) What is the kinetic energy of one proton in a frame in which the
other is at rest?
(Answer: I make it 535 TeV.)

The factor K (the kinetic energy in units of m c ) is the last of several factors used in this chapter to describe the speed at which a
0
2

particle is moving, and I take the opportunity here of summarising the formulas that have been derived in the chapter for combining
these several measures of speed. These are
β1 + β2
β1 ⊕ β2 = . (15.16.7)
1 + β1 β2

−−−−−−−−−−−−−
2 2
γ1 ⊕ γ2 = γ1 γ2 + √ (γ − 1)(γ − 1) . (15.16.14)
1 2

k1 ⊕ k2 = k1 k2 (15.18.11)

z1 ⊕ z2 = z1 z2 + z1 + z2 . (15.18.12)

−−−−−−−−−−−−−−−−− −
K = K1 ⊕ K2 = K1 + K2 + K1 K2 + √K1 K2 (K1 + 2)(K2 + 2) .
ϕ1
= ϕ1 + ϕ2 (15.16.11)
ϕ2

If the two speeds to be combined are equal, these become

2β1
β1 ⊕ β1 = . (15.25.3)
2
1 +β
1

2
γ1 ⊕ γ1 = 2 γ −1 (15.25.4)
1

k1
2
=k (15.25.5)
1
k1

z1 ⊕ z1 = z1 (z1 + 2) (15.25.6)

K1 ⊕ K1 = 2 K1 (K1 + 2). (15.25.7)

ϕ1 ⊕ ϕ1 = 2ϕ. (15.25.8)

These formulas are useful, but for numerical examples, if you already have a program for interconverting between all of these
factors, the easiest and quickest way of combinng two “speeds” is to convert them to ϕ . We have seen examples of how this works
in Sections 15.16 and 15.18. We can do the same thing with our example from the present section when combining two kinetic
energies. Thus we were combining two kinetic energies in laboratory space, each of magnitude K = 6.08881 (ϕ = 2.64665).
1 1

From this, ϕ = 5.29330, which corresponds to K = 98.5025.

This page titled 15.25: Addition of Kinetic Energies is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

15.25.2 https://phys.libretexts.org/@go/page/8478
15.26: Energy and Mass
The nonrelativistic expression for kinetic energy T = mu has just one term in it, a term which depends on the speed. The
1

2
2

relativistic expression which approximates to the nonrelativistic expression at low speeds) can be written T = mc − m c that is,
2
0
2

a speed-dependent term minus a constant term. The kinetic energy can be thought of as the excess over the energy over the constant
term m c . The expression m c is known as the rest-mass energy. The sum of the kinetic energy and the rest-mass energy is the
0
2
0
2

“total energy”, or just the “energy” E :

2 2
E = T + m0 c = mc (15.26.1)

This means that, if the kinetic energy of a particle is zero, the total energy of the particle is not zero – it still has its rest-mass energy
m c .
2
0

Of course, giving the name “rest-mass energy” to the constant term m c , and calling the speed-dependent term mc the “total
0
2 2

energy” and writing the famous equation E = mc , does not by itself immediately and directly tell us that “matter” can be
2

converted to “energy” or the other way round. Whether such conversion can in fact take place is a matter for experiment and
observation to determine. The equation by itself merely tells us how much mass is held by a given quantity of energy, or how much
energy is held by a given quantity of mass. That entities that we traditionally think of as “matter” can be converted into entities that
we traditionally think of as “energy” is well established with, for example, the “annihilation” of an electron and a positron
(“matter” and “antimatter”) to form photons (“energy”) as is the inverse process of pair production (production of an electron-
positron pair from a gamma ray in the presence of a third body).
It is unfortunate that the main (almost the only) example of application of the equation E = mc persistently presented to the
2

nonscientific public is the atom bomb, whose operation actually has nothing at all to do with the equation E = mc , nor, contrary
2

to the popular mind, is any “matter” converted to energy.

I have heard it said that you can find out on the Web how to build an atom bomb, so here goes – here is how an atom bomb works.
A uranium-235 nucleus is held together by strong attractive forces between the nucleons, which, at short femtometre ranges are
much stronger than the Coulomb repulsive forces between the protons. When the nucleus absorbs an additional neutron, the
resulting 236U nucleus is unstable and breaks up into two intermediate-mass nuclei plus two or three neutrons. The two
intermediate-mass nuclei are generally not of exactly equal mass; one is usually a bit less than half of the uranium nucleus and the
other a bit more than half, but that’s a detail. The potential energy required to bind the nucleons together in the uranium nucleus is
rather greater than the binding energy of the two resulting intermediate-mass nuclei; the difference is of order 200 MeV, and that
potential energy is converted into kinetic energy of the two resulting nuclei and, to a lesser extent, the two or three neutrons
released. That is all. It is merely the familiar conversion of potential binding energy (admittedly a great deal of energy) into kinetic
energy. No matter, no protons, no neutrons, are “destroyed” or “converted into energy”, and E = mc simply does not enter into it
2

anywhere! The rest-mass energy of a proton or a neutron is about 1 GeV, and that much energy would be released if a proton were
miraculously and for no cause converted into energy. Let us hope that no one invents a bomb that will do that – though we may rest
assured that that is rather unlikely.
Where the equation E = mc does come in is in the familiar observation that the mass of any nucleus other than hydrogen is a
2

little less than the sum of the masses of the constituent nucleons. It is for that reason that nuclear masses, even for pure isotopes, are
not integral. The mass of a nucleus is equal to the sum of the masses of the constituent nuclei plus the mass of the binding energy,
the latter being a negative quantity since the inter-nucleon forces are attractive forces. The equation E = mc tells us that energy
2

(such as, for example, the binding energy between nucleons) has mass.

This page titled 15.26: Energy and Mass is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.26.1 https://phys.libretexts.org/@go/page/8479
15.27: Energy and Momentum
A moving particle has energy arising from its momentum and also from its rest mass, and we need to find an expression relating
energy to rest mass and momentum. It is fairly easy and it goes like this:
2 2 4 2 2 2 2 2 2 2 2 2 2 2 2
E =m c = c (m c −m u + m u ) = c [ m (c −u )+p ]

2 2 2
m ( c −u )
2 0 2 2 2 2 2
=c ( + p ) = c (m c +p )
u
2 0
1−
2
c

Thus we obtain for the energy in terms of rest mass and momentum
2 2 2 2
E = (m0 c ) + (pc ) . (15.27.1)

If the speed (and hence momentum) is zero, the energy is merely m c . If the rest mass is zero (as, for example, a photon) and the 0
2

energy is not zero, then E  =  pc  =  muc . But also E = mc , so that, if the rest mass of a particle is zero and the energy is not,
2

the particle must be moving at the speed of light. This could be regarded as the reason why photons, which have zero rest mass,
travel at the speed of photons. If neutrinos have zero rest mass, they, too, will travel at the speed of light; if they are not massless,
they won’t.
In addition to Equation 15.27.1, which relates the energy to the magnitude of the momentum, it will be of interest to see how the
components of momentum transform between reference frames. As usual, we are considering frame Σ to be moving with respect ′

to Σ at a speed v with respect to Σ. There is no difficulty with the y - and z - components. We have merely p = p and p = p . ′
y
′ y
′
z
′ z

However:
′
m0 u
m0 ux
and p .
′
′ ′ ′ x
px = m ux = ′ =m u ′ =
1 x x 1
2 2
ux 2 u 2
′
x
(1− )
2 (1− )
c 2
c

1
1
2
1 u 2 2
x v 2
′2 (1− ) (1− )
u 2 2 2
ux −v
Also u , from which (1 − .
′ c c
′ x
= ) =
x′ u
2
c
2 ux v
x 1−
(1− ) c
2
2
c

After a little algebra, we obtain

m0 ( ux −v)
px = 1 1
.
2 2
u 2 v 2
(1− ) (1− )
2 2
c c

And this is
vE
px −
c
2 vE
′
p ′ = = γ ( px − ) (15.27.2)
x 1
c2
v2 2

(1 − 2
)
c

The inverse is found in the usual way:

′
vE
′
px = γ ( p ′
+ ) (15.27.3)
x 2
c

If we eliminate p from Equations 15.27.2 and 15.27.3, we’ll find E in terms of E and p :
′
x
′
′
x

′
E = γ(E − vpx ). (15.27.4)

Thus the transformations between energy and the three spatial components of momentum is similar to the transformation between
time and the three space coordinates, and are described by a similar 4-vector:
′
p px
⎛ x
′
⎞ γ 0 0 iβγ
⎛ ⎞⎛ ⎞
′
⎜ p ′ ⎟ py
⎜ y
⎟ ⎜  0 1 0 0 ⎟⎜ ⎟
⎜ ⎟  =  ⎜ ⎟⎜ ⎟. (15.27.5)
⎜ p′ ⎟ ⎜ ⎟
⎜ 0 0 1 0 ⎟ ⎜ pz ⎟
⎜ z′ ⎟
iE
′ ⎝ ⎠⎝ iE ⎠
⎝ ⎠ −iβγ 0 0 γ
c
c

15.27.1 https://phys.libretexts.org/@go/page/8480
The reader should multiply this out to verify that it does reproduce Equations 15.27.3 and 15.27.4.

This page titled 15.27: Energy and Momentum is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.27.2 https://phys.libretexts.org/@go/page/8480
15.28: Units
It is customary in the field of particle physics to express energy (whether total, kinetic or rest-mass energy) in electron volts (eV) or
in keV, MeV, GeV or TeV (103, 106, 109, or 1012 eV respectively). A electron volt is the kinetic energy gained by an electron if it is
accelerated through an electrical potential of 1 volt; alternatively it is the work required to move an electron through one volt.
Either way, since the charge on an electron is 1.602 % 10-19) C, 1eV = 1.602 % 10-19 J.
The use of such a unit may understandably dismay those who would insist always on expressing any physical quantity in SI units,
and I am much in sympathy with this view. Yet, to those who deal daily with particles whose charge is equal to or is a small
multiple or rational fraction of the electronic charge, the eV has its attractions. Thus if you accelerate a particle through so many
volts, you do not have to remember the exact value of the electronic charge or carry out a long multiplication every time you do so.
One might also think of a hypothetical question such as: An electron is accelerated through 3426.7189628471 volts. What is its
gain in kinetic energy? You cannot answer this in joules unless you know the value of the electronic charge to a comparable
precision; but of course you do know the answer in eV.
One situation that does require care is this. An a-particle is accelerated through 1000 V. What is the gain in kinetic energy? Because
the charge on an a-particle is twice that of an electron, the answer is 2000 eV.
Very often you know the energy of a particle (because you have accelerated it through so many volts) and you want to know its
momentum; or you know its momentum (because you have measured the curvature of its path in a magnetic field) and you want to
know its energy. Thus you will frequent occasion to make use of Equation 15.27.1:
E
2
= (m0 c )
2 2
+ (pc )
2
.
You have to be careful to remember how many c s there are, and what is the exact value of c . Particle physicists prefer to make life
easier for themselves (not necessarily for the rest of us!) by preferring not to state what the momentum of a particle is, or its rest
mass, but rather to give the values of pc or of m c – and to express E ,pc and m c all in eV (or keV, MeV or GeV). Thus one
0
2
0
2

may hear that

pc = 6.2 GeV
m0 c
2
= 0.938 GeV.
More often this is expressed, somewhat idiosyncratically and in somewhat doubtful use of English, as
p = 6.2 GeV/c
m0 = 0.938 GeV/c2
or in informal casual conversation (one hopes not for publication) merely as
p = 6.2 GeV
m0 = 0.938 GeV.
While this may puzzle some and raise the ire of others, it is not entirely without merit, because, provided one uses these units, the
relation between energy, momentum and rest mass is then simply
E
2
=m
2
0
+p
2
.
The practice is not confined to energy, momentum and rest mass. For example, the SI unit of magnetic dipole moment is N m T-1
(newton metre per tesla). Now N m (unit of torque) is not quite the same as a joule (unit of energy), although dimensionally similar.
Yet it is common practice to express the magnetic moments of subatomic particles in eV T-1. Thus the Bohr magneton is a unit of
magnetic dipole moment equal to 9.27 % 10-24 N m T-1, and this may be expressed as 5.77 % 10-5 eV T-1.
One small detail to be on guard for is this. One may hear talk of “a 500 MeV proton”. Does this mean that the kinetic energy is 500
MeV or that its total energy is 500 MeV? In this case the answer is fairly clear (although it would have been completely clear if the
speaker had been explicit). The rest-mass energy of a proton is 938 MeV, so he must have been referring to the kinetic energy. If,
however, he had said “a 3 GeV proton”, there would be no way of deducing whether he was referring to the kinetic or the total
energy. And if he had said “a 3 GeV particle”, there would be no way of telling whether he was referring to its total energy, its
kinetic energy or its rest-mass energy. It is incumbent on all of us – or at least those of us who wish to be understood by others –
always to make ourselves explicitly clear and not to suppose that others will correctly guess what we mean.

15.28.1 https://phys.libretexts.org/@go/page/8481
This page titled 15.28: Units is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.28.2 https://phys.libretexts.org/@go/page/8481
15.29: Force
Force is defined as rate of change of momentum, and we wish to find the transformation between forces referred to frames in
uniform relative motion such that this relation holds on all such frames.
Suppose that, in Σ , a mass has instantaneous mass m and velocity whose instantaneous components are u and
′ ′ ′
x′
′
u
y′
. If a force
acts on it, then the velocity and hence also the mass are functions of time. The x-component of the force is given by
d ′
′ ′
F ′ = (m u ′ ). (15.29.1)
x x
dt

We want to express everything on the right hand side in terms of unprimed quantities. Thus from Equation 15.21.8 and the inverse
of Equation 15.16.2, we obtain
′ ′
m u = mγ(ux − v). (15.29.2)
x′

Also
d dt d
= (15.29.3)
′ ′
dt dt dt

Let us first evaluate (mγ u − mγv) . In this expression, v and γ are independent of time (the frame
d

dt
x Σ
′
is moving at constant
velocity relative to Σ), and of mu is the x-component of the force in Σ, that is F . Thus
dt
d
x x

d dm
(mγ ux − mγv) = γ ( Fx − v ). (15.29.4)
dt dt

Now we need to evaluate d

dt
′
in terms of unprimed quantities. If we start with
′ ′
∂t ∂t
′
dt = ( ) dx + ( ) dt (15.29.5)
∂x ∂t
t x

and we’ll evaluate dt

dt
which, being a total derivative, is the reciprocal of dt
dt
′
. The partial derivatives are given by Equations
15.15.3j,k and l, while dx

dt
= ux . Hence we obtain
dt 1
= . (15.29.6)
′
dt ux v
γ (1 − 2
)
c

Thus we arrive at
dm
Fx − v ( )
dt
′
F ′ = (15.29.7)
x ux v
1− 2
c

The mass is not constant (i.e. is not zero) because there is a force acting on the body, and we have to relate the term
dm

dt
to the dm

dt

force. At some instant when the force and velocity (in Σ) are F and u , the rate at which F is doing work on the body is F ∗ u
= F u +F u +F u
x x y y and this is equal to the rate of increase of energy of the body, which is ṁc . (In Section 15.24, in deriving
z z
2

the expression for kinetic energy, I wrote that the rate of doing work was equal to the rate of increase of kinetic energy. Now I have
just written that it is equal to the rate of increase of (total) energy. Which is right?)
dm 1
= (Fx ux + Fy uy + Fz uz ). (15.29.8)
2
dt c

Substitute this into Equation 15.29.7 and, after a very little more algebra, we finally obtain the transformation for F : ′
x′

v
′
F ′ = Fx − (uy Fy + uz Fz ). (15.29.9)
x
c2 − ux v

The y ′
− and z ′
− components are a little easier, and I leave it as an exercise to show that

15.29.1 https://phys.libretexts.org/@go/page/9117
 −−−−−
2
v
√1 −
c2
′
F = Fy (15.29.10)
y′ ux v
1−
c

−−−−−
2
v
√1 − 2
c
′
F = Fz . (15.29.11)
z′ ux v
1−
c

As usual, the inverse transformations are found by interchanging the primed and unprimed quantities and changing the sign of v .
The force on a particle and its resultant acceleration are not in general in the same direction, because the mass is not constant.
(Newton’s second law is not F = ma ; it is F = ṗ ) Thus
d
F = (mu) = ma + ṁu. (15.29.12)
dt

Here
m0
m = (15.29.13)
1

u2 2

(1 − 2
)
c

and so
m0 ua
ṁ = . (15.29.14)
3

2 u2 2

c (1 − 2
)
c

Thus
m0 ua
F = (a + u) . (15.29.15)
1
2 2
2 2
c −u
u
(1 − )
c2

This page titled 15.29: Force is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.29.2 https://phys.libretexts.org/@go/page/9117
15.30: The Speed of Light
The speed of light is, by definition, exactly 2.997 924 58 x 108 ms−1, and is the same relative to all observers.
This seemingly simple sentence invites several comments.
First: Note that I have used the word “speed”. Some writers use the word “velocity” as if it were merely a more impressive and
scientific-sounding synonym for “speed”. I trust that all readers of these notes know the difference and will use the word “speed”
when they mean “speed”, and the word “velocity” when they mean “velocity – surely not an unreasonable demand. To say that the
“velocity” of light is the same for all observers means that the direction of travel of light is the same relative to all observers. This
is doubtless not at all what a writer who uses the word “velocity” intends to convey – but it is the literal (and of course quite
erroneous) meaning of the assertion.
Second: How can we possibly define the speed of light to have a certain exact value? Surely the speed of light is what we find it to
be, and we are not free to define its value. But in fact we are allowed to do this, and the explanation, briefly, is as follows.
Over the course of history, the metre has been defined in several different ways. At one time it was a specified fraction of the
circumference of Earth. Later, it was the distance between two scratches on a bar of platinum-iridium alloy held in Paris. Later still
it was a specified number of wavelengths of a particular line in the spectrum of mercury, or cadmium, or argon or krypton. In our
present state of technology it is far easier to measure and reproduce precise standards of frequency than it is to measure and
reproduce standards of length. Because of that, the current SI (Système International) unit of time is the SI second, which is based
on the frequency of a particular transition in the spectrum of caesium, and from there, the metre is defined as the distance travelled
by light in vacuo in a defined fraction of an SI second, the speed of light being assigned the exact value quoted above.
Detailed discussion of the exact definitions of the units of time, distance and speed is part of the subject of metrology. That is an
important and interesting subject, but it is only marginally relevant to the topic of relativity, and consequently, having quoted the
exact value of the speed of light, we leave further discussion of metrology here.
Third: How can the speed of light be the same relative to all observers? This assertion is absolutely central to the theory of special
relativity, and it may be regarded as its fundamental and most important principle. We shall discuss it further in the remainder of
the chapter.

This page titled 15.30: The Speed of Light is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum
via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.30.1 https://phys.libretexts.org/@go/page/7024
15.31: Electromagnetism
These notes are intended to cover only mechanics, and therefore I resist the temptation to cover here special relativity and
electromagnetism. I point out only that in many ways this misses many of the most exciting parts of special relativity, and indeed it
was some puzzles with electromagnetism that led Einstein to formulate the theory of special relativity. One proceeds as we have
done with mechanical quantities; that is, we have to define carefully what is meant by each quantity and how in principle it is
possible to measure it, and then see how it transforms between frames in such a manner that the laws of physics – in particular
Maxwell’s equations - are the same in each. One such transformation that is found, for example, is E' = γ(E + u × B) so that
what appears in one frame as an electric field appears in another at least in part as a magnetic field. The Coulomb force transforms
to a Lorentz force; Coulomb’s law transforms to Ampères law.
Although I do no more than mention this topic here, I owe it to the reader to say just a little bit more about the speedometer that I
designed in Section 15.4. It is indeed true that, as the train moves forward, the net repulsive force between the two rods does
diminish, although not quite as I have indicated, for one has to make the correct transformations between frames for force, current,
electric field, magnetic field, and so on. But it turns out that the weights of the rods – i.e. the downward forces on them – also
diminish in exactly the same ratio, and the angle between the strings remains stubbornly the same. Our trip to the patent office will
be in vain. The speedometer will not work, and it remains impossible to determine the absolute motion of the train.

This page titled 15.31: Electromagnetism is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum
via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

15.31.1 https://phys.libretexts.org/@go/page/8483
 CHAPTER OVERVIEW
16: Hydrostatics
This relatively short chapter deals with the pressure under the surface of an incompressible fluid, which in practice means a liquid,
which, compared with a gas, is nearly, if not quite, incompressible. It also deals with Archimedes’ principle and the equilibrium of
floating bodies. The chapter is perhaps a little less demanding than some of the other chapters, though it will assume a familiarity
with the concepts of centroids and radius of gyration, which are dealt with in Chapters 1 and 2.
16.1: Introduction to Hydrostatics
16.2: Density
16.3: Pressure
16.4: Pressure on a Horizontal Surface. Pressure at Depth
16.5: Pressure on a Vertical Surface
16.6: Centre of Pressure
16.7: Archimedes' Principle
16.8: Some Simple Examples
16.9: Floating Bodies

This page titled 16: Hydrostatics is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1
16.1: Introduction to Hydrostatics
This relatively short chapter deals with the pressure under the surface of an incompressible fluid, which in practice means a liquid,
which, compared with a gas, is nearly, if not quite, incompressible. It also deals with Archimedes’ principle and the equilibrium of
floating bodies. The chapter is perhaps a little less demanding than some of the other chapters, though it will assume a familiarity
with the concepts of centroids and radius of gyration, which are dealt with in Chapters 1 and 2.

This page titled 16.1: Introduction to Hydrostatics is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

16.1.1 https://phys.libretexts.org/@go/page/7030
16.2: Density
There is little to be said about density other than to define it as mass per unit volume. However, this expression does not literally
mean the mass of a cubic metre, for after all a cubic metre is a large volume, and the density may well vary from point to point
throughout the volume. Density is an intensive quantity in the thermodynamical sense, and is defined at every point. A more exact
definition of density, for which I shall usually use the symbol ρ, is
δm
ρ = lim . (16.2.1)
δV →0 δV

The awful term “specific gravity” was formerly used, and is still regrettably often heard, as either a synonym for density, or the
dimensionless ratio of the density of a substance to the density of water. It should be avoided. The only concession I shall make is
that I shall use the symbol s to mean the ratio of the density of a body to the density of a fluid in which it may be immersed or on
which it may be floating,
The density of water varies with temperature, but at 4 ºC is 1 g cm-3 or 1000 kg m-3, or 10 lb gal-1. The original gallon was the
volume of 10 pounds (lb) of water. These are Imperial (UK) gallons, and avoirdupois pounds - not the gallons (wet or dry) used in
the U.S., and not the pounds (troy) used in the jewellery trade.

This page titled 16.2: Density is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

16.2.1 https://phys.libretexts.org/@go/page/7031
16.3: Pressure
Pressure is force per unit area, or, more precisely,
δF
ρ = lim . (16.3.1)
δA→0 δA

There is no particular direction associated with pressure – it acts in all directions – and it is a scalar quantity. The SI unit is the
pascal (Pa), which is a pressure of one newton per square metre (N m-2). Blaise Pascal (1623-1662) was a French mathematician
and philosopher who contributed greatly to the theory of conic sections and to hydrostatics. He showed that the barometric pressure
decreases with height – hence the famous examination question: “Explain how you would use a barometer to measure the height of
a tall building” – to which the most accurate answer is said to be: “I would drop it out of the window and time how long it took to
reach the ground.”
The CGS unit of pressure is dyne cm-2, and 1 Pa = 10 dyne cm-2.
Some other silly units for pressure are often seen, such as psi, bar, Torr or mm Hg, and atm.
A psi or “pound per square inch” is all right for those who define a “pound” as a unit of force (US usage) but is less so for those
who define a pound as a unit of mass (UK usage). A psi is about 6894.76 Pa and a bar is 105 Pa or 100 kPa.

[The “British Engineering System”, as far as I know, is used exclusively in the U.S. and is
not and never has been used in Britain, where it would probably be unrecognized. In the
“British” Engineering System, the pound is defined as a unit of force, whereas in Britain
a pound is a unit of mass.]
A Torr is a pressure under a column of mercury 760 mm high. This may be convenient for casual conversational use where extreme
precision is not expected in laboratory experiments in which pressure is actually indicated by a mercury barometer or manometer.
To find out exactly what the pressure in Pa under 760 mm Hg is, one would have to know the exact value of the local gravitational
acceleration and also the exact density of mercury, which varies with temperature and with isotopic constitution. A Torr is usually
given as 133.322 Pa. Evangelista Torricelli (1608 – 1647) is regarded as the inventor of the mercury barometer. He succeeded
Galileo as professor of mathematics at the University of Florence.
An atm is 760 torr or about 14.7 psi or 101 325 Pa. That is to say, 1.013 25 bar
As usual, the use of a variety of different units, and knowing the exact definitions and conversion factors between all of them and
carrying out all the tedious multiplications, is an unnecessary chore that is inflicted upon all of us in all branches of physics.

This page titled 16.3: Pressure is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

16.3.1 https://phys.libretexts.org/@go/page/7032
16.4: Pressure on a Horizontal Surface. Pressure at Depth
Figure XVI.1 shows a horizontal surface of area A immersed at a depth z under the surface of a fluid at depth z . The force F on
the area A is equal to the weight of the superincumbent fluid.

This gives us occasion to use a gloriously pompous word. “Incumbent” means “lying down”, so that “superincumbent” is lying
down above the area. It is incumbent upon all of us to understand this. The weight of the superincumbent fluid is evidently its
volume Az times its density ρ times the gravitational acceleration g . Thus
F = ρgzA, (16.4.1)

and, since pressure is force per unit area, we find that the pressure at a depth z is

P = ρgz. (16.4.2)

This is, of course, in addition to the atmospheric pressure that may exist above the surface of the liquid.
The pressure is the same at all points at the same horizontal level within a homogeneous incompressible fluid. This seemingly
trivial statement may sometimes be worth remembering under the stress of examination conditions. Thus, let’s look at an example.

 Example 16.4.1
In Figure XVI.2, the vessel at the left is partly filled with a liquid of density 0.8 g cm-3, the upper part of the vessel being filled
with air. The liquid also fills the tube along to the point C. From C to B, the tube is filled with mercury of density 13.6 g cm-3.
Above that, from B to A, is water of density 1.0 g cm-3, and above that is the atmospheric pressure of 101 kPa.

The height of the four interfaces above the thick black line are
A: 200 cm
B: 160 cm
C: 140 cm
D: 155 cm

16.4.1 https://phys.libretexts.org/@go/page/7033
 With g = 9.8 m s-2, what is the pressure of the air in the closed vessel?

Solution
I’ll do the calculation in SI units.
Pressure at A = 101000 Pa
Pressure at B = 101000 + 1000 % 9.8 % 0.4 = 104920 Pa
Pressure at C = 104920 + 13600 % 9.8 % 0.20 = 131576 Pa
Pressure at D = 131576 - 800 % 9.8 % 0.15 = 130400 Pa,
and this is the pressure of the air in the vessel.

Rather boring so far, and the next problem will also be boring, but the problem after that should keep you occupied arguing about it
over lunch.

 Exercise 16.4.1

This problem is purely geometrical and nothing to do with hydrostatics – but the result will help you with the next problem
after this. If you do not want to do it, just use the result in the next problem.

Show that the volume of the frustrum of a cone, whose upper and lower circular faces are of radii r and r , and whose height
1 2

is h , is πh(r + r r + r )
1

3
2
1 1 2
2
2

 Exercise 16.4.2: Pascal’s Paradox

16.4.2 https://phys.libretexts.org/@go/page/7033
 Figure XVI.4 shows four vessels. The base of each is circular with the same radius, 10

√π
cm, so the area is 100 cm2. Each is
filled with water (density = 1 g cm-3) to a depth of 15 cm.
Calculate
1. The mass of water in each.
2. The pressure at the bottom of each vessel.
3. The force on the bottom of each vessel.
If the bottom of each vessel were made of glass, which cracked under a certain pressure, which would crack first if the vessels
were slowly filled up? If the bottom of each vessel were welded to the scale of a weighing machine, what weight would be
recorded?

I’ll leave you to argue about this for as long as you wish.

This page titled 16.4: Pressure on a Horizontal Surface. Pressure at Depth is shared under a CC BY-NC 4.0 license and was authored, remixed,
and/or curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is
available upon request.

16.4.3 https://phys.libretexts.org/@go/page/7033
16.5: Pressure on a Vertical Surface
Figure XVI.5 shows a vertical surface from the side and face-on. The pressure increases at greater depths. I show a strip of the
surface at depth z .

Suppose the area of that strip is dA . The pressure at depth z is ρgz , so the force on the strip is ρgzdA . The force on the entire area
is ρg ∫ zdA , and that, by definition of the centroid (see Chapter 1), is ρgz̄A where z̄ is the depth of the centroid. The same result
¯
¯ ¯
¯

will be obtained for an inclined surface.

Therefore:

The total force on a submerged vertical or inclined plane surface is equal to the area of
the surface times the depth of the centroid.

 Example 16.5.1
Figure XVI.6 shows a triangular area. The uppermost side of the triangle is parallel to the surface at a depth z . The depth of the
centroid is z + h , so the pressure at the centroid is ρg (z + h) . The area of the triangle is bh so the total force on the
1

3
1

3
1

triangle is ρgbh (z + h) .
1

2
1

This page titled 16.5: Pressure on a Vertical Surface is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

16.5.1 https://phys.libretexts.org/@go/page/7034
16.6: Centre of Pressure
“The centre of pressure is the point at which the pressure may be considered to act.” This is a fairly meaningless sentence, yet it is
not entirely devoid of all meaning. If you refer to the left hand side of Figure XVI.5, you will see an infinite number (I’ve drawn
only eight) of forces. If you were to replace all of these forces by a single force, where would you put it? Or, more precisely, if you
were to replace all of these forces by a single force such that the (first) moment of this force about a line through the surface of the
fluid is the same as the (first) moment of all the actual forces, where would you place this single force? You would place it at the
centre of pressure. The depth of the centre of pressure is a depth such that the moment of the total force on a vertical surface about
a line in the surface of the fluid is the same as the moment of all the hydrostatic forces about a line in the surface of the fluid. I shall
use the Greek letter ζ to indicate the depth of the centre of pressure. We can continue to use Figure XVI.5.

The force on the strip of area dA at depth z is, as we have seen, ρgzdA , so the first moment of that force is ρgz dA . The total
2

moment is therefore ρg ∫ z dA which is, by definition of radius of gyration k , (see Chapter 2), ρgk A. The total force, as we have
2 2

seen, is pgzA and the total moment is to be this times ζ . Thus the depth of the centre of pressure is
¯
¯¯

2
k
ζ = (16.6.1)
¯
¯¯
z

 Example 16.6.1
A semicircular trough of radius a is filled with water, density ρ. One semicircular end of the trough is freely hinged at its
diameter (the thick line in the Figure). What force must be exerted at the bottom of the trough to prevent the end from swinging
open?

Solution
The area of the semicircle is π a . The depth of the centroid is
1

2
2
so the total hydrostatic force is ρga . The square of the
4a

3π
2

3
2

radius of gyration is a , so the depth of the centre of pressure is ζ = a . The moment of the hydrostatic forces is therefore
1

4
2 3π

16

πρga . If the required force is F , this must equal F a, and therefore F = πρga .
1 3 1 2

8 8

This page titled 16.6: Centre of Pressure is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

16.6.1 https://phys.libretexts.org/@go/page/7035
16.7: Archimedes' Principle
The most important thing about Archimedes’ principle is to get the apostrophe in the right place and to spell principle correctly.
Archimedes was a Greek scientist who lived in Syracuse, Sicily. He was born about 287 BC and died about 212 BC. He made
many contributions to mechanics. He invented the Archimedean screw, he is reputed to have said “Give me a fulcrum and I shall
move the world”, and he probably did not set the Roman invading fleet on fire by focussing sunlight on them with concave mirrors
– though it makes a good story. The most famous story about him is that he was commissioned by King Hiero of Sicily to
determine whether the king’s crown was contaminated with base metal. Archimedes realized that he would need to know the
density of the crown. Measuring its weight was no problem, but – how to measure the volume of such an irregularly-shaped object?
One day, he went to take a bath, and he had filled the bath full right to the rim. When he stepped into the bath he was much
surprised that some of the water slopped over the edge of the bath on to the floor. Suddenly, he realized that he had the solution to
his problem, so straightway he raced out of the house and ran absolutely starkers through the streets of Syracuse shouting “
ϵυρηκα! ϵυρηκα!”, which is Greek for "Eureka, Eureka" meaning "I found it, I found it."

When a body is totally or partially immersed in a fluid, it experiences a hydrostatic upthrust equal to the weight of fluid displaced.
Figure XVI.8 is a drawing of some water or other fluid. I have outlined with a dashed curve an arbitrary portion of the fluid. It is
subject to hydrostatic pressure from the rest of the fluid. The small pressure of the fluid above it is pushing it down; the larger
pressure of the fluid below it is pushing it up. Therefore there is a net upthrust. The portion of the fluid outlined is in equilibrium
between its own weight and the hydrostatic upthrust. If we were to replace this portion of the fluid with a lump of iron, we
wouldn’t have changed the hydrostatic forces. Therefore the upthrust is equal to the weight of fluid displaced.

If a body is floating on the surface, the hydrostatic upthrust, as well as being equal to the weight of fluid displaced, is also equal to
the weight of the body.

This page titled 16.7: Archimedes' Principle is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum
via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

16.7.1 https://phys.libretexts.org/@go/page/8491
16.8: Some Simple Examples
As we pointed out in the introduction to this chapter, this chapter is less demanding than some of the others, and indeed it has been
quite trivial so far. Just to show how easy the topic is, here are a few quick examples.

 Example 16.8.1
A cylindrical vessel of cross-sectional area A is partially filled with water. A mass m of ice floats on the surface. The density
of water is ρ and the density of ice is ρ. Calculate the change in the level of the water when the ice melts, and state whether
0

the water level rises or falls.

 Example 16.8.2
A cork of mass m, density ρ, is held under water (density ρ0 ) by a string. Calculate the tension in the string. Calculate the
initial acceleration if the string is cut.

 Example 16.8.3
A lump of lead (mass m, density ρ) is held hanging in water (density ρ ) by two strings as shown. Calculate the tension in the
0

strings.

 Example 16.8.4
A hydrometer (for our purposes a hydrometer is a wooden rod weighted at the bottom for stability when it floats vertically)
floats in equilibrium to a depth z in water of density ρ . If salt is added to the water so that the new density is ρ , what is the
1 1 2

new depth z ?
2

 Example 16.8.5

A mass m, density ρ, hangs in a fluid of density ρ from the ceiling of an elevator (lift). The elevator accelerates upwards at a
0

rate a . Calculate the tension in the string.

16.8.1 https://phys.libretexts.org/@go/page/8492
 Example 16.8.6

A hydrometer of mass m and cross-sectional area A floats in equilibrium do a depth h in a liquid of density ρ. The hydrometer
is then gently pushed down and released. Determine the period of oscillation.

 Example 16.8.7

A rod of length l and density sρ (s < 1 ) floats in a liquid of density ρ. One end of the rod is lifted up through a height yl so that
a length xl remains immersed. I have drawn it with the rope vertical. Must it be?)

i. Find x as a function of s .
ii. Find θ as a function of y and s .
iii. Find the tension T in the rope as a function of m,  g and s .
Draw the following graphs:
a. x and T

(mg)
versus s .

b. θ versus y for several s .

c. θ versus s for several y .
d. x versus y for several s .
e. T

(mg)
versus y for several s .

 Answers

1. No, it does not.

ρ −ρ
2. T =(
0

ρ
) mg

ρ−ρ ρ−ρ
0 0
( )mg ( )mg

3. T
ρ ρ

1 = T2 =
s in θ s in θ
1 2
cos θ1 + cos θ2 +
t an θ t an θ
2 1

ρ1
4. z
2 =
ρ
z1
2

ρ−ρ0
5. T = m [a + g (
ρ
)]

16.8.2 https://phys.libretexts.org/@go/page/8492
 −−−
6. P = 2π √
m

ρAg

7.
−−−−
i. x = 1 − √1 − s
y
ii. sin θ = √1−s

√1−s−(1−s)
iii. T = mg (
s
)

a.

b.

c.

16.8.3 https://phys.libretexts.org/@go/page/8492
 d.

e.

This page titled 16.8: Some Simple Examples is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

16.8.4 https://phys.libretexts.org/@go/page/8492
16.9: Floating Bodies
This is the most grisly topic in hydrostatics.
We can start with an observation that we have already made in Section 16.7, namely that, if a body is freely floating, the hydrostatic
upthrust is equal to the weight of the body.
I also introduce here the term centre of buoyancy, which is the centre of mass of the displaced fluid. In a freely-floating body in
equilibrium, the centre of buoyancy is vertically below the centre of mass of the floating body. As far as calculating the moment
about some axis of the hydrostatic upthrust is concerned, the upthrust can be considered to act through the centre of buoyancy, just
as the weight of an object can be considered to act through its centre of mass. See Section 1.1 of Chapter 1, for example, for a
discussion of this point.
Also, before we get going, here is another small problem.

 Problem 8
The drawing shows a body, whose relative density (i.e. its density relative to that of the fluid that it is floating in) is s1 . The
dashed line is the water-line section.

Now, in the next drawing, a body of exactly the same size and shape (but not necessarily the same density) is floating upside
down, with the same water-line section.

What is the relative density of this second body?

Answer
Let us establish some notation.
V = total volume of each body
fV = volume of liquid displaced by the first body (i.e. volume below the waterline in the first drawing)
(1 − f )V = volume of liquid displaced by the second body (i.e. volume below the waterline in the second drawing)
ρ0 = density of liquid
ρ1 = density of first body = s
1 ρ0

ρ2 = density of first body = s

2 ρ0

g = gravitational acceleration
Now:
Weight of first body = weight of liquid displaced: V ρ 1g = f V ρ0 g I.e. s
1 =f

Weight of second body = weight of liquid displaced: V ρ 2g = (1 − f )V ρ0 g I.e. s

2 =f

Hence s 2 = f .
–
––
––––
––
––
––
–

I want to look now at the stability of equilibrium of a freely-floating body. While at first sight this may not be a very interesting
topic, if you ever happen to be a passenger on an ocean liner, you might then find it to be quite interesting, for you will be

16.9.1 https://phys.libretexts.org/@go/page/8493
interested to know, if the liner is given a small angular displacement from the vertical position, whether it will capsize and throw
you into the sea, or whether it will right itself. Under such circumstances it becomes a very interesting subject indeed.
Before I start, I just want to establish one small geometric result.

Figure XVI.9 shows a plane bilaterally-symmetric area. I have drawn a dashed line through the centroid of the area. The areas to
the left and right of this line are A and A , and I have indicated the positions of the centroids of these two areas. (I haven’t
1 2

calculated the positions of the three centroids accurately – I just drew them approximately where I thought they would be.) Note
that, since the dashed line goes through the centroid of the whole area, A x̄ = A x̄ . Now rotate the area about the dashed line
1
¯
¯
1 2
¯
¯
2

through an angle θ . By the theorem of Pappus (see Chapter 1, Section 1.6), the volume swept out by A is A × x̄ θ and the
1 1
¯
¯
1

volume swept out by A is A × x̄ θ . Thus we have established the geometrical result that I wanted, namely, that when a
2 2
¯
¯
2

bilaterally symmetric area is rotated about an axis perpendicular to its axis of symmetry and passing through its centroid, the areas
to left and right of the axis of rotation sweep out equal volumes.
We can now return to floating bodies, and I am going to consider the stability of equilibrium of a bilaterally symmetric floating
body to a rotational displacement about an axis lying in the water line section and perpendicular to the axis of symmetry.

I have drawn in figure XVI.10 the centre of mass C of the whole body, the centre of buoyancy H, and the centroid of the water-line
section. The body is bilaterally symmetric about the plane of the paper, and we are going to rotate the body about an axis through O
perpendicular to the plane of the paper, and we want to know whether the equilibrium is stable against such an angular
displacement. We are going to rotate it in such a manner that the volume submerged is unaltered by the rotation – which means that
the hydrostatic upthrust will remain equal to the weight of the body, and there will be no vertical acceleration. The geometrical
theorem that we have just established shows that, if we rotate the body about an axis through the centroid of the water-line section,
the volume submerged will be constant; conversely, our condition that the volume submerged is constant implies that the rotation is
about an axis through the centroid of the water-line section.
I am going to establish a set of rectangular axes, origin O, with the x-axis to the right, the y -axis towards you, and the z -axis
downwards. I’m going to call the depth of the centre H of buoyancy z̄ . Now let’s carry out the rotation about O through an angle θ .
¯
¯

16.9.2 https://phys.libretexts.org/@go/page/8493
 ′ ′
I have drawn the position of the new centre of buoyancy H' and I wish to find its coordinates (x̄ , z̄ ) relative to O. We shall find
¯
¯ ¯
¯

that it has moved a little horizontally compared with the original position of H, but its depth is almost unchanged. Indeed, for small
′ ′
θ , we shall find that z − z is of order θ , while x − x is of order θ . Thus, to first order in θ , I shall assume that the depth of the
¯
¯¯ ¯
¯¯ 2 ¯¯
¯ ¯¯
¯

centre of buoyancy has remained unchanged.

′
However, the coordinate x of the new centre of buoyancy will be of interest for the following reason. The weight of the body acts
¯¯
¯

at its centre of mass C while the hydrostatic upthrust acts at the new centre of buoyancy H' and these two forces form a couple and
exert a torque. You will understand from figure XVI.11 that if H' is to the left of C, the torque will topple the body over, whereas if
H' is to the right of C, the torque will stabilize the body. Indeed, the horizontal distance between C and H' is known as the righting
lever. The point on the line COH vertically above H' is called the metacentre. I haven’t drawn it on the diagram, in order to
minimise clutter, but I shall use the symbol M to indicate the metacentre. We can see that the condition for stability of equilibrium
is that HM > HC. This is why we are interested in finding the exact position of the new centre of buoyancy H'.

In the upper part of figure XVI.12 I have drawn the old and new water-line sections as seen from the side, and in the lower part I
have drawn the new water-line section seen from above. I have indicated an elemental volume of width δx of the displaced fluid at
a distance x from the centroid O of the water-line section. For small θ the depth of this element is xθ. Let’s call its area in the

16.9.3 https://phys.libretexts.org/@go/page/8493
water-line section δA, so that the volume element is xθδA. We’ll call the total volume of the displaced fluid (which is unaltered by
the rotation) V .
Consider the moments of volume about the x-axis. We have
′ ′
′ A 1 B 1
¯¯ ¯¯
V z̄ = V z̄  − ∫ xθ. xθδA  +   ∫ xθ. xθδA
O 2 O 2

′
B
′ 2
¯¯ ¯¯
V (x̄ − x̄) = θ∫ x dA. (16.9.1)
′
A

Thus, as previously asserted, the vertical displacement of the centre of buoyancy is of order 2
θ , and, to first order in θ may be
neglected.
Now consider the moments of volume about the y -axis. We have
′ ′
′ A A
¯¯
¯ ¯¯
¯
V x = V x  −   ∫ xθ dA x  +   ∫ xθ dA x
O O

′
B
′ 2
¯¯
¯ ¯¯
¯
V  (x − x)  =  θ ∫ x dA. (16.9.2)
′
A

But the integral on the right hand side of Equation 16.9.2 is Ak , where A is the area of the water-line section, and k is its radius
2

of gyration.
Thus
2
Ak θ
′
HH = . (16.9.3)
V

Now H H ′
  =  H M   sin  θ , where M is the metacentre, or, to first order in θ , HH' = HM % θ .
2
Ak
HM  =   . (16.9.4)
V

Therefore the condition for stability of equilibrium is that

2
Ak
> HC. (16.9.5)
V

Here, A and k refer to the water-line section, V is the volume submerged, and HC is the distance between centre of buoyancy and
2

centre of mass.
Example. Suppose that the body is a cube of side 2a and of relative density s . The water-line section is a square, and A = 4a and 2

2
2
k = . The volume submerged is 8a s . The distance between the centres of mass and buoyancy is a(1 − s) , and so the
a

3
3

condition for stability is

a
> a(1 − s) (16.9.6)
6s

The equilibrium is unstable if

2
6 s   −  6s  +  1 < 0. (16.9.7)

That is, the equilibrium is unstable if s is between 0.2113 and 0.7887. The cube will float vertically only if the density is less than
0.2113 or if it is greater than 0.7887.

 Problem 9

Here in British Columbia there is a large logging industry, and many logs float horizontally in the water. They gradually
become waterlogged, and, when the density of a log is nearly as dense as the water, the vertical position become stable and the
log tips to the vertical position, nearly all of it submerged, with only an inch or so above the surface. It then becomes a danger
to boats. If the length of the log is 2l and its radius is a , what is the least relative density for which the vertical position is
stable?

Answer

16.9.4 https://phys.libretexts.org/@go/page/8493
The condition for stability of equilibrium is that
2
Ak
> HC
V

Here, A and k refer to the water-line section, V is the volume submerged, and HC is the distance between centre of mass
2

and centre of buoyancy.

In the present case we have a log of radius a and length 2l. In this case
2
A = πa ,
2
k =
1

2
2
a ,
2
V = 2π a l .
2 2
Ak a
=
V 4l

Density of log = ρ
Density of water = ρ 0
ρ
Relative density s = ρ
0

Some distances:
AB = 2l
AC = l
SB = 2ls
AS = 2l(1 − s)
SH = ls
SC = AC - AS = 2ls − l 2ls-l\)
HC = SH - SC = l(1 − s)
2

The condition for stability is that a

4l
> l(1 − s)

That is: s > 1  − 1

4
(
diame te r

le ngth
) .

length/diameter
0.5 0.71 1 2 10 40
=

relative density > 0 0.50 0.75 0.9375 0.9975 0.9988

A flat log, whose length is less than half its diameter, floats with its axis vertical, whatever its density (provided, of course,
that it is less than that of water, when it will sink). If its length is equal to its diameter, it will float vertically provided that
its density is at least 0.75 that of water. A very long log floats horizontally until it is almost completely saturated with water,
and then it will tip over to a vertical position, almost completely submerged, when it is not readily visible and it is then a
danger to boats. The condition for vertically-floating stable equilibrium is illustrated in the two graphs below.

16.9.5 https://phys.libretexts.org/@go/page/8493
This page titled 16.9: Floating Bodies is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

16.9.6 https://phys.libretexts.org/@go/page/8493
 CHAPTER OVERVIEW
17: Vibrating Systems
Topic hierarchy
17.1: Introduction
17.2: The Diatomic Molecule
17.3: Two Masses, Two Springs and a Brick Wall
17.4: Double Torsion Pendulum
17.5: Double Pendulum
17.6: Linear Triatomic Molecule
17.7: Two Masses, Three Springs, Two brick Walls
17.8: Transverse Oscillations of Masses on a Taut String
17.9: Vibrating String
17.10: Water
17.11: A General Vibrating System
17.12: A Driven System
17.13: A Damped Driven System

This page titled 17: Vibrating Systems is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1
17.1: Introduction
A mass m is attached to an elastic spring of force constant k , the other end of which is attached to a fixed point. The spring is
supposed to obey Hooke’s law, namely that, when it is extended (or compressed) by a distance x from its natural length, the tension
(or thrust) in the spring is kx, and the equation of motion is mẍ = −kx . This is simple harmonic motion of period , where 2π

ω
2
ω =
k

m
. Most readers will have no difficulty with that problem. But now suppose that, instead of one end of the spring being
attached to a fixed point, we have two masses, m and m , one at either end of the spring. A diatomic molecule is much the same
1 2

thing. Can you calculate the period of simple harmonic oscillations? It looks like an easy problem, but it somehow seems difficult
to get a hand on it by conventional newtonian methods. In fact it can be done quite readily by newtonian methods, but this problem,
as well as more complicated problems where you have several masses connected by several springs and several possible modes of
vibration, is particularly suitable by lagrangian methods, and this chapter will give several examples of vibrating systems tackled
by lagrangian methods.

This page titled 17.1: Introduction is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

17.1.1 https://phys.libretexts.org/@go/page/7037
17.2: The Diatomic Molecule
Two particles, of masses m and m are connected by an elastic spring of force constant k . What is the period of oscillation?
1 2

Let’s suppose that the equilibrium separation of the masses – i.e. the natural, unstretched, uncompressed length of the spring – is a .
At some time suppose that the x-coordinates of the two masses are x and x . The extension q of the spring from its natural length
1 2

at that moment is q  =  x − x − a . We’ll also suppose that the velocities of the two masses at that instant are ẋ and ẋ . We
2 1 1 2

know from Chapter 13 how to start any calculation in lagrangian mechanics. We do not have to think about it. We always start with
T = ... and V = ... :

1 2
1 2
T = m1 ẋ1 + m2 ẋ2 , (17.2.1)
2 2

1 2
V = kq . (17.2.2)
2

We want to be able to express the equations in terms of the internal coordinate q. V is already expressed in terms of q. Now we
need to express T (and therefore ẋ and ẋ ) in terms of q. Since q  =  x − x − a we have, by differentiation with respect to time,
1 2 2 1

q̇ = ẋ2 − ẋ1 . (17.2.3)

We need one more equation. The linear momentum is constant and there is no loss in generality in choosing a coordinate system
such that the linear momentum is zero:
0 = m1 ẋ1 + m2 ẋ2 . (17.2.4)

From these two equations, we find that

m2
ẋ1 = q̇ (17.2.5a)
m1 + m2

and
m1
ẋ2 = q̇ . (17.2.5b)
m1 + m2

Thus we obtain
1
2
T = m q̇ (17.2.5)
2

and
1 2
V = kq
2

where
m1 m2
m = . (17.2.6)
m1 + m2

Now apply Lagrange’s equation

17.2.1 https://phys.libretexts.org/@go/page/7038
 d ∂T ∂T ∂V
− = . (13.4.13)
dt ∂q̇ ∂qj ∂qj
j

to the single coordinate q in the fashion to which we became accustomed in Chapter 13, and the equation of motion becomes

m q̈ = −kq, (17.2.7)

−−
which is simple harmonic motion of period 2π √ m

k
where m is given by Equation 17.2.6. The frequency is the reciprocal of this,
−−
and the “angular frequency” ω, also sometimes called the “pulsatance”, is 2π times the frequency, or √ k

m
.
m1 m2
The quantity ( m1 +m2 )
is usually called the “reduced mass” and one may wonder is what sense it is “reduced”. I believe the origin
of this term may come from an elementary treatment of the Bohr atom of hydrogen, in which one at first assumes that there is an
electron moving around an immovable nucleus – i.e. a nucleus of “infinite mass”. One develops formulas for various properties of
the atom, such as, for example, the Rydberg constant, which is the energy required to ionize the atom from its ground state. This
and similar formulas include the mass m of the electron. Later, in a more sophisticated model, one takes account of the finite mass
of the nucleus, with nucleus and electron moving around their mutual centre of mass. One arrives at the same formula, except that
m is replaced by , where M is the mass of the nucleus. This is slightly less (by about 0.05%) than the mass of the electron,
mM

(m+M)

and the idea is that you can do the calculation with a fixed nucleus provided that you use this “reduced mass of the electron” rather
than its true mass. Whether this is the appropriate term to use in our present context is debatable, but in practice it is the term
almost universally used.
It may also be remarked upon by readers with some familiarity with quantum mechanics that I have named this section “The
Diatomic Molecule” – yet I have ignored the quantum mechanical aspects of molecular vibration. This is true – in this series of
notes on Classical Mechanics I have adopted an entirely classical treatment. It would be wrong, however, to assume that classical
mechanics does not apply to a molecule, or that quantum mechanics would not apply to a system consisting of a cricket ball and a
baseball connected by a metal spring. In fact both classical mechanics and quantum mechanics apply to both. The formula derived
for the frequency of vibration in terms of the reduced mass and the force constant (“bond strength”) applies as accurately for the
molecule as for the cricket ball and baseball. Quantum mechanics, however, predicts that the total energy (the eigenvalue of the
hamiltonian operator) can take only certain discrete values, and also that the lowest possible value is not zero. It predicts this not
only for the molecule, but also for the cricket ball and baseball – although in the latter case the energy levels are so closely spaced
together as to form a quasi continuum, and the zero point vibrational energy is so close to zero as to be unmeasurable. Quantum
mechanics makes its effects evident at the molecular level, but this does not mean that it does not apply at macroscopic levels. One
might also take note that one is not likely to understand why wave mechanics predicts only discrete energy levels unless one has
had a good background in the classical mechanics of waves. In other words, one must not assume that classical mechanics does not
apply to microscopic systems, or that quantum mechanics does not apply to macroscopic systems.
Below leaving this section, in case you tried solving this problem by newtonian methods and ran into difficulties, here’s a hint.
Keep the centre of mass fixed. When the length of the spring is x, the lengths of the portions on either side of the centre of mass are
m2 x m1 x

m1 +m2
and m1 +m2
. The force constants of the two portions of the spring are inversely proportional to their lengths. Take it from
there.

This page titled 17.2: The Diatomic Molecule is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

17.2.2 https://phys.libretexts.org/@go/page/7038
17.3: Two Masses, Two Springs and a Brick Wall
The system is illustrated in Figure XVII.2, first in its equilibrium (unstretched) position, and then at some instant when it is not in
equilibrium and the springs are stretched. You can imagine that the masses are resting upon and can slide upon a smooth, horizontal
table. I could also have them hanging under gravity, but this would introduce a distracting complication without illustrating any
further principles. I also want to assume that all the motion is linear, so we could have them sliding on a smooth horizontal rail, or
have them confined in the inside of a smooth, fixed drinking-straw. For the present, I do not want the system to bend.

The displacements from the equilibrium positions are x and x , so that the two springs are stretched by x and
1 2 1 x2 − x1

respectively. The velocities of the two masses are ẋ and ẋ . We now start the lagrangian calculation in the usual manner:
1 2

1 1
2 2
T = m1 ẋ + m2 ẋ , (17.3.1)
1 2
2 2

1 1
2 2
V = k1 x + k2 (x2 − x1 ) . (17.3.2)
1
2 2

Apply Lagrange’s equation to each coordinate in turn, to obtain the following equations of motion:
m1 ẍ1 = −(k1 + k2 )x1 + k2 x2 (17.3.3)

and
m2 ẍ2 = k2 x1 − k2 x2 . (17.3.4)

Now we seek solutions in which the system is vibrating in simple harmonic motion at angular frequency ω ; that is, we seek
solutions of the form ẍ = −ω x and ẍ = −ω x .
1
2
1 2
2
2

When we substitute these in Equations 17.3.3 and 17.3.4, we obtain

2
(k1 + k2 − m1 ω )x1 − k2 x2 = 0 (17.3.5)

an\
2
k2 x1 − (k2 − m2 ω )x2 = 0. (17.3.6)

x2
Either of these gives us the displacement ratio x1
(and hence amplitude ratio). The first gives us
2
x2 −m1 ω + k1 + k2
= (17.3.7)
x1 k2

and the second gives us

x2 k2
= . (17.3.8)
2
x1 k2 − m2 ω

These are equal, and, by equating the right hand sides, we obtain the following equation for the angular frequencies of the normal
modes:
4 2
m1 m2 ω − (m1 k2 + m2 k1 + m2 k2 )ω + k1 k2   =  0. (17.3.9)

This equation can also be derived by noting, from the theory of equations, that Equations 17.3.5 and 17.3.6 are consistent only if
the determinant of the coefficients is zero.
The meaning of these equations and of the expression “normal modes” can perhaps be best illustrated with a numerical example.
Let us suppose, for example, that k   =  k   =  1 and m   =  3 and m   =  2 . In that case Equation 17.3.9 is
1 2 1 2

17.3.1 https://phys.libretexts.org/@go/page/7039
6ω
4
− 7ω
2
. This is a quartic equation in ω, but it is also a quadratic equation in ω , and there are just two positive
+1 = 0
2

solutions for ω. These are = 0.4082 (slow, low frequency) and 1 (fast, high frequency). If you put the low frequency ω into
1

√6

either of Equations 17.3.7 or 17.3.8 (or in both, to check for arithmetic or algebraic mistakes) you find a displacement ratio of
+1.5; but if you put the high frequency ω into either equation, you find a displacement ratio of -1.0. The first of these normal
modes is a low-frequency slow oscillation in which the two masses oscillate in phase, with m having an amplitude 50% larger2

than m . The second normal mode is a high-frequency fast oscillation in which the two masses oscillate out of phase but with
1

equal amplitudes.
So, how does the system actually oscillate? This depends on the initial conditions. For example, if you displace the first mass by
one inch to the right and the second mass by 1.5 inches to the right (this implies stretching the first spring by 1 inch and the second
by 0.5 inches), and then let go, the system will oscillate in the slow, in-phase mode. But if you start by displacing the first mass by
one inch to the right and the second mass by one inch to the left (this implies stretching the first spring by 1 inch and compressing
the second by 2 inches), the system will oscillate in the fast, out-of-phase mode. For other initial conditions, the system will
oscillate in a linear combination of the normal modes.
Thus, m1 might oscillate with an amplitude A in the slow mode, and an amplitude B in the fast mode:

x1 = A cos(ω1 t + α1 )  +  B cos(ω2 t + α2 ) , (17.3.10)

in which case the oscillation of m is given by 2

x2 = 1.5A cos(ω1 t + α1 )  −  B cos(ω2 t + α2 ). (17.3.11)

In our example, ω and ω are

1 2
1

√6
and 1 respectively.

Let’s suppose that the initial conditions are that, at t  =  0 , ẋ and ẋ are both zero. This means that α and α are both zero or π
1 2 1 2

(I’ll take them to be zero), so that

x1 = A cos ω1 t  +  B cos ω2 t (17.3.12)

and
x2 = 1.5A cos ω1 t  −  B cos ω2 t. (17.3.13)

Suppose further that at t = 0 , x and x are both +1, which means that we start by stretching both springs equally. Equations
1 2

17.3.12 and 17.3.13 then become 1  =  A  +  B and 1  =  1.5A  −  B . That is, A  =  0.8 and B  =  0.2 . I’ll leave you to draw

graphs of x and x versus time.

1 2

Here’s an exercise that might be useful if, perhaps, you wanted to construct a real system with two equal masses m and two equal
springs, each of constant k , to demonstrate the vibrations. Show that in that case, the angular frequency (which is, of course, 2π
times the actual frequency) of the slow, in phase, mode is
−− −−
1 – k k
ω1 = (√5 − 1) √   =  0.6180 √
2 m m

x2 –
with a displacement ratio x1
=
1

2
(√5  +  1)  =  1.6180 ;
and the angular frequency of the fast, out of phase, mode is
−− −−
1 – k k
ω2 = = (√5 + 1)√ = 1.6180 √
2 m m

–
with a displacement ratio, x 2 / x1 =−
1

2
(√5 − 1) = −0.6180 .
Knowing these displacements ratios will enable you to start with the appropriate initial conditions for each normal mode.
If you were to start at t = 0 with a displacements x = 1 and x = 2 which isn’t right for either normal mode, you can show that the
1 2

subsequent displacements would be

x1 = 1.170820 cos ω1 t − 0.170820 cos ω2 t

x2 = 1.894427 cos ω1 t − 0.105572 cos ω2 t .

That looks like

17.3.2 https://phys.libretexts.org/@go/page/7039
Although at first it looks like fast in-phase mode for both of them, you can see the influence of the slow mode, which has about 2.6
times the period of the last mode, in the slow amplitude modulation. If you look carefully at the modulation amplitudes of both
displacements, you will see that the amplitude of the x displacement is out of phase with the amplitude of the x displacement.
1 2

This page titled 17.3: Two Masses, Two Springs and a Brick Wall is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or
curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is
available upon request.

17.3.3 https://phys.libretexts.org/@go/page/7039
17.4: Double Torsion Pendulum
Here we have two cylinders of rotational inertias I and I hanging from two wires of torsion constants c and c .
1 2 1 2

At any instant, the top cylinder is turned through an angle θ from the equilibrium position and the lower cylinder by an angle θ
1 2

from the equilibrium position (so that, relative to the upper cylinder, it is turned by ). The equations and the description of the
motion are just the same as in the previous example, except that x , x , m , m , k , k are replaced by θ , θ , I , I , c , c . The
1 2 1 2 1 2 1 2 1 2 1 2

kinetic and potential energies are

1 2 1 2
˙ ˙
T = I1 θ 1 + I2 θ 2 , (17.4.1)
2 2

1 2
1 2
T = c1 θ1 + c2 (θ2 − θ1 ) . (17.4.2)
2 2

The equations for ω and the displacement ratios are just the same, and there is an in-phase and an out-of-phase mode.

This page titled 17.4: Double Torsion Pendulum is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

17.4.1 https://phys.libretexts.org/@go/page/7040
17.5: Double Pendulum
This is another similar problem, though, instead of assuming Hooke’s law, we shall assume that angles are small (
sin θ ≈ θ, cos θ ≈ 1 − θ ). For clarity of drawing, however, I have drawn large angles in Figure XVIII.4.
1

2
2

Because I am going to use the lagrangian equations of motion, I have not marked in the forces and accelerations; rather, I have
marked in the velocities. I hope that the two components of the velocity of m that I have marked are self-explanatory; the speed of
2

2 2
m2 is given by v
2
2
2 ˙ 2 ˙ ˙ ˙
= l θ 1 + l θ 2 + 2 l1 l2 θ 1 θ 2 cos(θ2 − θ1 )
1 2
. The kinetic and potential energies are
1 2 1 2 2
2 ˙ 2 ˙ 2 ˙ ˙ ˙
T = m1 l θ 1 + m1 [ l θ 1 + l θ 2 + 2 l1 l2 θ 1 θ 2 cos(θ2 − θ1 )], (17.5.1)
1 1 2
2 2

V = constant − m1 gl1 cos θ1 − m2 g(l1 cos θ1 + l2 cos θ2 ). (17.5.2)

If we now make the small angle approximation, these become

1 2 1 2
2 ˙ ˙ ˙
T = m1 l θ 1 + m2 (l1 θ 1 + l2 θ 2 ) (17.5.3)
1
2 2

and
1 1
2 2 2
V = contant + m1 gl1 θ + m1 g(l1 θ + l2 θ ) − m1 gl1 − m2 gl2 . (17.5.4)
1 1 2
2 2

Apply the lagrangian equation in turn to θ and θ : 1 2

2 ¨ ¨
(m1 + m2 )l θ 1 + m2 l1 l2 θ = −(m1 + m2 )gl1 θ1 (17.5.5)
1

and
¨ 2 ¨
m2 l1 l2 θ 1 + m2 l θ 2 = −m2 gl2 θ2 . (17.5.6)
2

Seek solutions in the form of θ˙ 1 = −ωθ1 and θ˙ 2 = −ω θ2

2
.
Then
2 2
(m1 + m2 )(l2 ω − g)θ1 + m2 l1 l2 ω θ2 = 0 (17.5.7)

and
2 2
l1 ω θ1 + (l2 ω − g)θ2 = 0. (17.5.8)

Either of these gives the displacement ratio θ /θ . Equating the two expressions for the ratio θ
2 1 2 / θ1 , or putting the determinant of
the coefficients to zero, gives the following equation for the frequencies of the normal modes:
4 2 2
m1 l1 l2 ω − (m1 + m2 )g(l1 + l2 )ω + (m1 + m2 )g = 0. (17.5.9)

As in the previous examples, there is a slow in-phase mode, and fast out-of-phase mode.

17.5.1 https://phys.libretexts.org/@go/page/7041
For example, suppose m = 0.01 kg, m = 0.02 kg, l = 0.3 m, l = 0.6 m, g = 9.8 m s−2.
1 2 1 2

Then 0.0018ω − 0.02446ω = 0. The slow solution is ω = 3.441 rad s−1 ( P = 1.826 s), and the fast solution is ω = 11.626 rad s−1
4 2

(P =0.540 s). If we put the first of these (the slow solution) in either of equations 17.5.7 or 8 (or both, as a check against mistakes)
we obtain the displacement ratio θ /θ = 1.319, which is an in-phase mode. If we put the second (the fast solution) in either
2 1

equation, we obtain θ /θ = −0.5689 , which is an out-of-phase mode. If you were to start withθ /θ = 1.319 and let go, the
2 1 2 1

pendulum would swing in the slow in-phase mode. If you were to start with θ /θ = −0.5689 and let go, the pendulum would
2 1

swing in the fast out-of-phase mode. Otherwise the motion would be a linear combination of the normal modes, with the fraction of
each determined by the initial conditions, as in the example in Section 17.3.

This page titled 17.5: Double Pendulum is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

17.5.2 https://phys.libretexts.org/@go/page/7041
17.6: Linear Triatomic Molecule
In Chapter 2, Section 2.9, we discussed a rigid triatomic molecule. Now we are going to discuss three masses held together by
springs, of force constants k and k . We are going to allow it to vibrate, but not to rotate. Also, for the time being, I do not want
1 2

the molecule to bend, so we’ll put it inside a drinking straw to that all the vibrations are linear. By the way, for real triatomic
molecules, the force constants and rotational inertias are such that molecules vibrate much faster than they rotate. To see their
vibrations you look in the near infra-red spectrum; to see their rotation, you have to go to the far infrared or the microwave
spectrum.

Suppose that the equilibrium separations of the atoms are a and a . Suppose that at some instant of time, the x-coordinates
1 2

(distances from the left hand edge of the page) of the three atoms are x , x , x . The extensions from the equilibrium distances are
1 2 3

then q = x − x − a , q = x − x − a . We are now ready to start:

1 2 1 1 2 3 2 2

1 2
1 2
1 2
T = m1 ẋ1 + m2 ẋ2 + m2 ẋ3 , (17.6.1)
2 2 2

1 1
2 2
V = k1 q + k2 q . (17.6.2)
1 2
2 2

We need to express the kinetic energy in terms of the internal coordinate, and, just as for the diatomic molecule (Section 17.2), the
relevant equations are

q̇ 1
= ẋ2 − ẋ1 , (17.6.3)

q̇ 2
= ẋ3 − ẋ2 , (17.6.4)

and

0 = m1 ẋ1 + m2 ẋ2 + m3 ẋ3 . (17.6.5)

These can conveniently be written

q̇ −1 1 0 ẋ1
⎛ 1 ⎞ ⎛ ⎞⎛ ⎞

⎜ q̇ 2
⎟ =⎜ 0 −1 1 ⎟ ⎜ ẋ2 ⎟ (17.6.6)

⎝ ⎠ ⎝ ⎠⎝ ⎠
0 m1 m2 m3 ẋ3

By one dexterous flick of the fingers (!) we invert the matrix to obtain
m2 +m3 m3 1

ẋ1 ⎛− M
−
M M
⎞ q̇
⎛ ⎞ ⎛ 1 ⎞
m1 m3 1
⎜ ⎟
⎜ ẋ2 ⎟ = − ⎜ q̇ 2 ⎟ , (17.6.7)
⎜ M M M ⎟
⎝ ⎠ m1 m1 +m2
⎝ ⎠
ẋ3 ⎝ 1 ⎠ 0
M M M

where M = m1 + m2 + m3 . On putting these into Equation 17.6.1, we now have

1 2 2
T = (a q̇ 1
+ 2h q̇ 1
q̇ 2
+ b q̇ 2
) (17.6.8)
2

and
1 1
2 2
V = k1 q + k2 q (17.6.9)
1 2
2 2

where
a = m1 (m2 + m3 )/M , (17.6.10)

17.6.1 https://phys.libretexts.org/@go/page/7042
 h = m3 m1 /M , (17.6.11)

b = m3 (m1 + m2 )/M (17.6.12)

and for future reference,

2 2
ab − h = m1 m2 m2 /M = m h. (17.6.13)

On application of Lagrange’s equation in turn to the two internal coordinates we obtain

a q̈ 1
+ h q̈ 2
+ k1 q1 = 0 (17.6.14)

and

b q̈ 2
+ h q̈ 2
+ k2 q2 = 0 (17.6.15)

Seek solutions of the form q̈ 1

2
= −ω q1 and q̈ 2
= −ω q2
2
and we obtain the following two expressions for the extension ratios:
2 2
q1 hω k2 − b ω
= = . (17.6.16)
2 2
q2 k1 − aω hω

Equating them gives the equation for the normal mode frequencies:
2 4 2
(ab − h )ω − (ak2 + b k1 )ω + k1 k2 = 0. (17.6.17)

For example, if k = k = l and m = m = m

1 2 1 2 3 =m , we obtain, for the slow symmetric (“breathing”) mode, q 1 / q2 = +1 and
ω = k/m . For the fast asymmetric mode, q / q and ω = 3k/m .
2 2
1 2 = −1

Example.
Consider the linear OCS molecule whose atoms have masses 16, 12 and 32. Suppose that the angular frequencies of the normal
modes, as determined from infrared spectroscopy, are 0.905 and 0.413. (I just made these numbers up, in unstated units, just for the
purpose of illustrating the calculation. Without searching the literature, I can’t say what they are in the real OCS molecule.)
Determine the force constants.
In Chapter 2 we considered a rigid triatomic molecule. We were given the moment of inertia, and we were asked to find the two
internuclear distances. We couldn’t do this with just one moment of inertia, so we made an isotopic substitution (18O instead of
16O) to get a second equation, and so we could then solve for the two internuclear distances. This time, we are dealing with

vibration, and we are going to use Equation 17.6.17 to find the two force constants. This time, however, we are given two
frequencies (of the normal modes), and so we have no need to make an isotopic substitution − we already have two equations.
Here are the necessary data.
16
OCS
Fast ω 0.905
Slow ω 0.413
m1 m2 m3 16 12 32
M 60
a 11.73
h 8.53
b 14.93
ab − h
2
102.4
Use equation 17.6.16 for each of the frequencies, and you’ll get two equations, in k and k . As in the rotational case, they are1 2

quadratic equations, but they are a bit easier to solve than in the rotational case. You’ll get two equations, each of the form of
A − Bk − C k + k k = 0 , where the coefficients are functions of a, b, h, ω. You’ll have to work out the values of these
1 2 1 2

coefficients, but, before you substitute the numbers in, you might want to give a bit of thought to how you would go about solving
two simultaneous equations of the form A − Bk − C k + k k = 0 . 1 2 1 2

You will find that there are two possible solutions:

17.6.2 https://phys.libretexts.org/@go/page/7042
 k1 = 2.8715 k2 = 4.9818

and
k1 = 3.9143 k2 = 3.6547

Both of these will result in the same frequencies. You would need some additional information to determine which obtains for the
actual molecule, perhaps with measurements on an isotopomer, such as 18OCS.
Note that in this section we considered a linear triatomic molecule that was not allowed either to rotate or to bend, whereas in
Chapter 2 we considered a rigid triatomic molecule that was not allowed either to vibrate or to bend. If all of these restrictions are
removed, the situation becomes rather more complicated. If a rotating molecule vibrates, the moving atoms, in a co-rotating
reference frame, are subject to the Coriolis force, and hence they do not move in a straight line. Further, as it vibrates, the rotational
inertia changes periodically, so the rotation is not uniform. If we allow the molecule to bend, the middle atom can oscillate up and
down in the plane of the paper (so to speak) or back and forth at right angles to the plane of the paper. These two motions will not
necessarily have either the same amplitude or the same phase. Consequently the middle atom will whirl around in a Lissajous
ellipse, giving rise to what has been called “vibrational angular momentum”. In a real triatomic molecule, the vibrations are usually
much faster than the relatively slow, ponderous rotation, so that vibration-rotation interaction is small – but is by no means
negligible and is readily observed in the spectrum of the molecule.

This page titled 17.6: Linear Triatomic Molecule is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

17.6.3 https://phys.libretexts.org/@go/page/7042
17.7: Two Masses, Three Springs, Two brick Walls
The three masses are equal, and the two outer springs are identical. Figure XVII.6 shows the equilibrium position.

Suppose that at some instant the first mass is displaced a distance x to the right and the second mass is displaced a distance y to the
right. The extensions of the first two springs are x and y − x respectively, and the compression of the third spring is y . If the speeds
of the masses are ẋ and ẏ , we have for the kinetic and potential energies:
1 1
2 2
T = m ẋ + m ẏ (17.7.1)
2 2

and
1 1 1
2 2 2
V = k1 x + k2 (y − x ) + k1 y . (17.7.2)
2 2 2

Apply Lagrange’s equation in turn to x and to y.

m ẍ + (k1 + k2 )x − k2 y = 0 (17.7.3)

and

m ÿ + (k1 + k2 )y − k2 x = 0. (17.7.4)

Seek solutions of the form ẍ = −ω 2

x and ÿ = −ω y
2
.
2
(−m ω + k1 + k2 ) − k2 y = 0 (17.7.5)

and
2
−k2 x + (−m ω + k1 + k2 )y = 0. (17.7.6)

On putting the determinant of the coefficients to zero, we find for the frequencies of the normal modes
k1 k1 + 2 k2
2 2
ω = and ω = , (17.7.7)
m m

corresponding to displacement ratios

x x
=1 and = −1. (17.7.8)
y y

In the first, slow, mode, the masses move in phase and there is no extension or compression of the connecting spring. In the second,
fast, mode, the masses move in antiphase and the compression or extension of the coupling spring is twice the extension or
compression of the outer springs.
The general motion is a linear combination of the normal modes:

x = A cos(ω1 t + α1 ) + B cos(ω2 t + α2 ), (17.7.9)

y = A cos(ω1 t + α1 ) − B cos(ω2 t + α2 ), (17.7.10)

ẋ = −A sin(ω1 t + α1 ) − B sin(ω2 t + α2 ), (17.7.11)

ẏ = −A sin(ω1 t + α1 ) + B sin(ω2 t + α2 ), (17.7.12)

Suppose that the initial condition is at t = 0, y = ẏ = 0, x = x , ẋ = 0 . That is, we pull the first mass a little to the right (keeping
0

the second mass fixed) and then we let go. The second two equations establish that α = α = 0 , and the first two equations tell us
1 2

17.7.1 https://phys.libretexts.org/@go/page/8501
that A = B = x 0 /2 . The displacements are then given by
1 1 1
x = x0 (cos ω1 t + cos ω2 t) = x0 cos (ω1 − ω2 )t cos (ω1 + ω2 )t (17.7.13)
2 2 2

and
1 1 1
y = x0 (cos ω1 t + cos ω2 t) = −x0 sin (ω1 − ω2 )t sin (ω1 + ω2 )t. (17.7.14)
2 2 2

Let us imagine, for example, that k is much less than k (but not negligible), so that we have two weakly-coupled oscillators. In
2 1

that case equations 17.7.7 tell us that the frequencies of the two normal modes are nearly equal. What equation 17.7.13 describes,
then, is a rapid oscillation of the first mass with angular frequency (ω + ω whose amplitude is modulated with a slow angular
1

2
1 2

frequency (ω − ω ) . Equation 17.7.14 describes the same sort of motion for the second mass, except that the modulation is out
1

2
1 2

of phase by 90o with the modulation of the motion of the first mass. For a while the first mass will oscillate with a large amplitude.
This will gradually decrease, while the amplitude of the motion of the second mass increases until the motion of the first mass
momentarily ceases. After that, the amplitude of the motion of the second mass starts to decrease, while the first mass starts up
again. And so the motion continues, with the first mass and the second mass alternately taking up the motion.

This page titled 17.7: Two Masses, Three Springs, Two brick Walls is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or
curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is
available upon request.

17.7.2 https://phys.libretexts.org/@go/page/8501
17.8: Transverse Oscillations of Masses on a Taut String
A light string of length 4a is held taut, under tension F between two fixed points.

Three equal masses m are attached at equidistant points along the string. They are set into transverse oscillation of small
amplitudes, the transverse displacements of the three masses at some time being y , y and y . 1 2 3

The kinetic energy is easy. It is just

1 2 2 2
T = m(ẏ 1 + ẏ 2 + ẏ 3 ). (17.8.1)
2

The potential energy is slightly more difficult.

In the undisplaced position, the length of each portion of the string is a .
In the displaced position, the lengths of the four portions of the string are, respectively,
−−−−−−−− − −−−−−−
−−−−−−−−−−− − −−−−−−−−−−− − 2
√y 2 + 1 + a2 √(y2 − y1 )2 + a2 √(y2 − y3 )2 + a2 √y +a
2
3

For small displacements (i.e. the ys much smaller than a ), these are, approximately (by binomial expansion),
2 2 2 2
y ( y −y ) ( y2 −y3 ) y
1 2 1 1
a+ a+ a+ a+
2a 2a 2a 2a

so the extensions are

2 2 2 2
y ( y2 −y1 ) ( y −y ) y
1 2 3 3

2a 2a 2a 2a

It is also supposed that the tension in the string is F and that the displacements are sufficiently small that this is constant. The work
done in displacing the masses, which is the elastic energy stored in the string as a result of the displacements, is therefore
V =
F

2a
[y
1
2
+ (y2 − y1 )
2
+ (y2 − y3 )
2
+y ] =
2
3
F

a
(y
1
2
+y
2
2
+y
3
2
− y1 y2 − y2 y3 ) .
We note with mild irritation the presence of the cross-terms y 1 y2 , y2 y3 .
Apply Lagrange’s equation in turn to the three coordinates:

am ÿ 1 + F (2 y1 − y2 ) = 0, (17.8.2)

am ÿ 2 + F (−y1 + 2 y2 − y3 ) = 0, (17.8.3)

am ÿ 3 + F (−y2 + 2 y3 ) = 0. (17.8.4)

Seek solutions of the form ÿ 1

2
= −ω y1 , ÿ 2 = −ω y2 , ÿ 2 = −ω y3
2 2
.
Then
2
(2F − am ω )y1 − F y2 = 0, (17.8.5)

2
−F y1 + (2F − am ω )y2 − F y3 = 0, (17.8.6)

2
−F y2 + (2F − am ω )y3 = 0. (17.8.7)

Putting the determinant of the coefficients to zero gives an equation for the frequencies of the normal modes. The solutions are:

Slow Medium Fast

(2−√2F 2 2F (2−√2F
2 ω = 2
ω = 2 am ω =
1 am 3 am

17.8.1 https://phys.libretexts.org/@go/page/8502
Substitution of these into equations 17.8.5 to17.8.7 gives the following displacement ratios for these three modes:
– –
y1 : y2 : y3 = 1 : √2 : 1 1 : 0 : −1 1 : √2 : 1

These are illustrated in Figure XVII.8.

As usual, the general motion is a linear combination of the normal modes, the relative amplitudes and phases of the modes
depending upon the initial conditions.
If the motion of the first mass is a combination of the three modes with relative amplitudes in the proportion ^ : q
q 1
^ : q
2
^
3
and
α : α : α with initial phases its motion is described by
1 2 3

^ sin(ω1 t + α1 ) + q
y1 = q ^ sin(ω2 t + α2 ) + q
^ sin(ω3 t + α3 ) (17.8.8)
1 2 3

The motions of the second and third masses are then described by

–
^ sin(ω1 t + α1 )
√2 q (17.8.9)
1

and
^1 sin(ω1 t + α1 ) − q
y3 = q ^2 sin(ω2 t + α2 ) + q
^3 sin(ω3 t + α3 ). (17.8.10)

These can be written

y1 = q1 + q2 + q3 , (17.8.11)

– –
y1 = √2q1 − √2q3 (17.8.12)

and

y1 = q1 − q2 + q3 , (17.8.13)

where the q , like the y , are time-dependent coordinates.

i i

We could, if we wish, express the q in terms of the y , by solving these equations:

i i

1 –
q1 = (y1 + √2y2 + y3 ), (17.8.14)
4

1
q2 = (y1 − y3 ) (17.8.15)
2

and

17.8.2 https://phys.libretexts.org/@go/page/8502
 1 –
q3 = (y1 − √2y2 + y3 ). (17.8.16)
4

We have hitherto described the state of the system as a function of time by giving the values of the coordinates y , y and y . We 1 2 3

could equally well, if we wished, describe the state of the system by giving, instead, the values of the coordinates q , q and q . 1 2 3

Indeed it turns out that it is very useful to do so, and these coordinates are called the normal coordinates, and we shall see that they
have some special properties. Thus, if you express the kinetic and potential energies in terms of the normal coordinates, you get
1
2 2 2
T = m(4 q̇ 1
+ 2 q̇ 2
+ 4 q̇ 3
) (17.8.17)
2

and
2F – –
2 2 2
V = [(2 − √2)q + q + (2 + √2)q ]. (17.8.18)
1 2 3
a

Note that there are no cross terms. When you apply Lagrange’s equation in turn to the three normal coordinates, you obtain
–
am q̈ 1
= −(2 − √2)F q1 , (17.8.19)

am q̈ 2
= −2F q2 , (17.8.20)

and
–
am q̈ 3
= −(2 + √2)F q3 . (17.8.21)

Notice that the normal coordinates have become completely separated into three independent equations and that each is of the form
q̈ = −ω q and that each of the normal coordinates oscillates with one of the frequencies of the normal modes. Much of the art of
2

solving problems involving vibrating systems concerns identifying the normal coordinates.

This page titled 17.8: Transverse Oscillations of Masses on a Taut String is shared under a CC BY-NC 4.0 license and was authored, remixed,
and/or curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is
available upon request.

17.8.3 https://phys.libretexts.org/@go/page/8502
17.9: Vibrating String
It is possible that the three modes of vibration of the three masses in Section 17.8 reminded you of the fundamental and first two
harmonic vibrations of a stretched string – and it is quite proper that it did. If you were to imagine ten masses attached to a
stretched string and to carry out the same sort of analysis, you would find ten normal modes, of which one would be quite like the
fundamental mode of a stretched string, and the remainder would remind you of the first nine harmonics. You could continue with
the same analysis but with a very large number of masses, and eventually you would be analysing the vibrations of a continuous
heavy string. We do that now, and we assume that we have a heavy, taut string of mass μ per unit length, and under a tension F .

I show in Figure XVII.9 a portion of length δx of a vibrating rope, represented by A B in its equilibrium position and by AB in a
0 0

displaced position. The rope makes an angle ψ A with the horizontal at A and an angle ψ B with the horizontal at B. The tension in
the rope is F . The vertical equation of motion is
2
∂ y
F (sin ψB − sin ψA ) = μδx . (17.9.1)
2
∂t

2
∂y ∂ y
If the angles are small, then sin ψ ≅ ∂x
, so the expression in parenthesis is ∂x
2
δx . The equation of motion is therefore
2 2
∂ y ∂ y
2
c = (17.9.2)
∂x2 ∂t2

where
−−
T
c =√ (17.9.3)
μ

As can be verified by substitution, the general solution to this is of the form

y = f (x − ct) + g(x + ct) (17.9.4)

This represents a function that can travel in either direction along the rope at a speed c given by Equation 17.9.3. Should the
disturbance be a periodic disturbance, then a wave will travel along the rope at that speed. Further analysis of waves in ropes and
strings is generally done in chapters concerned with wave motion. This section, however, at least establishes the speed at which a
disturbance (periodic or otherwise) travels along a stretched strong or rope.

This page titled 17.9: Vibrating String is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

17.9.1 https://phys.libretexts.org/@go/page/8503
17.10: Water
Water consists of a mass M (“oxygen”) connected to two smaller equal masses m (“hydrogen”) by two equal springs of force
constants k , the angle between the springs being 2θ. The equilibrium length of each spring is r. The torque needed to increase the
angle between the springs by 2δθ is 2cδθ. See Figure XVII.10. (θ is about 52°.)

At any time, let the coordinates of the three masses (from left to right) be
(x1 , y1 ), (x2 , y2 ), (x3 , y3 )

and let the equilibrium positions be

(x10 , y10 ), (x20 , y20 ), (x30 , y30 ),  wherey30 = y10

We suppose that these coordinates are referred to a frame in which the centre of mass of the system is stationary.
Let us try and imagine, in Figure XVII.11, the vibrational modes. We can easily imagine a mode in which the angle opens and
closes symmetrically. Let is resolve this mode into an x-component and a y -component. In the x-component of this motion, one
hydrogen atom moves to the right by a distance q while the other moves to the left by and equal distance q . In the y -component
1 1

of this symmetric motion, both hydrogens move upwards by a distance q , while, in order to keep the centre of mass of the system
2

unmoved, the oxygen necessarily moves down by a distance 2mq /M. We can also imagine an asymmetric mode in which one
2

spring expands while the other contracts. One hydrogen moves down to the left by a distance q , while the other moves up to the
3

left by the same distance. In the meantime, the oxygen must move to the right by a distance (2mq sin θ)/M , in order to keep the
3

centre of mass unmoved.

We are going to try to write down the kinetic and potential energies in terms of the internal coordinates q 1, q2 and q .
3

It is easy to write down the kinetic energy in terms of the (x, y) coordinates:

17.10.1 https://phys.libretexts.org/@go/page/8504
 1 2 2
1 2 2
1 2 2
T  =   m(ẋ1   +  ẏ 1 )  +   M (ẋ2   +  ẏ 2 )  + m(ẋ3   +  ẏ 3 ). (17.10.1)
2 2 2

From geometry we have:

ẋ1 =  q̇ 1
− q̇ 3
sin θ (17.10.2)

ẏ = q̇ 2
− q̇ 3
cos θ (17.10.3)
1

2m q̇ 3
sin θ
ẋ2 = (17.10.4)
M

2mq̇ 2
ẏ 2 = − (17.10.5)
M

ẋ3 = −q̇ 1
− q̇ 3
sin θ (17.10.6)

ẏ 3 = q̇ 2
  +  q̇ 3
cos θ (17.10.7)

On putting these into equation 17.10.1 we obtain

2
2m (2m sin θ)
2 2 2
T   =  m q̇   +  m (1 + ) q̇   +  m (1 + ) q̇ (17.10.8)
1 2 3
M M

For short, I am going to write this as

2 2 2
T = a11 q̇ 1
+ a22 q̇ 2
+ a33 q̇ 3
(17.10.9)

Now for the potential energy.

The extension of the left hand spring is
2m q2 cos θ 2m q3 sin θ cos θ
δr1 = −q1 sin θ − q2 cos θ − + q3 + (17.10.10)
M M

2
1 + 2m (2m sin θ)
= −q1 sin θ − q2 ( ) cos θ + q3 (1 + )
M M

The extension of the right hand spring is

2
2m q2 cos θ 2m q3 sin θ
δr2 = −q1 sin θ − q2 cos θ − − q3 − (17.10.11)
M M

2
1 + 2m (2m sin θ)
= −q1 sin θ − q2 ( ) cos θ − q3 (1 + ).
M M

The increase in the angle between the springs is

2m
2 q1 cos θ 2(1 + )q2 sin θ
M
2δθ = −  +  . (17.10.12)
r r

The potential energy (above the equilibrium position) is

1 2
1 2
1 2
V = k(δr1 )   +   k(δr2 )   +   c(2δθ) . (17.10.13)
2 2 2

On substituting Equations 17.10.10, 17.10.11 and 17.10.12 into this, we obtain an equation of the form
2 2 2
V = b11 q   +  2b q12 q1 q2   +  b22 q   +  b33 q , (17.10.14)
1 2 3

where I leave it to the reader, if s/he wishes, to work out the detailed expressions for the coefficients. We still have a cross term, so
we can’t completely separate the coordinates, but we can easily apply Lagrange’s equation to Equations 17.10.9 and 17.10.14, and
then seek simple harmonic solutions in the usual way. Setting the determinant of the coefficients to zero leads to the following
equation for the angular frequencies of the normal modes:

17.10.2 https://phys.libretexts.org/@go/page/8504
 2
b11 − ω a11 b12 0
⎡ ⎤
2
⎢ b12 b22 − ω a22 0 ⎥   =  0. (17.10.15)

⎣ 2 ⎦
0 0 b33 − ω a33

Thus, given the masses and r, θ, k and c , one can predict the frequencies of the normal modes. Can one calculate k and c given the
frequencies? I do not know, to tell the truth. Can I leave it to the reader to investigate further?

This page titled 17.10: Water is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

17.10.3 https://phys.libretexts.org/@go/page/8504
17.11: A General Vibrating System
For convenience, I'll refer to a collection of masses connected by springs as a "molecule", and the individual masses as "atoms". In
a molecule with N atoms, the number of degrees of vibrational freedom (the number of normal modes of vibration) n = 3N − 6
for nonlinear molecules, or n = 3N − 5 for linear molecules. Three equations are needed to express zero translational motion, and
three (or two) are needed to express zero rotational motion.

While reading this Section, it might be worthwhile for the reader to follow at the same time the treatment given to the OCS
molecule in Section 17.6. Bear in mind, however, that in that section we did not consider the possibility of the molecule
bending. Indeed we treated the molecule as if it were constrained inside a drinking straw, and it remained linear at all times.
That being the case only N coordinates (rather than 3N ) suffice to describe the state of the molecule. Only one equation is
needed to express zero translational motion, and none are needed to express zero rotational motion. Thus there are N − 1
internal coordinates, and hence N − 1 normal vibrational modes. In the case of OCS, N = 3 , so there are two normal
vibrational modes.

A molecule with n degrees of vibrational freedom can be described at some instant of time by n internal coordinates q . A typical i

such coordinate may be related to the external coordinates of two atoms, for example, by some expression of the form
q = x − x − a , as we saw in our example of the molecule OCS. Its potential energy can be written in the form
2 1

2
2V   = κ11 q + κ12 q1 q2 +. . . +κ1n q1 qn (17.11.1)
1

2
+ κ21 q2 q1 + κ22 q . . . +κ2n q2 qn
2

+. . .

+κn1 qn q1 + κ12 qn q2 . . . +κnm qn qn .

Unless the q are the judiciously chosen "normal coordinates" (see our example of the transverse vibrations of three masses on an
elastic string), there will in general be cross terms, such as q q . If both qs of a term are linear displacements, the corresponding κ
1 2

is a force constant (dimensions MT-2). If both qs are angles, κ is a torsion constant (dimensions ML2T-2. If one is a linear
displacement and he other is an angular displacement, κ will be a coefficient of dimensions MLT-2.
The matrix is symmetric, so that Equation 17.11.1 could also be written
2
2V   = κ11 q + κ12 q1 q2 +. . . +2 κ1n q1 qn (17.11.2)
1

2
+ κ22 q +. . . +2 κ2n q2 qn
2

+. . .

+κnm qn qn .

In matrix notation, the Equation (i.e. Equations 17.11.1 or 17.11.2) could be written:
2V   =  q̃κq. (17.11.3)

or in vector/tensor notation,
2V   =  q ⋅ κq. (17.11.4)

The kinetic energy can be written in terms of the time rates of change of the external coordinates x : i

2 2 2
2T   =  m1 ẋ   +  m2 ẋ  +. . .   +  m3N ẋ . (17.11.5)
1 2 3N

To make use of the Lagrangian equations of motion, we need to express V and T in terms of the same coordinates, and it is usually
advantageous if these be the n internal coordinates rather than the 3N external coordinates – so that we have to deal with only n
rather than 3N lagrangian equations. (Recall that n = 3N   −  6 or 5.) The relations between the external and internal coordinates
are given as a set of equations that express a choice of coordinates such that there is no pure translation and no pure rotation of the
molecule. These equations are of the form
q  =  Ax. (17.11.6)

Here q is an n × 1 column matrix, x a 3N × 1 column matrix, and A is a matrix with n rows and 3N columns, and it may need a
little trouble to set up. We could then use this to express V in terms of the external coordinates, so we would then have both V and

17.11.1 https://phys.libretexts.org/@go/page/8505
T in terms of the external coordinates. We could then apply Lagrange’s equation to each of the 3N external coordinates and arrive
at 3N simultaneous differential equations of motion.
A better approach is usually to set up the equations connecting q̇ and ẋ:

q̇  =  Bẋ. (17.11.7)

(These correspond to Equations 17.6.3 and 17.6.4 in our example of the linear triatomic molecule in Section 17.6.) We then want to
invert Equations 17.11.7 in order to express ẋ in terms of q̇ . But we can’t do this, because B is not a square matrix. ẋ has 3N
elements while q̇ has only n . We have to add an additional six (or five for linear molecules) equations to express zero pure
translational and zero pure rotational motion. This adds a further 6 or 5 rows to B, so that B is now square (this corresponds to
Equation 17.6.6 ), and we can then invert Equation 17.11.7:
−1
ẋ  =  B q̇ (17.11.8)

(This corresponds to Equation 17.6.7.)

By this means we can express the kinetic energy in terms of the time rates of change of only the n internal coordinates:
2
2T   = μ11 q̇ + μ12 q̇ 1
q̇ 2
+. . . +μ1n q̇ 1
q̇ n
(17.11.9)
1

2
+ μ21 q̇ 2
q̇ 1
+ μ22 q̇ 2
... +μ2n q̇ 2
q̇ n

+. . .

+μn1 q̇ n
q̇ 1
+ μ12 q̇ n
q̇ 2
... +μnm q̇ n
q̇ n

Since the matrix is symmetric, the equation could also be written in a form analogous to Equation 17.11.2. The equation can also
be written in matrix notation as
~
2T = q̇μq̇. (17.11.10)

or in vector/tensor notation,

2T = q̇ ⋅ μq̇. (17.11.11)

Here the μ are functions of the masses. If both qs in a particular term have the dimensions of a length, the corresponding μ and κ
ij

will have dimensions of mass and force constant. If both qs are angles, the corresponding μ and κ will have dimensions of
rotational inertia and torsion constant.. If one q is a length and the other is an angle, the corresponding μ and κ will have
dimensions ML and MLT-2.
Apply Lagrange's equation successively to q 1  , . . . , qn to obtain n equations of the form
μ11 q̈ 1
 +. . .   +  μ1n q̈ n
  +  κ11 q1  +. . .   + κ1n qn   =  0. (17.11.12)

That is to say
μq̈  =   − κq. (17.11.13)

Seek simple harmonic solutions of the form q̈   =   − ω 2

q

and we obtain n equations of the form

2 2
(κ11 − μ11 ω )c q1  +. . .   +  (κ1n − μ1n ω )qn   =  0. (17.11.14)

The frequencies of the normal modes can be obtained by equating the determinant of the coefficients to zero, and hence the
displacement ratios can be determined.
If N is large, this could be a formidable task. The work can be very much reduced by making use of symmetry relations of the
molecule, in which case the determinant of the coefficients may be factored into a number of much smaller subdeterminants.
Further, if the configuration of the molecule could be expressed in terms of normal coordinates (combinations of the internal
coordinates) such that the potential energy contained no cross terms, the equations of motion for each normal coordinate would be
in the form q̈   =   − ω q .
2

This page titled 17.11: A General Vibrating System is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

17.11.2 https://phys.libretexts.org/@go/page/8505
17.12: A Driven System
It would probably be useful before reading this and the next section to review Chapters 11 and 12.

Figure XVII.12 shows the same system as figure XVII.2, except that, instead of being left to vibrate on its own, the second mass is
subject to a periodic force F   =  F^ sin ωt . For the time being, we’ll suppose that there is no damping. Either way, it is not a
conservative force, and Lagrange’s equation will be used in the form of Equation 13.4.12. As in Section 17.2, the kinetic energy is
1 1
2 2
T = m1 ẋ   +   m2 ẋ (17.12.1)
1 2
2 2

Lagrange’s equations are

d ∂T ∂T
 −    =  P1 (17.12.2)
dt ∂ẋ1 ∂x1

and
d ∂T ∂T
 −    =  P2 . (17.12.3)
dt ∂ẋ2 ∂x2

We have to identify the generalized forces P and P . 1 2

In the nonequilibrium position, the extension of the left hand spring is x and so the tension in that spring is f   =  k x . The
1 1 1 1

extension of the right hand spring is x   −  x and so the tension in that spring is f   =  k (x − x ) . If x were to increase by
2 2 2 2 2 1 1

δx , the work done on m

1 would be (f − f )δx and therefore the generalized force associated with the coordinate x is
1 2 1 1 1

P   =  k (x − x ) − k x
1 2 2 1 1 1 . If x were to increase by δx , the work done on m would be (F − f )δx and therefore the
2 2 2 2 2

generalized force associated with the coordinate x is P = F^ sin ωt − k (x − x ) . The lagrangian equations of motion therefore
2 2 2 2 1

become
m1 ẍ1   +  (k1 + k2 )x1   − k2 x2   =  0 (17.12.4)

and
^ sin ωt.
m2 ẍ2   +  k2 (x2 − x1 )  =  F (17.12.5)

Seek solutions of the form ẍ 1 = −ω x1

2
and ẍ 2 = −ω x2
2
. The equations become
2
(k1 + k2 − m1 ω )x1   −  k2 x2   =  0 (17.12.6)

and
2 ^
−k2 x1   +  (k2   −  m2 ω )x2   =  F sin ωt. (17.12.7)

We do not, of course, now equate the determinants of the coefficients to zero (why not?!), but we can solve these equations to
obtain
^
k2 F sin ωt
x1   =   (17.12.8)
2 2 2
(k1 + k2 − m1 ω )(k2 − m2 ω ) − k
2

and

17.12.1 https://phys.libretexts.org/@go/page/8506
 2 ^
(k1 + k2 − m1 ω )F sin ωt
x2   =   . (17.12.9)
2 2 2
(k1 + k2 − m1 ω )(k2 − m2 ω ) − k
2

The amplitudes of these motions (and how they vary with the forcing frequency ω) are
^
k2 F
^1   =  
x (17.12.10)
4 2
m1 m2 ω   −  (m1 k2 + m2 k1 + m2 k2 )ω + k1 k2

and
2 ^
(k1 + k2 − m1 ω )F
^2   =  
x (17.12.11)
m1 m2 ω4   −  (m1 k2 + m2 k1 + m2 k2 )ω2 + k1 k2

where I have re-written the denominators in the form of a quadratic expression in ω . 2

For illustration I draw, in figure XVII.13, the amplitudes of the motion of m (continuous curve, in black) and of m (dashed curve,
1 2

in blue) for the following data:

^
F = 1,  k1 = k2 = 1,  m1 = 3,  m2 = 2,

when the equations become

1 1
^1 =
x = (17.12.12)
4 2 2
6ω − 7ω +1 (6 ω − 1)(ω2 − 1)

and
2 2
2 − 3ω 2 − 3ω
^1 =
x = (17.12.13)
4 2
6ω − 7ω +1 (6 ω2 − 1)(ω2 − 1)

Where the amplitude is negative, the oscillations are out of phase with the force F . The amplitudes go to infinity (remember we are
assuming here zero damping) at the two frequencies where the denominators of Equations 17.12.10 and 17.12.11 are zero. The
amplitude of the motion of m is zero when the numerator of Equation 17.12.11 is zero. This is at an angular frequency of
2
−−−−−
( k1 +k2 )
√
m1
, which is just the angular frequency of the motion of m1 held by the two springs between two fixed points. In our
−
−
numerical example, this is ω  =  √ 2

3
  =  0.8165 . This is an example of antiresonance.

17.12.2 https://phys.libretexts.org/@go/page/8506
This page titled 17.12: A Driven System is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

17.12.3 https://phys.libretexts.org/@go/page/8506
17.13: A Damped Driven System
I’ll leave the reader to add some damping to the system described in Section 17.12. Let us here try it with the system described in
Section 17.7. We’ll apply a periodic force to the left hand mass, and we’ll suppose that the damping constant for each mass is
γ =
b

m
. We could write the periodic force as F = F^ sin ωt , but the algebra will be easier if we write it as F = F^e . If the initial iωt

condition is such that F = 0 when t = 0 , then we choose just the imaginary part of this in subsequent expressions.
The equations of motion are
mẍ = - the damping force bẋ
- the tension in the left hand spring k 1x

+ the force F
+ the tension in the middle spring k 2 (y − x)

(this last is a thrust whenever y < x )

and
mÿ - the damping force bẏ
- the thrust in the right hand spring k 1y

- the tension in the middle spring k 2 (y − x)

That is,
^ iωt
m ẍ  +  b ẋ  +  (k1 + k2 )x  −  k2 y  =  F e (17.13.1)

and

m ÿ   +  b ẏ   +  (k1 + k2 )y  −  k2 y  =  0. (17.13.2)

For the steady-state motion, seek solutions of the form

ẍ = −ω x
2
, ÿ = −ω y
2
, so that ẋ = iωx and ẏ = iωy .
The equations then become
2 ^ iωt
(k1   +  k2   − m ω   +  ibω)x  −  k2 y  =  F e (17.13.3)

and
2
− k2 x  +  (k1   +  k1   − m ω   +  ibω)y  =  0. (17.13.4)

There is now a little algebra to be carried out. Solve these equations for x and y , and when, in doing so, there is a complex number
in the denominator, multiply top and bottom by the conjugate in the usual way, so as to get x and y in the forms x   +  i x and ′ ′′

′
y   +  i y . Then find expressions for the amplitudes x
′′
^ and y^ . After some algebra, the amount of which depends on one’s skill,

experience and luck (it is not always obvious how to gather terms in the most economical way, and you need some luck in this) you
eventually get, for the amplitudes of the motion
2
2 2 2 ^
2
((k1   +  k2   − m ω )  +  b ω )F
^
x = (17.13.5)
2 2 2 2 2 2 2 2
((k1 − m ω ) +  b ω )((k1 + 2 k2 − m ω ) +  b ω )

and

17.13.1 https://phys.libretexts.org/@go/page/8507
 2
2 ^
k F
2 2
^
y = . (17.13.6)
2 2 2 2 2 2 2 2
((k1 − m ω ) +  b ω )((k1 + 2 k2 − m ω ) +  b ω )

There are many variables in these expressions, but in order to see qualitatively what the steady state motion is like, I’m going to put
F , m and k = 1 . I think if I also put b = 1 , this will give light damping in the sense described in Chapter 11. As for k , I am
^
1 2

k2
going to introduce a coupling coefficient α defined by α  =  
k1 +k2
or k2   =   (
α

1−α
) k1 . This coupling constant will be close to
zero if the middle spring is very weak, and 1 if the middle connector is a rigid rod. The equations now become
1 2 2 2
((   − ω )   +  ω )
2 1−α
x
^ = . (17.13.7)
1+α
2 2 2 2 2 2
((1 − ω ) +  ω )( −ω ) +  ω )
1−α

and
α

2 (1−α)
^
y = (17.13.8)
2 2 2 1+α 2 2 2
((1 − ω ) +  ω )( −ω ) +  ω )
1−α

For computational efficiency you might want to rewrite these equations a little. For example you could write (1 − ω )   +  ω as 2 2 2

1  −  Ω(1 − Ω) , where Ω  =  ω . In any case, figure XVII.15 shows the amplitudes of the motions of the two masses as a function
2

of frequency, for α  =  0.1,  0.5 and 0.9. The continuous black curves are for the left hand mass; the dashed blue curve is for the
right hand mass.

This page titled 17.13: A Damped Driven System is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

17.13.2 https://phys.libretexts.org/@go/page/8507
 CHAPTER OVERVIEW
18: The Catenary
If a flexible chain or rope is loosely hung between two fixed points, it hangs in a curve that looks a little like a parabola, but in fact
is not quite a parabola; it is a curve called a catenary, which is a word derived from the Latin catena, a chain.
18.1: Introduction
18.2: The Intrinsic Equation to the Catenary
18.3: Equation of the Catenary in Rectangular Coordinates, and Other Simple Relations
18.4: Area of a Catenoid

This page titled 18: The Catenary is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1
18.1: Introduction
If a flexible chain or rope is loosely hung between two fixed points, it hangs in a curve that looks a little like a parabola, but in fact
is not quite a parabola; it is a curve called a catenary, which is a word derived from the Latin catena, a chain.

This page titled 18.1: Introduction is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

18.1.1 https://phys.libretexts.org/@go/page/7044
18.2: The Intrinsic Equation to the Catenary
We consider the equilibrium of the portion AP of the chain, A being the lowest point of the chain (Figure XVIII.1).

It is in equilibrium under the action of three forces: The horizontal tension T at A; the tension T at P, which makes an angle ψ
0

with the horizontal; and the weight of the portion AP. If the mass per unit length of the chain is μ and the length of the portion AP
is s , the weight is μsg. It may be noted than these three forces act through a single point.
Clearly,

T0 = T cos ψ (18.2.1)

and

μsg =  T sin ψ, (18.2.2)

from which
2 2 2
(μsg ) +T =  T (18.2.3)
0

and
μgs
tan ψ = (18.2.4)
T0

Introduce a constant a having the dimensions of length defined by

T0
a = . (18.2.5)
μg

Then Equations 18.2.3 and 18.2.4 become

−−−−−−−
2 2
T   =  μg √ s   +  a (18.2.6)

and
s  =  a tan ψ. (18.2.7)

Equation 18.2.7 is the intrinsic equation of the catenary.

This page titled 18.2: The Intrinsic Equation to the Catenary is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated
by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

18.2.1 https://phys.libretexts.org/@go/page/7045
18.3: Equation of the Catenary in Rectangular Coordinates, and Other Simple
Relations
2
dy d y
The slope at some point is y
′
=
dx
= tan ψ =
a
s
from which ds

dx
  =  a
dx
2
. But, from the usual pythagorean relation between
1

intrinsic and rectangular coordinates ds = (1 + y ′2

) 2 dx , this becomes
′
1 dy
′2
(1 + y ) 2 =a . (18.3.1)
dx

On integration, with the condition that y ′

=0 where x = 0 , this becomes
x
′
y = sinh( ), (18.3.2)
a

and, on further integration,

x
y  =  a cosh( )  +  C (18.3.3)
a

If we fix the origin of coordinates so that the lowest point of the catenary is at a height a above the x-axis, this becomes
x
y  =  a cosh( ) (18.3.4)
a

This, then, is the x, y Equation to the catenary. The x-axis is the directrix of this catenary.
The following additional simple relations are easily derived and are left to the reader:
x
s  =  a sinh( ) (18.3.5)
a

2 2 2
y   =  a   +  s  , (18.3.6)

y  =  a sec ψ. (18.3.7)

x = a ln  (sec ψ + tan ψ (18.3.8)

T = μgy (18.3.9)

Equations 18.3.7 and 18.3.8 may be regarded as parametric Equations to the catenary.
If one end of the chain is fixed, and the other is looped over a smooth peg, Equation 18.3.9 shows that the loosely hanging vertical
portion of the chain just reaches the directrix of the catenary, and the tension at the peg is equal to the weight of the vertical portion.

 Exercise 18.3.1
By expanding Equation 18.3.4 as far as x , show that, near the bottom of the catenary, or for a tightly stretched catenary with a
2

small sag, the curve is approximately a parabola. Actually, it does not matter what Equation 18.3.4 is – if you expand it as far
as x , provided the x term is not zero, you’ll get a parabola – so, in order not to let you off so lightly, show that the semi latus
2 2

rectum of the parabola is a .

 Exercise 18.3.2
Expand Equation 18.3.5 as far as x . 3

Now: let 2s= total length of chain, 2k = total span, and d = sag. Show that for a shallow catenary 3
s − k = k /(6 a )
3
and
k
2
= 2ad hence that length − span = sag2/span.8

 Example 18.3.1
A cord is stretched between points on the same horizontal level. How large a force must be applied so that the cord is no longer
a catenary, but is accurately a straight line?
Answer:

18.3.1 https://phys.libretexts.org/@go/page/7046
 There is no force however great
Can stretch a string however fine
Into a horizontal line
That shall be accurately straight.

I am indebted to Hamilton Carter of Texas A & M University for drawing my attention to a note by C. A. Chant in J. Roy. Astron.
Soc. Canada 33, 72, (1939), where this doggerel is attributed to the early nineteenth century Cambridge mathematician William
Whewell.

 Exercise 18.3.3

And here’s something for engineers. We, the general public, expect engineers to built safe bridges for us. The suspension chain
of a suspension bridge, though scarcely shallow, is closer to a parabola than to a catenary. There is a reason for this. Discuss.

This page titled 18.3: Equation of the Catenary in Rectangular Coordinates, and Other Simple Relations is shared under a CC BY-NC 4.0 license
and was authored, remixed, and/or curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts
platform; a detailed edit history is available upon request.

18.3.2 https://phys.libretexts.org/@go/page/7046
18.4: Area of a Catenoid
A theorem from the branch of mathematics known as the calculus of variations is as follows. Let y = y(x) with y ′
= dy/dx and
let f (y, y , x) be some function of y, y and x. Consider the line integral of f from A to B along the route y = y(x).
′ ′

B
′
∫ f (y, y , x)dx (18.4.1)
A

In general, and unless f is a function of x and y alone, and not of y , the value of this integral will depend on the route (i.e.
′

y = y(x) ) over which this line integral is calculated. The theorem states that the integral is an extremum for a route that satisfies

d ∂f ∂f
= (18.4.2)
′
dx ∂y ∂y

By "extremum" we mean either a minimum or a maximum, or an inflection, though in many – perhaps most – cases of physical
interest, it is a minimum. It can be difficult for a newcomer to this theorem to try to grasp exactly what this theorem means, so
perhaps the best way to convey its meaning is to start by giving a simple example. Following that, I give an example involving the
catenary. There will be another example, involving a famous problem in dynamics, in Chapter 19, and in fact we have already
encountered an application of it in Chapter 14 in the discussion of Hamilton's variational principle.
Let us consider, for example, the problem of calculating the distance, measured along some route y(x) between two points; that is,
we want to calculate the arc length ∫ ds From the usual pythagorean relation between ds , dx and dy , this is ∫ (1 + y ) dx . The ′ 1/2

variational principle says that this distance – measured along y(x) – is least for a route y(x) that satisfies Equation 18.4.2, in which
in this case f = (1 + y )′ 1/2

′
df df y
For this case, we have dy
=0 and dy
=
′ 1/2
. Thus integration of Equation 18.4.2 gives
(1+y )

′ ′2 1/2
y = c(1 + y ) , (18.4.3)

where c is the integration constant. If we solve this for y

′
, we obtain which is just another constant, which I'll write as a , so that
y = a . Integrate this to find
′

y = ax + b (18.4.4)

This probably seems rather a long way to prove that the shortest distance between two points is a straight line – but that wasn't the
point of the exercise. The aim was merely to understand the meaning of the variational principle.
Let's try another example, in which the answer will not be so obvious.
Consider some curve y = y(x), and let us rotate the curve through an angle ϕ (which need not necessarily be a full (\2 \pi \)
−−− −−−−−
radians) about the y -axis. An element ds of the curve can be written as √1 + y dx and the distance moved by the element ds ′2

(which is at a distance x from the y -axis) during the rotation is ϕx. Thus the area swept out by the curve is
−−−−−−
′2
A =ϕ∫ x√ 1 + y dx. (18.4.5)

For what shape of curve, y = y(x) , is this area least? The answer is – a curve that satisfies Equation 18.4.2 , where
−−−−− − ∂f ∂f xy
′

f = x √1 = y
′2
. For this function, we have ∂y
=0 and ∂y
=
√1+y ′2
.

Therefore the required curve satisfies

′
xy
−−−−− − = a. (18.4.6)
′2
√1 + y

That is,
dy a
= − −− −−−. (18.4.7)
dx √ x2 − a2

On substitution of x = a cosh θ and looking up everything we have forgotten about hyperbolic functions, and integrating, we
obtain

18.4.1 https://phys.libretexts.org/@go/page/7047
 y = a cosh(x/a). (18.4.8)

Thus the required curve is a catenary.

If a soap bubble is formed between two identical horizontal rings, one beneath the other, it will take up the shape of least area,
namely a catenoid.

This page titled 18.4: Area of a Catenoid is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

18.4.2 https://phys.libretexts.org/@go/page/7047
 CHAPTER OVERVIEW
19: The Cycloid

A cycloid generated by a rolling circle. (CC BY-SA 3.0; Zorgit).

19.1: Introduction to Cycloids
19.2: Tangent to the Cycloid
19.3: The Intrinsic Equation to the Cycloid
19.4: Variations
19.5: Motion on a Cycloid, Cusps Up
19.6: Motion on a Cycloid, Cusps Down
19.7: The Brachystochrone Property of the Cycloid
19.8: Contracted and Extended Cycloids
19.9: The Cycloidal Pendulum
19.10: Examples of Cycloidal Motion in Physics

This page titled 19: The Cycloid is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1
19.1: Introduction to Cycloids
Let us set up a coordinate system Oxy, and a horizontal straight line y = 2a . We imagine a circle of diameter 2a between the x-
axis and the line y = 2a , and initially the lowest point on the circle, P, coincides with the origin of coordinates O. We now allow
the circle to roll counterclockwise without slipping on the line y = 2a , so that the centre of the circle moves to the right. As the
circle rolls on the line, the point P describes a curve, which is known as a cycloid.

When the circle has rolled through an angle 2θ, the centre of the circle has moved to the right by a horizontal distance 2aθ, while
the horizontal distance of the point P from the centre of the circle is a sin 2θ and the vertical distance of the point P below the
centre of the circle is a cos 2θ . Thus the coordinates of the point P are

x = a(2θ + sin 2θ) (19.1.1)

and

y = a(1 − cos 2θ). (19.1.2)

Equations 19.1.1 and 19.1.2 are the parametric equations of the cycloid. Using a simple trigonometric identity, Equation 19.1.2

can also be written

2
y = 2a sin θ. (19.1.3)

 Example 19.1.1

When the x-coordinate of P is 2.500a , what (to four significant figures) is its y -coordinate?

Solution
We have to find 2θ by solution of 2θ + sin 2θ . By Newton-Raphson iteration or otherwise, we find 2θ = 0.931 599 201
radians, and hence y = 0.9316a .

This page titled 19.1: Introduction to Cycloids is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

19.1.1 https://phys.libretexts.org/@go/page/7051
19.2: Tangent to the Cycloid
The slope of the tangent to the cycloid at P is dy/dx , which is equal to dy/dθ , and these can be obtained from Equations 19.1.1
and 19.1.2.

 Exercise 19.2.1
Show that the slope of the tangent at P is tan θ . That is to say, the tangent at P makes an angle θ with the horizontal.

Having done that, now consider the following:

y
Let A be the lowest point of the circle. The angle ψ that AP makes with the horizontal is given by tan ψ = x−2aθ

 Exercise 19.2.2
Show that ψ = θ . Therefore the line AP is the tangent to the cycloid at P; or the tangent at P is the line AP.

This page titled 19.2: Tangent to the Cycloid is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

19.2.1 https://phys.libretexts.org/@go/page/7052
19.3: The Intrinsic Equation to the Cycloid
An element ds of arc length, in terms of dx and dy , is given by the theorem of Pythagoras: ds = ((dx ) + (dy ) ))2
or, since x
2 1/2

and y are given by the parametric Equations 19.1.1 and 19.1.2, by And of course we have just shown that the intrinsic coordinate ψ
(i.e. the angle that the tangent to the cycloid makes with the horizontal) is equal to θ .

 Exercise 19.3.1

Integrate ds (with initial condition s = 0, θ = 0) to show that the intrinsic equation to the cycloid is
s = 4a sin ψ (19.3.1)

Also, eliminate ψ (or θ ) from Equations 19.3.1 and 19.1.2 to show that the following relation holds between arc length and
height on the cycloid:
2
s = 4ay. (19.3.2)

This page titled 19.3: The Intrinsic Equation to the Cycloid is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

19.3.1 https://phys.libretexts.org/@go/page/7053
19.4: Variations
In Sections 19.1,2,3, we imagined that the cycloid was generated by a circle that was rolling counterclockwise along the line
y = 2a . We can also imagine variations such as the circle rolling clockwise along y = 0 , or we can start with P at the top of the

circle rather than at the bottom. I summarise in this section four variations. The distinction between ψ and θ is as follows. The
angle that the tangent to the cycloid makes with the positively-directed x-axis is ψ ; that is to say, dx/dy = tanψ .The circle rolls
through an angle 2θ. There is a simple relation between ψ and θ , which is different for each case.
In each figure, x and y are plotted in units of a . The vertical height between vertices and cusps is 2a , the horizontal distance
between a cusp and the next vertex is πa, and the arc length between a cusp and the next vertex is 4a.
I. Circle rolls counterclockwise along y = 2a . P starts at the bottom. The cusps are up. A vertex is at the origin.

x = a(2θ + sin 2θ)

2
y = 2a sin θ (19.4.2)

s = 4a sin θ (19.4.3)

2
= 8ay (19.4.4)

ψ = θ. (19.4.5)

II. Circle rolls clockwise along y = 0 . P starts at the bottom. The cusps are down. A cusp is at the origin.
x = a(2θ − sin2θ) (19.4.6)

2
y = 2asi n θ (19.4.7)

s = 4a(1 − cosθ) (19.4.8)

2
s = 8a(y − s) (19.4.9)

ψ = 90 ∘ −θ. (19.4.10)

III. Circle rolls clockwise along y = 0 . P starts at the top. The cusps are down. A vertex is at x = 0 .
x = a(2θ + sin 2θ (19.4.11)

2
y = 2a cos θ (19.4.12)

s = 4a sin θ (19.4.13)

2
s = 8a(2a − y) (19.4.14)

ψ = 180 ∘ −θ. (19.4.15)

IV. Circle rolls counterclockwise along y = 2a . P starts at the top. The cusps are up. A cusp is at x = 0 .

19.4.1 https://phys.libretexts.org/@go/page/7054
 x = a(2θ − sin 2θ) (19.4.16)

2
y = 2a cos θ (19.4.17)

s = 4a(1 − cos θ) (19.4.18)

2
s − 8as + 8a(2a − y) = 0 (19.4.19)

ψ = 90 ∘ +θ (19.4.20)

This page titled 19.4: Variations is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

19.4.2 https://phys.libretexts.org/@go/page/7054
19.5: Motion on a Cycloid, Cusps Up
We shall imagine either a particle sliding down the inside of a smooth cycloidal bowl, or a bead sliding down a smooth cycloidal
wire, Figure XIX.6.

We shall work in intrinsic coordinates to obtain the tangential and normal Equations of motion. These Equations are, respectively:
s̈ = −g sin ψ (19.5.1)

and
2
mv
= R − mg cos ψ. (19.5.2)
ρ

Here R is the normal (and only) reaction of the bowl or wire on the particle and ρ is the radius of curvature. The radius of curvature
is ds/dψ , which, from Equation 19.3.1, (or Equations 19.4.3 and 19.4.5) is
ρ = 4a cos ψ (19.5.3)

From Equations 19.3.1 and 19.5.1 we see that the tangential Equation of motion can be written, without approximation:
g
s̈ = − s. (19.5.4)
4a

−−−
This is simple harmonic motion of period 4π √a/g, independent of the amplitude of the motion. This is the isochronous property
−−−
of the cycloid. Likewise, if the particle is released from rest, it will reach the bottom of the cycloid in a time π √a/g whatever the
starting position.
Let us see if we can find the value of R where the generating angle is ψ . Let us suppose that the particle is released from rest at a
height y above the x-axis (generating angle = ψ ); what is its speed v when it has reached a height y (generating angle ψ )?
0 0

Clearly this is given by

1
2
mv = mg(y0 − y), (19.5.5)
2

and, following Equation 19.3.2, and recalling that θ = ψ , this is

2
v = 2ga(cos2ψ − cos2 ψ0 ). (19.5.6)

On substituting this and Equation 19.5.3 into Equation 19.5.2, we find for R:
mg
R = (1 + 2 cos 2ψ − cos 2 ψ0 ) (19.5.7)
2 cos ψ

This page titled 19.5: Motion on a Cycloid, Cusps Up is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

19.5.1 https://phys.libretexts.org/@go/page/7055
19.6: Motion on a Cycloid, Cusps Down
We imagine a particle sliding down the outside of an inverted smooth cycloidal bowl, or a bead sliding down a smooth cycloidal
wire. We shall suppose that, at time t = 0 , the particle was at the top of the cycloid and was projected forward with a horizontal
velocity v . See Figure XIX.7.
0

This time, the equations of motion are

s̈ = gsinψ (19.6.1)

and
2
mv
= mg cos ϕ − R. (19.6.2)
ρ

By arguments similar to those made in Section 19.5, we find that

gs
s̈ = (19.6.3)
4a

The general solution to this is

pt −pt
s = Ae + Be , (19.6.4)

where
−−−−−
p = √g/(2a). (19.6.5)

With the initial condition given (at t = 0, s = 0, ṡ = v0 ), we can find A and B and hence:
−
−
a
pt −pt
s = v0 √ (e −e ) (19.6.6)
g

Again proceeding as in Section 19.5, we find for R :

m
2
R = (4ga cos 2ψ − v ). (19.6.7)
0
4 cos ψ

So – what happens?
If the constraint is two-sided (bead sliding on a wire) R becomes zero when cos 2π = v 2
0
/(2/ga), and thereafter R is in the opposite
direction.
If the constraint is one-sided (particle sliding down the outside of a smooth cycloidal bowl):
1. If v   > 4ga , the particle loses contact at the moment of projection.
2
0

2. If If v   < 4ga the particle loses contact as soon as cos 2π = v /(2ga), is very small (i.e. very much smaller than √(2ga) ),
2
0
2
0

this will happen when ψ = 45∘ ; for faster initial speeds, contact is lost sooner.

19.6.1 https://phys.libretexts.org/@go/page/7056
  Example 19.6.1

A particle is projected horizontally with speed v0 = 1 m s−1 from the vertex of the smooth cycloidal hill
x = a(2θ + sin 2θ

2
y = 2a cos θ,

−2
where a = 2 m. Assuming that g = 9.8 m s , how long does it take to get halfway down the hill (i.e. to y = a )?

Solution
We have to use Equation 19.6.6. With the numerical data given, this is
1.565248t −1.565248t
s = 0.451754(e −e ).

We can find s from Equation 19.4.12, which gives us s = 2.828427 m. If we let we now have to solve 6.26099 = ξ − 1/ξ , or
ξ − 6.26099ξ − 1 = 0 . From this, ξ = 6.41683 and hence t = 1.19 s.
2

I leave it to the reader to calculate R at this time – and indeed to see whether the particle loses contact with the hill before then.
Perhaps the fact that I got a positive real root for ξ means that we are all right and the particle is still in contact – but I wouldn't
be sure of that. I leave it to the reader to investigate further.

This page titled 19.6: Motion on a Cycloid, Cusps Down is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

19.6.2 https://phys.libretexts.org/@go/page/7056
19.7: The Brachystochrone Property of the Cycloid
A small point. The word is sometimes spelled brachistochrone, and I have no recommendation one way or the other. For what it's
worth, the only dictionary within easy reach of my desk has brachiopod and brachycephalic. In any case, the word is derived from
Greek, and means shortest time.
The famous brachystochrone problem problem is this: A smooth wire, which can be of any desired length, is to connect two points
O and P; P is at a lower level than O, but is not vertically below O. The wire is to be bent to a shape, and cut to a length, such that
the time taken for a bead to slide down the wire from O to P is least.
It is not easy to prove that the required curve is a cusps-up cycloid; but it is quite reasonable to speculate or to guess that this might
be so. And, having speculated that it might be a cycloid, it is easy to verify that the required curve is indeed a cusps-up cycloid, the
bead starting from rest at a cusp of the cycloid.
A speculation might go something like this. Generally one would expect that the further P is from O, the longer it will take for the
bead to slide from O to P. But, if O and P are connected with a cycloidal wire, the time taken to go from O to P does not increase
with distance. (See the isochronous property of the cycloid discussed in Section 19.5.) Thus, as you increase the distance between
O and P, the time taken to travel by any route other than the cycloidal one must take longer than the cycloidal route. This argument
may not sound like a rigorous proof, though it is enough to arouse our suspicions and to test whether it is correct.
Since I am going to deal with a bead sliding downwards under gravity, I am going to find it convenient to set up our coordinate
axes such that x increases to the right, and y increases downwards. In that case, the parametric equations to a cusps-up cycloid,
with the origin at a cusp, are
x = a(2θ − sin 2θ) (19.7.1)

and
2
y = 2a sin θ (19.7.2)

− and these are the equations that we shall be testing.

The time taken for the bead to travel a distance ds along the wire, while it is moving at speed v is ds/v . In (x, y) coordinates, ds
− −−−−−
is √1 + y dx where y = dy/dx . Also, the speed reached is related (by equating the gain in kinetic energy to the loss of potential
′2 ′

−−
−
energy) to the vertical distance y dropped by v = √2gy .Thus the time taken to go from O to P is
−−−
P ′2 P
1 √1y 1 ′
– =∫ dx = ∫ f (y, y )dx. (19.7.3)
√2g 0 √y 2g O

This is least (see Chapter 18 for a discussion of this theorem from the calculus of variations) for a function y(x) that satisfies
d ∂f ∂f
= . (19.7.4)
′
dx ∂y ∂y

We have:
′2
1 +y
f = , (19.7.5)
√y

′2 1/2
∂f (1 + y )
=− (19.7.6)
3/2
∂y 2y

′
∂f y
=− (19.7.7)
∂y 1/2 ′2 1/2
y (1 + y )

It is left for the reader to see whether equations 19.7.1 and 19.7.2 satisfy equation 19.7.4. You should find that both sides of the
equation are equal to Thus our speculation is confirmed, and a cusps-up cycloid is indeed the curve that offers passage from O to P
in the shortest time.

This page titled 19.7: The Brachystochrone Property of the Cycloid is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or
curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is

19.7.1 https://phys.libretexts.org/@go/page/8520
available upon request.

19.7.2 https://phys.libretexts.org/@go/page/8520
19.8: Contracted and Extended Cycloids
As in Section 19.1, we consider a circle of radius a rolling to the right on the line y = 2a . The point P is initially below the centre
of the circle, but, instead of being on the rim of the circle, its distance from the centre of the circle is r. If r < a , the path described
by P will be a contracted cycloid; if r > a , the path is an extended cycloid. (I think there’s a case for using this nomenclature the
other way round, but most authors seem to use “contracted” for r < a and “extended” forr > a .) It should not take long to be
convinced, by arguments similar to those in Section 19.1, that the parametric equations to a contracted or extended cycloid are
x = 2aθ + r sin 2θ (19.8.1)

and

y = a − r cos 2θ (19.8.2)

These are illustrated in Figures XIX.8 and XIX.9 for a contracted cycloid with r = 0.5a and an extended cycloid with r = 1.5a .

This page titled 19.8: Contracted and Extended Cycloids is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

19.8.1 https://phys.libretexts.org/@go/page/8521
19.9: The Cycloidal Pendulum
Let us imagine building a wooden construction in the shape of the cycloid

x = a(2θ − sin 2θ) (19.9.1)

2
y = 2a cos θ (19.9.2)

shown with the thick line in Figure XIX.10. Now suspend a pendulum of length 4a from the cusp, and allow it to swing to and fro,
partially wrapping itself against the wooden frame as it does so. If the arc length from the cusp to P is s , then the length of the
“free” string is 4a − s , and so the coordinates of the bob at the end of the pendulum are
∘
x = a(2θ − sin 2θ) + (4a − s) cos(180 − ψ)
(19.9.3)
= a(2θ − sin 2θ) + (4a − s) sin θ.

and
2 ∘
y = 2a cos θ − (4a − s) sin(180 − ψ)
(19.9.4)
2
= 2a cos θ − (4a − s) cos θ

(You will need to remind yourself of the exact meaning of ψ and also make use of Equation 19.4.20.) Now Equation 19.4.18 tells
us that , and, on substitution of this in equations 19.9.3 and 19.9.4, we find (after a very little algebra and trigonometry) for the
parametric equations to the path described by the bob of the pendulum:
x = a(2θ + sin 2θ (19.9.5)

and
2
y = −2a cos θ. (19.9.6)

Thus the path of the pendulum bob (shown as a dashed line in Figure XIX.10) is a cycloid, and hence its period is independent of
its amplitude. (Recall Section 19.5.) Thus the pendulum is isochronous or tautochronous. It is astonishing to learn that Huygens
constructed just such a pendulum as long ago as 1673.

Figure 19.9.1 : Five isochronous cycloidal pendula with different amplitudes. All of them are isochronous, meaning they have the
same frequency regardless of their amplitudes. Notice the two upper cycloidal arcs which make the bobs describe cycloidal
trajectories. (CC BY-SA 4.0; Rem088roy).

19.9.1 https://phys.libretexts.org/@go/page/8522
This page titled 19.9: The Cycloidal Pendulum is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

19.9.2 https://phys.libretexts.org/@go/page/8522
19.10: Examples of Cycloidal Motion in Physics
Several examples of cycloidal motion in physics come to mind. One is the nutation of a top, which is described in Section 4.10 of
Chapter 10. Earth’s axis nutates in a similar fashion. Another well known example is the motion of an electron in crossed electric
and magnetic fields. This is described in Chapter 8 of the Electricity and Magnetism section of these notes. In cosmology, if the
mean density of the Universe is low, the Universe expands indefinitely, but, if the density is higher than a certain critical density,
the (dimensionless) scale factor R of the Universe expands and contracts with time t according to the following parametric
cycloidal equations:
Ω0
R = (1 − cos 2θ), (19.10.1)
2(Ω0 − 1)

Ω0
t = (2θ − sin 2θ). (19.10.2)
3/2
2(Ω0 − 1 )

Here t is expressed in units of the reciprocal of the present Hubble constant, and Ω0 is the ratio of the present density of the
Universe to the density required to “close” the Universe.
A less well known example concerns the propagation of sound in the atmosphere. In the troposphere, which is the lower part of the
atmosphere up to about 11 km, the temperature decreases roughly linearly with height. In that case sound travels through the
troposphere in a cycloidal path. The speed of sound in a gas is proportional to the square root of the temperature. (If you are
wondering how it depends on pressure P and density ρ, the answer is that it depends on the ratio P/ρ - and this ratio is proportional
to the temperature.) In any case, if the temperature decreases linearly with height, the sound speed v varies with height y as
− −−− −
v = v0 √ 1 − cy (19.10.3)

where c is a constant, equal to about 0.023 km−1.Now to trace a sound ray through the atmosphere, we have to understand how the
direction of propagation changes as the sound passes through layers of air of different temperature. This is governed, as with light,
by Snell’s law (see figure XIX.11):
dv
= − tan ψdψ (19.10.4)
v

Snell’s law states that when sound (or light) enters a slower medium (i.e. one in which the speed of propagation is slower) it is bent
towards the normal. I have drawn figure XIX.11 to represent the situation in the troposphere where the temperature (and hence the
sound speed v ) decreases with height. That is, dv/dy is negative. In other words dv in figure XIX is negative, and Equation
19.10.4 indicates that dψ is positive, as drawn. In case you do not recognize this differential form of Snell’s law, try integrating it

from v to v and from ψ to ψ , and it should assume its more familiar integral form. If you now eliminate v between Equations
1 2 1 2

19.10.3 and 19.10.4, you will get a differential relation between y and ψ , which, upon integration, becomes

2
cos ψ
cy = 1 − (19.10.5)
2
cos ψ0

where ψ is the ground-level value of ψ . If we introduce

0

1
a = , (19.10.6)
2
2c cos ψ0

19.10.1 https://phys.libretexts.org/@go/page/8523
equation 19.10.5 can be conveniently re-written
2 2 2 2
y = 2a(sin ψ −s ∈ ψ0 ) = 2a(cos ψ0 − cos ψ (19.10.7)

Now tan ψ = dy/dx , and elimination of y between this and Equation 19.10.7 will give a differential relation between x and ψ ,
which, upon integration, becomes
x = a[2(ψ − ψ0 ) + sin 2ψ − sin 2 ψ0 ]. (19.10.8)

Equations 19.10.7 and 19.10.8 are the parametric equations of the sound path through the troposphere, and describe a cycloid.

 Exercise 19.10.1: Problem for a Rainy Day

If x = 2.0 and y = 1.6, what are ψ and ψ ? 0

Solution
I make it ψ = 69 ∘ ∘
, ψ0 = 15 52
′
.

This page titled 19.10: Examples of Cycloidal Motion in Physics is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or
curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is
available upon request.

19.10.2 https://phys.libretexts.org/@go/page/8523
 CHAPTER OVERVIEW
20: Miscellaneous
This chapter is a miscellany of diverse and unrelated topics – namely surface tension, shear modulus and viscosity – discussed only
for the purpose of presenting a few more examples of elementary problems in mechanics. It is not intended in any way to substitute
for a comprehensive course in any of the vast and interesting fields of surface chemistry, elasticity or hydrodynamics. All of these
subjects have a huge and specialized literature, each worthy of a full-length course, and I am not remotely competent to offer one.
Nevertheless, the few simple problems chosen in this chapter are suitable for a bit more practice in classical mechanics.

Topic hierarchy
20.1: Introduction
20.2: Surface Tension
20.2.1: Excess Pressure Inside Drops and Bubbles
20.2.2: Angle of Contact
20.2.3: Capillary Rise
20.3: Shear Modulus and Torsion Constant
20.4: Viscosity
20.4.1: Poiseuille's Law
20.4.2: The Couette Viscometer

This page titled 20: Miscellaneous is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1
20.1: Introduction
This chapter is a miscellany of diverse and unrelated topics – namely surface tension, shear modulus and viscosity – discussed only
for the purpose of presenting a few more examples of elementary problems in mechanics. It is not intended in any way to substitute
for a comprehensive course in any of the vast and interesting fields of surface chemistry, elasticity or hydrodynamics. All of these
subjects have a huge and specialized literature, each worthy of a full-length course, and I am not remotely competent to offer one.
Nevertheless, the few simple problems chosen in this chapter are suitable for a bit more practice in classical mechanics.

This page titled 20.1: Introduction is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

20.1.1 https://phys.libretexts.org/@go/page/7058
 SECTION OVERVIEW
20.2: Surface Tension
The cause of surface tension is often explained roughly as follows. Molecules within a liquid are subject to intermolecular forces
whose exact nature and origin need not concern us other than to say that they are principally van der Waals forces and they hold the
liquid together and prevent it from evaporating. A molecule deep within the liquid is surrounded in all directions by other
molecules, and so the net force on it averages zero. But a molecule on the surface experiences forces from beneath the surface, and
consequently it tends to get dragged beneath the surface. This results in as few molecules as possible remaining on the surface; i.e.
it results in the surface contracting to as small an area as possible consistent with whatever other geometrical constraints may exist.
That is, the surface appears to be in a state of tension causing it to contract to the least possible area.

This tension can be described qualitatively thus. In Figure XX.1, the dashed line is an imaginary line drawn in the surface of a
liquid. The liquid to the left of the line is being pulled to the right as indicated by the red arrows; the liquid to the right of the line is
being pulled equally to the left as indicated by the green arrows. The force per unit length perpendicular to a line drawn in the
surface of the liquid is the surface tension. Its SI unit is newtons per metre, and its CGS unit is dynes per centimetre. The
dimensions are MT−2.
I have seen various symbols, such as T , S and γ used for surface tension. The first two of these symbols are already heavily
worked in thermodynamics, so I shall use the symbol γ (although, it must be admitted, γ is heavily worked in thermodynamics,
too.) Not everyone is comfortable with a definition involving forces perpendicular to an imaginary line drawn in the surface, and an
alternative approach may be more palatable to some. The idea of a molecule beneath the surface being surrounded on all sides by
other molecules and hence experiencing zero net average force, while a molecule on the surface is pulled asymmetrically by the
molecules beneath it, remains. But instead of drawing an imaginary line on the surface, we reason that it requires work to move a
molecule from within the liquid to the surface, and it requires a lot of work to move many molecules from beneath to the surface.
That is, it requires work to create new surface. Thus we can define surface tension as the work required to create unit area of new
surface. The conditions under which this work is done have to be carefully defined in any precise definition, and, from a
thermodynamical point of view, the strict definition is the increase in the Gibbs free energy per unit area of new surface created
under conditions of constant temperature and pressure. That is:
∂G
γ =( ) (20.2.1)
∂A
T ,P

This is consistent with the definition of the Gibbs free energy as a quantity whose increase is equal to the work , other than P dV

work, done on a system in a reversible, isothermal isobaric process.

Such a nicety will be of interest to those versed in thermodynamics (and I have added a bit about the thermodynamics of surface
energy in Chapter 12 of Thermodynamics), but for those not so versed, you may, without any serious prejudice to understanding
most of the matter in this section, think of surface tension either as the force per unit length perpendicular to an imaginary line in
the surface, or as the work required to create unit area of new surface. You may express surface tension either in newtons per metre
or in joules per square metre (or, if you are of CGS persuasion, dynes per centimetre or ergs per square centimetre). These are

20.2.1 https://phys.libretexts.org/@go/page/7059
dimensionally equivalent.metre or in joules per square metre (or, if you are of CGS persuasion, dynes per centimetre or ergs per
square centimetre). These are dimensionally equivalent.

Topic hierarchy

20.2.1: Excess Pressure Inside Drops and Bubbles

20.2.2: Angle of Contact

20.2.3: Capillary Rise

This page titled 20.2: Surface Tension is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

20.2.2 https://phys.libretexts.org/@go/page/7059
20.2.1: Excess Pressure Inside Drops and Bubbles
The pressure inside a spherical drop is greater than the pressure outside. The way in which the excess pressure P depends on the
radius a of the drop, and the surface tension γ and density ρ of the liquid is amenable to dimensional analysis. One can suppose
that P ∝ a γ ρ , after which I leave it to the reader to show that α = −1, β = 1, δ = 0 , and therefore P ∝ γ/a .
α β δ

However, it is also quite easy to calculate the excess pressure (other than as a mere proportionality) in terms of the surface tension
and the radius of the drop. In Figure XX.2 I have divided a spherical drop of radius a into two hemispheres, and we are going to
consider the equilibrium of the upper hemisphere.

The upper hemisphere is being pulled down by surface tension all round the base of the hemisphere, and this downward force is
equal to the circumference of the base times the surface tension, or 2πγa. If the excess pressure inside the drop is P , the upward
component of the force due to this pressure is equal to P times the area of the base, πa . In case this is not obvious, consider an
2

elemental area dA as shown, at a spherical angle θ from the top of the drop. The force on this element is equal to P dA . The
upward component of this force is P cos θdA , and this is equal to P times the horizontal projection of dA . Now you are welcome
to do a nice double integration over the hemisphere, but since this (i.e " this is equal to P times the horizontal projection of dA ")
is true for every elemental area over the surface of the hemisphere, the total upward force must be equal to P times the area of the
base. Thus 2πγa = πa2P , and so the excess pressure inside the drop is
2γ
P = (20.2.2)
a

The smaller the drop, the greater the excess pressure. You may regard this as an explanation as to why droplets cannot form from a
vapor unless there is a dust nucleus of finite size for them to condense upon. Of course, two molecules colliding with each other
cannot in any case coalesce unless there is something to remove or absorb the kinetic energy.
The case of a nonspherical drop might be mentioned in passing. It is a well known result in geometry (or at least it is well known to
those who already know it) that if z = z(x, y) is a nonspherical surface, and you take two vertical planes at right angles to each
other, and if a and a are the radii of curvature of the intersections of the two planes with the surface, then
1 2 +
1

a1
is 1

a2

independent of the orientations of the two planes, as long as they remain perpendicular to each other. In other words, a and a do
1 2

not have to be the maximum and minimum radii of curvature. The excess pressure inside a nonspherical drop is
1 1
P =γ( + ) (20.2.3)
a1 a2

What about the pressure inside a spherical bubble of air (or other gas) under water (or other liquid)? If we are hasty, we might
suggest that, since this is the opposite situation to a liquid drop in air, maybe the pressure is less inside an underwater bubble. This
would be a very hasty conclusion, and quite wrong. If you go through exactly the same argument as we did for a drop, considering
the equilibrium of one hemisphere, you will see immediately that there is (as for the drop) an excess pressure inside the bubble
given again by Equation 20.2.2. And exactly the same would apply to a spherical drop of one liquid under the surface of a second
liquid, if the two liquid are immiscible. But, rather than just repeat the identical derivation, let's try a different approach.

20.2.1.1 https://phys.libretexts.org/@go/page/8527
Let us imagine that we have a bubble of radius a in a liquid of surface tension γ, and suppose that we are able, by means of a fine
syringe, to inject some more air inside so as to increase the radius of the bubble by da at constant pressure and temperature. The
surface area of a sphere of radius a is A = 4πa , so, if we increase the radius by da we increase the surface area by 8πada, and
2

we increase the volume by 4πa2da. The work done against the surface tension is 8πγada, and this must also be equal to
4πP a da, where P is the excess pressure inside the bubble. Equating these two expressions leads again to Equation 20.2.2.
2

What about a hollow spherical soap bubble in air? Here the soap has two surfaces – inside and out. If you repeat either of the above
derivations to this case, you will see that the excess pressure inside a hollow spherical soap bubble is
4γ
P = (20.2.4)
a

This page titled 20.2.1: Excess Pressure Inside Drops and Bubbles is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or
curated by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is
available upon request.

20.2.1.2 https://phys.libretexts.org/@go/page/8527
20.2.2: Angle of Contact
When a static quantity of liquid is in contact with an impermeable solid surface, it generally rests so that there is a characteristic
angle (measured in the liquid) between the surface of the liquid and the surface of the solid. This angle is the angle of contact, and
is shown as the angle θ is Figure XX.3. Figure XX.3(a) shows an acute angle of contact, in which the liquid spreads out a little and
"wets" the surface. Figure XX.3(b) shows an obtuse angle of contact, in which the liquid "bunches up", and does not wet the
surface, rather like drops of mercury on most surfaces, or drops of water on the surface of a car that has been freshly waxed. In
many cases the angle of contact is close to either 0° or 180°, although it will be appreciated that if θ were exactly zero, the liquid
would spread

out in an infinitesimally thin layer to cover or "wet" the entire surface; and if it were exactly 180°, the liquid, in the absence of
other forces (such as its weight!), would form a spherical globule in contact with the surface only at a single point. The angle of
contact is determined by the nature of both surfaces, and is very sensitive to any surface contamination. In order to wet a surface,
water may need to be helped by a small amount of wetting agent or detergent; only a small amount is necessary, because only the
surface, and not the bulk, of the liquid is involved. The chemical nature of wetting agents and detergents is beyond the scope of
these notes (i.e. it is beyond my scope!), but how the angle of contact depends on the surface tension provides a useful example of
the technique of virtual work (see Section 9.4 of Chapter 9).

Figure XX.4 represents a liquid, L, (e.g. water) in contact with a solid, S, (e.g. glass) and a gas, G, (e.g. air). The angle of contact is
θ , and the surface tensions (energy per unit area, or, for those who are versed in thermodynamics, the Gibbs free energy per unit

area) of the three interfaces are as shown. We'll suppose that the three media extend for a distance l at right angles to the plane of
the paper (or computer screen). The three phases are in equilibrium. Now, if we move the SLG boundary to the left by a distance
δx, we create a new area lδx of SL interface and a new area l cos θδx of LG interface, while we lose an area lδx of GS interface.

The work done on the system is therefore γ lδx + γ l cos θδx − γ lδx . By the principle of virtual work, this is zero, and
SL LG GS

therefore
γGS − γSL
cos θ = . (20.2.5)
γLG

The angle of contact is acute or obtuse, according to whether γ GS is greater than or less than γ
SL .

This page titled 20.2.2: Angle of Contact is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum
via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

20.2.2.1 https://phys.libretexts.org/@go/page/8528
20.2.3: Capillary Rise
When the lower end of a narrow capillary tube is immersed in a liquid, the liquid inside the tube rises a little above the level of the
liquid outside. If is then very simple to calculate how far the liquid rises in terms of the surface tension, the angle of contact and the
inside radius of the tube. See Figure XX.5.

The upward force due to surface tension is 2πaγ cos θ where a is the inside radius of the tube, and, if we neglect the very small
mass of the liquid in the meniscus (the curved surface at the top of the liquid column), the weight of the liquid column is π a hρg, 2

and therefore
2γ cos θ
h = . (20.2.6)
ρga

Of course if θ is obtuse (as with mercury in contact with glass), h will be negative, and the level of the mercury in the tube will be
below the outside level.

This page titled 20.2.3: Capillary Rise is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

20.2.3.1 https://phys.libretexts.org/@go/page/8529
20.3: Shear Modulus and Torsion Constant
Imagine that we have a rectangular block of solid material, as shown on the left hand side of Figure XX.6. We now apply a couple
of tangential forces F as shown on the right hand side. (I have not decided to go all chatty and informal by saying “a couple” of
forces; far from it – I am using the word “couple” in its formal sense in mechanics.) The material will undergo an angular
deformation, and the ratio of the tangential force per unit area to the resulting angular deformation is called the shear modulus or
the rigidity modulus. Its SI unit is N m−2 rad−1 and its dimensions are ML−1T−2θ−1. (I’d advise against using “pascals” per radian.
The unit “pascal” is best restricted to pressure, which is normal force per unit area, and is not quite the same thing as the tangential
force per unit area that we are discussing here.) You should convince yourself that the definition must specify the force F, not the
torque provided by the couple. If the block were twice as thick, and the forces were the same, you’d still get the same angular
deformation.

If you hold one end of a wire or rod fixed and apply a torque to the other end, this end will twist through an angle, and the ratio of
the applied torque to the angle through which the wire twists is the torsion constant, c , of the wire. You can see how the torsion
constant depends on the shear modulus η of the metal and the radius a and length l of the wire by the method of dimensions. You
can start by supposing that
α β γ
c ∝η a l ,

but you will soon find yourself in difficulty because a and l are each of dimension L. However, you will probably have no
difficulty with making the assumption that γ = −1 (the longer the wire, the easier it is to twist), and dimensional analysis will soon
show that α = 1 and β = 4 - which, being interpreted, means that it is much more difficult to twist a thick wire than a thin wire.
But can we do better and get an expression other than a mere proportionality for the torsion constant? Can we find the
proportionality constant? Let’s try some simpler problems first, and see how things go.
Let us consider a long, thin strip or ribbon of metal. By long and thin I mean that its length is much greater than its width, and its
width is much greater than its thickness. I can use any symbol I like to represent any quantity I like, so I could, if I wished, use Ξ
for the length, m for the width, and G for the thickness. Instead, the symbols that I shall choose to represent the length, width
α 2

and thickness of the strip are going to be, respectively, l, 2πr and δr. This seems silly at the moment, but in the end you’ll be glad
that I made this choice. The strip is shown at the left hand side of Figure XX.7.

I am now going to fix the upper end of the strip and apply a force F to the lower end, as shown in the right hand side of Figure
XX.7, and I can use any symbol I like to represent the displacement of the lower end, and I choose the symbol rϕ. This means that
the angular displacement θ is equal to rϕ/l.The tangential force per unit area is F /(2πrδr), and therefore

20.3.1 https://phys.libretexts.org/@go/page/7060
 Fl
η = , (20.3.1)
2
2πϕr δr

or
2
2πηϕr δr
F = . (20.3.2)
l

Now I’m going to restore the strip to its original shape, and then I’m going to roll it into a hollow cylindrical tube, so that it now
looks like a metal drinking straw. The circumference of the straw is 2πr, its radius is r and its thickness is δr (Figure XX.8). (Now
my notation is beginning to make some sense!)

I shall hold the upper end of the tube fixed and I shall apply a torque τ = Fr to the lower end. The tube will evidently twist
through an azimuthal angle ϕ given by
3
2π η δr
τ = ϕ. (20.3.3)
l

The torsion constant of the hollow tube is therefore

3
2πη r δr
c = . (20.3.4)
l

The torsion constant of a long solid cylinder (a wire) of radius a is the integral of this from 0 to a , which is
4
πηa
c = (20.3.5)
2l

This page titled 20.3: Shear Modulus and Torsion Constant is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

20.3.2 https://phys.libretexts.org/@go/page/7060
 SECTION OVERVIEW
20.4: Viscosity
Consider a river flowing over a smooth bed, as in Figure XX.9.

There will be a transverse velocity gradient dv/dz, with the liquid stationary at the river bottom, and the speed becoming faster as
we ascend from the bottom. As a consequence of the transverse velocity gradient, the liquid below the dashed line will be dragged
forward by the tangential force of the faster liquid above it, and the liquid above the dashed line will be dragged backward by the
tangential force of the more sluggish liquid below it. The ratio of the tangential force per unit area to the transverse velocity
gradient is called the coefficient of dynamic viscosity, for which the usual symbol is η . The dimensions of dynamic viscosity are
ML−1T−1. The CGS unit of dynamic viscosity is the poise. The abbreviation for the unit is P – though it would be well to define it if
you use it, since not everyone will recognize it. The unit is named after a nineteenth century French doctor, Jean Poiseuille, who
was interested in blood pressure and hence in the rate of flow of liquids through tubes. That is to say, if, for a transverse velocity
gradient of 1 cm s−1 per cm, the tangential force per unit area is 1 dyne cm−2, the dynamic viscosity is one poise. The SI (MKS)
unit is the decapoise (also spelled dekapoise), though the SI unit the pascal second (Pa s), which is dimensionally correct, is also
seen. If, for a transverse velocity gradient of 1 m s−1 per cm, the tangential force per unit area is 1 N m−2, the dynamic viscosity is
one decapoise. The dynamic viscosity of water varies from about 1.8 centipoise at 0∘ C to about 0.3 centipoise at 100∘C.
The ratio of the coefficient of dynamic viscosity to the density is the coefficient of kinematic viscosity, for which the usual symbol
is the Greek letter ν. The dimensions of kinematic viscosity are L2T−1. The CGS unit of kinematic viscosity is the stokes
(abbreviation St). It is named after nineteenth century British physicist, Sir George Stokes, who made major contributions to
diverse areas of physics. The SI unit of kinematic viscosity is usually given simply as m2 s−1, and 1 m2 s−1 = 104 stokes. The
kinematic viscosity of water varies from about 1.8 centistokes (1.8 - 10−6 m2 s−1) at 0∘ C to about 0.3 centistokes (3 - 10−7 m2 s−1)
at 100∘C.
Hydrodynamics is a huge and very difficult subject (at least I think it is), but there are a couple of simple problems that, if nothing
else, make good homework problems. These are Poiseuille’s law and the Couette viscometer.

Topic hierarchy

20.4.1: Poiseuille's Law

20.4.2: The Couette Viscometer

This page titled 20.4: Viscosity is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

20.4.1 https://phys.libretexts.org/@go/page/7061
20.4.1: Poiseuille's Law
Poiseuille’s law tells you how the rate of nonturbulent flow of a liquid through a cylindrical pipe depends on the viscosity of the
liquid, the radius of the pipe, and the pressure gradient. If all else fails, you can at least try dimensional analysis. Assume that the
γ
rate of flow of liquid (in cubic metres per second) is proportional to η a ( ) , and show by dimensional analysis that
α β dP

dx

α = −1, β = −4 and γ = 1 , which shows that the rate of flow is very sensitive to the radius of the pipe. That β = −4 tells you

that if your arteries are at all constricted, even by a little bit, you had better watch out. Gas flow is more complicated because gases
are compressible, (so are liquids, but not by much), but β = −4 tells you that the rate at which you can pump out gas from a
system depends a lot on the size of the smallest tube you have between the volume that you are trying to evacuate and the pump.
Now let’s try and analyse it further.

Figure XX.10 represents a pipe of radius a with liquid flowing to the right. At a distance r from the axis of the pipe the speed of
the liquid is v . The length of the pipe is l, and there is a pressure gradient along the length of the pipe, the pressure at the left end
being higher than the pressure at the right by P . There is a velocity gradient in the pipe. The speed of the liquid along the axis of
the pipe is v0, and the speed at the circumference of the pipe is zero. That is, the speed decreases from axis to circumference, so
that the velocity gradient (dv/dr) is negative.
Now consider the equilibrium of the liquid inside radius r. (It is in equilibrium because it is moving at constant speed.) It is being
pushed forward by the pressure gradient. This rightward force is π r P . It is being dragged back by the viscous force acting on the
2

area 2πrl. This leftward force is −2πηlr(dv/dr), this expression for the leftward force being positive.
Therefore
dv
−2ηl = P r. (20.4.1)
dr

Integrate from the axis (r = 0, v = v 0) to r:

2
Pr
v = v0 − . (20.4.2)
4ηl

Thus the speed decreases quadratically (parabolically) as you move away from the axis. The speed is zero at the circumference, and
hence the speed on the axis is
2
Pr
v0 = . (20.4.3)
4ηl

Verify the dimensions.

3

Now the volume flow through a cylindrical shell of radii r and r + dr is the speed times the area 2πrdr,, which is πr dr

2ηl
, and if you
integrate that through the whole pipe, from 0 to a , you find that the rate of flow of liquid through the pipe (cubic metres per
second) is
4
πa P
. (20.4.4)
8ηl

This is Poiseuille’s Law.

20.4.1.1 https://phys.libretexts.org/@go/page/8532
This page titled 20.4.1: Poiseuille's Law is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

20.4.1.2 https://phys.libretexts.org/@go/page/8532
20.4.2: The Couette Viscometer
A cylindrical vessel of radius b contains the liquid whose viscosity is to be measured. A smaller, solid cylinder of radius a and
length l is suspended from a torsion wire, whose torsion constant c is known, and is immersed in the liquid in the vessel, the two
cylinders being coaxial. The vessel containing the liquid is spun about its axis at an angular speed Ω, thus setting the liquid in
motion. This causes a viscous torque on the inner cylinder, which is therefore pulled round through an angle ϕ . When the restoring
torque of the torsion wire cϕ is equal to the viscous torque, the system will be in equilibrium, and one can then calculate the
viscosity η of the liquid. We shall refer to Figure XX.11. In the simple analysis given below, we suppose that the angular and linear
speed and gradients are sufficiently small that the flow is nonturbulent. We also neglect the effects of viscous drag on the flat ends
of the cylinder. Thus the diameter of the cylinder, in our analysis, must be much less than its length.

Incidentally, for a long time I thought that the word “couette” must be French for
something. It is – it’s French for “feather bed” or for “pigtail”. But the Couette
viscometer is actually named after a little-known nineteenth century French scientist,
Maurice Couette.

Let us calculate the viscous torque on the liquid within radius r. Notice that, since we have a steady-state situation, this torque is
independent of r; in particular the torque on the liquid within radius r is the same as the torque (which we can measure with the
torsion wire) on the inner cylinder. The area of the curved surface of the liquid within radius r is 2πlr. The viscous torque on this
surface is r times η times the area times the transverse velocity gradient. But we have to be careful about this last term. If the whole
body of the liquid were rotating as a solid body with angular speed ω, the speed at radius r would be rω and hence there would be a
transverse velocity gradient equal to ω − but no viscous drag! But the liquid is not, of course, rotating as a solid, and ω (as well as
v ) is a function of r . Since v = rω , the velocity gradient is + ω and the only part of this that goes into the expression for
dv dω
=r
dr dr

the viscous torque is the part r .Thus the expression for the torque on the liquid within radius r (and hence also on the inner
dω

dr

cylinder) is
dω
τ = r. η. 2πrl. r (20.4.5)
dr

That is,
dω τ
= . (20.4.6)
3
dr 2πηlr

Integration from r = a, ω = 0 to r = b, ω = Ω gives

20.4.2.1 https://phys.libretexts.org/@go/page/8533
 2 2
4πηlΩa b
τ = (20.4.7)
2 2
b −a

In equilibrium, this is equal to cϕ, where c is the torsion constant of the suspension and ϕ is the angle through which the inner
cylinder has turned, and hence the viscosity can be determined. You should, as usual, check the dimensions of Equation 20.4.7.

This page titled 20.4.2: The Couette Viscometer is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

20.4.2.2 https://phys.libretexts.org/@go/page/8533
 CHAPTER OVERVIEW
21: Central Forces and Equivalent Potential
21.1: Introduction to Central Forces
21.2: Motion Under a Central Force
21.3: Inverse Square Attractive Force
21.4: Hooke’s Law
21.5: Inverse Fourth Power Attractive Force
21.6: A General Central Force
21.7: Inverse Cube Attractive Force

This page titled 21: Central Forces and Equivalent Potential is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated
by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

1
21.1: Introduction to Central Forces
When a particle is in orbit around a point under the influence of a central attractive force (i.e. a force F (r) which is directed
towards a central point, with no transverse component) it experiences, when referred to an inertial reference frame, a centripetal
acceleration. If, however, the system is described with respect to a co-rotating reference frame, there is no centripetal acceleration;
rather, it appears as though an additional force, the centrifugal force, is pushing it away from the centre of attraction. In the co-
rotating frame, this force depends only on the distance of the particle from the centre of attraction, and it is therefore a conservative
force – and, like any conservative force, it can be described by the negative of the derivative of a potential energy function. When
describing the motion with respect to the co-rotating frame, we must add this potential to any additional “real” potentials (such as
originate from the gravitational fields of other bodies), to form an equivalent potential which constrains the motion of the particle.
An excellent example of this method is the analysis of the restricted three-body problem given in some detail in Chapter 16 of the
Celestial Mechanics notes. But I deal first, by way of example, with some simpler problems involving central forces, in which we
shall be able, by simple arguments, to deduce some basic characteristics of the motion.

This page titled 21.1: Introduction to Central Forces is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

21.1.1 https://phys.libretexts.org/@go/page/8538
21.2: Motion Under a Central Force
I consider the two-dimensional motion of a particle of mass m under the influence of a conservative central force F (r), which can
2
be either attractive or repulsive, but depends only on the radial coordinate r. Recalling the formula r̈ − rθ˙ \)for acceleration in
polar coordinates (the second term being the centripetal acceleration), we see that the equation of motion is
2
˙
m r̈ − mr θ = F (r). (21.2.1)

This describes, in polar coordinates, two-dimensional motion in a plane. But since there are no transverse forces, the angular
2
momentum m 2 ˙
θ is constant and equal to L, say. Thus we can write Equation 21.2.1 as
2
L
m r̈ = F (r) + . (21.2.2)
mr3

This has reduced it to a one-dimensional equation; that is, we are describing, relative to a co-rotating frame, how the distance of the
particle from the centre of attraction (or repulsion) varies with time. In this co-rotating frame it is as if the particle were subject not
2

only to the force F (r), but also to an additional force . In other words the total force on the particle (referred to the co-rotating
L

mr
3

frame) is
2
′
L
F (r) = F (r) + . (21.2.3)
3
mr

Now F (r), being a conservative force, can be written as minus the derivative of a potential energy function, F =−
dV

dr
. Likewise,
2 2
L

m
is minus the derivative of
3
. Thus, in the co-rotating frame, the motion of the particle can be described as constrained by the
L

2mr
2

potential energy function V , where ′

2
L
′
V =V + . (21.2.4)
2
2mr

This is the equivalent potential energy. If we divide both sides by the mass m of the orbiting particle, this becomes
2
h
′
Φ =Φ+ . (21.2.5)
2
2r

Here h is the angular momentum per unit mass of the orbiting particle, Φ is the potential in the inertial frame, and Φ
′
is the
equivalent potential in the corotating frame.

This page titled 21.2: Motion Under a Central Force is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

21.2.1 https://phys.libretexts.org/@go/page/8539
21.3: Inverse Square Attractive Force
This is dealt with in detail in Chapter 9 of Celestial Mechanics. Here we investigate some general properties of the motion.
If F =−
GMn

r
2
then V =
GMn

r
, and hence
2
′
GM n L
V =− + . (21.3.1)
r 2mr2

I sketch this in Figure XXI.1. The total energy (potential + kinetic) is constant (independent of r) and is greater than (or equal to)
the potential energy. If the total energy is less than zero, you can see from the graph that r has a lower (perihelion) and upper
(aphelion) limit; this corresponds to an elliptic orbit. But if the total energy is positive, r has a lower limit, but no upper limit; this
corresponds to a hyperbolic orbit. If the total energy is equal to the minimum of V , only one value of r is possible, and the orbit is
′

a circle.

This page titled 21.3: Inverse Square Attractive Force is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

21.3.1 https://phys.libretexts.org/@go/page/8540
21.4: Hooke’s Law
We imagine a particle whirling around on the end of a spring, oscillating in and out as it does so. The force constant of the spring is
k , the force on the particle is −kr and the potential (elastic) energy is V = kr . This is akin to a non-rigid rotor.
1 2

Figure 21.4.1 : In the absence of the spring, the particles would fly apart. However, the force exerted by the extended spring pulls
the particles onto a periodic, oscillatory path. (CC BY-SA 3.0; Cleonis).
The effective potential energy is therefore
2
1 L
′ 2
V = kr + . (21.4.1)
2
2 2mr

and is sketched in Figure XVI.2. The total energy (potential + kinetic) is constant (independent of r) and is greater than (or equal
to) the potential energy. The distance of the particle from the centre of attraction is bounded above and below. The motion is a
Lissajous ellipse, with the centre of attraction at the centre (not the focus) of the ellipse. The lower bound is the semi minor axis
and the upper bound is the semi major axis.

An inverse square force (e.g. a gravitational force, or a Coulomb’s law electrostatic force) and a Hooke’s law force (kx) are
obvious examples of real forces in nature. In what follows we shall investigate the behavior of a particle under the influence of
other force laws, such as inverse fourth power and inverse cube forces. It is difficult to imagine whether such forces actually exist
in nature (the field of an electric dipole falls off as the cube of the distance - but the field is not radial, and the force is not a central
force), and to that extent much of what follows is an exercise in mathematics more than in physics. But inverse square and Hooke’s
law forces are certainly not the only forces to operate in nature. What is the force law, for example, for the residual strong
interactions between nucleons in an atomic nucleus, or the force law between the quarks within a nucleon? It will be worthwhile
investigating the simpler hypothetical forces to be discussed here in order to understand the principles and methods that may be
applicable to a more difficult problem.

This page titled 21.4: Hooke’s Law is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

21.4.1 https://phys.libretexts.org/@go/page/8541
21.5: Inverse Fourth Power Attractive Force
If F =−
3a

r4
then V =−
a

r3
, and hence
2
′
a L
V =− + (21.5.1)
3 2
r 2mr

I sketch this in Figure XXI.3. The total energy (potential + kinetic) is constant (independent of r) and is greater than (or equal to)
the potential energy. If the total energy is negative, the distance r has an upper limit, but the only lower limit is the origin, or the
centre of attraction, and particle will eventually end there. If the total energy is greater than the maximum of V , the motion is
′

completely unbounded. If the total energy is positive but less than V , the motion depends on the initial value of r. For small r
′
max

the motion is bounded above, and the particle will eventually end at the origin. For large r, there is a minimum distance to which
the particle can approach the origin, and the particle will eventually wander off to infinity. For total energy in this range, there is a
range of r that is not possible.

This page titled 21.5: Inverse Fourth Power Attractive Force is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated
by Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

21.5.1 https://phys.libretexts.org/@go/page/8542
21.6: A General Central Force
Let us suppose that we have a particle that is moving under the influence of a central force F (r). The equations of motion are
Radial:
2
˙
m(r̈ − r θ ) = F (r) (21.6.2)

Transverse:
˙
r θ̈ + 2 ṙ θ = 0. (21.6.3)

These can also be written

2
˙
r̈ − θ = a(r) (21.6.4)

2˙
r θ = h. (21.6.5)

Here a is the radial force per unit mass (i.e. the radial acceleration) and h is the (constant) angular momentum per unit mass. [If
you are unsure of why Equations 21.6.3 and 21.6.5 are the same, differentiate equation 21.6.5 with respect to time.]
These are two simultaneous equations in r, θ, t . In principle, if we could eliminate t between them, we would obtain a relation
between r and θ , which would tell us the shape of the path pursued by the particle. In Chapter 9 of my Celestial Mechanics notes
we do this for the gravitational case, and we find that the path is an ellipse of the form r = . Or perhaps we could eliminate l

1+e cos θ

r and hence find out how the angle θ changes with time. Or again we might be able to eliminate θ and hence get a relation telling

us how r varies with the time. Yet again we might be told the shape of the path r(θ) , and asked to find the force law F (r). Or
again, rather than the force, we might be given the form of the potential energy V (r), which is related to the force by
F = −dV /dr . The potential Φ is the potential energy per unit mass, and −dΦ/dr is the radial force per unit mass - i.e. it is the

radial acceleration a(r) of the orbiting particle. The angular momentum of the particle, which is constant, is L − mr θ˙ , and the 2

angular momentum per unit mass is h = r θ˙ , which is twice the rate at which the radius vector sweeps out area.
2

We might also remember that, if we are given the potential energy V or the potential Φ in an inertial frame, we might also want to
2

work in a co-rotating frame, making use of the equivalent potential energy V

′
=V +
2mr
L
2
or the equivalent potential
2
′
Ω =Ω+
h
2
.
2

One last thing to bear in mind before starting any problems of this class. It turns out that, very often, a change of variable u = 1/r
turns out to be useful. Conservation of angular momentum then takes the form θ˙/u = h Also 2

dr du dr dr du θ̇ du du
ṙ = = =− = −h (21.6.6)
2
du dt dθ dt dθ u dθ dθ

and
2 2
d du d du dθ d du 2
d u 2 2
d u
r̈ = (−h ) = −h = −h = −h. h u . = −h u . (21.6.7)
2
dt dθ dt dθ dt dθ dθ dθ dθ2

Equations 21.5.4 and 21.6.5 now become

2
2 2
d u 2 3
h u +h u = −a(r). (21.6.8)
dθ2

and
˙ 2
θ = hu . (21.6.9)

We can now easily eliminate the time which was one of our aims:
2
d u
2 2 2 3
h u +h u = −a(r). (21.6.10)
2
dθ

21.6.1 https://phys.libretexts.org/@go/page/8543
[As ever, check the dimensions.] This equation, which does not contain the time, when integrated will give us the equation to the
path.
With these remarks in mind, let us try a few problems. For example:

This page titled 21.6: A General Central Force is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

21.6.2 https://phys.libretexts.org/@go/page/8543
21.7: Inverse Cube Attractive Force
A particle moves in a field such that the attractive force on it varies inversely as the cube of the distance from a centre of attraction.
What is the shape of the path? How does the angle θ vary with time?
Let’s suppose that the radial acceleration is a(r) = −k /r = −k u . (I want the coefficient of 1/r to be negative, so that the
3 3 3 3 3

force is attractive, which is why I have written the coefficient as −k . Besides, the dimensions of k are then L2T−1, which are the
2

same as those of h , the angular momentum per unit mass, which helps to make the algebra simple.) The differential equation to the
2

path (Equation 21.6.10) is then h 2 2 d

u
u
2
2
+h u
3 2
=k u
3
or
dθ

2
2
d u 2 3
h + h u = k u. (21.7.1)
dθ2

That is,
2 2 2
d u k −h
= u. (21.7.2)
2 2
dθ h

The form of the motion evidently depends on whether k > h (a strongly attractive force, or a small angular momentum), or if
2 2

2
k <h (a weak force, or a large angular momentum.) If we start the particle rolling with just the right amount of angular
2

momentum (k = h ) , there will evidently be zero radial acceleration, and the particle will move in a circle.
2 2

Before integrating Equation 21.7.2, let us look at the equivalent potential. For a(r) = −k /r , the potential in the inertial frame is 2 3

Ω = − k /r
1

2
2
provided we take the potential at infinity to be zero. The equivalent potential is then (see equation 21.2.5)
2

2 2
k h
′
Ω =− + . (21.7.3)
2 2
2r 2r

We see that, if k = h , the potential is zero and independent of distance. If h < k , the equivalent potential is negative,
2 2 2 2

increasing to zero as r → ∞, and the particle accelerates towards the centre of attraction. If h > k , the potential is positive, 2 2

decreasing to zero as r → ∞, and the particle accelerates away from the centre of attraction. This sounds like a contradiction, but
what is happening is h > k , that means that the particle has initially been given a large angular momentum, and, in the corotating
2 2

frame, the centrifugal force is larger than the attractive force.

If h 2
<k
2
, the equation of motion (Equation 21.7.2) is
2
d u
2
= c u, (21.7.14)
2
dθ

where
2 2
k −h
2
c = . (21.7.15)
2
h

The general solution is

cθ −cθ
u = Ae + Be (21.7.16)

If the initial conditions are that at t = 0, r = r , u = u , = 0 (this last condition means that the particle was launched in a
0 0
du

dθ

direction at right angles to the radius vector, this solution becomes

u = y0 cos hcθ. (21.7.17)

That is,

r = r0 sec hcθ (21.7.18)

I have drawn this below for c = 0.1; that is, for k ≈ 1.05h And for c = 0.5; that is k ≈ 1.22h, for a smaller angular momentum.

21.7.1 https://phys.libretexts.org/@go/page/8544
We also need to consider the case h > k , in which case the general solution is of the form
2 2
u = A cos cθ + B sin cθ . Alas, I
haven’t had the energy to do this yet. Perhaps some viewer can beat me to it, and let me know.

This page titled 21.7: Inverse Cube Attractive Force is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

21.7.2 https://phys.libretexts.org/@go/page/8544
 CHAPTER OVERVIEW
22: Dimensions
22.1: Mass, Length and Time
22.2: Table of Dimensions
22.3: Checking Equations
22.4: Deducing Relationships
22.5: Dimensionless Quantities
22.6: Different Fundamental Quantities
22.7: Appendix A
22.8: Appendix B

This page titled 22: Dimensions is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via source
content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

1
22.1: Mass, Length and Time
Any mechanical quantity can be expressed in terms of three fundamental quantities, mass, length and time. For example, speed is a
length divided by time. Force is mass times acceleration, and is therefore a mass times a distance divided by the square of a time.
We therefore say that [Force] = MLT−2. The square brackets mean: “The dimensions of the quantity within”. The equations indicate
how force depends on mass, length and time. We use the symbols MLT (not in italics) to indicate the fundamental dimensions of
mass, length and time. In the above equation, MLT−2 are not enclosed within square brackets; it would make no sense to do so.
We distinguish between the dimensions of a physical quantity and the units in which it is expressed. In the case of MKS units
(which are a subset of SI units), the units of mass, length and time are the kg, the m and the s. Thus we could say that the units in
which force is expressed are kg m s−2, while its dimensions are MLT−2.
For electromagnetic quantities we need a fourth fundamental quantity. We could choose, for example, quantity of electricity Q, in
which case the dimensions of current are QT−1. We do not deal further with the dimensions of electromagnetic quantities here.
Further details are to be found in my notes on Electricity and Magnetism, http://orca.phys.uvic.ca/~tatum/elmag.html
To determine the dimensions of a physical quantity, the easiest way is usually to look at the definition of that quantity. Most readers
will have no difficulty in understanding that, since work is force times distance, the dimensions of work (and hence also of energy)
are ML2T−2 . A more challenging one would be to find [dynamic viscosity]. One would have to refer to its definition (see Chapter
20) as tangential force per unit area per unit transverse velocity gradient.
−2

Thus [dynamic viscosity] = [  force  distance

area velocity
] =
MLT
−2
L
−1
= MLT
−1 −1
L .
L LT

This page titled 22.1: Mass, Length and Time is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

22.1.1 https://phys.libretexts.org/@go/page/8546
22.2: Table of Dimensions
I supply here a table of dimensions and MKS units of some mechanical quantities. Some are obvious and trivial. Others might be
less so, and readers to whom this topic is new are encouraged to derive some of them from the definitions of the quantities
concerned. Let me know (jtatum at uvic.ca) if you detect any mistakes.
I do not know whether angle is a dimensionless or a dimensioned quantity. I can convince you that it is dimensionless by reminding
you that it is defined as a ratio of two lengths. I can convince you that it is dimensioned by pointing out that it is necessary to state
the units (e.g. radians or degrees) in which it is expressed. This might make for an interesting lunchtime conversation

This page titled 22.2: Table of Dimensions is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum
via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

22.2.1 https://phys.libretexts.org/@go/page/8547
22.3: Checking Equations
When you are doing a complicated calculation involving difficult equations connecting several physical quantities, you must,
routinely, check the dimensions of every line in your calculation. If the equation does not balance dimensionally, you know
immediately that you have made a mistake, and the dimensional imbalance may even give you a hint as to what the mistake is. If
the equation does balance dimensionally, this, of course, does not guarantee that it is correct - you may, for example, have missed a
dimensionless constant in the equation.
−
−
Suppose that you have deduced (or have read in a book) that the period of oscillations of a torsion pendulum is P = 2π √
C
I
, where
I is the rotational inertia and c is the torsion constant. You have to check to see whether the dimensions of the right hand side are
indeed that of time. We have
−−−−− −−−
−− 2
I ML
[√ ] =√ .
2 −2
C ML T

This does indeed come to T , and so the equation balances dimensionally.

This page titled 22.3: Checking Equations is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum
via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

22.3.1 https://phys.libretexts.org/@go/page/8548
22.4: Deducing Relationships
i. We may suppose that the period P of a simple pendulum depends upon its mass m, its length l, and the gravitational acceleration
g . In particular we suppose that the period is proportional to some power α of the mass, some power β of the length, and some

power γ of the gravitational acceleration. That is

p ∝m
α β
l g
γ
.
Both sides must have the same dimension - namely T.
That is
α β γ
[m l g ] =T

That is
α β −2 γ
M L (LT ) =T

We equate powers of M, L and T to get three equations in α, β, γ:

α = 0, β + γ = 0, −2γ = 1,

with solutions α = 0, β = 1

2
,γ =−
1

2
, which shows that
1 1
−
−
P ∝m l
0
2
g
−
2
, or P ∝√
l

ii. Here’s another: The torque τ required to twist a solid metallic cylinder through an angle θ is proportional to θ : τ = cθ .
c is the torsion constant. How does c depend upon the length l and radius a of the cylinder, its density ρ and its shear
modulus η ? There is an immediate difficulty, in that we have four quantities to consider − l, a, ρ and η , yet we have
only three dimensions L, M , T to deal with. Hence we shall have three equations in four unknowns. Further, two of
the quantities, l and a have similar dimensions, which adds to the difficulties.
In cases like this we may have to make a sensible assumption about one of the quantities. We may, for example, find it
easy to accept that, the longer the cylinder, the easier it is to twist, and we may make the assumption that the torsion
constant is inversely proportional to the first power of its length. Then we can suppose that
α β γ
cl ∝ a ρ η

in which case
α β γ
[cl] h [a ρ η ]

That is
2 −2 α −3 β −1 −2 γ
ML T L h L (ML ) (ML T )

Equate the powers of M, L and T:

1 = β + γ; 3 = α − 3β − γ; −2 = −2γ

4
ηa
This gives α = 4, β = 0, γ = 1, and hence c ∝ l
.
iii. How does the orbital period P of a planet depend on the radius of its orbit, the mass M of the Sun, and the gravitational
constant G?
Assume
α β γ
P ∝G M a

−−−
It is left to the reader to show that P .
3
a
∝√
GM

iv. A sphere of radius a moves slowly at a speed v through a fluid of density ρ and dynamic viscosity η . How does the viscous drag
F depend upon these four variables?

Four variables, but only three dimensions, and hence three equations! What to do? If you have better insight than I have, or if you
already know the answer, you can assume that it does not depend upon the density. I haven’t got such clear insight, but I’d be

22.4.1 https://phys.libretexts.org/@go/page/8549
willing to suppose that the viscous drag is proportional to the first power of the dynamic viscosity. In which case I’d be happy to
assume that
F α β γ
∝a ρ v
η

Then
−2
MLT α −3 β −1 γ
−1 −1
=L (ML ) (LT )
ML T

Equate the powers of M, L and T:

′
0 = β; 2 = α − 3β + γ − 1 = −γ

This gives α = 1, β = 0, γ = 1 , and hence F ∝ ηav .

This page titled 22.4: Deducing Relationships is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

22.4.2 https://phys.libretexts.org/@go/page/8549
22.5: Dimensionless Quantities
Dimensionless Quantities are used extensively in fluid dynamics. For example, if a body of some difficult shape, such as an
aircraft, is moving through a fluid at speed V , it will experience all sorts of forces, external and internal. The ratio of the internal
forces to the external forces will depend upon its speed, and the viscosity of the fluid, and the size of the body. By “size” of a body
of “difficult” shape we could take the distance between two defined points on the body, such as its top and bottom, or its front and
back, or its greatest width, or whatever. Call that distance l. But the ratio of the internal to the viscous forces is dimensionless, so it
must depend on some combination of the viscosity, speed V and linear size l that is dimensionless. Since V and l do not contain M
in their dimensions, the viscosity concerned must be the kinematic viscosity ν , which is the ratio of dynamic viscosity to density
and does not have M in its dimensions. So, what combination of ν , V and l is dimensionless?
It is easy to see that VI

V
- or any power of it, positive, negative, zero, integral, nonintegral - is dimensionless. is called the
VI

Reynolds number, and is usually given the symbol Re. It is supposed that if you make a small model of the aircraft (or whatever the
body is) and move it through some fluid and some speed, the ratio of internal to viscous forces in the model will be the same as in
the real thing provided that the Reynolds numbers in the model and in the real thing are the same.
There are oodles of similar dimensionless numbers used in fluid dynamics, such as Froude’s number and Mach number, but this
example of Reynolds number should give the general idea.

This page titled 22.5: Dimensionless Quantities is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy
Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

22.5.1 https://phys.libretexts.org/@go/page/8550
22.6: Different Fundamental Quantities
We stated at the beginning of this chapter that any mechanical quantity could be expressed in terms of three fundamental quantities,
mass, length and time. But there is nothing particularly magic about these quantities. For example, we might decide that we could
express any mechanical quantity in terms of, say, energy E, speed V and angular momentum J. We might then say that the
dimensions of area could be expressed as E-2V2J2 .(Verify this!)
While agreeing that such a system might be possible, you might feel that it would be totally absurd and there is no point in reading
further.
But stop! Such a system is not only possible, but it is normally and routinely used in the field of high-energy particle physics. That,
perhaps, is a surprise, but, if you are thinking of taking an interest in particle physics, read on.
The units generally used in particle physics to express the fundamental quantities energy, speed and angular momentum are GeV
(or MeV, or TeV , etc) for energy, the speed of light c for speed, and the modified Planck constant ħ for angular momentum. There
are often referred to as “natural” units, the speed of light being a “natural” unit of speed and ħ being a “natural” unit for angular
momentum, whereas metre, kilogram and second are not so “natural” in this sense as they are “man-made”. It is true that a GeV is
not particularly “natural”, but at least a system with GeV, c and ħ as fundamental quantities is certainly more “natural” than metre-
kilogram-second.
In any case, the dimensions of mass in this system are EV−2. (You can see this immediately, for example from Einstein’s famous
equation E = mc2.) The units used in this system are GeV/c2. Thus the rest mass of a proton is 0.9383 GeV/c2, and the rest mass of
an electron is 0.5110 MeV/c2. One way to interpret this, if you like, is to say that the rest-mass energy of a proton (i.e. its m0c2) is
0.9383 GeV.
Likewise the dimensions of linear momentum are EV−1, and units in which it is expressed are GeV/c. (You can see this, for
example, if you look at the energy and momentum of a photon: E = hv, p = h/γ, from which = = )
P

E
1

vγ
1

Torque (which has the same dimensions as energy) is equal to rate of change of angular momentum, from which we see that time
has dimensions E−1J and could be expressed in units of ħ/GeV. Alternatively you can see that [time] = ħ/GeV immediately from
Planck’s equation . E = hω And speed is distance over time, so that we see that distance, or length, has dimensions E−1VJ, and
hence units ħc/GeV.
Using data from the 2010 Particle Physics Booklet, I calculate as follows.

Mass: 1 GeV/c2 = 1.782 661 76 ~ 10−27 kg

Length: 1 ħc/GeV = 1.973 269 63 ~ 10−16 m
Time: 1 ħ/GeV = 6.582 118 99 ~ 10−26 s
Energy: 1 GeV = 1.602 176 49 ~ 10−10 J
Linear Momentum: 1 GeV/c = 5.344 285 50 ~ 10−19 kg m s−1

I give here a table of the dimensions (in terms of EVJ) of the same quantities as in the table of page 2. I dare say some of them are
never likely to be needed, but some certainly will be needed, and, rather than predict which will be useful and which not, I might as
well give them all. The dynamic viscosity of water at room temperature is about 10−3 kg m−1 s−1, or 10−3 dekapoise. I cannot
imagine anyone needing to know that the dynamic viscosity of water at room temperature is about 7.3 ~ 10−18 (GeV)3/(c3ħ2) , or
that its surface tension is so many (GeV)3/(cħ)2 − but you never know

22.6.1 https://phys.libretexts.org/@go/page/8551
This page titled 22.6: Different Fundamental Quantities is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by
Jeremy Tatum via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon
request.

22.6.2 https://phys.libretexts.org/@go/page/8551
22.7: Appendix A
Miscellaneous Problems
In this Appendix I offer a number of random problems in classical mechanics. They are not in any particular order – they come just
as I happen to think of them, and they are not necessarily related to any of the topics discussed in any of the chapters. They are
intended just to occupy you on those dull, rainy days when you have nothing better to do. Solutions will be in Appendix B – except
that whenever I add any new problems to Appendix A, which I shall from time to time, I shall wait a few days before posting the
solutions in Appendix B.
_________________________

 Exercise 22.7.1
No book on classical mechanics is complete without a problem of a ladder leaning against a wall. Here, then, is a ladder
problem – except that it is nothing whatever to do with mechanics, and it is put here just for fun. It is a problem only in
geometry, yet it is one which some people at first find difficult. It even seems difficult to try to find an approximate solution by
trying to draw it accurately to scale, and I have deliberately not drawn it to scale, so you can’t find the answer merely by taking
a ruler and measuring it!

Two ladders, of lengths 8 m and 10 m, are leaning against two walls as shown. Their point of intersection is 3 m above the
ground. What is the distance between the walls?

 Exercise 22.7.2

A pendulum of length l , is set into motion so that it describes a cone as shown of semi-vertical angle α , the bob describing a
0

horizontal circle at angular speed Ω

Show that
g
cos α = 2
.
l0 Ω

This, of course, is a very trivial problem not worthy of your mettle. It is given only as an introduction to the next problem.

22.7.1 https://phys.libretexts.org/@go/page/9091
 Exercise 22.7.3a

A string of a pendulum passes through a board as shown in the figure below, in such a manner that, by lowering or raising the
board, the length of the string below the board can be varied. The part below the board is initially of length l , and it is set into
0

motion as a conical pendulum so that the angular speed and the semi vertical angle are related by
g
cos  α = 2
.
l0 Ω

As the board is raised or lowered (or alternatively the pendulum is lowered or raised) and consequently the length l below the
board is varied, the semivertical angle θ will change and so will the angular speed ω. (The symbols l , α and Ω represent the
0

initial values of these quantities.)

Show that
i. l sin θ tan θ is constant,
3 3

ii. ω cot θ is constant,

3 2

iii. ω (ωl   −  Ω l sin α) is constant.

3 2 2
0
2

 Exercise 22.7.3b

Start with the following initial conditions:

−1
l0   =  50 cm Ω  =  5 rad s

and assume that g  =  9.8 m s =2

, so that α  =  38
∘ ′
 22 .
i. Plot a graph of θ (vertically) versus l (horizontally), for l  =  0 to 1 m. When l = 40 cm, what is θ correct to one arcmin?
ii. Plot a graph of ω (vertically) versus θ (horizontally), for θ = 0 to 70o.
iii. Plot a graph of ω (vertically) versus l (horizontally) for l = 16 cm to 1 m. When l = 60 cm, what is ω correct to four
significant figures?
________________
The next few problems involve a rod with its lower end in contact with a horizontal table and the rod falling over from an
initial vertical (or inclined) position. There are several versions of this problem. The table could be smooth, so that the rod
freely slips over the table. Or the lower end could be freely hinged at the table, so that the lower end does not move as the rod
falls over. Or the table might be rough, so that the rod might or might not slip.

22.7.2 https://phys.libretexts.org/@go/page/9091
 Exercise 22.7.4

A uniform rod of mass m and length 2l is initially vertical with its lower end in contact with a smooth horizontal table. It is
given an infinitesimal angular displacement from its initial position, so that it falls over. When the rod makes an angle θ with
the vertical, find:
The angular speed of the rod;
The speed at which the centre of the rod is falling;
The speed at which the lower end of the rod is moving;
Show that the speed of the lower end is greatest when θ =  37 ∘ ′
 50 .
If the length of the rod is 1 metre, and g = 9.8 m s-2, what is the angle θ when the speed of the lower end is 1 m s-1?

 Exercise 22.7.5

A uniform rod is initially vertical with its lower end smoothly hinged to a horizontal table. Show that, when the rod falls over,
the reaction of the hinge upon the rod is vertical when the rod makes an angle 48o 11' with the vertical, and is horizontal when
the rod makes an angle 70o 31' with the vertical.

 Exercise 22.7.6

A uniform rod of length 1 metre, with its lower end smoothly hinged to a horizontal table, is initially held at rest making an
angle of 40o with the vertical. It is then released. If g = 9.8 m s-1, calculate its angular speed when it hits the table in a
horizontal position (easy) and how long it takes to get there (not so easy).

 Exercise 22.7.7

A uniform rod is initially vertical with its lower end in contact with a rough horizontal table, the coefficient of friction being μ .
Show that:

 Exercise 22.7.7a

If μ < 0.3706, the lower end of the rod must slip before the rod makes an angle θ with the vertical of 35o 05'.

 Exercise 22.7.7b

If μ > 0.3706, the rod will not slip before θ = 51o 15', but it will certainly slip before θ = 70o 31' .

 Exercise 22.7.7c

If μ = 0.25, at what angle θ will the lower end slip? If μ = 0.75?

 Exercise 22.7.8

It is time for another ladder problem. Most ladders in elementary mechanics problems rest on a rough horizontal floor and lean
against a smooth vertical wall. In this problem, both floor and wall are smooth. The ladder starts making an angle of a with the
vertical, and then it is released. It immediately starts to slip, of course. After a while it will cease contact with the smooth
vertical wall. Show that, at the moment when the upper end of the ladder loses contact with the wall, the angle q that the ladder
makes with the vertical is given by cos θ  = cos α .
2

22.7.3 https://phys.libretexts.org/@go/page/9091
 Exercise 22.7.9
If you managed that one all right, this one, which is somewhat similar, should be easy. Maybe.
A uniform solid semicylinder of radius a and mass m is placed with its curved surface against a smooth vertical wall and a
smooth horizontal floor, its base initially being vertical.

It is then released. Find the reaction N of the floor on the semicylinder and the reaction
1 N2 of the wall on the semicylinder
when its base makes an angle θ with the vertical.
Show that the semicylinder loses contact with the wall when θ = 90o, and that it then continues to rotate until its base makes an
angle of 39o 46' with the vertical before it starts to fall back.
_________________________
Many problems in elementary mechanics involve a body resting upon or sliding upon an inclined plane. It is time to try a few
of these. The first one is very easy, just to get us started. The two following that might be more interesting.

 Exercise 22.7.10

A particle of mass m is placed on a plane which is inclined to the horizontal at an angle α that is greater than tan μ , where
−1

μ is the coefficient of limiting static friction. What is the least force required to prevent the particle from sliding down the

plane?

 Exercise 22.7.11

A cylinder or mass m, radius a , and rotational inertia ka rolls without slipping down the rough hypotenuse of a wedge on
2

mass M , the smooth base of which is in contact with a smooth horizontal table. The hypotenuse makes an angle α with the
horizontal, and the gravitational acceleration is g. Find the linear acceleration of the wedge as it slips along the surface of the
table, in terms of m,  M ,  g,  a,  k and α .
[Note that by saying that the rotational inertia is ka , I am letting the question apply to a hollow cylinder, or a solid cylinder, or
2

even a hollow or solid sphere.]

 Exercise 22.7.12

22.7.4 https://phys.libretexts.org/@go/page/9091
A particle is placed on a rough plane inclined at an angle a to the horizontal. It is initially in limiting static equilibrium. It is
given an initial velocity V along the x-axis. Ignoring the small difference between the coefficients of moving and limiting
0

static friction, show that at a point on the subsequent trajectory where the tangent to the trajectory makes an angle ψ with the
x-axis, the speed V is given by

V0
V =
1+cos ψ

What is the limiting speed reached by the particle after a long time?

 Exercise 22.7.13

Calculate the moment of inertia of a hollow sphere, mass M , outer radius a , inner radius xa. Express your answer in the form
I =
2

5
2
M a   ×  f (x) .
What does your expression become if x = 0 ? And if x → 1 ?

 Exercise 22.7.14

Calculate the moment of inertia of a spherical planet of outer radius a , consisting of a dense core of radius xa surrounded by a
mantle of density s times the density of the core. Express your answer in the form
I =
2

5
Ma
2
× f (x, s) .
Make sure that, if the density of the core is zero, your expression reduces to the answer you got for Exercise 13.

22.7.5 https://phys.libretexts.org/@go/page/9091
Draw graphs of 2
I
versus x (x going from 0 to 1), for s = 0.2, 0.4, 0.6 and 0.8.
( M a2 )
5

Show that, for a given mass M and density ratio s , the moment of inertia is least for a core size give by the solution of
5 2
2(I − s)x + 15 x −9 = 0

For a mantle-to-density ratio of 0.6, calculate the core size for which the moment of inertia is least and calculate (in units of
M a ) the moment of inertia for that core.
2 2

Now let’s see if we can determine the core size from a knowledge of the moment of inertia. It is sometimes asserted that one
can determine the moment of inertia (and hence the core size) of a planet from the rate of precession of the orbit of a satellite. I
am not sure how this would work with a planet such as Mercury, which has never had a satellite in orbit around it. (Mariner 10,
while in orbit around the Sun, made three fly-bys past Mercury). Unless a planet departs from spherical symmetry, the orbit of
a satellite will not precess, since the gravitational planet is then identical with that from a point mass. And, even if a planet
(C −A)
were dynamically oblate, the rate of precession allows us to determine the dynamical ellipticity C
, but not either moment
of inertia separately.
Nevertheless, let’s suppose that the moment of inertia of a planet is (0.92 ± 1%) % M a ; specifically, let’s suppose that the
2

5
2

moment of inertia has been determined to be between 0.911 and 0.929 % M a , and that the mantle-to-core density ratio is
2

5
2

known (how?) to be 0.6. Calculate the possible range in the value of the core radius x.

 Exercise 22.7.15

A rectangular brick of length 2l rests (with the sides of length 2l vertically) on a rough semicylindrical log of radius R . The
drawing below shows three such bricks. In the first one, 2l is quite short, and it looks is if it is stable. In the second one, 2l is
rather long, and the equilibrium looks decidedly wobbly. In the third one, we’re not quite sure whether the equilibrium is stable
or not. What is the longest brick that is stable against small angular displacements from the vertical?

22.7.6 https://phys.libretexts.org/@go/page/9091
 Exercise 22.7.16
A Thing with a semicylindrical (or hemispherical) base of radius a is balanced on top of a rough semicylinder (or hemisphere)
of radius b as shown. The distance of the centre of mass of the Thing from the line (or point) of contact is l. Show that the
equilibrium is stable if
1

l
  > 
1

a
+
1

b
.

If a = b , is the equilibrium stable if the Thing is

1. A hollow semicylinder?
2. A hollow hemisphere?
3. A uniform solid semicylinder?
4. A uniform solid hemisphere?

 Exercise 22.7.17

A log of square cross-section, sides 2a, rests on two smooth pegs a distance 2ka apart, one of the diagonals making an angle θ
with the vertical.
Show that, if k  <  
1
  =  0.354 the only equilibrium position possible is θ = 90o, but that this position is unstable;
√8

consequently, following a small displacement, the log will fall out of the pegs. Show that if the pegs are farther apart, with
0.354  <  k  <  0.500 , three equilibrium positions are possible. Which of them are stable, and which are unstable? If

k = 0.45 , what are the possible equilibrium values of θ ? Show that, if 0.500  <  k  <  1.414 , only one equilibrium position is

possible, and that it is stable.

 Exercise 22.7.18

22.7.7 https://phys.libretexts.org/@go/page/9091
A uniform solid hemisphere of radius a rests in limiting static equilibrium with its curved surface in contact with a smooth
vertical wall and a rough horizontal floor (coefficient of limiting static friction μ ). Show that the base of the hemisphere makes
an angle θ with the floor, where
8μ
sin  θ  =  
3
.
Calculate the value of θ if (a) μ  =   1

4
and μ  =   .
3

What happens if μ  >   ?3

 Exercise 22.7.19
A uniform rod of length 2l rocks to and fro on the top of a rough semicircular cylinder of radius a . Calculate the period of
small oscillations.

A uniform solid hemisphere of radius a with its curved surface in contact with a rough horizontal table rocks through a small
angle. Show that the period of small oscillations is
−−
−
P   =  2π √
26a

15g
.

 Exercise 22.7.21
The density ρ of a solid sphere of mass M and radius a varies with distance r from the centre as
ρ  =  ρ0 (1 −
r

a
) .
Calculate the (second) moment of inertia about an axis through the centre of the sphere. Express your answer in the form of
constant % M a .2

22.7.8 https://phys.libretexts.org/@go/page/9091
 Exercise 22.7.22

Two identical particles are connected by a light string of length 2aα . The system is draped over a cylinder of radius a as
shown, the coefficient of limiting static friction being μ . Determine the angle θ when the system is in limiting equilibrium and
just about to slide.

 Exercise 22.7.23

A mass M hangs from a light rope which passes over a rough cylinder, the coefficient of friction being μ and the angle of lap
being α . What is the least value of F , the tension in the upper part of the rope, required to prevent the mass from falling?

 Exercise 22.7.24

A wooden cube floats on water. One of its faces is freely hinged to an axis fixed in the surface of the water. The hinge is fixed
at a distance from the top of the face equal to x times the length of a side. The opposite face is submerged to a distance y times
the length of a side. Find the relative density s (that is, relative to the density of the water) of the wood in terms of x and y .

22.7.9 https://phys.libretexts.org/@go/page/9091
 Exercise 22.7.25
A uniform solid sphere sits on top of a rough semicircular cylinder. It is given a small displacement so that it rolls down the
side of the cylinder. Show that the sphere and cylinder part company when the line joining their centres makes an angle 53o 58'
with the vertical.

 Exercise 22.7.26

A student has a triangular sandwich of sides 9 cm, 12 cm, 15 cm. She takes a semicircular bite of radius 3 cm out of the middle
of the hypotenuse. Where is the centre of mass of the remainder? Is it inside or outside the bite?

 Exercise 22.7.27

An rubber elastic band is of length 2πa and mass m; the force constant of the rubber is k . The band is thrown in the air,
spinning, so that it takes the form of a circle, stretched by the centrifugal force. (This takes much practice, skill and manual
dexterity.) Find a relation between its radius and angular speed, in terms of a,  m and k .

 Exercise 22.7.28

Most of us have done simple problems on friction at high school or in first year at college or university. You know the sort – a
body lies on a rough horizontal table. A force is applied to it. What happens? Try this one.
Find a uniform rod AB. A ruler will do as long as it is straight and not warped. Or a pencil of hexagonal (not circular) cross-
section, provided that it is uniform and does not have an eraser at the end. Place it on a rough horizontal table. Gradually apply
a horizontal force perpendicular to the rod at the end A until the rod starts to move. The end A will, of course, move forward.
Look at the end B – it moves backward. There is a point C somewhere along the rod that is stationary. I.e., the initial motion of
the rod is a rotation about the point C. Calculate – and measure – the ratio AC/AB. What is the force you are exerting on A
when the rod is just about to move, in terms of its weight and the coefficient of friction?

22.7.10 https://phys.libretexts.org/@go/page/9091
 Exercise 22.7.29

Some of the more dreaded friction problems are of the “Does it tip or does it slip?” type. This and the following four are
examples of this type.

A uniform solid right circular cone of height h and basal radius a is placed on an inclined plane whose inclination to the
horizontal is gradually increased. The coefficient of limiting static friction is μ . Does the cone slip, or does it tip?

 Exercise 22.7.30

A cubical block of side 2a rests on a rough horizontal table, the coefficient of limiting static friction being μ . A gradually
increasing horizontal force is applied as shown at a distance x above the table. Will the block slip or will it tip? Show that, if
μ  <
1

2
, the block will slip whatever the value of x.

 Exercise 22.7.31

A cylindrical log of diameter 2a and mass m rests on two rough pegs (coefficient of limiting static friction μ ) a distance 2ka
apart. A gradually increasing torque τ is applied as shown. Does the log slip (i.e. rotate about its axis) or does it tip (about the
right hand peg)?
When you’ve done that one, you can try a variant (which I haven’t worked out and haven’t posted a solution) in which a
cylinder of radius a is resting against a kerb (or curb, if you prefer that spelling) of height h , and a torque is applied. Will it tip
or will is slip?

22.7.11 https://phys.libretexts.org/@go/page/9091
 Exercise 22.7.32

This problem reveals the severe limitations of my artistic abilities, but the drawing above, believe it or not, represents a motor
car seen from behind. You can see the driver and passenger. The height of the centre of mass is h and the distance between the
wheels is 2d. The car is travelling on a horizontal road surface, coefficient of friction μ , and is steering to the left in a circle of
radius R , the centre of curvature being way off to the left of the drawing. As they gradually increase their speed, will the car
slip to the right, or will it tip over the right hand wheel, and at what speed will this disaster take place? Fortunately, driver and
passenger were both wearing their seat belts and neither of them was badly hurt, and never again did they drive too fast round a
corner.

 Exercise 22.7.33

A uniform rod of length 2l rests on a table, with a length l − a in contact with the table, and the remainder, l + a sticking over
the edge – that is, a is the distance from the edge of the table to the middle of the rod. It is initially prevented from falling by a
force as shown. When the force is removed, the rod turns about A. Show that the rod slips when it makes an angle θ with the
horizontal, where
2μ
tan  θ  =   a 2
2 + 9( )
l

Here m is the coefficient of limiting static friction at A.

 Exercise 22.7.34
A flexible chain of mass m and length l is initially at rest with one half of it resting on a smooth horizontal table, and the other
half dangling over the edge:

22.7.12 https://phys.libretexts.org/@go/page/9091
It is released, so that it starts to slide off the table. At a subsequent time t , a length 1

2
l−x remains in contact with the table, the
remaining length l + x hanging vertically, and the speed of the chain is v .
1

Show that
g
2
v   =  gx  +  
l
x
2
,
g
√
2
lt
l( e −1 )
x  =   g
√
lt
4e

4g
√
lt
√gl( e −1)
v  =  
4g
√
lt
4e

 Exercise 22.7.35a

Four books, each of width 2w, are stacked on top of each other in a heap, thus:

What is the maximum possible overhang, D?

 Exercise 22.7.35b
How many books would be needed to achieve an overhang of 10w?

 Exercise 22.7.35c

Given an unlimited supply of books, what is the maximum overhang achievable?

 Exercise 22.7.36

The Man and the Dog.

At time t = 0 , the Man is at the origin of coordinates, and he starts to walk up the y -axis at constant speed v . The Dog starts at
(a, 0) and runs at constant speed Av(v > 1) towards the Man. The velocity of the Dog is always directed straight towards the

22.7.13 https://phys.libretexts.org/@go/page/9091
Man. Find an equation for the path pursued by the Dog, and draw a graph of this path. How far has the Man walked when the
Dog reaches the Man, and how long does this take?

 Exercise 22.7.37

A particle A of mass m is attached by a light string to a second particle, B, also of mass m. A rests on a smooth horizontal
table, while B hangs vertically through a hole in the table. At time zero, the length of the horizontal portion of the string (i.e.
the distance of A from the hole) is a, and A is moving on the table in a horizontal circle of radius a with initial angular speed
ω .
0

At some subsequent time the length of the horizontal portion of the string is r and the angular speed of A is ω. Let us denote by
the rate of increase of r with time, which will evidently be negative if B is falling.

 Exercise 22.7.37a
Show that ṙ is given by
2 2 −−
−
ṙ aω ω ω
0
=1+ (1 − ) −√ . (22.7.1)
ga 2g ω0 ω0

 Exercise 22.7.37b

Show that, if aω 2
0
=g , where Ω = ω ω .
0

2
ṙ 3 1 −
−
= − Ω − 1/ √Ω , (22.7.2)
ga 2 2

 Exercise 22.7.37c
Show that there is only one value of Ω, namely 1, for which there is a real solution for ṙ , namely ṙ = 0 . This implies that the
system remains in equilibrium, with the radius of the circle, the angular speed of A and the height of B remaining constant,
with the centrifugal force on A remaining equal to the weight of B.

 Exercise 22.7.37d
Show that if aω 2
0
= 2g ,
ṙ
2
−
−
= 2 − Ω − 1/ √Ω
ga

and that A moves outwards (its angular speed decreasing) and B moves upwards,
reaching a maximum speed of ṙ −−
= 0.331841 √ga

where ṙ = 1.259921a

when ω = 0.629961ω 0,

and it reaches an equilibrium when ṙ =0

22.7.14 https://phys.libretexts.org/@go/page/9091
 where ṙ = 1.618034a

when ω = 0.381966ω 0.

 Exercise 22.7.37e

Show that if aω 2
0
=
1

2
g ,
ṙ
2
5 1
−
−
= − Ω − 1/ √Ω ,
ga 4 4

and that A moves inwards (its angular speed increasing) and B moves downwards,
reaching a maximum speed of ṙ −−
= −0.243822 √ga

where r = 0.793701a
when ω = 1.587401ω 0,

and it reaches an equilibrium when ṙ =0

where r = 0.640338a
when ω = 2.438447ω 0.

 Exercise 22.7.38

This problem – the bifilar torsion pendulum − was suggested to me by Claude Plathey, who used the method in a practical
application to determine the rotational inertia (moment of inertia) of a real nonuniform rod. He also drew my attention to an
interesting paper on the determination of the moments of inertia of bodies (such as aircraft!) by the method:
naca.larc.nasa.gov/digidoc/re...ACA-TR-467.PDF

A symmetric but not necessarily uniform rod of mass m and moment of inertia I is suspended from the ceiling by two light
threads each of length L a distance D apart (D << L) .The rod is twisted about a vertical axis through its midpoint through a
small angle and then released. Find the period of small oscillations in the horizontal plane.

 Exercise 22.7.39

A yo-yo is of mass M and rotational inertia I . The radius of its axle is a , and it falls in the usual way with a length of string
wrapped around the axle.
How that its linear acceleration downwards is

22.7.15 https://phys.libretexts.org/@go/page/9091
 2
Ma
2
×g
M a +I

and that the tension in the string i

I
2
× Mg
M a +1

 Exercise 22.7.40a

A yo-yo, mass M , axle radius a , outer radius b , rests on a horizontal table as shown. The string, wrapped around the axle, is
held vertically as shown, and a force P is applied. The coefficient of friction between yo-yo and table is μ . Show that, if
MabP
μ > 2
,
(Mg−P )(I+Mb

the initial motion of the yo-yo will be to roll to the left without slipping, with an initial linear acceleration
abP
;
I+Mab

but that if
MabP
μ < 2
,
(Mg−P )(I+Mb

the yo-yo will rotate counterclockwise without rolling, with an initial angular acceleration about C of
P (a−μ)−μMg
.
I

 Exercise 22.7.40b

A yo-yo, mass M , axle radius a , outer radius b , rests on a horizontal table as shown. The string, wrapped around the axle, is
held horizontally as shown, and a force P is applied. The coefficient of friction between yo-yo and table is μ .
(i) Show that, if I > M ab :
If
I−Mab P
μ  >   ( 2
)( )
I+Mb Mg

the initial motion of the yo-yo will be to roll to the right without slipping, with an initial linear acceleration
P b(a+b)

2
;
I+Mb

22.7.16 https://phys.libretexts.org/@go/page/9091
but that, if

μ  <   (
I−Mab

2
)(
P

Mg
) ,
I+Mb

the yo-yo simultaneously accelerates to the right with a linear acceleration of

P −μMg

while undergoing a clockwise angular acceleration about C of

P a + μMgb

I
.
(ii) Show that, if I > M ab :
If

μ  >   (
Mab−I
2
)(
P

Mg
) ,
M b +I

the initial motion of the yo-yo will be to roll to the right without slipping, with an initial linear acceleration
P b(a+b)

2
;
I+Mb

but that, if

μ  <   (
Mab−I

2
)(
P

Mg
) ,
M b +I

the yo-yo simultaneously accelerates to the right with a linear acceleration of

P  + μMg

while undergoing a clockwise angular acceleration about C of

P a − μMgb

I
.
(iii) Show that, if I > M ab :

 Exercise 22.7.40c

(After Problem 40(b) this one is a good deal easier and a welcome relief.)

A yo-yo, mass M , axle radius a , outer radius b , rests on a horizontal table as shown. The string, wrapped around the axle, is
held horizontally as shown, and a force P is applied. The coefficient of friction between yo-yo and table is m.
Show that, if

μ  >   (
I+Mab
2
)(
P

Mg
) ,
I+Mb

the initial motion of the yo-yo will be to roll to the right without slipping, with an initial linear acceleration
P b(b−a)

2
I+Mb

and angular acceleration

22.7.17 https://phys.libretexts.org/@go/page/9091
 P (b−a)

2
I+Mb

but that, if
I+Mab P
μ  <   ( 2
)( )
I+Mb Mg

P −μMg
the yo-yo slips at A. C accelerates to the right at a rate of M
, while the yo-yo spins around C with a counterclockwise
P a−μMgb
angular acceleration of I
.

 Exercise 22.7.40d

A yo-yo, mass M , axle radius a , outer radius b , rests on a horizontal table as shown. The string, wrapped around the axle, is
held at an angle θ to the horizontal as shown, and a force P is applied. The coefficient of friction between yo-yo and table is μ .
The complete analysis of this problem is similar to that of Problem 40(b), except that a factor of cos θ appears in many of the
equations. No new phenomena appear, and the analysis is tedious without any new points of interest. For that reason I limit this
problem to asking you to show that the direction of the frictional force of the table on the yo-yo at A depends upon whether the
cos θ is less than or greater than .
Mab

 Exercise 22.7.40e

A yo-yo, mass M , axle radius a , outer radius b , rests on a horizontal table as shown. The string, wrapped around the axle, is
held at an angle θ to the horizontal as shown, and a force P is applied. The coefficient of friction between yo-yo and table is μ .
Show that, provided there is no slipping, the yo-yo rolls to the right if cos θ >
a

b
and to the left if cos θ <
a

b
. Describe what
happens if cos θ = .a

22.7.18 https://phys.libretexts.org/@go/page/9091
 Exercise 22.7.41

A uniform plane lamina of mass 3m is in the form of a truncated square like the one above.
Find the position of the centre of mass, the principal moments of inertia with respect to the centre of mass, and the eccentricity
and inclination of the momental ellipse.

 Exercise 22.7.42

A mass M sits on a smooth horizontal table. A second mass, m, hangs from the first by a light inextensible string. A slot in the
table allows m and the string to swing as a pendulum.

The system is then set in motion with the pendulum swinging, and the mass M sliding back and forth on the table. At some
instant when the horizontal displacement of M from its equilibrium position is x the string makes an angle θ with the vertical.

Show that the equations of motion are

2
¨ ˙
(M + m)ẍ  +  ml(θ cos θ  −  θ sin θ)  =  0 ,
¨ ˙
I θ   +  ẍ cos θ  −  ẋ sin θθ   =   − g sin θ .
2
Show that for small oscillations ˙ ¨ ˙
(cos θ ≈ 1,   sin θ ≈ θ,  θ θ << θ ,  ẋθθ << g) the period the motion is approximately
−−−−−− −
−
2π √
Ml

(M+m)g
. Note that, if m << M , this reduces to 2π √ as expected.
g
l

22.7.19 https://phys.libretexts.org/@go/page/9091
 Exercise 22.7.43

A gun projects a shell, in the absence of air resistance, at an initial angle to the horizontal. The speed of projection varies with
the angle α of projection and is given by
initial speed = V 0 cos
1

2
α .
Show that, in order to achieve the greatest range on the horizontal plane, the shell should be projected at an angle to the
horizontal whose cosine c is given by the solution of the equation
3 2
3c + 2c − 2c − 1 = 0

and determine this angle to the nearest arcminute.

 Exercise 22.7.44

The length of a cylindrical log is L times its diameter, and its density is s times that of water (0 < s < 1) . Show that the log
can float vertically in stable equilibrium, whatever its density, provided that L < 0.707. and that, if its length is greater than
this, it can float vertically in stable equilibrium only if
L <
1
.
√8s(1−s)

Show that, if the length is equal to the diameter, it can float in stable equilibrium with its cylindrical axis vertical only if its
density is less than 0.146 or greater than 0.854 times that of water.

 Exercise 22.7.45

A uniform heavy rod of length 6 hangs from a fixed point C by means of two light strings of lengths 4 and 5. What angle does
the rod make with the horizontal?
Incidentally, a (4, 5, 6) triangle has the interesting property that one of its angles is exactly twice one of the other ones.

 Exercise 22.7.46

22.7.20 https://phys.libretexts.org/@go/page/9091
 A uniform rod rests on two smooth (frictionless) planes inclined at 30º and 45º to the horizontal. What angle does the rod make
with the horizontal?

 Exercise 22.7.47

A uniform rod of length 2l rests on the inside of a circular cylindrical pipe of radius a . The coefficient of limiting static friction
(often known for short, if with less precision, as “the”coefficient of friction) μ . What is the maximum angle θ that the rod can
make with the horizontal in static equilibrium?

This page titled 22.7: Appendix A is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

22.7.21 https://phys.libretexts.org/@go/page/9091
22.8: Appendix B
Solutions to Miscellaneous Problems

 Exercise 22.8.1

By proportions, h

k
=
n

x
and h

l
=
m

x
and therefore h

k
+
h

l
=1 .
Therefore by Pythagoras:

h(
1
+
1
) = 1. .
√a2 −x2 √b2 −x2

Everything but x is known in this equation, which can therefore be solved for x. There are several ways of solving it; here’s a
suggestion. If we put in the numbers, the equation becomes

1 1
3( + ) − 1 = 0.
√64−x2 √100−x2

Put X = 100 − x , and the equation becomes

2

1 1
3( + ) − 1 = 0.
√X−36 √X

This can be written f (X) = 3(A + B) − 1 = 0 , where A and B are obvious functions of X. Differentiation with respect to X
f
gives f (X) = − (A + B ) and Newton-Raphson iteration (X = X − ) soon gives X, from which it is then found that
′ 3

2
3 3
′
f

x = 6.326 182 m.

 Exercise 22.8.2

In the corotating frame the bob is in equilibrium under the action of three forces – its weight, the tension in the string and the
centrifugal force. (If you do not like rotating reference frames and centrifugal force, it will be easy for you to do it “properly”.)

22.8.1 https://phys.libretexts.org/@go/page/9092
Resolve the forces perpendicular to the string: ml 0
2
sin α Ω . cos α  =  mg sin α and the problem is finished.

 Exercise 22.8.3a
Raising or lowering the board does not apply any torques to the system, so the angular momentum L is conserved. That is,
2 2
L = ml sin θ. ω is constant. (1)

We also have that

g  =  l cos θ. ω. (2)

i. Eliminate w from these equations. This gives:

2
3 3
L
l sin θ tan θ = (3)
gm2

which is constant.
ii. Eliminate l from equations (1) and (2). This gives:
2
3 2
mg
ω cot θ = , (4)
L

which is constant.
iii. Eliminate θ from equations (1) and (2). This gives:
L
3 2 2
ω (ωl − ) =g . (5)
m

(Check the dimensions of all the equations.) Then we can get L

m
from equation (1) and hence
3
ω (ωl
2
− Ωl
2
0
sin
2
α) = g
2
,
which is constant.

 Exercise 22.8.3b

i.
3 3 3 3 3
l sin θ tan θ = l sin α tan α = 0.023675 m .
0

Although we are asked to plot θ vertically versus l horizontally, it is easier, when working out numerical values, to calculate l
as a function of θ . That is,
l  =  
0.287142
3
.
sin θ√tan θ

(The number in the numerator is the cube root of 0.023675.)

For l = 40 cm = 0.4 m, the semivertical angle is given by

sin
3
θ tan θ = 0.369 923 .

22.8.2 https://phys.libretexts.org/@go/page/9092
The solution to this is
θ = 45  31
∘ ′
.
(See Section 1.4 of Celestial Mechanics if you need to know how to solve the equation f (x) = 0 .)
ii.
3 2 3 2
ω cot θ =Ω cot α

With the given data, this is ω 3

= 199.385 tan
2
θ .

iii. ω3 2
(ωl − Ωl
2
0
2
sin
3 2
α) =  Ω (Ω l   −  Ω l
0
2
0
sin
2
α) = Ω l
4 2
0
cos
2
α .
That is, ω 3 2
(ωl   − a) =  g
2
where, with the given initial data,
a = 0.48168 m s
2 −1
and g 2
  =  96.04 m s
2 −4
.
Although we are asked to plot ω vertically versus l horizontally, it is easier, when working out numerical values, to calculate l
as a function of ω. That is,
2
g
2
l   = 
ω
4
+
a

ω
.

To solve the above equation for ω might be slightly easier with the substitution of u for 1

ω
:
2 4
g u   +  au  − l
2
=0 .
With l = 0.6 m, this gives u= 0.226121 rad s, and hence ω = 4.422 rad s-1. As in part (b) i, it is necessary to know how to
-1

solve the equation f (x) = 0 . See Section 1.4 of Celestial Mechanics if you need to know how.

22.8.3 https://phys.libretexts.org/@go/page/9092
 Exercise 22.8.4
There are no horizontal forces, because the table is smooth. Therefore the centre of mass of the rod falls vertically.centre of
mass of the rod falls vertically.

From energy considerations

1 1 1 2
2 2 ˙
m ẏ  +  ( m l ) θ   +  mgy  = constant. (1)
2 2 3

But y  =  l cos θ and therefore ẏ ˙

= −l sin θ. θ .
2
2 ˙
(3 sin θ  +  1)l θ   +  6g sin θ = C . (2)

Initially θ = θ˙ = 0, ∴ C = 6g .
2 6g(1 − cos θ)
θ̇   =  (3)
2
l(3 sin θ + 1)
–––––––––––––––––––
–––––––––––––––––––– –
2 2
Also, since ẏ 2 2
=l
2
sin
˙
θθ and ẋ 2
=l
2
cos
2 ˙
θθ , we obtain
2
6gl sin θ(1 − cos θ)
2
ẏ   =  (4)
2
3 sin θ + 1
–––––––––––––––––––––––––
––––––––––––––––––––––––––
–

and
2
6gl cos θ(1 − cos θ)
2
ẋ   =   (5)
2
3 sin θ + 1
–––––––––––––––––––––––––
––––––––––––––––––––––––––
–

Of course ˙
θ and ẏ increase monotonically with θ ; but ẋ starts and finishes at zero, and must go through a maximum. With
c  =   cos θ , Equation (5) can be written
6glc(1 − c)
2
ẋ   =   , (6)
2
4 − 3c

and by differentiating ẋ with respect to c , we see that ẋ is greatest at an angle θ given by

2 2

2
3 c   − 12c + 8 = 0, (7)

the solution of which is θ  =  37  50

∘ ′
.
–––––––––––
––––––––––––
–

If the length of the rod is 1 m (l = 0.5 m) and ẋ = 1 m s-1, Equation 4 becomes

2
26.4 c   − 29.4c  +  4  =  0, (8)

22.8.4 https://phys.libretexts.org/@go/page/9092
and the two solutions are –
θ  =  17  15 and 80  52 .
–––––––––––
∘

–––––––
′ ∘ ′

–––––––––––– –––––––

The reader who has done all the problems so far will be aware of the importance of being able instantly to solve the equation
f (x) = 0 . If you have not already done so, you should write a computer or calculator program that enables you to do this

instantly and at a moment’s notice. See Section 1.4 of Celestial Mechanics if you need to know how.
If you want to find the normal reaction N of the table on the lower end of the rod, you could maybe start with the vertical
equation of motion mÿ = N − mg . Differentiate Equation (4): 2ẏ ÿ = whatever, and the use equation
4 again for ẏ . This looks like rather heavy and uninteresting algebra to me, so I shan’t pursue it. There may be a better way...

 Exercise 22.8.5

In the figure below I have marked in red the forces on the rod, namely its weight mg and the horizontal and vertical
components X and Y of the reaction of the hinge on the rod. I have also marked, in green, the transverse and radial
components of the acceleration of the centre of mass. The transverse component is lθ̈ and the radial component is the
2

centripetal acceleration ˙
lθ .centre of mass. The transverse component is ¨
lθ and the radial component is the centripetal
2
acceleration lθ˙ .

From consideration of the moment of the force mg about the lower end of the rod, it is evident that the angular acceleration is
3g sin θ
¨
θ = (1)
4l

˙ ˙
and by writing ¨
θ as θ dθ

dθ
and integrating (with initial conditions ˙
θ  =  θ   =  0 ), or from energy considerations, we obtain the
angular speed:

2 3g(1 − cos θ)
˙
θ = . (2)
2l

The horizontal and vertical equations of motion are:

2
X = ml(θ̈ cos θ − θ̇ sin θ) (3)

and
2
˙
mg  −  Y   =  ml(θ̈ sin θ  +  θ cos θ). (4)

(As ever, check the dimensions - and count the dots!)

2
After substitution for θ¨ and θ˙ we find
3
X  =   mg sin θ(3 sin θ − 2) (5)
4

and

22.8.5 https://phys.libretexts.org/@go/page/9092
 1 2
Y   =  mg(1 − 3 cos θ) . (6)
4

The results follow immediately.

 Exercise 22.8.6

Call the length of the rod 2l. Initially the height above the table of its centre of mass is l cos 40°, and its gravitational potential
energy is mg l  cos 40 . When it hits the table at angular speed w, its kinetic energy is I ω   =   ( ml ) ω   =   ml ω .
∘ 1

2
2 1

2
4

3
2 2 2

3
2 2

Therefore,
−−−−−−∘
3g cos 40
ω  =  √
2l
  =  4.746 rad s
−1
  = 271.9 deg s
−1
.
––––––––––––––––––––––––––––
––––––––––––––––––––––––––––––

To find the time taken, you can use Equation 9.2.10:

−− 90
∘

I dθ
t  =  √ ∫ . (7)
−−− −−−−−
2 40
∘
√ E − V (θ)

Here, I   =   3

4
2 ∘
m l ,  E = mgl cos 40 ,  V (θ)  =  mgl cos θ and therefore
−−
−
90
2l dθ
t  =  √ ∫ −−−−− −−−−− −. (8)
∘
3g 40 √ cos 40 − cos θ

The magnitude of the quantity before the integral sign is 0.184428 s. To find the value of the integral requires either that you be
an expert in elliptic integrals or (more likely and more useful) that you know how to integrate numerically (see Celestial
Mechanics 1.2.) I make the value of the integral 2.187314, so that the time taken is 0.4034 seconds. When integrating, note that
the value of the integrand is infinite at the lower limit. How to deal with this difficulty is dealt with in Celestial Mechanics 1.2.
It cannot be glossed over.

 Exercise 22.8.7

Here is the diagram. The forces are the weight mg of the rod, and the force of the table on the rod. However, I have resolved
the latter into two components – the normal reaction N of the table on the rod, and the frictional force F , which may be either
to the left or the right, depending on whether rod is tending to slip towards the right or the left. The magnitude of F is less than
μN as long as the rod is not jus about the slip. When the rod is just about to slip, F = μN ,  μ being the coefficient of limiting

static friction.

Just as in Problem 5, the equations of motion, as long as the rod does not slip, are
3
F  =   mg sin θ(3 cos θ − 2) (1)
4

and
1
2
N  =   mg(1 − 3 cos θ) . (2)
4

22.8.6 https://phys.libretexts.org/@go/page/9092
 F 3 sin θ(3 cos θ − 2)
= . (3)
2
N (1 − 3 cos θ)

The figure below shows F

N
as a function of θ . One sees that, as the rod falls over, increases, and, as soon as it attains a F

value of μ , the rod will slip. We see, however, that reaches a maximum value, and by calculus we can determine that it
F

N
15 √10
reaches a maximum value of 128
  =  0.3706 when θ  =  θ ( ) = 35 06 . If μ < 0.3706, the bottom of the rod will slip
−1

11
9 ∘ ′

before θ = 35 06 . If, however, μ > 0.3706, the rod will not have slipped by the time θ = 35 06 , and it is safe for a while as
∘ ′ ∘ ′

N
starts to decrease. When θ reaches cos ( ) = 48 11 , the frictional force changes sign and thereafter acts to the left. (The
−1 2

3
∘ ′

frictional force of the table on the rod acts to the left; the frictional force of the rod on the table acts to the right.) We know by
now (since the rod survived slipping before θ = 35 06 that the magnitude of
∘ ′
can be at least as large N
F

as 0.3706, and it does not reach this until θ = 51 15 . Therefore, if the rod hasn’t slipped by θ = 51 15 it won’t slip before
∘ ′ ∘ ′

θ = 51 15 . But after that it is in danger again of slipping. becomes infinite (N = 0 ) when θ = cos ( ) = 70 32 , so it
∘ ′ F −1 1 ∘ ′

N 3

will certainly slip (to the right) before then.

If μ = 0.25, the rod will slip to the left when
3 sin θ(3 cos θ−2)

2
  = 
1

4
,  or θ = 19 39
∘ ′
.
(1−3 cos θ) –––––––––
––––––––––
–

If μ = 0.75, the rod will slip to the right when

3 sin θ(3 cos θ−2)
  =  −
3
,  or θ = 53 07
∘ ′
.
(1−3 cos θ)
2
4 –––––––––
––––––––––
–

Again, it is very necessary that you prepare for yourself a program that will instantly solve the equation f (x) = 0 .

 Exercise 22.8.8

22.8.7 https://phys.libretexts.org/@go/page/9092
Let the length of the ladder be 2l. By geometry, the distance OC remains equal to l throughout the motion; therefore C
describes a circle of radius l, centre O. I have marked in, in green, the radial and transverse components of the acceleration of
2
C, namely lθ˙ and lθ¨. The angular speed of the ladder is θ˙ and the linear speed of the centre of mass C is lθ¨. I have also
marked, in red, the three forces acting on the ladder, namely its weight and the reactions of the floor and the wall on the
2
ladder.centre O. I have marked in, in green, the radial and transverse components of the acceleration of C, namely lθ˙ and lθ¨.
The angular speed of the ladder is θ˙ and the linear speed of the centre of mass C is lθ¨. I have also marked, in red, the three
forces acting on the ladder, namely its weight and the reactions of the floor and the wall on the ladder.
The angular speed θ˙ can be obtained from energy considerations. That is, the loss of potential energy in going from angle a to
the vertical to angle θ is equal to the gain in translational and rotational kinetic energies:
2
mgl(cos α − cos θ) =
1

2
˙ 2
m(l θ )   +  
1

2
(
1

2
2 ˙
ml ) θ .
2 3g
˙
θ = (cos α − cos θ). (1)
2l

The angular acceleration θ¨ can be obtained from the following equation:

4
2 ¨
mgl sin θ = ml θ (2)
3

The derivation of Equation (2) raises some points of interest, and I discuss it in an Appendix at the end of the problem.
The vertical and horizontal equations of motion are:
2
¨ ˙
N2   =  m(l θ cos θ − l θ sin θ) (3)

and
¨ ˙
mg  −  N1   = m(l θ sin θ  +  l θ cos θ), (4)

although we need only the first of these, because we wish to find out when N 2  =  0 .
2
On substitution for θ¨ and θ˙ we find that
3
N2   =   mg sin θ(3 cos θ − 2 cos α) (5)
4

and

22.8.8 https://phys.libretexts.org/@go/page/9092
 1 2
N1   =   mg(1 − 6 cos α cos θ + 9 cos θ) (6)
4

We need only the first of these to see that N2 becomes zero (and hence the upper end loses contact with the wall) when
cos α .
2
cos θ =
3

Appendix: Derivation of equation (2).

In my original posting of this solution I had derived Equation (2) by considering that the total moment of all forces about Q is
mgl sin θ, and the rotational inertia with respect to Q is m l . I then equated mgl sin θ to m l θ . I am indebted to correspondent
4 2 4 ¨ 2

3 3

Amin Rezaee Zadeh for pointing out a flaw in this argument, and for supplying a correct derivation. The flaw is that I am applying
the equation τ = L˙
to a moving point Q. In Section 3.12 of Chapter 3 of these notes it is pointed out that τ = L˙
can be applied to a
moving point only if the moving point satisfies one or more of three conditions, and it is evident in this problem that Q satisfies
none of these conditions. I present Mr Rezaee’s correct derivation of Equation (2 ) below.
I shall be making use of Equations 3.12.1 and 3.12.2:
′
L̇Q   =  τQ   +  M rq   ×  r̈ Q . (3.12.1)

LQ   =  Σ(ri − rQ ) × [ mi (vi − vQ )]. (3.12.2)

I shall also be making use of the notation used in Section 3.12, and I reproduce here Figure III.7 from that Section, and I also draw
the relevant vectors appropriate to this ladder problem.

In the figure below, I have indicated an elemental portion ds of the ladder at a distance s from the upper end of the ladder. Its mass
is evidently dm = mds

2l
. I have drawn the position vectors r and r of ds and of Q. This notation corresponds to the same
i O

notation used in Section 3.12. From the geometry of the figure, we can determine that

ri   =  s sin θi  +  (2l − s) cos θj (A1)

22.8.9 https://phys.libretexts.org/@go/page/9092
and
rQ   =  2l sin θi  +  2l cos θj, (A2)

where i and j are the unit vectors in the x− and y− directions respectively.

On differentiation with respect to time, we find the following expressions for the velocities of the element ds and the point Q, in
which I again retain the notation used in Section 3.12:
˙ ˙
vi   =  sθ cos θi  −  (2l − s)θ sin θj (A3)

and
˙ ˙
vQ   =  2l θ cos θi  −  2l θ sin θj (A4)

On making use of Equation 3.12.2, we obtain for the angular momentum of the element ds with respect to Q:
m
dLQ   =   (ri − rQ ) × [(vi − vQ )]ds. (A5)
2l

The instantaneous angular momentum of the entire ladder about Q is therefore

2l
m
LQ   =   ∫ (ri − rQ ) × [(vi − vQ )]ds. (A6)
2l 0

On substitution of Equations (A1) – (A4) into equation (A6) and a modest amount of algebra, we obtain
2l
mθk 2
2 ˙
LQ   =   ∫ s(s − 2l)ds  =   − m l θ k, (A7)
2l 0
3

where k is the unit vector in the z− direction. (The z− direction is out of the plane of the “paper”, and therefore L is into theQ

plane of the “paper”. It is worth spending a moment or two trying to imagine this. The ladder is rotating counterclockwise about C,
while C and Q are moving in clockwise trajectories. It may not be immediately obvious to decide whether one would expect L to Q

be directed into or out of the plane of the “paper”. Equation (A7) answers this question.)
We now make use of Equation 3.12.1:
˙ ′
LQ   =  τQ   +  m r Q × r̈ Q . (A8)

Let us find expressions for the four vector quantities in this equation.
By differentiation of Equation (A7) with respect to time, we obtain
2 2 ¨
˙
LQ   =   − m l θ k. (A9)
3

22.8.10 https://phys.libretexts.org/@go/page/9092
The torque about Q is
τQ   =  mgl sin θk (A10)

We can see from the geometry of the figure (see especially the second of our figures, in which we see that r and r are the same in ′
Q
¯
¯¯

magnitude and direction) that

′
r   =  l sin θi  +  l cos θj. (A11)
Q

Finally, by differentiation of Equation (A4) (in which ), we obtain

2 2
¨ ˙ ¨ ˙
r̈Q   =  2l[(θ cos θ − θ sin θ)i  −  (θ sin θ − θ cos θ)j]. (A12)

Substitution of Equations (A9) to (A12) into Equation (A8) gives, after some algebra,
4
2
mgl sin θ  =   m l θ̈ . (A13)
3

This is Equation (2), quod erat demonstrandum.

 Exercise 22.8.9

It will, I think, be agreed that the point O remains fixed in space as long as the semicylinder remains in contact with wall and
floor. Therefore the centre of mass C moves in a circle around O. We’ll call the radius of the circle, which is the distance
between O and C, b , which, for a semicylinder, equals centre of mass C moves in a circle around O. We’ll call the radius of the
circle, which is the distance between O and C, b , which, for a semicylinder, equals (see Chapter 1), where a is the radius
4a

(3π)

of the semicylinder. I have marked, in red, the three forces on the semicylinder, and also, in green, the radial and transverse
components of the acceleration.
The angular speed ˙
θ can be obtained from energy considerations. The gain in kinetic energy in going from rest to an angular
2
speed θ˙ is (mk ) θ˙ and the gain in potential energy when the centre of mass drops through a vertical distance
1

2
2
b  sin θ is
mgb sin θ . Here k is the radius of gyration about O, which, for a semicylinder, is given by k = a .
2 1 2

[I have left b and k as they are in the equations, so that the analysis could easily be adapted, if needed, for a hollow
semicylinder, or a solid hemisphere, or a hollow hemisphere. From Chapters 1 and 2 we recall:
2

Solid semicylinder: b =
4a

3π
2
k   =  
1

2
a
b
2
  =  
32

9π 2
k

Hollow semicylinder: b =
2a

π
2
k   =  a
2 b
2
  =  
4

π2
k

Solid hemisphere: b =
3a

8
2
k   =  
2

5
a
2 b
2
  =  
45

128
k

22.8.11 https://phys.libretexts.org/@go/page/9092
 2

Hollow hemisphere: b =
1

2
a
2
k   =  
2

3
a
2 b
2
  =  
3

8
k

On equating the gain in kinetic energy to the loss in potential energy, we obtain
2 2bg
˙
θ   =  sin θ. (1)
2
k

The angular acceleration θ¨ can be obtained from applying τ   =  I θ¨ about O:

2 ¨
mgb cos θ  =  mk θ ,
from which
bg
θ̈   =   cos θ (2)
2
k

The horizontal and vertical equations of motion are

2
N2   =  mb(θ̇ cos θ  + θ̈ sin θ) (3)

and
2
˙
N2   =  mb(θ cos θ  + θ̈ sin θ) (4)

We do not really need Equation (4), because we are trying to determine when N 2 =0 .
On substitution from Equations (1) and (2), Equation (3) becomes
2
6m b g
N1 = sin θ cos θ. (5)
2
a

This is zero when θ  =  0 (which was the initial condition) or when

∘
θ  =  0
∘
, at which point contact with the wall is lost,
which it was required to show.
−−
−
2bg
At this instant, the rotational velocity is √ 2
counterclockwise.
k

−−
−
2bg
and the linear velocity of C is b√ 2
horizontally to the right.
k

−−
−
2bg
The rotational kinetic energy is 1

2
Iω
2
where ω  =  √ 2
, and I is the rotational inertia about the centre of masscentre of
k

mass, which is m(k 2 2

−b ) .
2 2
mbg( k −b )
Krot = 2
.
k

−−
−
2bg
The translational kinetic energy is 1

2
mv
2
where v  =  √ 2
.
k

3
mb g
Ktr   =   2
k

The sum of these is mbg, which is just equal to the loss of the original potential energy, which serves as a check on the
correctness of our algebra.
There are now no horizontal forces, so the horizontal component of the velocity of C remains constant. The semicylinder
continues to rotate, however, until the rotational kinetic energy is converted to potential energy and C rises to its maximum
height. If the base then makes an angle ϕ with the vertical, the gain in potential energy is mbg sin ϕ, and equating this to the
rotational kinetic energy gives
2

sin ϕ  =  1  −  
b
2
.
k

This gives the following results:

Solid semicylinder: ∘
ϕ  =  39  46
′

Hollow semicylinder: ∘
ϕ  =  36  30
′

22.8.12 https://phys.libretexts.org/@go/page/9092
 Solid hemisphere: ∘
ϕ  =  40  25
′

Hollow hemisphere: ∘
ϕ  =  38  41
′

 Exercise 22.8.10

Add text here. For the automatic number to work, you need to add the "AutoNum" template (preferably at the end) to the page.

It is well known that if α  >   tan μ the particle will slide down the plane unless helped by an extra force. I have drawn the
−1

three forces acting on the particle. Its weight mg. The reaction R of the plane on the particle; if the particle is in limiting static
equilibrium, this reaction will make an angle λ (“the angle of friction”) with the plane such that tan λ  =  μ . It therefore
makes an angle α − θ with the vertical. Finally, the additional force P needed; we do not initially know the direction of this
force.
When three (or more) coplanar forces are in equilibrium and are drawn head-to-tail, they form a closed triangle (polygon). I
draw the triangle of forces below.

It will be clear from the triangle that P is least when the angle between P and R is 90 : ∘

22.8.13 https://phys.libretexts.org/@go/page/9092
 μ
The least value of P is therefore mg(sin α cos λ  −   cos α sin λ) . But tan λ = μ and therefore sin λ =
2
and
√1+μ

cos λ =
1
2
.
√1+μ

mg(sin α−μ cos α)

Pmin =
2
√1−μ

and P then makes an angle \( \lambda\) with the plane.

You may, if you wish, go further, and show that when P makes an angle β with the plane, it must have magnitude
sin α−μ cos α
P =
μ sin β +  cos β
mg .

You can then differentiate this with respect to β (you need only differentiate the denominator) and show that this is a minimum
when β  =  λ . That is just a harder way of finding what we already found by using the triangle of forces.
For α  =  70 and μ  =  0.8 , P varies with β like this:
∘

This goes through a minimum of P min = 0.520mg at β  =   tan −1

0.8 = 38 .7
∘
.
_________________________

 Exercise 22.8.11
Add text here. For the automatic number to work, you need to add the "AutoNum" template (preferably at the end) to the page.

22.8.14 https://phys.libretexts.org/@go/page/9092
As the cylinder rolls down the plane, the wedge, because its base is smooth, will slide towards the left. Since there are no
external horizontal forces on the system, the centre of mass of the system will not move horizontally (or, rather, it won’t
accelerate horizontally.)centre of mass of the system will not move horizontally (or, rather, it won’t accelerate horizontally.)
As usual, we draw a large diagram, using a ruler , and we mark in the forces in red and the accelerations in green, after which
we’ll apply F   =  ma to the cylinder, or to the wedge, or to the system as a whole, in two directions. It should be easy and
straightforward.
I have drawn the linear acceleration s̈ of the cylinder down the slope, and its angular acceleration θ¨ . I have drawn the linear
acceleration ẍ of the wedge, which is also shared with the cylinder. I have drawn the gravitational force mg on the cylinder.
There is one more force on the cylinder, namely the reaction of the wedge on the cylinder. But I’m not sure in which direction
to draw it. Is it normal to the plane? That would mean there is no frictional force between the cylinder and the plane. Is that
correct (remembering that both the cylinder and the wedge are accelerating)? Of course I could calculate the moment of the
force mg about the point of contact of the cylinder with the plane, and then I wouldn’t need to concern myself with any forces
at that point of contact.
But then that point of contact is not fixed. Oh, dear, I’m getting rather muddled and unsure of myself.
This problem, in fact, is ideally suited to a lagrangian rather than a newtonian treatment, and that is what we shall do. Lagrange
proudly asserted that it was not necessary to draw any diagrams in mechanics, because it could all be done analytically. We are
not quite so talented as Lagrange, however, so we still need a large diagram drawn with a ruler. But, instead of marking in the
forces and accelerations in red and green, we mark in the velocities in blue.

No frictional or other nonconservative forces do any work, so we can use Lagrange’s equations of motion for a conservative
holonomic system; d

dt
(
∂T

˙
∂q
)−(
∂T

∂q
) = −(
∂V

∂q
) .

The speed of the wedge is ẋ and the speed of the centre of mass of the cylinder is centre of mass of the cylinder is
−−−−−−−−−−−−−− −
√ṡ + ẋ − 2 ṡ ẋ cos α and the angular speed of the cylinder is .
2 2 ṡ

The kinetic energy of the system is

T  =  
1

2
m(ṡ
2
+ ẋ
2
− 2 ṡ ẋ cos α)  +  
1

2
2
(m k ) (
ṡ

a
) + 
1

2
2
M ẋ ,

22.8.15 https://phys.libretexts.org/@go/page/9092
or
2

T  =  
1

2
m (1 +
k

a2
2
) ṡ   −  m ṡ ẋ cos α  +  
1

2
2
(m  +  M )ẋ ,

and the potential energy is

V   =  constant  −  mgs sin α

Application of Lagrange’s equation to the coordinate x gives us

2

(1 +
k

a2
) s̈ = ẍ cos α  +  g sin α .

and application of Lagrange’s equation to the coordinate s gives us

m s̈ cos α  =  (m + M )ẍ

Elimination of s̈ from these two equations gives us

mg sin α cos α
ẍ = 2
k
2
(m+M)(1+ )−m cos α
2
a

You can also easily find an expression for s̈ is you wish.

 Exercise 22.8.12

22.8.16 https://phys.libretexts.org/@go/page/9092
There is no acceleration normal to the plane, and therefore N   =  mg cos α . The frictional force F acts along the tangent to
the path and is equal to μN , or μmg cos α, where μ is the coefficient of moving friction. We are told to ignore the difference
between the coefficients of moving and limiting static friction. Since the particle was originally at rest in limiting static
friction, we must have μ  =   tan α . Therefore F   =  mg sin α . The tangential equation of motion is
m s̈   =   − F  + whatever the component of mg is in the tangential direction in the sloping plane.
The component of mg down the plane would be (look at the left hand drawing) mg sin α, and so its tangential component
(look at the right hand drawing) is mg sin α sin ψ. So we have, for the tangential equation of motion,
m s̈ = −mg sin α  +  mg sin α sin ψ ,
or
s̈ = −g sin α(1 − sin ψ) .
We are seeking a relation between V and ψ , so, in the now familiar fashion, we write V dV

ds
for s̈ , so the tangential equation of
motion is
dV
V = −g sin α(1 − sin ψ). (1)
ds

We also need the equation of motion normal to the trajectory. The component of mg in that direction is mg sin α cos ψ, and so
the normal equation of motion is
2
mV

ρ
  =  mg sin α cos ψ .

Here ρ is the radius of curvature of the path, which is the reciprocal of the curvature ds

dψ
. The normal equation of motion is
therefore
dψ
2
V   =  g sin α cos ψ. (2)
ds

Divide Equation (1) by Equation (2) to eliminate s and thus get a desired differential equation between V and ψ :
1 dV (1 − sin ψ)
  =  − . (3)
V dψ cos ψ

This is easily integrated; a convenient (not the only) way is to multiply top and bottom by 1  +   sin ψ . In any case we soon
arrive at
ln V   =   − ln(1 + sin ψ)  +  constant, (4)

and with the initial condition V   =  V when ψ  =  0 , this becomes

0

V0
V = . (5)
1 + sin ψ
–––––––––––––
–––––––––––––––

In the limit, as ψ → 90 ,  V → V . The particle is then moving at constant velocity and is in equilibrium under the forces
∘ 1

2
0

acting upon it just when it was initially at rest.

 Exercise 22.8.13

13.
M1 = mass of complete sphere of radius a .
M1 = mass of missing inner sphere of radius xa.
M = mass of given hollow sphere.
M2
We have M   =  M 2 − M1 and M1
=x
3
and therefore
3

M1   =  
M

1−x
3
and M 2  = 
Mx

1−x
3
.

22.8.17 https://phys.libretexts.org/@go/page/9092
 Also I =
2

5
2
M1 a   −  
2

5
2
M2 x a   =  
2 2

5
2
a (M1 − M2 x )
2
.
5
1−x
Hence I =
2

5
Ma
2
×
1−x
3
.

If x = 0,  I   =   2

5
Ma
2
, as expected. If x → 1 , you may have to use de l’Hôpital’s rule to show that I →
2

5
Ma
2
as expected.

 Exercise 22.8.14

M1 = mass of mantle.
M2 = mass of core.
M = mass of entire planet.
3
M1 s(1−x )
We have M = M1 + M2 and M2
=
x
3
and therefore
3
3 s(1−x )
M2   =  M × 3
x

x  + s(1−x )
3
and M 1  =  M × 3
x  + s(1−x )
3

Also
5

I   =  Icore   +  Imantle  =  
2

5
2
M2 x a   +  
2 2

5
M1 a
2
×
1−x

1−x3
,

where I have made use of the result from the previous problem. On substitution of the expressions for M and M , we quickly 1 2

obtain
5
2 s  +  (1 − s)x
2
I  =   Ma × (1)
3
5 s  +  (1 − s)x

A hollow planet would correspond to = 0 . Divide top and bottom by s and it is immediately seen that the expression for a
1

hollow planet would be identical to the expression obtained for the previous problem.

Note that both x  =  0 and x  =  1 correspond to a uniform sphere, so that in either case, I   =   2

5
Ma
2
for all other cases, the
moment of inertia is less than M a . 2

5
2

The core size for minimum moment of inertia is easily found by differentiation of the above expression for I , and the required
expression follows after some algebra. For s  =  0.6 , the equation becomes 9 − 15x − 4x   =  0 , of which the only positive 2 5

real root is x  =  0.73682, which corresponds to a moment of inertia of 0.90376 % M a . Note that. for s  =  0.6 , the 2

5
2

moment of inertia, expressed in units of M a varies very little as the core size goes from 0 to 1, so that measurement of the
2

5
2

moment of inertia places very little restriction on the possible core size.
The inverse of Equation (1) is
5 3
(1 − s)x   −  I (1 − s)x   +  (1 − I )s  =  0, (2)

22.8.18 https://phys.libretexts.org/@go/page/9092
where Iis expressed in units of M a . For I   =  0.911 , there are two positive real roots (look at the graph); they are
2

5
2

x  =  0.64753 and 0.81523. For I   =  0.929 , the roots are 0.55589 and 0.87863. Thus the core size could be anything between

0.55589 and 0.64753 or between 0.81523 and 0.87863 a rather large range of uncertainty. Even if I were known exactly
(which does not happen in science), there would be two solutions for x.

 Exercise 22.8.15

This is just a matter of geometry. If, when you make a small angular displacement, you raise the centre of mass of the brick the
equilibrium is stable. For, while the brick is in its vertical position, it is evidently at a potential minimum, and you have to do
work to raise the centre of mass. If, on the other hand, your action in making a small angular displacement results in a lowering
of the centre of mass, the equilibrium is unstable.centre of mass of the brick the equilibrium is stable. For, while the brick is in
its vertical position, it is evidently at a potential minimum, and you have to do work to raise the centre of mass. If, on the other
hand, your action in making a small angular displacement results in a lowering of the centre of mass, the equilibrium is
unstable.
When the brick is in its vertical position, the height h of its centre of mass above the base of the semicylinder is justcentre of
0

mass above the base of the semicylinder is just

h0   =  R  +  l .
When it is displaced from the vertical by an angle θ , the point of contact between brick and semicylinder is displaced by a
distance Rθ , and, by inspection of the drawing, the new height h is
h  =  R cos θ  +  Rθ sin θ  + l cos θ .
h − h0   =  Rθ sin θ  −  (R + l)(1 − cos θ) .
If you Maclaurin expand this as far as θ , you arrive at
2

h − h0 ≈
1

2
(R − l)θ
2
.
This is positive, and therefore the equilibrium is stable, if l  <  R , or 2l  <  2R , i.e. if the length of the brick is less than the
diameter of the semicylinder.

22.8.19 https://phys.libretexts.org/@go/page/9092
 Exercise 22.8.16

As in the previous question, it is just a matter of geometry. If rolling the Thing results in raising its centre of mass, the
equilibrium is stable. Initially, the height of the centre of mass is H   =  b  + l .centre of mass, the equilibrium is stable.
0

Initially, the height of the centre of mass is H   =  b  + l .

0

After rolling, the dashed line, which joins the centres and is of length a + b , makes an angle θ with the vertical. The short line
joining the centre of mass of the Thing to the centre of curvature of its bottom is of length l − a and it makes an angle θ + with
phi

the vertical. The height of the centre of mass is therefore nowcentre of mass of the Thing to the centre of curvature of its
bottom is of length l − a and it makes an angle θ + with the vertical. The height of the centre of mass is therefore now
phi

h  =  (a + b) cos θ + (l − a) cos(θ + ϕ)

The centre of mass has therefore rise through a heightcentre of mass has therefore rise through a height
h  −  h0   = (a + b) cos θ  +  (l − a) cos(θ + ϕ)  − b − l .
Also, the two angles are related by aϕ  =  bθ , so that
b
h  −  h0   = (a + b) cos θ  +  (l − a) cos[{1 + ( )} θ]  − b − l
a

h  −  h0   = −
1

2
2
θ [a + b  +  (l − a)(1 +
b

a
2
) ] .
For stability this must be positive, and hence 1

l
  > 
1

a
 + 
1

b
.
If a  =  b , this becomes l  <   1

2
a .

For a hollow semicylinder, l  =  (1 −

2
π
)a = 0.363a □ Stable

For a hollow hemisphere, l = 0.5a □ Borderline stable

4
For a solid semicylinder, l  =  [1 −
(3π )
]a = 0.576a □ Unstable

For a solid hemisphere, l =

5

8
a = 0.625a □ Unstable

 Exercise 22.8.17

We need to find the height h of the centre of mass above the level of the pegs as a function of θ . See drawing on next
page.centre of mass above the level of the pegs as a function of θ . See drawing on next page.
Angels:
∘
BAC  =  45 −θ

∘
ABX  =  45 +θ

Distances:
AB  =  2ka

∘
AC  =  2ka  cos(45 − θ)

22.8.20 https://phys.libretexts.org/@go/page/9092
 ∘ ∘
EF  =  2ka  cos(45 − θ) cos(45 + θ) = ak cos 2θ

–
DC  =  a√2

–
DF  =  a√2 cos θ

–
h  =  DF  −  EF  =  a(√2 cos θ − k cos 2θ)

–
h0 = height of centre of mass above pegs when θ = 0 centre of mass above pegs when θ  =  0∘
  =  a(√2 − k)

h √2 cos θ−k cos 2θ

y  =     =
h0 √2−k

22.8.21 https://phys.libretexts.org/@go/page/9092
 Exercise 22.8.18

22.8.22 https://phys.libretexts.org/@go/page/9092
There are three forces acting on the hemisphere: Its weight mg. The reaction N of the wall, which is perpendicular to the wall
since the wall is smooth. The reaction R of the floor, which acts at an angle λ to the floor, where μ  =   tan λ . Three forces in
equilibrium must act through a point; therefore all three forces act through the point P . It is thus clear that
aμ 8μ
sin  θ  =  
OP

OC
  = 
3
  = 
3
.
a
8

If μ  =   ,  θ  =  41 48 . If μ  =  
1

4
∘ ′ 3

8
∘
,  θ  =  90 . If μ  >   3

8
the hemisphere can rest in any position, the equilibrium not being
limiting static equilibrium.

 Exercise 22.8.19

This solution uses the same method that Professor Marsh (Warwick University) showed me for Problem 20. I believe it to be
clearer than an earlier solution that I had posted.

At an instant when the rod is tilted at angle θ , the coordinates of C with respect to the fixed point O are:
¯¯
¯
x  =  a(sin θ  −  θ cos θ), (1)

¯
¯¯
y   =  a(cos θ  +  θ sin θ) (2)

and so its velocity components are

˙
¯¯
x̄  =  aθ sin θθ̇ (3)

and
˙
¯¯ ˙
ȳ   =  aθ cos θθ . (4)

The moment of inertia of the rod about the centre of mass is 1

3
ml
2
.
The kinetic energy T is the sum of the translational kinetic energy and the rotational kinetic energy about the centre of
mass:centre of mass:
1 2 2
1 2
2
˙
T  =  ( a θ  +  l ) mθ , (5)
2 6

22.8.23 https://phys.libretexts.org/@go/page/9092
and the potential energy V is
V   =  mga(cos  θ  +  θ sin  θ). (6)

One can now get the equation of motion either by Lagrangian means or by equating the derivative with respect to θ of the total
energy to zero, since there are no nonconservative forces and hence the total energy is independent of θ . In carrying out the
2 ˙
differentiation, note that d

dθ
˙ ˙
θ   =  2 θ
dθ

dθ
¨
  =  2 θ . We obtain, for the equation of motion:

2
2
2 2
2 2
˙ ¨
a θθ   +  (a θ + l )θ   +  gaθ cos θ  =  0 (7)
3

For small θ (neglecting second and higher powers of θ ), cos θ → 1 and a 2

θ
2
is negligible compared with l , so the equation of
2

3ga
motion becomes, approximately, θ¨ = − 2
, and so the period is P   =   2πl
.
l √3ga
––––––––––
–––––––––––
–

 Exercise 22.8.20

I am much indebted to Professor T. R. Marsh of Warwick University not only for finding a mistake in an earlier posted solution
to this problem, but for providing the following solution.

2
7 3
I   =  m a (  −  cos θ) . (2)
5 4

3aθ
x  =  
8

We are going to refer the motion to a fixed point Q, which is the point of contact between hemisphere and table when the
hemisphere is in its equilibrium position.
At an instant when the hemisphere is tilted at an angle θ , the distance between A and Q is aθ , and the coordinates of C relative
to Q are
3
¯¯
¯
x  =  aθ  −   a sin θ, (1)
8

3
¯
¯¯
y   =  a −   a cos θ. (2)
8

Therefore the velocity components of C are

22.8.24 https://phys.libretexts.org/@go/page/9092
 3
˙
¯¯ ˙
x̄  =  a (1 − cos θ) θ , (3)
8

3
˙
¯
¯¯ ˙
y  =   a sin θθ . (4)
8

By the parallel axes theorem, the moment of inertia around the centre of mass iscentre of mass is
2
2 2
3 83 2
I  =   m a   −  m ( a)   =   ma (5)
5 8 320

The kinetic energy T is the sum of the translational kinetic energy and the rotational kinetic energy about the centre of
mass:centre of mass:
2 2
1 2
3 3 83 2
2
7 3 2
˙ ˙
T  =   ma [ (1 − cos θ)   +  ( sin θ)   +   ] θ   =  m a ( − cos θ) θ (6)
2 8 8 230 10 8

The potential energy V is

3
V   =  mga(1 − cos θ). (7)
8

We can the get the equation of motion either by using the Lagrangian equations, or by calculating the derivative with respect to
θ of the total energy T   +  V . The derivative is zero, because there are no nonconservative forces and total energy is constant.
2 ˙
Note that (as in Problem 19) the derivative of θ˙ with respect to θ is 2θ˙ dθ

dθ
, which is 2θ¨. Either method results in the equation
of motion:
7 3 3 2 3
¨ ˙
(  −  cos θ) θ   + sin θθ   + g sin θ  = 0. (8)
5 4 8 8

In the small angle limit, cos θ → 1 , and sin θ → θ is negligible compared with g, so the equation of motion becomes
15g
¨
θ  =  − θ (9)
26a

−−−−
26a
P   =  2π √ . (10)
15g

––––––––––––––
–––––––––––––––
–

 Exercise 22.8.21

The (second) moment of inertia with respect to the centre (see centre (see Section 2.19 of Chapter 2) is
a
Icentre = 4π ρ0 ∫
0
4
(r
5
− r /a)dr =
2

15
π ρ0 a .
5
.
The moment of inertia with respect to an axis through the centre is 2/3 of this:centre is 2/3 of this:
Iaxis   =  
4

45
π ρ0 a
5
.
∴ Iaxis   =  
15
4
Ma
2
.
––––––––––––––
–––––––––––––––
–

 Exercise 22.8.22

22.8.25 https://phys.libretexts.org/@go/page/9092
Left-hand particle:T   =  mg[μ cos(α − θ)  +   sin(α − θ)] .
Right-hand particle:T   =  mg[sin(α + θ)  −  μ cos(α + θ)] .
∴ μ[cos(α − θ)  +   cos(α + θ)]  =   sin(α + θ)  −   sin(α − θ) ,
and, by the “sum and difference” trigonometrical formulae, we obtain
2μ cos α cos θ  =  2 cos α sin θ,

from which
tan  θ  =  μ.
–––––––––––
–––––––––––––

 Exercise 22.8.23

Consider a portion of the rope between θ and δθ. There are four forces on this portion. The tension T at θ . The tension
T   +  δT at θ  +  δθ (δT is negative). The normal reaction δN of the cylinder on the rope. The frictional force μδN of the

cylinder on the rope. Note that the rope is about to slip downwards, so the friction force is upwards as shown.
We have
1
δN   =  (2T   +  δT ) sin( θ)
2

and
(T   +  δT ) cos(
1

2
δθ)  +  μδN   =  T cos(
1

2
δθ) .
To first order, these become
δN   =  T δθ

δT   =   − μδN

and
δT   =   − μT δθ

and hence by integration

22.8.26 https://phys.libretexts.org/@go/page/9092
 −μα
F   =  M ge .
–––––––––––––
––––––––––––––
–

 Exercise 22.8.24

Area of square = 4a 2

Area of rectangle = 4a 2
(1 − x)

Area of triangle = 2a 2
(x + y − 1)

Area of trapezoid = 2a 2
(1 − x + y)

The weight of the cube is 8a ρsg, and it acts downward through C, the centre of mass. The hydrostatic upthrust is
3

4 a (1 − x + y)ρg and it acts upward through the centre of buoyancy H. Here ρ is the density of the fluid, and ρs is the density
3

of the wood. We evidently must find the centre of mass. The hydrostatic upthrust is 4a (1 − x + y)ρg and it acts upward
3

through the centre of buoyancy H. Here ρ is the density of the fluid, and ρs is the density of the wood. We evidently must find
the X − coordinate of C and of H. Let’s first of all find the X− and Y − coordinates (see the next figure).
′

The X− and Y −coordinates of C are trivial and quite easy respectively:

XC   =  a YC   =  a(1 − 2x)

You are going to have to work quite hard at it to find the X− and Y − coordinates of H, the centre of buoyancy, which is the
centroid of the trapezoid. “After some algebra” you should findcentre of buoyancy, which is the centroid of the trapezoid.
“After some algebra” you should find
2 2
2(1−x+2y)a 2(2−4x+2y+2 x −2xy−y )a
XH   =   YH   =  
3(1−x+y) 3(1−x+y)

22.8.27 https://phys.libretexts.org/@go/page/9092
To find the ′
X − coordinates of C and of H, we use the usual formulas for rotation of axes, being sure to get it the right way
round:
′
X cos θ − sin θ X
(
′
)   =  ( )( ) ,
Y sin θ cos θ Y

together with
tan  θ  =  x  +  y − 1 .
Take moments about the axle (origin):
3
8 a ρsg X
C
′ 3
  =  4 a ρg(1 − x + y) .
After a little more algebra, you should eventually arrive at
2 2 3 3 2
3−7x+2y+6 x −3 y −2 x +y +3xy
s = 2
3(2−3x−y+2 x +2xy)
–––––––––––––––––––––––––––––
––––––––––––––––––––––––––––––
–

 Exercise 22.8.25

Let the radii of the cylinder and sphere be a and b respectively, and the mass of the sphere be M . The angles θ and ϕ are
related by aθ  =  bϕ . I have drawn the three forces on the sphere, namely its weight, the normal reaction of the cylinder on the

22.8.28 https://phys.libretexts.org/@go/page/9092
sphere, and the frictional force on the sphere. The transverse acceleration of the centre of the sphere is (a + b)θ̈ and the
2

centripetal acceleration is ˙
(a + b)θ . The equations of motion are:centre of the sphere is ¨
(a + b)θ and the centripetal
2
acceleration is (a + b)θ˙ . The equations of motion are:
¨
M g sin θ  −  F   =  M (a + b)θ (1)

and
2
˙
M g cos θ  −  N   =  M (a + b)θ (2)

The angular acceleration of the sphere about its centre is θ¨  +  ϕ̈   =   (1 + ) θ¨ , and its rotational inertia is
a

b
2Mb

5
. The torque
that is causing this angular acceleration is F b, and therefore the rotational equation of motion is
2 a
2 ¨
Fb =   Mb (1 + )θ (3)
5 b

Elimination of F between Equations (1) and (3) yields

\ddot{\theta} = \frac{5\text{g}}{7(a+b)}\sin\theta. \tag{4}\label{20.4}

˙
Write θ¨ as θ˙ dθ

dθ
in the usual way and integrate with initial conditions θ = θ˙ = 0 or from energy considerations:

\dot{\theta}^{2} =\frac{10\text{g}}{7(a+b)}(1-\cos\theta) \tag{5}\label{20.5}

2

Substitute for θ¨ and θ˙ into Equation (2) to obtain

N =\ M\text{g}(17\cos\theta-10). \tag{6}\label{20.6}

This is zero, and the sphere leaves the cylinder, when cos θ  =   10

17
, θ  =  53 ∘ ′
58 .

 Exercise 22.8.26

Surface density = σ g cm-2

Original sandwich:
Mass = 54σ g
x -coordinate of centre of mass = 3 cm centre of mass = 3 cm
y -coordinate of centre of mass = 4 cm centre of mass = 4 cm
Bite:

22.8.29 https://phys.libretexts.org/@go/page/9092
Mass = 1

2
2
π3 σ =14.137 166 94σ g
Distance of centre of mass from hypotenuse = centre of mass from hypotenuse= 4

3π
× 3  =  
4

π
= 1.273 239 545 cm
x-coordinate of centre of mass = 4.5 -centre of mass = 4.5  −   4

π
sin θ = 4.5 −
16

5π
= 3.481 408 364 cm
y-coordinate of centre of mass = 6- centre of mass = 6  −   4

π
cos θ  =  6 −
12

5π
= 5.236 056 273 cm
Remainder:
Mass = (54 - 14.137 166 94)σ = 39.862 833 06σ g
x -coordinate of centre of mass = xcentre of mass = x
¯¯
¯ ¯¯
¯

y -coordinate of centre of mass = ȳ centre of mass = ȳ

¯
¯ ¯
¯

Moments:
39.862 833 06x + 14.137 166 94 × 3.481 408 364 = 54 × 3. –
¯¯
¯
x = 2.829 270 780 cm
––––––––––––––––––––
¯¯
¯

–––––––––––––––––––––

39.862 833 06ȳ + 14.137 166 94 × 5.236 056 273 = 53 × 4. ȳ  =  3.561 638 436 cm
¯
¯ ¯
¯

–––––––––––––––––––––
––––––––––––––––––––––
–

This point is very close to the edge of the bite. The centre of the bite is at (4.5, 6), and its radius is 3. Its equation is
thereforecentre of the bite is at (4.5, 6), and its radius is 3. Its equation is therefore
2 2
(x − 4.5 )   +  (y − 6 )   = 9, or x 2 2
  +  y   − 9x  − 12y  +  47.25 = 0 .
The line x = 2.829 270 780 cuts the circle where y − 12y  +  29.791 336 13  =  0 . The lower of the two points of
2

intersection is at y = 3.508 280 941 cm. The centre of mass is slightly higher than this and is therefore just inside the
bite.centre of mass is slightly higher than this and is therefore just inside the bite.

 Exercise 22.8.27

Consider a portion of the band within the angle δθ. Its mass is mδθ

2π
When the band is spinning at angular speed ω and its radius
2

is r, the centrifugal force on that portion is δF   =   . (I leave it to the philosophers and the schoolteachers to debate as to
mrω δθ

2π

whether there “really” is “such thing” as centrifugal force – I want to get this problem done, and I’m referring to a co-rotating
2

frame.) The y -component of this force is . Also, the tension in the band when its radius is r is T = 2πk(r − a) .
mrω δθ

2π

π
2 + 2

Consider the equilibrium of half of the band. The y -component of the centrifugal force on it is mrω

2π
∫
−
π
2
cos θdθ =
mrω

π
.
2
2
4π k(r−a)
The opposing force is 2T   =  4πk(r − a) . Equating these gives ω 2
  = 
mr
.
––––––––––––––
–––––––––––––––
–

22.8.30 https://phys.libretexts.org/@go/page/9092
 Exercise 22.8.28
Let the distance AB be l and the distance AC be c . Let the mass of the rod be m.

Consider an elemental portion δx of the rod at P at a distance x from A. Its weight is mδx

l
. When the rod is about to move, it
μmgδx
will experience a frictional force δf   =   l
, which will be in the direction shown if P is to the left of C, and in the opposite
direction if P is to the right of C. When the rod is just about to move (but has not yet done so) it is still in equilibrium. Consider
the moment about A of the frictional forces on the rod. The clockwise moment of the frictional forces on AC must equal the
counterclockwise moment of the frictional forces on CB. Thus
μmg c μmg l

l
∫
0
x dx =
l
∫
c
x dx .
∴ c =
l
.
√2
––––––
–––––––
–

The net force on the rod is

μmg c μmg l
F − 
l
∫
0
dx  +  
l
∫
c
dx ,
and this is zero, and therefore
μmg(2c−l) –
F  =   = (√2 − 1)μmg.
l
––––––––––––––––––––––––––––––
–––––––––––––––––––––––––––––––
–

 Exercise 22.8.29
The cone slips when tan θ  >  μ .
It tips when C (the centre of mass) is to the left of M. centre of mass) is to the left of M.
The distance OC is h

4
. (See Chapter 1, Section 1.7). Therefore it tips when tan θ > 4 . a

Thus it slips if μ < 4 a

h
and it tips if μ < 4 .
a

22.8.31 https://phys.libretexts.org/@go/page/9092
 Exercise 22.8.30

When the block is just about to tip, the reaction of the table on the block acts at A and it is directed towards the point K,
because, when three coplanar forces are in equilibrium they must act through a single point. The angle λ is given by
tan λ  =   . However, by the usual laws of friction, the block will slip as soon as tan λ  =  μ . Thus the block will slip if
a

μ <
a

x
, and it will tip if μ > . Expressed otherwise, it will slip if x < and it will tip if x > . The greatest possible value
a

x
a

μ
a

of x is 2a; therefore the block will inevitably slip if μ < 1

 Exercise 22.8.31

+
When or if the cylinder is just about to tip, it is about to lose contact with the left hand peg. The only forces on the cylinder are
the torque, the weight, and the reaction R of the right hand peg on the cylinder, which must be vertical and equal to mg. But
the greatest possible angle that the reaction R can make with the surface of the cylinder is the angle of friction λ given by
tan  λ  =  μ . From geometry, we see that sin  θ  =  k , or tan  θ  =   . Thus the cylinder will slip before it tips if
k
2
√1−k

μ  <  
k

2
and it will tip before it slips if μ  >   k

2
.
√1−k √1−k

If the cylinder tips (which it will do if μ  >  

k

2
), the clockwise torque t at that moment will equal the counterclockwise
√1−k

torque of the couple (R and mg), which is mgka. Thus the torque when the cylinder tips is
TIP:

τ   =  mgak. (1)

22.8.32 https://phys.libretexts.org/@go/page/9092
When or if the cylinder is just about to slip, the forces are as shown above, in which I have resolved the reactions of the pegs
on the cylinder into a normal reaction (towards the axis of the cylinder) and a frictional force, which, when slipping is about to
occur, is equal to m times the normal reaction. The equilibrium conditions are
μ(N1   +  N2 ) cos θ  +  (N1 − N2 ) sin θ  =  0 ,
μ(N1   −  N2 ) sin θ  −  (N1 + N2 ) cos θ  +  mg =  0

and
μ(N1   +  N2 )a  =  τ .
−−−− −
We can find N   +  N by eliminating
1 2 N1   −  N2 from the first two equations, and then, writing √1 − k2 for cos θ, we find
that, when slipping is about to occur,
SLIP:
μ 1
τ   =  mga  ×    ×  − −−− −. (2)
2
1 +μ √ 1 − k2

I have drawn below the functions

τ τ μ 1
  =  k (tip) and    =  2
×  (slip)
mga mga 1 + μ √1−k
2

for k  =  0.1,  0.3,  0.5,   1

√2
and 0.9. The horizontal lines are the tip functions, and the curves are the slip functions. As long as
μ  <  
k

2
the cylinder will slip. As soon as μ  <   k

2
the cylinder will tip.
√1−k √1−k

22.8.33 https://phys.libretexts.org/@go/page/9092
 Exercise 22.8.32

We’ll leave to the philosophers the question as to whether centrifugal force “really exists”, and we’ll work in a co-rotating
reference frame, so that the car, when referred to that frame, is in static equilibrium under the six forces shown. Clearly, N 1
2

and N 2   =  mg and F 1   +  F2   =  

mv

R
.
−−−−
The car slips when F 1  +  F2   =  μ(N1 + N2 ) ; that is, when v  =  √μgR .
−−−
2
dgR
The car tips when mv h

R
  =  mgd; that is, when v  =  √ h
.

That is, it will slip or tip according as to whether μ  < d

h
or > d

h
.
For example suppose d  =  60 cm,  h  =  60 cm,  g  =  9.8 m s −2
,  R  =  30 m,  μ  =  0.8 .
In that case, d

h
  =  0.75 , so it will tip at v  =  14.8 m s −1
  =  53.5 km hr
−1
.
But if it rains, reducing μ to 0.7, it will slip at v  =  14.3 m s −1
  =  51.6 km hr
−1
.

22.8.34 https://phys.libretexts.org/@go/page/9092
 Exercise 22.8.33

I have drawn in green the radial and transverse components of the acceleration of the centre of mass aθ˙ and aθ¨ respectively. I
have drawn in red the weight of the rod and the normal and frictional components of the force of the table on the rod at A , N
2
and F respectively.centre of mass aθ˙ and aθ¨ respectively. I have drawn in red the weight of the rod and the normal and
frictional components of the force of the table on the rod at A , N and F respectively.
The following are the equations of motion:
Normal:
¨
maθ   =  mg cos θ  −  N . (1)

Lengthwise:
2
˙
maθ   =   − mg sin θ  +  F . (2)

Rotation:
2 ¨
k θ   =  ga cos θ (3)

Here k is the radius of gyration about A, given by

2
1 2 2
k   =  l   +  a . (4)
3

From Equations (1), (3) and (4), we obtain

2
l
N   =  mg cos θ. ( ). (5)
2 2
l   +  3 a

The space integral (see Chapter 6, Section 6.2) of Equation (3), with initial condition θ˙  =  0 when θ  =  0 , results in
2 2ga
˙
θ   =  sin θ. (6)
2
k

This can also be obtained by equation the loss of potential energy, mga sin θ to the gain in kinetic energy, 1

2
2 ˙
mk θ .
Combining this with Equations (2) and (4) leads to
2 2
l   +  9 a
F   =  mg sin θ. ( ). (7)
l   +  3 a2
2

At the instant of slipping, F   =  μN , and hence, from Equations (5) and (7) we find
μ
tan θ  =   a 2
.
1 + 9( )
l

22.8.35 https://phys.libretexts.org/@go/page/9092
 Exercise 22.8.34
g
I derive v   =  gx  +   x by two different methods – one from energy considerations, the other from angular momentum
2

l
2

considerations. First, energy.

If the table top is taken to be the zero level for potential energy, the initial potential energy was − 1

2
m. g.
1

4
l  =   −
1

8
mgl .
When the length of the dangling portion is 1

2
l  +  x the potential energy is
1
2
l + x mg 2 mgx
−(
2

l
) m. g.
1

2
(
1

2
l  +  x)   =   −
2l
(
1

2
l  +  x)   =  −
1

8
mgl  −  
1

2
mgl  −  
2l
.
2
mgx
The loss of potential energy is therefore 1

2
mgx  +  
2l
.
This is equal to the gain in kinetic energy 1

2
mv
2
, and therefore
g
2
v   =  gx  +  
l
2
x .
Another method:

22.8.36 https://phys.libretexts.org/@go/page/9092
Consider a point A. Anywhere will do, but I have chosen it to be a distance l below the level of the table and l to the left of the
table edge. The moment of momentum (= angular momentum) of the chain about this point is mlv  =  mlẋ and its rate of
change is therefore mlv  =  mlẋ . The torque about A is \left(\mlv\ =\ ml\dot{x} and its rate of change is therefore
1
l + x

mlv  = mlẋ . The torque about A is (

2

l
) mgl  =   (
1

2
l  +  x) mg . These are equal, and so lẍ  =  g (
1

2
l  +  x) . Write
g
ẍ  =  v
dv

dx
in the usual way, and integrate (with v = 0 when x = 0 ) and the result v 2
  =  gx  +  
l
2
x follows.
To find the relation between x and t we can use the energy Equation 9.2.9 for conservative systems
−− x
t  =  √
m

2
∫
x0 √E−V (x)
dx
.
mg mgx
Here x   =  0 and we have already seen that
0 E  −  V (x)  =  
2l
2
x  + 
2
. Upon integrating this expression, we obtain,
after a little algebra and calculus,
(1)

The converse of this is the required expression

(2)

Differentiation of this with respect to time produces the third required expression:

(3)

g
You may verify from these last two equations, if you wish, that v 2
  =  gx  +  
l
2
x .
The chain falls completely off the table when x  =  
1

2
l . That is (by using Equation (1)), at time
−
− −−
–
√
l

g
ln(2  +  √3)  =  1.317 √
l

g
.
−
− −
−
If we express distances in units of l, time in units of √ l

g
and therefore necessarily speeds in units of √gl, Equations (2) and (
3 ) become
t 2
(e − 1 ) 1 1
t −t
x  =     =  (e   +  e − 2)  =   (cosh t − 1) (4)
t
4e 4 2

2t
e −1 1
v  =     =  sinh  t (5)
t
4e 2

and we can get the acceleration by a further differentiation:

1 1
t −t
a  =   (e   +  e )  =   cosh  t. (6)
4 2

We are pleased to note that, by the time that x  =  

1

2
l [i.e. when the chain completely leaves the table at time
−− −−−
– −
−
t  =   ln(2  +  √3)√
l

g
], the acceleration is g. The speed is then √ 3

4
lg  =  0.866 √lg .

 Exercise 22.8.35a

The maximum overhang of book 1 is d 1  =  w .

The centre mass of 1  +  2 centre mass of 1  +  2 is at 3w

2
from the left hand side (LHS) of 2 , so d3  =   w

3
.
The distance of the centre of mass of 1 + 2 + 3 centre of mass of 1 + 2 + 3 is at 5w

2
from the LHS of 3 , so d3  =   w

3
.

22.8.37 https://phys.libretexts.org/@go/page/9092
Thus D  =  d 1  +  d2   +  d3   =   (1  +  
1

2
 + 
1

3
) w  = 1.8 3̇w .

 Exercise 22.8.35b

In a similar manner we find that, given n  +  1 books, the maximum overhang is

D  =   (1 +
1

2
+
1

3
. . .  . . . +
1

n
)w .
I do not know if there is a simple expression for the sum to n terms of this harmonic series. Please let me know if you know of
one or can find one. Therefore I used a computer to solve
1 1 1
1+ + . . .  . . . +   =  10
2 3 n

by brute force. I got n = 12367, so you would need 12368 books.

 Exercise 22.8.35c

The harmonic series is divergent and has no finite limit, so there is no finite limit to the possible overhang.
You might wish to speculate on any practical limitations on constructing such a pile of books. For example, we have been
assuming a uniform gravitational field – but this will no longer be valid once the overhang becomes comparable to the radius
of Earth. This will, however, need quite a large number of books.

 Exercise 22.8.36
dy
In the solution that follows, a prime will be used to denote differentiation with respect to x, and p  =  y   =   . I shall also ′

dx

make use of an auxiliary variable ϕ  =   sinh p . The initial conditions are y = 0,  x = a,  p = 0,  ϕ = 0 . The final
−1

conditions are x = 0,  p = −∞, ϕ = −∞, y to be determined.

At time t , the y−coordinate of the Man is v t . If (x,  y) are the coordinates of the Dog at that time, the slope of the path taken
by the Dog is
vt − y
p  =   − , (1)
x

so that
v t  =  y − px. (2)

22.8.38 https://phys.libretexts.org/@go/page/9092
The speed of the Dog is
−−−−−
ds 2
dx
Av =   −   =   − √1 + p . (3)
dt dt

−−−−−−−−
2
dy
[This comes from ds  =  √1 + (
dx
) dx . The minus sign is necessary because (
dx

dt
) is negative, and Av , the speed (not
velocity!) of the Dog is necessarily positive.]
Now ( dx

dt
)  =  −
1
′
t
so Equation (??? ) can be written
−−−−−
′ 2
Avt   =   − √ 1 + p (4)

If we can eliminate t between Equations (2) and (4), we will obtain a relation between the slope p and x, and hence potentially
a relation between y and x.
Differentiate Equation (??? with respect to x (recalling that y ′
=p ):
′ ′
vt   =   − p x (5)

It is now simple to eliminate t from Equations (4) and (5):

′

−−−−−
′ 2
Ap x  =  √ 1 + p (6)

On separating the variables and integrating, we obtain

dp dx
A∫ −−− −−  =  ∫ . (7)
2 x
√1 + p

With initial conditions p = 0 when x = a , this gives us

x
−1
A sinh p  =   ln( ), (8)
a

or
Aϕ
x  =  ae , (9)

where
−1
ϕ  =   sinh p. (10)

Equation (9), with (10), gives us the relation between x and the slope, p. Note that p and hence f are negative, so that equation
says that x < a .
Our next task will be to find a relation between y and p (or between y and ϕ ).
From Equation (10) we have
dy  =   sinh ϕ dx, (11)

and from Equation (9) we have

Aϕ
dx  =  aAe dϕ. (12)

From these we obtain the differential relation between y and ϕ :

Aϕ
dy  =  aAe sinh ϕdϕ, (13)

or
1
(A+1)ϕ (A−1)ϕ
dy  =   aA(e   −  e )dϕ. (14)
2

Integrate this, with initial condition ϕ  =  0 when y  =  0 , to obtain

22.8.39 https://phys.libretexts.org/@go/page/9092
 x 1+1/A
(A−1)ϕ
1 ⎛( ) e 2 ⎞
a
y  =   aA − + . (15)
2
2 ⎝ A+1 A−1 A −1 ⎠

Equation (9) and (15) are parametric equations to the path of the Dog, though it is easy to eliminate ϕ and write y explicitly as
a function of x:
(A+1)ϕ (A−1)ϕ
1 e e 2
y  =   aA ( − + ). (16)
2
2 A+1 A−1 A −1

The figure was drawn for a = 1,  A = 2, for which Equation (16) reduces to
1 1

y  =   [x 2 (x − 3)  +  2]. (17)

3

The distance walked by the Man is found by putting ϕ = −∞ in Equation 15. Thus
aA
y  =   , (18)
A2 − 1

and the time taken is

aA
t  =   (19)
2
v(A − 1)

 Exercise 22.8.37

Let l be the length of the string.

 Exercise 22.8.37a

Kinetic energy of the upper mass = 1

2
2
(m r )ω   +
2 1

2
m ṙ
2
.
Kinetic energy of the lower mass = 1

2
m ṙ
2

Potential energy of the lower mass = mg(l − r) .

Total energy of the system =
1 2
2 2
(m r )ω   + m ṙ   −  mg(l − r). (1)
2

Initial total energy of the system =

1 2
2 2
(m a )ω   + m ṙ   −  mg(l − a). (2)
0
2

Energy is conserved and therefore, by equating (1) and (2), we obtain

2
1 2
1 2 2
2
ṙ   =  g(a − r)  +   a ω  −  r ω . (3)
0
2 2

Angular momentum is also conserved, and therefore

2 2
r ω  =  a ω0 (4)

On elimination of r between Equations (3) and (4) we obtain, after some algebra,
2 2 −−
−
ṙ aω ω ω0
0
  =  1  +   (1 − )  −√ . (5)
ga 2g ω0 ω

22.8.40 https://phys.libretexts.org/@go/page/9092
 Exercise 22.8.37b

If aω 2
0
  =  g it is trivial to show that
2
ṙ 3 1 1
  =   − Ω  − −.
− (6)
ga 2 2 √Ω

 Exercise 22.8.37c

Algebra and calculus show that 3

2
 −
1

2
Ω  −
√Ω
1
is negative for all positive Ω except for Ω  =  1 , when it reaches a maximum
value of zero.

 Exercise 22.8.37d

If aω 2
0
  =  2g and Ω  =   ω0
ω
it is trivial to show that
2
ṙ 1
  =  2  −  Ω  − −
− (7)
ga √Ω

Algebra and calculus show that 2  −  Ω  − √Ω

1
reaches a maximum value for Ω  =   ω0
ω
  = 
1
2
  =  0.629 961 at which time
2 3

2
ṙ

(ga)
  =  0.110 118 . That is, when −−
ṙ   =  0.331 841 √ga Equation (4) (conservation of angular momentum) shows that
−−
r  =  
a 3
  =  √2a  =  1.259 921a .
√Ω

Solution of 2  −  Ω  −   √Ω
1
=0 shows that the speed is zero when Ω  =  1 (the initial condition) and when (the equilibrium
value). Equation (4) (conservation of angular momentum) shows that r  =   a
  =  1.618 034a .
√Ω

 Exercise 22.8.37e

If aω 2
0
  = 
1

2
g and Ω  = ω0
ω
it is trivial to show that
2
ṙ 5 1 1
  =   − Ω  − −
− (8)
ga 4 4 √Ω

Algebra and calculus show that 5

4
 −
1

4
Ω  −
√Ω
1
reaches a maximum value for Ω  =   ω

ω0
  =  2 3
  =  1.587 401 at which time
2
ṙ

(ga)
= 0.059449 . That is, when −−
ṙ   =   − 0.243 822 √ga Equation (4) (conservation of angular momentum) shows that
r  =  
a
  = 
a
3
  =  0.793 701a .
√Ω √2

Solution of 5

4
 −
1

4
Ω  −
1
shows that the speed is zero when Ω  =  1 (the initial condition) and when
√Ω

Ω  =  
ω

ω0
  =  2.438 447 (the equilibrium value). Equation (4) (conservation of angular momentum) shows that
r  =  
a
  =  0.640 388a .
√Ω

How much further can we go with this question? By elimination of r between Equations (3) and (4) we obtained a relation
between ṙ and ω. By elimination of ω between Equations (3) and (4) we can get a relation between ṙ and r. It will be of the
form
−−−−−−−−−−−−
B
ṙ   =  √ A  −  gr  −   , (9)
2
r

where A  =  ga  +   a ω and B  =  a ω . If you can integrate this, you then get a relation between r and t . I haven’t given
1

2
2 2
0
4 4
0

much though as to whether you can get integrate Equation (9) analytically (if anyone manages it, please let me know), but at
least a numerical integration will certainly be possible.
In another variation of the question, you can start with an equilibrium situation in which aω   =  g and then add an extra mass 2
0

m (or M , if you want to make it more general) and then follow the motion from there. I leave that to you.

22.8.41 https://phys.libretexts.org/@go/page/9092
 Exercise 22.8.38

Let’s look at the rod from above when it is twisted in the horizontal plane through a small angle θ .

Each of the points where the threads are attached to the rod is displaced horizontally through a distance Dθ (Since θ is small 1

and D  << L we can neglect the slight vertical rise in the position of the rod.) Each thread is now displaced from the vertical
by an angle ϕ given by

The tension T in each thread is 1

2
mg cos ϕ which, to first order in ϕ , is just 1

2
.
mg

The horizontal component of each of these forces is 1

2
mg sin ϕ which, to first order in ϕ , is 1

2
mgϕ.

1
Dθ

Therefore the rod experiences a restoring torque equal to 1

2
mgDϕ . But ϕ  =  
2

L
and therefore the restoring torque is
2
mg D θ

4L
.
The equation of motion is therefore
2
mgD
¨
Iθ  =   − θ
4L

and consequently the period P of small oscillations is

22.8.42 https://phys.libretexts.org/@go/page/9092
 −−−− −
−−
P   =  2π √
4LI
2
  = 
4π

D
√
LI

mg
.
mgD

If the rod is uniform and of length 2l, its moment of inertia is 1

3
ml
2
and in that case the period of small oscillations is
−−
P  =  
4πl

D
√
L

3g
.

There is no need to remind the reader to check the dimensions of these equations.

 Exercise 22.8.39

When the yo-yo has fallen through a distance x, it has lost potential energy M gh , and it has gained translational kinetic energy
2
1

2
mv
2
and gained rotational kinetic energy 1

2
Iω
2
where ω  =   v

a
. Therefore M gx  =   1

2
2
Mv   +  
1

2
 I (
v

a
) from which
2
Ma g
2
v   =  2  ⋅  
M a  + I
2
  ⋅  x .

Thus, from the usual equations for constant linear acceleration, the acceleration is
2
Ma

M a2  + I
×g .

The net downward force is Mg − P where P is the tension in the string. This is equal to M times the acceleration, from
which we obtain
1
P  =   2
× M g.
M a +1

 Exercise 22.8.40a

I have drawn four forces on the yo-yo. Its weight M g. The tension P in the string. The normal reaction N of the table on the
yo-yo. And the frictional force F of the table on the yo-yo. As long as the yo-yo is in contact with the table and there is no
vertical acceleration, we must have P   +  N   =  M g .
Let us suppose that there is no slipping between the yo-yo and the table, so that the yo-yo rolls to the left. We note that there is
a net force F to the left, and a net counterclockwise torque about C equal to P a − F b . Thus the yo-yo accelerates to the left at
a rate it and experiences a counterclockwise angular acceleration (P a − F b) . If there is no slipping, these must be related
F

M
a

by M
F
=b× . Thus, if there is no slipping,
P a−F b

M ab
F  =  ( )P. (1)
2
I   +  M b

The linear acceleration to the left must be F

M
, or

22.8.43 https://phys.libretexts.org/@go/page/9092
 abP
(2)
2
I + Mb
––––––––
–––––––––
–

_________________________
Alternative derivation:
There is a net counterclockwise torque about A equal to P a. The moment of inertia with respect to A is I   +  M b . Therefore 2

there is an angular acceleration about A equal to . Therefore the linear acceleration of C to the left is
Pa
2
and the abP
2
I + Mb I + Mb
––––––
–––––––
–

frictional force F is M times this, or F   =   ( Mab

2
)P .
I + Mb

End of Alternative Derivation.

__________________________
However, if the yo-yo is just about to slip, F   =  μN   =  μ(M g − P ) . Upon substitution of this into Equation (1), we see that
the yo-yo will just slip if
2
μM g(I   +  M b )
P  =   . (3)
μ(I   +  M b2 )  +  M ab

That is, the yo-yo will roll to the left without slipping if
M abP
μ  >   (4)
2
(M g − P )(I + M b )
–––––––––––––––––––––––––
–––––––––––––––––––––––––––

Its linear acceleration is then given by Equation (2), namely

abP
(2)
2
I + Mb
––––––––
–––––––––
–

On the other hand, the yo-yo will rotate counterclockwise with no rolling if
M abP
μ  <   (5)
2
(M g − P )(I + M b )
–––––––––––––––––––––––––
–––––––––––––––––––––––––––

The sum of the counterclockwise moments of the forces about C is then Pa−Fb where F = μN = μ(M g − P ) . The
counterclockwise angular acceleration about C is
Pa−Fb P (a − μb) − μM g
  =  . (6)
I I
–––––––––––––––––
––––––––––––––––––
–

 Exercise 22.8.40b

22.8.44 https://phys.libretexts.org/@go/page/9092
I have drawn the four forces on the yo-yo. Its weight M g. The normal reaction of the table on the yo-yo, which is also of
magnitude M g. The tension P in the string. And the frictional force F of the table on the yo-yo.
At this point it may not be immediately obvious whether F acts to the left or the right. For example, let us suppose that the
coefficient of friction is zero. The force P will result in a translation of the yo-yo to the right together with a clockwise rotation
of the yo-yo. So, in which direction does the point A on the circumference of the yo-yo move - to the left or the right? It is hard
to say, but one might suppose, qualitatively, that, if the moment of inertia is large, the induced rotation will be sluggish, so that
A moves to the right. Whereas if I is small, the induced rotation will be rapid, and A will move to the left in spite of the
translational motion of the centre of mass to the right. From this we might conclude that, if μ ≠ 0 , F will act to the right if I is
small, and F will act to the left if I is large. The following analysis shows that this qualitative expectation is correct.centre of
mass to the right. From this we might conclude that, if μ ≠ 0 , F will act to the right if I is small, and F will act to the left if I
is large. The following analysis shows that this qualitative expectation is correct.
(The reader might find some of the Problems in Section 8.2 of Chapter 8 to be helpful at this point, particularly Problem 2.5.)
For the time being, I have drawn F as if acting towards the left.
Let us suppose there is no slipping and that the yo-yo rolls.
The sum of the clockwise moments of the forces about A is P (a  +  b) and the moment of inertia about A is I   +  M b . The 2

P (a + b)
yo-yo therefore undergoes an initial clockwise angular acceleration about A equal to 2
and, therefore (if there is no
I+Mb

slipping), an initial linear acceleration of C to the right equal to

P b(a  +  b)
. (1)
2
I   +  M b
–––––––––––
–––––––––––– –

(P −F )
The above linear acceleration must equal M
from which we obtain

I − M ab
F  =  ( )P (2)
I + M b2

This shows that the frictional force F acts to the left, as shown, if I   >  M ab but if I < M ab the frictional force F acts to the
right. This is in agreement with our qualitative expectations, namely that F will act to the left if I is large, and to the right if I
is small.
Let is consider three cases in turn: I > M ab,  I < M ab and I = M ab .
(i) I > M ab

In this case, F acts to the left, as drawn. Provided F < μM g there will be no slipping at A, and the yo-yo will roll to the right
without slipping. On recalling Equation (2), we see that the yo-yo will roll to the right without slipping, with a linear
acceleration given by Equation (1) if

22.8.45 https://phys.libretexts.org/@go/page/9092
 I − M ab P
μ >( )( ). (3)
2
I + Mb Mg
––––––––––––––––––––––––
–––––––––––––––––––––––––
–

The linear acceleration to the right is given by Equation (1), namely

P b(a  +  b)
. (1)
2
I   +  M b
–––––––––––
–––––––––––– –

However, if
I − M ab P
μ <( )( ). (4)
2
I + Mb Mg
––––––––––––––––––––––––
–––––––––––––––––––––––––
–

slipping occurs at A. The frictional force is the no longer given by Equation (2), but is given by

F = μM g, (5)

and it acts to the left.

(We are concerned in this problem with the initial motion. Once motion is underway, μ has to be replaced with the smaller
coefficient of kinetic friction.)
The net force to the right is then P − μM g so the linear acceleration of C to the right is
P − μM g
. (6)
M
––––––––––
––––––––––

Because of condition (4), this is necessarily positive.

The net clockwise moment of the forces about the centre of mass C is P a  +  μM gb . The yo-yo therefore undergoes a
clockwise angular acceleration about C ofcentre of mass C is P a  +  μM gb . The yo-yo therefore undergoes a clockwise
angular acceleration about C of
P a  +  μM gb
. (7)
I
––––––––––––––
––––––––––––––

P  − μMg P a + μMgb
The linear acceleration to the right of the point A on the circumference of the yo-yo is M
  −  b  ×  
I
, and,
because of condition (4), some algebra will show that this is necessarily positive, as expected.
(ii) I < M ab

(P  + F )
In this case, F acts to the right, and the linear acceleration is M
. Provided that

M ab − I P
μ  >   ( )( ), (8)
2
Mb + I Mg
––––––––––––––––––––––––––
–––––––––––––––––––––––––––

the yo-yo will roll to the right with a linear acceleration given by Equation (1), namely
P b(a  +  b)
. (1)
I   +  M b2
–––––––––––
–––––––––––– –

However, if
M ab − I P
μ  >   ( )( ), (9)
2
Mb + I Mg
––––––––––––––––––––––––––
–––––––––––––––––––––––––––

slipping occurs at A. The frictional force is then given by

F   =  μM g, (10)

and it acts to the right.

The net force to the right is then P   +  μM g , so the linear acceleration to the right is

22.8.46 https://phys.libretexts.org/@go/page/9092
 P   +  μM g
. (11)
M
–––––––––––
––––––––––––

The net clockwise moment of the forces about the centre of mass C is . The yo-yo therefore undergoes a clockwise angular
acceleration about C ofcentre of mass C is . The yo-yo therefore undergoes a clockwise angular acceleration about C of
P a  −  μM gb
. (12)
I
–––––––––––––
––––––––––––––
–

P a−μMgb P +μMg
The linear acceleration to the left of the point A on the circumference of the yo-yo is b × I
−
M
, and, because of
condition (8), some algebra will show that this is necessarily positive, as expected.
I = M ab

In this case, F is zero. Whatever the coefficient of friction, even zero, the yo-yo will undergo a linear acceleration to the P

right (Verify that this is consistent with Equation (1) ), and a clockwise angular acceleration about C equal to . The linearPa

acceleration to the right of the point A on the circumference of the yo-yo is

P Pa
 −b ×
M I

which is zero. The initial linear velocity of the point A is therefore zero.

 Exercise 22.8.40c

I have drawn the four forces on the yo-yo. Its weight M g. The normal reaction of the table on the yo-yo, which is also of
magnitude M g. The tension P in the string. And the frictional force F of the table on the yo-yo. On this occasion (unlike in
Problem 40 (b)) there is no question about the direction of F , which is towards the left.
Let us suppose there is no slipping.
P (b−a)
The sum of the clockwise moments of the forces about A is 2
and the moment of inertia about A is I   +  M b . The yo-
2

I + Mb
P (b−a)
yo therefore undergoes an initial clockwise angular acceleration about A equal to 2
and therefore (if there is no slipping)
I + Mb

an initial linear acceleration to the right equal to

P b(b − a)
. (1)
2
I   +  M b
–––––––––
–––––––––– –

Additional string therefore becomes wrapped around the axle. (Yes, it really does! I tried it!)
(P −F )
The above linear acceleration must equal M
from which we obtain

I   +  M ab
F  =  ( )P. (2)
2
I   +  M b

22.8.47 https://phys.libretexts.org/@go/page/9092
Provided F < μM g ,there will be no slipping at A, and the yo-yo will roll to the right without slipping. Thus the yo-yo will
roll to the right without slipping, with a linear acceleration given by equation (1) if
I+Mab
( 2
)
I+Mb
μ > . (2)
P
( )
Mg

–––––––––––––––

I+Mab
( 2
)
I+Mb
μ < , (22.8.1)
P
( )
Mg

––––––––––––––

F = μM g, (2)

P − μM g
. (22.8.2)
M
––––––––––

P a − μM gb
(22.8.3)
I
––––––––––––

 Exercise 22.8.40d

P b(a + b cos θ
(1)
2
I + Mb

I cos θ − M ab
F =( )P. (2)
2
I + Mb

 Exercise 22.8.40E

P (b cos θ − a)
. (1)
2
I + Mb

P (b − cos θ − a)
. (2)
2
I + Mb

I cos θ + M ab
F =( )P, (3)
2
I + Mb

22.8.48 https://phys.libretexts.org/@go/page/9092
 P (a − b cos θ
. (4)
2
I + Mb

P b(a − b cos θ
(5)
2
I + Mb

F = \left(\frac{I\cos \theta + Mab}{I + Mb^2}\right)P, } \label{40e.6}

 Exercise 22.8.41

¯¯ 4
3m x̄ = 2ma + m × a
3

¯¯ 1
3m ȳ = 2m × a + m × 43a.
2

¯¯ 10 ¯¯ 7
x̄ = a ȳ = a.
9 9

x1 x2 x3
⎡ ⎤
1
⎢ y1 y2 y3 ⎥ .
2
⎣ ⎦
1 1 1

1 1 2 1 2
(2m)( a) = ma .
3 2 6

1 2 2 2
= (2m)a = ma .
3 3

= 0.
1 2 5 2 28 2
Arect = ma + 2m( a) = ma .
6 18 81

2 2 1 2 65 2
Brect = ma + 2m( a) = ma
3 9 81

1 5 5 2
Hrect = 0 + 2m( a)( a) = ma .
9 18 81

1 2 1 2 1 2
= ma − m( a) = ma .
6 3 18

1 2 1 2 2 2
= (m)(2a) − m( × 2a) = ma .
6 3 9

1 1 2
+ m(2a)(a) = + ma .
36 18

1 2 5 2 59 2
Atria = ma + m( a) = ma .
18 8 162

2 2 2 2 22 2
Btria = ma + m( a) = ma .
9 9 81

1 2 2 5 29 2
Htria = ma + m(− a)(− a) = ma .
18 9 9 162

22.8.49 https://phys.libretexts.org/@go/page/9092
 26 2 59 2 37 2
A = ma + ma = ma .
81 162 51

56 2 22 2 26 2
B = ma + ma = ma .
81 81 27

5 2 29 13 2
H = ma + m = ma .
81 162 54

θ = 30º.009 180 and 120º.009 180

2 2
A cos θ − 2H sin θ cos θ + b sin θ.

A0 = 0.546142ma , B = 1.102006ma2. 2
0

 Exercise 22.8.42

2
1 2 1 2 2 ˙ ˙
T = Mx + m(ẋ +l θ + 2l θ ẋ cos θ
2 2

V = constant − mgl cos θ.

2
¨ ˙
(M + m)ẍ + ml(θ cos θ − θ sin θ) = 0,

¨
l θ + ẍ + gθ = 0.

(M+m)g
¨
θ =− θ
Ml

−−−−−
Ml
2π = √ g.
M+m

–––––––––––––
––––––––––––––
–

 Exercise 22.8.43
1 dr 1 2 1
=− sin α sin 2α + 2 cos α cos 2α
k dα 2 2

2 2
= − sin α cos α + (1 + cos α)(2 cos α − 1) = 0.

2 2
3 c + 2 c − 2c − 1 = 0
–––––––––––––––––––––
–––––––––––––––––––––– –

 Exercise 22.8.44

The condition for stability, from Chapter 16 Section 16.9, Equation 16.9.5 is that
2
Ak

V
> HC .

22.8.50 https://phys.libretexts.org/@go/page/9092
k
2
for a filled circle of radius a is 1

4
a
2
. If the length of the cylinder is l, the volume immersed is Asl, so the left hand side of
2

the inequality is a

4sl
.
The depth of the centre of mass is l(s −
1

2
) and the depth of the centre of buoyancy is 1

2
, so that
ls HC  =  
1

2
l(1 − s) . The
2

condition for stability is, then, 4sl

a
>
1

2
l(1 − s) .
With L  = l

(2a)
, this gives, for the condition for stability,
1
L  <  
√8s(1−s)
–––––––––––––
––––––––––––––
–

This function is least for s = 1

2
,  L  <  
√2
1
= 0.707 . For any length less than this, the system is stable for any density. With
L =1 , the inequality can be written 8s 2
− 8s + 1 > 0 , so that s must be less than 0.146 or greater than 0.854.

 Exercise 22.8.45

Before doing the problem, let’s just have a look at the “interesting” property of a (4, 5, 6) triangle

Calculate A by the cosine rule: 16  =  25  +  36  − 60 cos A , hence cos A  =   . 3

Calculate C by the cosine rule: 36  =  16  +  25  − 40 cos C , hence cos C   =   . 1

But cos 2A  =  2 cos 2

 A  − 1  =
1

8
. Therefore C   =  2A .
The external angle at B is 3A.

22.8.51 https://phys.libretexts.org/@go/page/9092
The angles are A  =  41 ∘
.4096
∘
C   =  82 .8192
∘
B  =  55 .7711 (cos B =
9

16
)

Supplement of B  =  124 ∘
.2289

It is not the case that a triangle with one angle equal to twice another one is necessarily a (4, 5, 6) triangle.
After that diversion, let’s move on to the given problem - except that we’ll generalize it to make the length of the rod 2l, and
the lengths of the strings a and b .

The only physics involved is to recall that, if three coplanar forces are in equilibrium, they must be concurrent at a point - in
this case the point C. This means that C must be vertically above the mid-point of the rod.
After that, there is no more physics; the rest is “just” geometry. All we have to do is to find θ in terms of a and b .

If C is to be directly above the mid-point, then AN = MB. That is:

b cos(A + θ)  =  a cos(B − θ) .
This quickly results in
tan θ =
b cos A − a cos B

b sin A + a sin B

.

In our particular example, a

b
  = 
4

5
∘
  =  0.8,  B  =  180   −  3A , and A  =  41 ∘
.4096 .
Thus
tan θ  =  
cos A + 0.8 cos 3A

sin A + 0.8 sin 3A

,

and

22.8.52 https://phys.libretexts.org/@go/page/9092
 ∘
θ  =  12 .78
––––––––––––

If the weight of the rod is mg, I’ll leave it to you to work out the tensions in the strings.

 Exercise 22.8.46

The only physics involved is to recall that, if three coplanar forces are in equilibrium, they must be concurrent at a point - in
this case the point C. This means that C must be vertically above the mid-point. Also, since the planes are smooth, the forces at
A and B are perpendicular to the planes.
The rest is geometry - almost the same as in Problem 45, except that in this problem we are given the angles α and β rather
than the lengths a and b . Start by convincing yourself that the two angles at C are indeed α and β, as marked. Now all that is
required is to express θ in terms of α and β.

Since the mid-point of the rod must be vertically below C, we must have AN = MB. That is:
b sin α = a sin β .
By the Sine Rule, b

a
=
sin B

sin A
, so that sin α sin B = sin β sin A
But A = 90 ° −α + θ and B = 90 ° −β − θ , so
sin α = cos(β − θ) = sin β cos(α + θ) ,
–
which quickly yields tan θ = 1

2
(cot α − cot β) . In our particular example, this is tan θ = 1

2
(√3 − 1), θ = 20.1°. If you wish,
––
–––
––
–
––––––––––––––––––––––
–––––––––––––––––––––––
–
you could work out the forces at A and B in terms of the weight of the rod.

 Exercise 22.8.47

22.8.53 https://phys.libretexts.org/@go/page/9092
I have drawn above the three forces on the rod. The forces at the ends of the rod each make an angle λ to the normal to the
surface, where tan λ = μ , and the three coplanar forces, being in static equilibrium, are concurrent at a point. I have also
introduced the angle ϕ , given by cos ϕ = l/a . All we have to do is to find θ in terms of λ and ϕ - that is to say, in terms of μ
and l/a
Fortunately I found the following formula for a triangle in an old geometry book:
1
cot γ = (cos α − cot β)
2

I’ll leave you to see if you can derive it. The book actually gave a formula for a more general case in which the base of the
triangle isn’t divided equally. For the case

the formula is (1 + x) cot γ = cot α − x cot β .

You can use this in various problems in geometric optics, where you are trying to find relations between object distance, image
distance and radius of curvature or focal length. However, for this problem, we need only the simpler formula, where the base
of the triangle is equally divided.
On applying the simpler formula to our present problem we obtain
1
tan θ = [cot(ϕ − λ) − cot(ϕ + λ)] ,
2
–––––––––––––––––––––––––––––––––
––––––––––––––––––––––––––––––––––
–

and the problem is solved.

Below, I illustrate some examples. Going from left to right we have short l/a = 0.2, medium l/a = 0.4, long l/a = 0.6 rod.
Going from top to bottom we have a slippery μ = 0.5 , a medium /mu = 1.0, and a sticky /mu = 1.5 surface.
Inside each drawing, I tabulate

22.8.54 https://phys.libretexts.org/@go/page/9092
 l/a

/mu

θ °
The long (l/a = 0.6) rod will rest vertically for any /mu > 1.33. But, while it will not slip, the equilibrium is no longer
stable, and the rod will tip after an infinitesimal counterclockwise displacement.

This page titled 22.8: Appendix B is shared under a CC BY-NC 4.0 license and was authored, remixed, and/or curated by Jeremy Tatum via
source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.

22.8.55 https://phys.libretexts.org/@go/page/9092
Index
A corotating reference frame generalized momentum conjugate to the
air resistance 4.9: Centrifugal and Coriolis Forces generalized coordinate
7.2: Air Resistance Proportional to the Speed Couette Viscometer 13.4: The Lagrangian Equations of Motion
Angular momentum 20.4.2: The Couette Viscometer gnorable coordinate
3.7: Angular Momentum cycloid 13.4: The Lagrangian Equations of Motion
Archimedes' Principle 19: The Cycloid
19.1: Introduction to Cycloids
16.7: Archimedes' Principle 19.3: The Intrinsic Equation to the Cycloid
H
asymmetric top 19.5: Motion on a Cycloid, Cusps Up Hamilton's Equations of Motion
2.16: Rotation of Axes - Three Dimensions cylinders 14.3: Hamilton's Equations of Motion
4.6: Force-free Motion of a Rigid Asymmetric Top 2.6: Three-dimensional Solid Figures. Spheres, Hamilton's Variational Principle
Cylinders, Cones. 13.9: Hamilton's Variational Principle
B Hamiltonian Mechanics
body cone D 14: Hamiltonian Mechanics
4.8: Force-free Motion of a Rigid Symmetric Top d’Alembert’s principle holonomic constraint
brachistochrone 4.9: Centrifugal and Coriolis Forces 13.3: Holonomic Constraints
19.7: The Brachystochrone Property of the Cycloid Directional cosines Hooke’s law
Brachystochrone Property 2.16: Rotation of Axes - Three Dimensions 21.4: Hooke’s Law
19.7: The Brachystochrone Property of the Cycloid Doppler effect Hydrostatics
buoyancy 15.18: Doppler Effect 16: Hydrostatics
16.7: Archimedes' Principle Double Pendulum
16.9: Floating Bodies 17.5: Double Pendulum I
ignorable coordinate
C E 14.3: Hamilton's Equations of Motion
calculus of variations electromagnetism inductor
18.4: Area of a Catenoid 15.31: Electromagnetism 11.6: Electrical Analogues
Capillary Rise ellipse inertia tensor
20.2.3: Capillary Rise 2.20: Ellipses and Ellipsoids 2.17: Solid Body Rotation and the Inertia Tensor
catenary Ellipsoid 2.18: Determination of the Principal Axes
18: The Catenary 2.20: Ellipses and Ellipsoids intrinsic equation of the catenary
18.3: Equation of the Catenary in Rectangular
Coordinates, and Other Simple Relations Euler angles 18.2: The Intrinsic Equation to the Catenary

Catenoid 4.8: Force-free Motion of a Rigid Symmetric Top isochronous property

18.4: Area of a Catenoid Euler’s equations 19.5: Motion on a Cycloid, Cusps Up
19.9: The Cycloidal Pendulum
center of buoyancy 4.5: Euler's Equations of Motion
extended cycloid
16.9: Floating Bodies
19.8: Contracted and Extended Cycloids
K
center of mass
kinematic viscosity
1.1: Introduction and Some Definitions
Center of Pressure F 20.4: Viscosity
first time integral kinetic energy (rotation)
16.6: Centre of Pressure
4.3: Kinetic Energy of Rigid Body Rotation
centre of buoyancy 6.3C: Body thrown vertically upwards, initial speed \
(v_{0}\)
16.9: Floating Bodies
centrifugal force Floating Bodies L
4.9: Centrifugal and Coriolis Forces
16.9: Floating Bodies Lagrange's Equations of Motion
coefficient of restitution Force 4.4: Lagrange's Equations of Motion
5.2: Bouncing Balls
3.2: Moment of Force lagrangian
5.3: Head-on Collision of a Moving Sphere with an Force (special relativity) 13.4: The Lagrangian Equations of Motion
Initially Stationary Sphere 15.29: Force Lagrangian Acceleration
complementary function 13.5: Acceleration Components
12.1: More on Differential Equations G Lagrangian Equation
12.2: Forced Oscillatory Motion
Galilean transformation 4.10: The Top
cones 15.5: The Lorentz Transformations Lagrangian Mechanics
2.6: Three-dimensional Solid Figures. Spheres,
Cylinders, Cones.
generalized coordinates 4.4: Lagrange's Equations of Motion
13: Lagrangian Mechanics
4.4: Lagrange's Equations of Motion
conservative force 13.2: Generalized Coordinates and Generalized law of conservation of angular
13.4: The Lagrangian Equations of Motion Forces momentum
Contact Angle generalized forces 3.11: Torque and Rate of Change of Angular
20.2.2: Angle of Contact 4.4: Lagrange's Equations of Motion Momentum
contracted cycloid 13.2: Generalized Coordinates and Generalized linear momentum
19.8: Contracted and Extended Cycloids Forces
3.5: Linear Momentum
Coriolis Forces generalized momentum
Linear Triatomic Molecule
4.9: Centrifugal and Coriolis Forces 14.3: Hamilton's Equations of Motion
2.9: Linear Triatomic Molecule

1 https://phys.libretexts.org/@go/page/47869
Lissajous ellipse Poiseuille's Law Spheres
21.4: Hooke’s Law 20.4.1: Poiseuille's Law 2.6: Three-dimensional Solid Figures. Spheres,
Lorentz transformations Poisson Brackets Cylinders, Cones.
15.5: The Lorentz Transformations 14.5: Poisson Brackets spherical top
principal axes 2.16: Rotation of Axes - Three Dimensions
M 2.18: Determination of the Principal Axes superelastic collision
principal moments of inertia 5.2: Bouncing Balls
Minkowski diagram
15.10: Time Dilation 2.12: Rotation of Axes surface tension
15.14: Order of Events, Causality and the product moment of inertia 20.2: Surface Tension
Transmission of Information 2.11: Plane Laminas. Product Moment. Translation system of conservative forces
Moment of Force of Axes (Parallel Axes Theorem) 4.10: The Top
3.2: Moment of Force Projectiles
Moment of Momentum 7: Projectiles T
3.3: Moment of Momentum prolate spheroid tautochronous
Moments of Inertia 4.8: Force-free Motion of a Rigid Symmetric Top 19.9: The Cycloidal Pendulum
2.1: Definition of Moment of Inertia prolate symmetric top terminal speed
2.3: Moments of Inertia of Some Simple Shapes 2.16: Rotation of Axes - Three Dimensions 6.3B: Body falling under gravity in a resisting
proper length medium, resistive force proportional to the speed
N 15.10: Time Dilation The Lorentz Transformation
nutation proper time 15.7: The Lorentz Transformation as a Rotation
4.10: The Top 15.10: Time Dilation time dilation
19.10: Examples of Cycloidal Motion in Physics 15.10: Time Dilation
R Top
O radius of gyration 4.10: The Top
oblate symmetric top 2.4: Radius of Gyration torque
2.16: Rotation of Axes - Three Dimensions 16.6: Centre of Pressure 3.2: Moment of Force
4.8: Force-free Motion of a Rigid Symmetric Top 3.8: Torque
Resistive Force
oblique collisions 6.3A: Resistive Force Only
3.11: Torque and Rate of Change of Angular
5.4: Oblique Collisions Momentum
rigidity modulus torsion constant
Oblique Doppler Effect 20.3: Shear Modulus and Torsion Constant
15.19: The Transverse and Oblique Doppler Effects 2.10: Pendulums
rocket equation Torsional Oscillator
10.3: The Rocket Equation
P Rotation of Axes
11.3: Torsion Pendulum

paraboloid of safety torus

2.12: Rotation of Axes
2.8: Torus
7.1: No Air Resistance Rotational Inertia
Parallel Axes Theorem transient solution
3.7: Angular Momentum
12.2: Forced Oscillatory Motion
2.5: Plane Laminas and Mass Points distributed in a
Plane Transverse Doppler Effect
2.11: Plane Laminas. Product Moment. Translation
S 15.19: The Transverse and Oblique Doppler Effects
of Axes (Parallel Axes Theorem) second moment of inertia twin paradox
particular integral 2.1: Definition of Moment of Inertia 15.11: The Twins Paradox
12.1: More on Differential Equations 3.13: The Virial Theorem
Pascal’s Paradox second time integral V
16.4: Pressure on a Horizontal Surface. Pressure at 6.3C: Body thrown vertically upwards, initial speed \
Depth (v_{0}\) virial theorem
shear modulus 3.13: The Virial Theorem
Pendulums
2.10: Pendulums 20.3: Shear Modulus and Torsion Constant virtual work
space integral 9.3: Virtual Work
Perpendicular Axes Theorem
2.5: Plane Laminas and Mass Points distributed in a 6.3C: Body thrown vertically upwards, initial speed \ Viscometer
Plane (v_{0}\) 20.4.2: The Couette Viscometer

2 https://phys.libretexts.org/@go/page/47869
Detailed Licensing
Overview
Title: Classical Mechanics (Tatum)
Webpages: 230
Applicable Restrictions: Noncommercial
All licenses found:
CC BY-NC 4.0: 96.1% (221 pages)
Undeclared: 3.9% (9 pages)

By Page
Classical Mechanics (Tatum) - CC BY-NC 4.0 2.12: Rotation of Axes - CC BY-NC 4.0
Front Matter - Undeclared 2.13: Momental Ellipse - CC BY-NC 4.0
TitlePage - Undeclared 2.14: Eigenvectors and Eigenvalues - CC BY-NC 4.0
InfoPage - Undeclared 2.15: Solid Body - CC BY-NC 4.0
Table of Contents - Undeclared 2.16: Rotation of Axes - Three Dimensions - CC BY-
Licensing - Undeclared NC 4.0
2.17: Solid Body Rotation and the Inertia Tensor - CC
1: Centers of Mass - CC BY-NC 4.0
BY-NC 4.0
1.1: Introduction and Some Definitions - CC BY-NC 2.18: Determination of the Principal Axes - CC BY-
4.0 NC 4.0
1.2: Plane Triangular Lamina - CC BY-NC 4.0 2.19: Moment of Inertia with Respect to a Point - CC
1.3: Plane Areas - CC BY-NC 4.0 BY-NC 4.0
1.4: Plane Curves - CC BY-NC 4.0 2.20: Ellipses and Ellipsoids - CC BY-NC 4.0
1.5: Summary of the Formulas for Plane Laminas and 2.21: Tetrahedra - CC BY-NC 4.0
Curves - CC BY-NC 4.0
3: Systems of Particles - CC BY-NC 4.0
1.6: The Theorems of Pappus - CC BY-NC 4.0
3.1: Introduction to Systems of Particles - CC BY-NC
1.7: Uniform Solid Tetrahedron, Pyramid and Cone -
4.0
CC BY-NC 4.0
3.2: Moment of Force - CC BY-NC 4.0
1.8: Hollow Cone - CC BY-NC 4.0
3.3: Moment of Momentum - CC BY-NC 4.0
1.9: Hemispheres - CC BY-NC 4.0
3.4: Notation - CC BY-NC 4.0
1.S: Centers of Mass (Summary) - CC BY-NC 4.0
3.5: Linear Momentum - CC BY-NC 4.0
2: Moments of Inertia - CC BY-NC 4.0
3.6: Force and Rate of Change of Momentum - CC
2.1: Definition of Moment of Inertia - CC BY-NC 4.0 BY-NC 4.0
2.2: Meaning of Rotational Inertia - CC BY-NC 4.0 3.7: Angular Momentum - CC BY-NC 4.0
2.3: Moments of Inertia of Some Simple Shapes - CC 3.8: Torque - CC BY-NC 4.0
BY-NC 4.0 3.9: Comparison - CC BY-NC 4.0
2.4: Radius of Gyration - CC BY-NC 4.0 3.10: Kinetic energy - CC BY-NC 4.0
2.5: Plane Laminas and Mass Points distributed in a 3.11: Torque and Rate of Change of Angular
Plane - CC BY-NC 4.0 Momentum - CC BY-NC 4.0
2.6: Three-dimensional Solid Figures. Spheres, 3.12: Torque, Angular Momentum and a Moving
Cylinders, Cones. - CC BY-NC 4.0 Point - CC BY-NC 4.0
2.7: Three-dimensional Hollow Figures. Spheres, 3.13: The Virial Theorem - CC BY-NC 4.0
Cylinders, Cones - CC BY-NC 4.0
4: Rigid Body Rotation - CC BY-NC 4.0
2.8: Torus - CC BY-NC 4.0
2.9: Linear Triatomic Molecule - CC BY-NC 4.0 4.1: Introduction to Rigid Body Rotation - CC BY-NC
2.10: Pendulums - CC BY-NC 4.0 4.0
2.11: Plane Laminas. Product Moment. Translation of 4.2: Angular Velocity and Eulerian Angles - CC BY-
Axes (Parallel Axes Theorem) - CC BY-NC 4.0 NC 4.0

1 https://phys.libretexts.org/@go/page/65342
 4.3: Kinetic Energy of Rigid Body Rotation - CC BY- 10.3: The Rocket Equation - CC BY-NC 4.0
NC 4.0 10.E: Rocket Motion (Exercises) - CC BY-NC 4.0
4.4: Lagrange's Equations of Motion - CC BY-NC 4.0 11: Simple and Damped Oscillatory Motion - CC BY-NC
4.5: Euler's Equations of Motion - CC BY-NC 4.0 4.0
4.6: Force-free Motion of a Rigid Asymmetric Top - 11.1: Simple Harmonic Motion - CC BY-NC 4.0
CC BY-NC 4.0 11.2: Mass Attached to an Elastic Spring - CC BY-NC
4.7: Nonrigid Rotator - CC BY-NC 4.0 4.0
4.8: Force-free Motion of a Rigid Symmetric Top - 11.3: Torsion Pendulum - CC BY-NC 4.0
CC BY-NC 4.0 11.4: Ordinary Homogeneous Second-order
4.9: Centrifugal and Coriolis Forces - CC BY-NC 4.0 Differential Equations - CC BY-NC 4.0
4.10: The Top - CC BY-NC 4.0 11.5: Damped Oscillatory Motion - CC BY-NC 4.0
4.11: Appendix - CC BY-NC 4.0
11.5i: Light damping- \( \gamma < 2\omega_{0}\)
5: Collisions - CC BY-NC 4.0
- CC BY-NC 4.0
5.1: Introduction - CC BY-NC 4.0 11.5ii: Heavy damping- \( \gamma >
5.2: Bouncing Balls - CC BY-NC 4.0 2\omega_{0}\) - CC BY-NC 4.0
5.3: Head-on Collision of a Moving Sphere with an 11.5iii: Critical damping- \( \gamma =
Initially Stationary Sphere - CC BY-NC 4.0 2\omega_{0}\) - CC BY-NC 4.0
5.4: Oblique Collisions - CC BY-NC 4.0
11.6: Electrical Analogues - CC BY-NC 4.0
5.5: Oblique (Glancing) Elastic Collisions,
12: Forced Oscillations - CC BY-NC 4.0
Alternative Treatment - CC BY-NC 4.0
12.1: More on Differential Equations - CC BY-NC 4.0
6: Motion in a Resisting Medium - CC BY-NC 4.0
12.2: Forced Oscillatory Motion - CC BY-NC 4.0
6.1: Introduction - CC BY-NC 4.0 12.3: Electrical Analogue - CC BY-NC 4.0
6.2: Uniformly Accelerated Motion - CC BY-NC 4.0
13: Lagrangian Mechanics - CC BY-NC 4.0
6.3: Uniformly Accelerated Motion - CC BY-NC 4.0
13.1: Introduction to Lagrangian Mechanics - CC BY-
6.3A: Resistive Force Only - CC BY-NC 4.0
NC 4.0
6.3B: Body falling under gravity in a resisting
13.2: Generalized Coordinates and Generalized
medium, resistive force proportional to the speed -
Forces - CC BY-NC 4.0
CC BY-NC 4.0
13.3: Holonomic Constraints - CC BY-NC 4.0
6.3C: Body thrown vertically upwards, initial
13.4: The Lagrangian Equations of Motion - CC BY-
speed \(v_{0}\) - CC BY-NC 4.0
NC 4.0
6.4: Motion in which the Resistance is Proportional to
13.5: Acceleration Components - CC BY-NC 4.0
the Square of the Speed - CC BY-NC 4.0
13.6: Slithering Soap in Conical Basin - CC BY-NC
6.4A: Resistive Force Only - CC BY-NC 4.0 4.0
7: Projectiles - CC BY-NC 4.0 13.7: Slithering Soap in Hemispherical Basin - CC
7.1: No Air Resistance - CC BY-NC 4.0 BY-NC 4.0
7.2: Air Resistance Proportional to the Speed - CC 13.8: More Lagrangian Mechanics Examples - CC
BY-NC 4.0 BY-NC 4.0
7.3: Air Resistance Proportional to the Square of the 13.9: Hamilton's Variational Principle - CC BY-NC
Speed - CC BY-NC 4.0 4.0
8: Impulsive Forces - CC BY-NC 4.0 14: Hamiltonian Mechanics - CC BY-NC 4.0
8.1: Introduction - CC BY-NC 4.0 14.1: Introduction to Hamiltonian Mechanics - CC
8.2: Problem - CC BY-NC 4.0 BY-NC 4.0
9: Conservative Forces - CC BY-NC 4.0 14.2: A Thermodynamics Analogy - CC BY-NC 4.0
14.3: Hamilton's Equations of Motion - CC BY-NC
9.1: Introduction - CC BY-NC 4.0
4.0
9.2: The Time and Energy Equation - CC BY-NC 4.0
14.4: Hamiltonian Mechanics Examples - CC BY-NC
9.3: Virtual Work - CC BY-NC 4.0
4.0
9.E: Conservative Forces (Exercises) - CC BY-NC 4.0
14.5: Poisson Brackets - CC BY-NC 4.0
10: Rocket Motion - CC BY-NC 4.0
15: Special Relativity - CC BY-NC 4.0
10.1: Introduction - CC BY-NC 4.0
10.2: An Integral - CC BY-NC 4.0

2 https://phys.libretexts.org/@go/page/65342
 15.1: Introduction to Special Relativity - CC BY-NC 17.2: The Diatomic Molecule - CC BY-NC 4.0
4.0 17.3: Two Masses, Two Springs and a Brick Wall -
15.2: Preparation - CC BY-NC 4.0 CC BY-NC 4.0
15.3: Preparation - CC BY-NC 4.0 17.4: Double Torsion Pendulum - CC BY-NC 4.0
15.4: Speed is Relative - The Fundamental Postulate 17.5: Double Pendulum - CC BY-NC 4.0
of Special Relativity - CC BY-NC 4.0 17.6: Linear Triatomic Molecule - CC BY-NC 4.0
15.5: The Lorentz Transformations - CC BY-NC 4.0 17.7: Two Masses, Three Springs, Two brick Walls -
15.6: But This Defies Common Sense - CC BY-NC CC BY-NC 4.0
4.0 17.8: Transverse Oscillations of Masses on a Taut
15.7: The Lorentz Transformation as a Rotation - CC String - CC BY-NC 4.0
BY-NC 4.0 17.9: Vibrating String - CC BY-NC 4.0
15.8: Timelike and Spacelike 4-Vectors - CC BY-NC 17.10: Water - CC BY-NC 4.0
4.0 17.11: A General Vibrating System - CC BY-NC 4.0
15.9: The FitzGerald-Lorentz Contraction - CC BY- 17.12: A Driven System - CC BY-NC 4.0
NC 4.0 17.13: A Damped Driven System - CC BY-NC 4.0
15.10: Time Dilation - CC BY-NC 4.0 18: The Catenary - CC BY-NC 4.0
15.11: The Twins Paradox - CC BY-NC 4.0 18.1: Introduction - CC BY-NC 4.0
15.12: A, B and C - CC BY-NC 4.0 18.2: The Intrinsic Equation to the Catenary - CC BY-
15.13: Simultaneity - CC BY-NC 4.0 NC 4.0
15.14: Order of Events, Causality and the 18.3: Equation of the Catenary in Rectangular
Transmission of Information - CC BY-NC 4.0 Coordinates, and Other Simple Relations - CC BY-NC
15.15: Derivatives - CC BY-NC 4.0 4.0
15.16: Addition of Velocities - CC BY-NC 4.0 18.4: Area of a Catenoid - CC BY-NC 4.0
15.17: Aberration of Light - CC BY-NC 4.0
19: The Cycloid - CC BY-NC 4.0
15.18: Doppler Effect - CC BY-NC 4.0
15.19: The Transverse and Oblique Doppler Effects - 19.1: Introduction to Cycloids - CC BY-NC 4.0
CC BY-NC 4.0 19.2: Tangent to the Cycloid - CC BY-NC 4.0
15.20: Acceleration - CC BY-NC 4.0 19.3: The Intrinsic Equation to the Cycloid - CC BY-
15.21: Mass - CC BY-NC 4.0 NC 4.0
15.22: Momentum - CC BY-NC 4.0 19.4: Variations - CC BY-NC 4.0
15.23: Some Mathematical Results - CC BY-NC 4.0 19.5: Motion on a Cycloid, Cusps Up - CC BY-NC 4.0
15.24: Kinetic Energy - CC BY-NC 4.0 19.6: Motion on a Cycloid, Cusps Down - CC BY-NC
15.25: Addition of Kinetic Energies - CC BY-NC 4.0 4.0
15.26: Energy and Mass - CC BY-NC 4.0 19.7: The Brachystochrone Property of the Cycloid -
15.27: Energy and Momentum - CC BY-NC 4.0 CC BY-NC 4.0
15.28: Units - CC BY-NC 4.0 19.8: Contracted and Extended Cycloids - CC BY-NC
15.29: Force - CC BY-NC 4.0 4.0
15.30: The Speed of Light - CC BY-NC 4.0 19.9: The Cycloidal Pendulum - CC BY-NC 4.0
15.31: Electromagnetism - CC BY-NC 4.0 19.10: Examples of Cycloidal Motion in Physics - CC
BY-NC 4.0
16: Hydrostatics - CC BY-NC 4.0
20: Miscellaneous - CC BY-NC 4.0
16.1: Introduction to Hydrostatics - CC BY-NC 4.0
16.2: Density - CC BY-NC 4.0 20.1: Introduction - CC BY-NC 4.0
16.3: Pressure - CC BY-NC 4.0 20.2: Surface Tension - CC BY-NC 4.0
16.4: Pressure on a Horizontal Surface. Pressure at 20.2.1: Excess Pressure Inside Drops and Bubbles
Depth - CC BY-NC 4.0 - CC BY-NC 4.0
16.5: Pressure on a Vertical Surface - CC BY-NC 4.0 20.2.2: Angle of Contact - CC BY-NC 4.0
16.6: Centre of Pressure - CC BY-NC 4.0 20.2.3: Capillary Rise - CC BY-NC 4.0
16.7: Archimedes' Principle - CC BY-NC 4.0 20.3: Shear Modulus and Torsion Constant - CC BY-
16.8: Some Simple Examples - CC BY-NC 4.0 NC 4.0
16.9: Floating Bodies - CC BY-NC 4.0 20.4: Viscosity - CC BY-NC 4.0
17: Vibrating Systems - CC BY-NC 4.0 20.4.1: Poiseuille's Law - CC BY-NC 4.0
17.1: Introduction - CC BY-NC 4.0 20.4.2: The Couette Viscometer - CC BY-NC 4.0

3 https://phys.libretexts.org/@go/page/65342
21: Central Forces and Equivalent Potential - CC BY-NC 22.2: Table of Dimensions - CC BY-NC 4.0
4.0 22.3: Checking Equations - CC BY-NC 4.0
21.1: Introduction to Central Forces - CC BY-NC 4.0 22.4: Deducing Relationships - CC BY-NC 4.0
21.2: Motion Under a Central Force - CC BY-NC 4.0 22.5: Dimensionless Quantities - CC BY-NC 4.0
21.3: Inverse Square Attractive Force - CC BY-NC 22.6: Different Fundamental Quantities - CC BY-NC
4.0 4.0
21.4: Hooke’s Law - CC BY-NC 4.0 22.7: Appendix A - CC BY-NC 4.0
21.5: Inverse Fourth Power Attractive Force - CC BY- 22.8: Appendix B - CC BY-NC 4.0
NC 4.0 Back Matter - Undeclared
21.6: A General Central Force - CC BY-NC 4.0 Index - Undeclared
21.7: Inverse Cube Attractive Force - CC BY-NC 4.0 Glossary - Undeclared
22: Dimensions - CC BY-NC 4.0 Detailed Licensing - Undeclared
22.1: Mass, Length and Time - CC BY-NC 4.0

4 https://phys.libretexts.org/@go/page/65342