Chapter 1

Vectors

1.1 Introduction

Physical quantities such as volume, area, length, temperat-

ure and time only have magnitude and each is completely
characterized by a single real number (with an appropriate
unit of measurement). These quantities are often referred
to as scalars.
Quantities such as position, velocity, acceleration or force,
each has a magnitude and a direction, and they are represen- y
ted mathematically in a component-by-component fashion,
or by means of an arrow, where the arrow points in a cer-
tain direction. The length of the arrow may be interpreted
as the size of the relevant quantity. The term vector is given

}
h
gt
to such quantities. le n
y -com pon en t
}}
n
d ir e c t io

1.2 Vector notation x-com p on e n t

Let us first only view vectors in 2D, i.e. in a plane. A vector x

in 2D may be described by its magnitude and its direction.
(The direction is always taken relative to some fixed axis,
e.g., the x-axis). Alternatively, given an xy coordinate (or
Cartesian) system the vector may also be described by its
two components in the x and y directions respectively. See Figure 1.1. A vector as an arrow, showing its
Figure 1.1. two components, as well as its magnitude and its
It is the purpose of this chapter to establish the notation and direction. The direction is an angle.
rules by which vectors are manipulated mathematically.
In these notes vectors will be printed in boldface symbols

1
 2

such as a, b, v, F, etc. There are also other notations used to de-

note vectors, for example:
We shall discuss and use mostly vectors in 2D (a plane)
and vectors in 3D (space). We shall often alternate between a or a or ~a or
*
a .
working with vectors in 2D or in 3D, depending on the prob-
lem at hand, and it will usually be clear from the problem If the tail of the arrow is at A and the
statement whether 2D or 3D vectors are required. arrow head is at B, it may also be denoted
A vector in 2D, has two components. Consider the vector a by
! *
with first component 4, and second component 3. We may AB or AB .
write it in column form as follows:
" #
4
a= .
3

2D Vectors as displacement arrows:

The most straightforward use of a vector is to denote po-
sition or displacement. For example, the point in the 2D
Cartesian plane with x-coordinate 4, and y-coordinate 3,
a

}
may be viewed as a displacement from the origin given by
the vector a above. In other words, the point lies at the y
arrow head of the vector. 3
a
If the vector a denotes position it is therefore an arrow start-
}

}
3 4
ing at the origin and ending at the point (4, 3). With this a
interpretation of a vector, coordinates of the point in xy-
}
}
space correspond to components of a vector.[See Figure 1.2] 4 3

The vector starting at the origin is often used to describe } x

position in 2D space and is then called a position vector. 4
Vectors may however start at any coordinate point — its
components denote shifts or displacements in the particular Figure 1.2.  Graphical representation of the
directions. With this notion it is clear that the arrow repres- vector a = 4 in three di↵erent positions.
3
entation of a vector may be an arrow positioned anywhere
in 2D space as long as it keeps the same two components.
Consequently this means that the arrow also keeps its dir-
ection and length, irrespective of where it is placed on the
xy-plane.
3D Vectors as displacement arrows:
In three dimensions, there are three position coordinates, x,
y, and z. A position vector in 3D must therefore have three
components, where each component represents a shift (or
displacement) in each of the three directions.
Consider the vector a with first component 3, and second
component 1, and third component 2. We may write it in
 3

column form as follows:

2 3
3
6 7
a = 4 1 5.
2

Although more difficult to visualize, we shall still often try

to depict such vectors as arrows in a system of three axes
(x, y, and z), by adding a rectangular box with sides equal
to the respective components of the vector. [See Figure 1.3]
In this graphical representation it is recommended to keep
the scales in the various directions relatively correct.
Subscript notation for components:
In 3D we shall label the components of a vector with sub-
script indices, x, y, and z. The vector a will then have as Figure 1.3.
" #Graphical representation of the
first component ax , and its second component is written as 3
vector a = 1 .
ay , etc. This means that every vector can be written out 2
fully showing all its components, or may be referred to by
only using its boldface symbol, for example Checkpoint 1.1:
If a = 2b, b1 = 3, a2 = 8, and a3 = 12,
2 3
a find b.
x
6 7
a = 4 ay 5 .
az

We may also label the component rather with numbers, for

example we may write
2 3
v1
6 7
v = 4 v2 5 .
v3
[Do Checkpoint 1.1]

1.3 Basic vector algebra

We shall state the rules for vector algebra here for vectors
in 3D. However, in Figure 1.4 the corresponding geometrical
interpretation will be depicted only for 2D vectors.
Vector equality:
Vectors are equal if (and only if) each of the corresponding
components are equal. Let
2 3 2 3
a1 b1
6 7 6 7
a = 4 a2 5 b = 4 b2 5 ,
a3 b3
 4

then:

a = b, implies aj = bj , j = 1, 2, 3. (1.1)
2
(From equation (1.1) it means that a=b, if a1 =b1 , a2 =b2 , b
a3 =b3 .)
As mentioned already, this implies that in the geometrical
2
a
interpretation of a vector, the two arrows representing the
two equal vectors, have the same corresponding x, y, and z- 1 1
components. This means that the vectors are parallel (also
pointing in the same direction) and have the same mag-
nitudes. See Figure 1.4(a). Note that: Equality does not
necessarily mean that the two equal vectors start at the same
point. 2
a
Negative vectors: 2
c=-a
The negative (additive inverse) of a vector is simply a vector
where each component has changed sign:
1 1

c= a, implies cj = aj , j = 1, 2, 3. (1.2)

Using the arrow visualization, c is just a with its direction

reversed. See Figure 1.4(b).
Vector addition: 2
c b
2
Vectors are added by adding each component:
2
a
c = a + b, implies cj = aj + bj , j = 1, 2, 3. (1.3)
1 1
1
Using the geometrical interpretation of a vector as an arrow,
vector addition may be viewed as joining the two vectors a
and b head to tail. The result c, is the arrow running from
the free tail to the free head. See Figure 1.4(c). Note that
the tail of c is where the free tail is, and its head is where
2 a2
the free head is. Check for yourself that adding a to b, gives
2 a 2a
the same result as adding b to a — i.e. vector addition is
commutative.
Combining (1.2) and (1.3) enables us to subtract vectors: 1 2 a1

a b = a + ( b). (1.4) Figure 1.4. Graphical representation of (a)

vector equality, (b) the negative of a vector, (c)
vector addition, and (d) scalar multiplication of a
vector.
 5

Scaling a vector:
A vector multiplied by a scalar is a new vector where every
component has been multiplied by that scalar. See Figure
1.4(d).

c = a, implies cj = aj , j = 1, 2, 3. (1.5)

We shall also use this opportunity to introduce another con-

vention used in these notes: scalars will be denoted by Greek
lower case letters such as ↵, , , µ, etc.
A vector divided by a scalar is just another way of doing
scalar multiplication:
a 1
= a.

Here is a summary of the definitions and rules of vector

operations, where we have generalized the definitions of the
vector operations for vectors in n-dimensions.

DEFINITIONS: VECTOR OPERATIONS

Equality: a=b when aj = bj for all j
a and b are both in n dimensions
Addition: c = a + b when cj = aj + bj for all j
a, b and c are all in n dimensions
Scalar multiplication: b= a when bj = a j for all j

RULES: VECTOR OPERATIONS:

VECTOR ADDITION:
Commutativity: a+b=b+a
Associativity: a + (b + c) = (a + b) + c
MULTIPLICATION BY A SCALAR:
Commutativity: a=a
Associativity: (µa) = ( µ)a
Distributivity over vector addition: (a + b) = a + b
Distributivity over scalar addition: ( + µ)a = a + µa
 6

1.4 Magnitude and direction Checkpoint 1.2:

Magnitude " #
5
The magnitude or length of vector a (more precisely the 1. If a = , calculate a.
12
Euclidean length), also called the norm of the vector, is ex-
2 3
pressed as kak or just a. Note that the letter ‘a’ is now em- 1
ployed in three di↵erent ways relating to the same vector: 6 7
2. If b = 4 2 5, calculate b.
The vector is a, its j-th component is aj and the magnitude 2
of the vector is a.
3. What must cx2be, so3 that the mag-
For vectors in 2D it is easy to use Pythagoras’ theorem to
2
show that the magnitude of the vector is given by 6 7
nitude of c = 4 6 5 is 7?
q cx
kak = (ax )2 + (ay )2 . (1.6)
4. Is a b the same as ka bk? Sup-
For vectors in 3D, Pythagoras’ theorem must be applied ply an example or give a counter ex-
twice to show that the magnitude of the vector is given by ample.

q
kak = (ax )2 + (ay )2 + (az )2 . (1.7)
[Do Checkpoint 1.2]

Unit vectors of Cartesian coordinates

A unit vector is vector of magnitude 1. Unit vectors may be
used to represent the axes of a Cartesian coordinate system.
The unit vectors in the direction of the x, y, and z axes of a
three dimensional Cartesian coordinate system are defined,
respectively, as: z
The unit vectors:
k
2 3 2 3 2 3
1 0 0
6 7 6 7 6 7
i=4 0 5
0
j=4 1 5
0
k = 4 0 5.
1
y
j
When working in 2D (in Cartesian coordinates), we shall
use only i and j, and then the definitions are of course
" # " #
1 0
i= j= . i
0 1
x
Using the notion of vector addition, any vector in 3D can be
Figure 1.5. Unit vectors of Cartesian coordin-
expressed in terms of these unit vectors, for example
ates i, j and k.
 7

2 3 2 3 2 3 2 3
1 0 0 4
6 7 6 7 6 7 6 7
4i + 3j 2k = 4 4 0 5 + 3 4 1 5 24 0 5 = 4 3 5.
0 0 1 2

The two notations (i.e. using a column of components, or

expressing it as a sum of scaled unit vectors) are equivalent.
The direction of a vector in 2D:
For the purpose of these notes, we shall use the convention
that the direction of a vector in 2D is given by the angle
(labeled ✓ here) between the vector and the positive x-axis.
[See Figure 1.6] The angle is positive if it is taken counter

clockwise from the positive x-axis, and it is negative if it is

in a clockwise direction. y

}
By the use of trigonometry, the relationship between the
components of a vector and its magnitude and direction is a a
ax = a cos ✓ a sin
ay = a sin ✓.

This means that a position vector that lies in the first quad-
rant, i.e. its direction satisfies ✓ 2 (0, 90 ), has both com-
ponents positive. A position vector that lies in the second
quadrant, i.e. its direction satisfies ✓ 2 (90 , 180 ), has ax
negative and ay positive. A position vector that lies in the
third quadrant, i.e. its direction satisfies ✓ 2 (180 , 270 ),
}} a cos
Figure 1.6. The relationship between the
x

direction (✓) and the two components of a vector

in 2D.
has both components negative. Lastly, a position vector a
that lies in the fourth quadrant, i.e. its direction satisfies
✓ 2 (270 , 360 ), has ax positive and ay negative.
The direction of a vector in 3D:
For vectors in 3D the direction cannot be associated with a
single angle, and often direction vectors, i.e. a unit vector
that points in the desired direction, are used. Describing the
direction of a 3D vector is discussed elsewhere.
The null vector:
What is a a? If we use the arrow interpretation, it must be
a vector with no magnitude at all — it is just a point. This
vector will be called the null vector, and will be denoted by
the symbol 0. Written out this vector (in 3D) is
2 3
0
6 7
0 = 4 0 5.
0
 8

The null vector is the only vector that has no particular

direction. All other vectors have specific magnitudes and
directions. Geometrically the null vector describes the point
at the origin of the axes.

Example 1.1

Two vectors a and b are shown. [See Figure 1.7]. First add
the vectors graphically. Call the result c, and find the mag-
nitude of c by measuring it, and also find the direction of c
using a protractor.
Then write down the components of a and b, and calculate y
c = a + b. Find the direction of c as well as its magnitude
using the formulae.

Graphical Solution: a
The figure [See Figure 1.8] shows the graphical addition of
c = a + b. Vector b is positioned so that its tail touches
the arrow head of a. Remember that when moving a vector b
about, you must keep its magnitude and orientation con- x
stant. Measuring its length yields c ⇡ 5 and the direction
can be measured as ✓ ⇡ 217 .

Calculated solution (column notation):

Reading the components o↵ the figure yields
" # " #
2 2
a= , b= .
2 5 Figure 1.7. Vectors a and b.

Then
" # " # " #
2 2 4
c=a+b= + =
2 5 3
 9

The length of c is
q
c= ( 3)2 + ( 4)2 = 5,

and its direction is

✓ ◆
3
✓ = 180 + arctan = 216.87 .
4 y
Calculated solution (unit vector notation):
Using the unit vectors i and j, the solution is as follows:
a
a= 2i + 2j, b= 2i 5j

}
217

Then c is as follows b
c x
c = ( 2i + 2j) + ( 2i 5j), -3
... after removing the brackets and rearaging the terms
= 2i 2i + 2j 5j
=( 2 2)i + (2 5)j

}
= 4i 3j. -4
Then the length is
q Figure 1.8. Adding a and b graphically.
c= ( 3)2 + ( 4)2 = 5,

and the direction of c is

✓ ◆
3
✓ = 180 + arctan = 216.87 .
4

1.5 The dot product

The dot product also called the scalar product or the inner
product is a special operation between two vectors that re-
turns a scalar as answer. The dot product a · b may either
be defined geometrically or algebraically. In these notes we
shall use the algebraic approach.
 10

The dot product:

2 3 2 3
a1 b1
6 7 6 7
The dot product a · b of a = 4 a2 5 and b = 4 b2 5
a3 b3
is given by
a · b = a 1 b 1 + a 2 b2 + a 3 b 3 .

In 2D the dot product is just a · b = a1 b1 + a2 b2 .

The dot product with unit vectors
It is fundamentally important to first determine the dot
products of the unit vectors, i, j and k, with each other.
This information is used to calculate dot products of gen-
eral vectors (especially when we shall later work in coordin-
ate systems other than Cartesian). Let us for example check
the result of i · i as well as i · j:
2 3 2 3
1 1
6 7 6 7
i · i = 4 0 5 · 4 0 5 = 1 + 0 + 0 = 1,
0 0
2 3 2 3
1 0
6 7 6 7
i · j = 4 0 5 · 4 1 5 = 0 + 0 + 0 = 0.
0 0
It should be clear that in general the dot product of any unit
vector with itself is one, and the dot product between two
orthogonal unit vectors is zero. (This is only the case in spe-
cial coordinate systems that are called orthogonal, however,
we shall not further explain this technicality at this stage.)
The full set of rules for dot products of the unit vectors is
as follows:

Dot products of i, j and k:

i·i =1 i·j =0
j·j =1 j·k =0
k·k =1 k·i =0

Distributivity of the dot product over addition:

The dot product is distributive over vector addition,

a · (b + c) = a · b + a · c. (1.8)

Confirm this yourself with a simple example or do the check-

point. [Do Checkpoint 1.3]
 11

When expressing a vector in the unit vector notation, these Checkpoint 1.3:
identities together with the distributivity rule are used to Let
calculate the dot product. " # " # " #
a1 b1 c1
Angles between vectors a= , b= , c=
a2 b2 c2
There is a close relationship between the dot product and
the angle between two vectors. Consider the two vectors a Add b and c first and then multiply out
and b in Figure 1.9. If we join the arrow heads of a and b
a · (b + c).
then it completes a triangle. This is true in both 2D and
3D. The completing line segment is itself a vector — it is Alternatively, multiply out a · b as well as
a b. We can calculate the lengths of all three vectors. The a · c and then add them. Is the result the
cosine rule for triangles may then be used to find the angle same?
between two sides if we know the lengths of all three sides
of the triangle.
Let the angle between a and b be ✓. Then
ka bk2 = a2 + b2 2ab cos ✓
(ax bx )2 + (ay by )2 + (az bz )2 = a2x + a2y + a2z + b2x + b2y + b2z 2ab cos ✓
....... Cancelling similar terms on both sides.
2(ax bx + ay by + az bz ) = 2ab cos ✓
ax bx + ay by + az bz = ab cos ✓

Notice that the expression on the left hand side is a scalar

quantity that we have seen before — it is the dot product
between a and b:
a · b = a x b x + a y by + a z b z . (1.9)

We therefore have the following convenient relationship between

the dot product of two vectors and the angle between them:

Dot product of two vectors:

Figure 1.9. Calculating the angle ✓ between a
Let ✓ be the angle between a and b (both are in 3D). Then and b.

a · b = ab cos ✓.

In some texts the above statement is used as the definition of

Checkpoint 1.4:
the dot product as it has an interesting geometric interpret-
ation. The disadvantage of this definition is that it difficult
to extend to more than three dimensions and also it is cum- 1. Find the angle between a = i + j
bersome to use it to prove some of the properties of a dot and b = i. Also check your answer
product. geometrically.
[Do Checkpoint 1.4]
2. Find the angle between a = i + j
The dot product provides a convenient notation to express and b = i. Also check your answer
the length of a vector. Since the angle between a and itself geometrically.
 12

is 0 , and cos 0 = 1,

a · a = a2 .

When calculating expressions that involve the length of a

vector, it is often convenient to rather work with the length
squared and to express it as the dot product of the vector
with itself.
Therefore the length of a vector may be expressed as
p
a = a · a. (1.10)

It now makes sense that, for example i · i = 1, since this

product simple gives the length-squared of the unit vector.
Also is makes sense that the dot product between two dif-
ferent (orthogonal) unit vectors must be zero because they
have a 90 angle between them.
 13

Example 1.2

Calculate the dot product between a and b and also calcu-

late the angle between them.
2 3 2 3
3 2
6 7 6 7
a=4 2 5 b=4 1 5
6 2

Solution (Column notation):

2 3 2 3
3 2
6 7 6 7
a·b=4 2 5 · 4 1 5 = (3)(2) + ( 2)(1) + (6)(2) = 16
6 2
q
kak = 32 + ( 2)2 + 62 = 7
p
kbk = 22 + 12 + 22 = 3.
Therefore
3 ⇥ 7 ⇥ cos ✓ = 16
and therefore ✓ = 40.37 .
Solution (Unit vector notation):

a · b = (3i 2j + 6k) · (2i + j + 2k)

= (3)(2)i · i + (3)(1)i · j + (3)(2)i · j + ( 2)(2)j · i + ( 2)(1)j · j
+( 2)(2)j · k + (6)(2)k · i + (6)(1)k · j + (6)(2)k · k
.... remember that i · j = 0, etc, therefore most of the terms are multiplied by zero.
= 6 + 0 + 0 + 0 + ( 2) + 0 + 0 + 0 + 12
= 16

The rest follows as above.

c
Orthogonality:
An important special case is when the dot product between b
two vectors is zero – the vectors are said to be orthogonal.
A zero dot product means the angle between the vectors is a
90 , i.e. the two vectors are perpendicular in space.
Can three vectors in 3D be mutually perpendicular? Geo-
metrically the vectors are like three sides of a cube meeting
at the vertex of the cube.[See Figure 1.10] When a number of
vectors are mutually perpendicular, we say that they form
an orthogonal set.

Figure 1.10. A set of three orthogonal vectors

in 3D.
 14

Expressed in symbols, this is Checkpoint 1.5:

Which of the following vectors are ortho-
gonal?
Orthogonal vectors:
" # " #
{a, b, c} is an orthogonal set, if 1 10
a · b = 0, (1.11) (a) a = , and b =
2 5
b · c = 0,
" # " #
c · a = 0. 1 1
(b) c = , and d =
1 2
Think of the unit vectors i, j, and k. They are certainly 2 3 2 3
mutually perpendicular. Check for yourself by calculating 2 2
6 7 6 7
the dot products between them. [Do Checkpoint 1.5] (c) p = 4 3 5 , and q = 4 1 5
1 7
Rules for dot products
Let us formally define the dot product in n dimensions: (c) Find any
" vector
# that is perpendicu-
5
DEFINITION: DOT PRODUCT lar to .
2
n
X
Dot product: a·b= a j bj
j=1
a and b are both in n-dimensions

Check for yourself that the following rules are correct:

RULES: DOT PRODUCT

Commutativity: a·b=b·a
Associativity with scalar multiplication: a · ( b) = (a ) · b
Distributivity over addition: a · (b + c) = a · b + a · c
 15

Example 1.3

Vector a has length a, and vector b has twice the length of

a, p
but they point in di↵erent directions. The length of a + b
is 8a. Find the angle between a and b.
Solution:
We can write kak = a and kbk = 2a.
The length squared of a + b is

(a + b) · (a + b) = a · a + a · b + b · a + b · b
= a2 + 2ab cos ✓ + b2
= a2 + 2a(2a) cos ✓ + (2a)2
= 5a2 + 4a2 cos ✓
p
But it is given that the length squared of a + b, is ( 8a)2 ,
therefore
5a2 + 4a2 cos ✓ = 8a2
and then
3
cos ✓ = , i.e. ✓ = 41.1 .
4

1.6 Points and vectors

The position of any point with respect to a predefined co-

ordinate system, may be denoted by a position vector point-
ing from the origin of the coordinate system to the point.
The position vectors of two points A and B with coordinates
A(xA , yA ) and B(xB , yB ) are given by rA = xA i + yA j and
rB = xB i + yB j, respectively. In the more general case, a B
z r AB
vector may be directed from point A to point B designated
y
by rAB . From the head-to-tail vector addition, A rB
rB = rA + rAB . rA

Solving for rAB and expressing rA and rB as Cartesian vec- x

tors yield

rAB = rB rA = (xB i + yB j) (xA i + yA j) Figure 1.11. The coordinates of points A and

B can be used to find the vector rAB , directed
or, more conveniently, from A to B.

rAB = (xB xA )i + (yB yA )j). (1.12)

 16

The components of the position vector, rAB from A to B

may be formed by taking the coordinates of the tail of the
vector A(xA , yA ) and subtracting them from the correspond-
The position vectors of points A and B
ing coordinates of the head of the vector B(xB , yB ). This
can also be denoted by rOA and rOB , but
applies equally to vectors in 3D.
the O is often omitted when the starting
point of the vector is at the origin.
1.7 Directions in 3D

For vectors in 2D, the direction is given by a single angle. As

mentioned before, a single angle is not sufficient for describ-
ing the direction of a vector in 3D. Although the direction
of a in 3D may be given by two angles, it is often more con-
venient to describe the direction of a vector in 3D by using
a normalized vector that points in the same direction as a.
Such a normalized vector is called a unit vector.
Arbitrary unit vectors
We have used the term unit vectors to refer to the vectors
i, j and k. However, the term unit vector is also is used to
describe any vector whose length is one.
In order to specify a certain direction in 3D, the unit vector
pointing in the desired direction is the best way to capture
the direction. Let a be a vector in 3D. We may use a and Checkpoint 1.6:
construct a unit vector, denoted by a b, pointing in the same 2 3 2 3
1 1
direction as a, as follows 6 7 6 7
Let a = 4 2 5 and b = 4 1 5.
a 2 1
b=
a . (1.13)
a
b
b and b.
1. Construct a
The vector a b points in the same direction as a, but is is
scaled so that its length is 1. Scaling a vector so that its 2. Construct a vector c that points
length is 1, is called normalizing the vector. In these notes in the same direction as a but has
the convention will be used to denote normalized vectors by length 15.
the same symbol as the vector from which it was calculated,
but with a hat over the symbol. [Do Checkpoint 1.6]
Other vectors with the same direction as a but possibly dif-
ferent lengths may now be expressed easily by using a b. If b
is a vector of length µ pointing in the direction of a, then
b = µab.
Direction cosines:
The three components of the unit vector â are called the
direction cosines of the vector a. Let us briefly explain where
this naming convention comes from.
Consider the unit vector n (whose length is 1) shown in
Figure 1.12. Its three components are nx , ny and nz , and
 17

the angles between the vector and the primary axes x, y,

and z are respectively given by ↵, , and .

}
By dropping a perpendicular line segment from the tip of n
n to the x-axis, we can complete the right angled triangle
shown in light blue. Its hypotenuse has length 1, and the
length of the side along the x-axis is nx . The angle between nz
these two edges is ↵. It is then obvious that cos ↵ = nx . A

}
similar relationship concerning the angle between the vector
and the primary axes can be derived. We therefore have

}
x ny
that
nx = cos ↵
ny = cos

}
nz = cos . 1

}
1.8 Forces in 3D

}
ny
Up to now we have used vectors to describe positions in nx

} }
space. Another important use of vectors is to describe forces.
This section is included in order to illustrate how 3D unit
vectors are used in real applications.
nz

}
At this stage we shall treat a force as an abstract concept.
It “pulls” with a certain strength in a certain direction. It
is invisible, but its e↵ects are not. It is measured in newtons
with unit N.
In some applications of statics, cables or strings are used to Figure 1.12. Explanation of the direction
apply forces to an object to keep it in place. Think for ex- cosines associated with a unit vector n.
ample of a telecommunications tower anchored to the ground
with cables. A fundamental property of a cable is that the
force in the cable always acts along the cable. The spatial
vector along the cable therefore supplies its direction, but its
magnitude is obtained in a di↵erent way. We shall explain
this concept with some examples.
 18

Example 1.4

Figure 1.13 shows a flat horizontal plank held in place by

two cables. The tension in cable OA is 100 N, and in cable
z
OB is 150 N. Find the vector form of the combined force on O
the peg at O as well as its direction.
x
Solution: aa 3m y
Denote the position vectors from O to A by a and from O bb
to B by b. Using the dimensions given in the figure,
2 3 2 3
4 4
6 7 6 7
a=4 0 5 and b=4 2 5 A
3 3 B
4m
4m
2m
Then the directions are respectively
2 3 2 3
4 4 Figure 1.13. The setup for Example 1.4.
16 7 1 6 7
â = 4 0 5 and b̂ = p 4 2 5
5 29
3 3
 19

Let FA be the force in cable OA and FB be the force in

cable OB. Then
2 3 2 3
4 80
16 7 6 7
FA = 100 ⇥ 4 0 5=4 0 5
5
3 60

and 2 3 2 3
4 111.42
1 6 7 6 7
FB = 150 ⇥ p 4 2 5 = 4 55.71 5
29 3 83.56
Let the combined force be F, then
2 3 2 3 2 3
80 111.42 191.42
6 7 6 7 6 7
F = FA + FB = 4 0 5 + 4 55.71 5 = 4 55.77 5
60 83.56 143.56

The magnitude of F is
p
F = 191.422 + 55.772 + 143.562 = 245.69 N,

and the direction of F is

2 3 2 3
191.42 0.78
1 6 7 6 7
4 55.77 5 = 4 0.23 5 .
245.69
143.56 0.58

1.9 The cross product

The dot product of vectors was introduced in section 1.5.

We shall now introduce another product between vectors
that is very useful. It is called the cross product, and it pro-
duces a vector as result (unlike the dot product that pro-
duces a scalar). It is therefore also sometimes called the
vector product.
Definition of the cross product:
The cross product is defined only for vectors with three com-
ponents. (You may recall that the dot product is defined for
vectors with any number of components). Consider the vec-
tors a and b below.
2 3 2 3
a1 b1
6 7 6 7
a = 4 a2 5 , b = 4 b2 5 .
a3 b3
 20

The cross product is defined as follows:

2 3
a 2 b3 a 3 b2
6 7
c = a ⇥ b = 4 a 3 b1 a 1 b3 5 .
a 1 b2 a 2 b1

Orthogonality of the cross product:

The vector c = a⇥b has a unique (and very useful) property:
c is perpendicular to both a and b. One may also view it
as follows: c is perpendicular to the plane in which both a
and b lie.
This is not difficult to verify: Just confirm that the dot
products a · c and b · c are both zero.

a · c = a1 c 1 + a2 c 2 + a3 c 3
= a1 (a2 b3 a3 b2 ) + a2 (a3 b1 a1 b3 ) + a3 (a1 b2 a 2 b1 )
=0
and

b · c = b1 c 1 + b2 c 2 + b3 c 3
= b1 (a2 b3 a3 b2 ) + b2 (a3 b1 a1 b3 ) + b3 (a1 b2 a 2 b1 )
=0 axb

In space we should therefore view c as a vector that is normal

to the plane defined by a and b. Figure 1.14 shows the
b
configuration of a, b and a ⇥ b in space.
The vector c = a ⇥ b is not the only vector perpendicular a
to the plane of a and b — in fact, any multiple of c, includ-
ing negative multiples, are also perpendicular to this plane.
Figure 1.14.
There are now two questions that arise: (1) in which direc-
tion must c point, up or down, and (2) what is the length
of c?
The right hand rule: c
The answer to the first question is given by the so called
right hand rule. It states that if you curl the fingers of your
c
right hand from a (the first factor in the product) to b (the
second factor in the product), then your thumb points in the b
direction of c. This is illustrated in Figure 1.15
This immediately implies a new caution that must be taken: a
the cross product is not commutative, i.e. the order in which
we take a ⇥ b or b ⇥ a matters. In fact it is easy to show
that
Figure 1.15.
b ⇥ a = a ⇥ b.

The length of a ⇥ b:
 21

It will be shown later that, if the angle between a and b is

✓ then the length of a ⇥ b is as follows:

ka ⇥ bk = ab sin ✓.

The presence of the sin ✓ factor in this result ought to make

sense. If we consider colinear vectors, i.e. if the angle
between a and b is zero, then such vectors do not define
a plane, and there cannot be a unique vector perpendicular
to both a and b. The cross product for two colinear vec-
tors therefore produces the zero vector, regardless of their
lengths.
Calculating the cross product (column method):
It may be difficult to remember the specific components of
the cross product – but there is a cyclic pattern to it. Here
is a quick way to do the cross product: Write down a ⇥ b in
column form:
2 3 2 3 2 3
a1 b1
6 7 6 7 6 7
a ⇥ b = 4 a 2 5 ⇥ 4 b2 5 = 4 5
a3 b3

Cover the first row in the cross product, so that only the
next two rows are visible. Then do a ’cross-over product’
as follows: multiply a2 with b3 and a3 with b2 , and then
subtract them. Note that the first element to pick (i.e. a2 ),
is the one in the leftmost column just below the cover.

Then move the cover down so that it covers the second row.
Start with the element immediately below the cover on the
left (i.e. a3 ) and do the ’cross-over product’, i.e. a3 b1 a1 b3 .

Then move the cover down again so that it covers the third
row. There is no element on the left just below the cover,
but if one follows the cyclic nature of the cross product, this
element is the first element on the left, i.e. a1 . Once again
do the ’cross-over product’, i.e. a1 b2 a2 b1 .
 22

Calculation of the cross product (determinant method):

Most handbooks recommend a di↵erent method for calculat-
ing the cross product. It, however, requires that you already
know how to calculate the determinant of a 3 ⇥ 3 matrix. It
is as follows: Set up a ‘matrix’ containing i, j and k in the
first row, write a (as a row) in the second row, and b in the
third row,
2 3
i j k
6 7
4 a1 a2 a3 5 .
b1 b 2 b 3

(We shall require of you to ignore the inconsistency of a

‘matrix’ containing both vectors and scalars as elements –
consider it to be just a convenient method for establishing Checkpoint 1.7:
a recipe for the cross product.) Calculate the following:

The cross product between a and b then is the ‘determinant’ (a)

of this matrix: 2 3 2 3
1 4
6 7 6 7
i j k 4 2 5⇥4 1 5
a1 a2 a3 = (a2 b3 a3 b2 )i (a1 b3 a3 b1 )j+(a1 b2 a2 b1 )k. 3 5
b1 b2 b3
(b)
[Do Checkpoint 1.7]
2 3 2 3
a 2a
6 7 6 7
4 b 5⇥4 b 5
1.10 Rules for the cross product c c

2 3 2 3 (c)
a1 b1
6 7 6 7
a = 4 a2 5 , b = 4 b2 5 . i⇥j
a3 b3
Ans: (a) [7, -17, 9], (b) [0, ac, ab], (c) k
The cross product is defined as follows:
2 3
a 2 b3 a 3 b2
6 7
c = a ⇥ b = 4 a 3 b1 a 1 b3 5
a 1 b2 a 2 b1

Rules for the cross product

Let us formally define the cross product in 3 dimensions:
 23

DEFINITION: CROSS PRODUCT

2 3 2 3 2 3
a1 b1 a 2 b3 a 3 b2
6 7 6 7 6 7
(DEF) Cross product: 4 a 2 5 ⇥ 4 b2 5 = 4 a 3 b1 a 1 b3 5
a3 b3 a 1 b2 a 2 b1
a and b are both in 3-dimensions only
Check for yourself that the following rules are correct:

RULES: CROSS PRODUCT

(RULE 1) Non-commutativity: a⇥b= b⇥a
(RULE 2) Associativity with scalar multiplication: a ⇥ ( b) = (a ) ⇥ b
(RULE 3) Distributivity over addition: a ⇥ (b + c) = a ⇥ b + a ⇥ c
(RULE 4) Relationship to ✓ (angle between a and b): ka ⇥ bk = ab sin ✓
(RULE 5) Vector triple product: a ⇥ (b ⇥ c) = (a · c)b (a · b)c

1.11 The cross product with Cartesian

unit vectors

Using the definition of a cross product it is easy to show the

following cyclic pattern:
i ⇥ j = k, j ⇥ k = i, k ⇥ i = j.

Of course
i ⇥ i = 0, j ⇥ j = 0, k ⇥ k = 0,
since the cross product between two colinear vectors is the
zero vector.
Let us summarize:

Cross products of i, j and k:

i⇥j =k j⇥i = k i⇥i =0

j⇥k =i k⇥j = i j⇥j =0
k⇥i =j i⇥k = j k⇥k =0

This result may also be used together with RULE 3 to cal-

culate the cross product.

1.12 Derivation of the cross product

Where did the formula for the cross product come from? The
following section shows one way in which it can be derived
 24

from first principles.

Consider 2 3 2 3
a1 b1
6 7 6 7
a = 4 a2 5 , b = 4 b2 5 .
a3 b3

Orthogonality:
We shall first derive the formula for the cross product by
assuming that (1) both a and b are perpendicular to c, and
(2) the formula must not contain fractions of the elements
of a and b.
Let us write c = a ⇥ b as
2 3
c1
6 7
c = 4 c2 5 .
c3

Note that we consider the components as a and b as given

and known, but we consider the c1 , c2 and c3 as unknown.
If c is perpendicular to both a and b then

a · c = 0, and b · c = 0.

Let us write out these two equations:

a1 c1 + a2 c2 + a3 c3 = 0, .................... (1)
b1 c1 + b2 c2 + b3 c3 = 0, .................... (2)

This is a system of two equations in three unknowns, c1 ,

c2 and c3 . Let us start by eliminating c3 from (1) and (2).
Multiply (1) by b3 and multiply (2) by a3 , and then subtract
the resulting equations:

b3 a1 c1 + b3 a2 c2 + b3 a3 c3 = 0, .................... (10 )
a3 b1 c1 + a3 b2 c2 + a3 b3 c3 = 0, .................... (20 )

(20 ) (10 ) : a3 b1 c1 + a3 b2 c2 b3 a 1 c 1 b3 a 2 c 2 = 0
or
a 3 b1 a 1 b3
c2 = c1
a 2 b3 a 3 b2
One may choose c1 to be any arbitrary nonzero number,
however, if you do not want fractions in the formula, it is
best to choose
c 1 = a 2 b3 a 3 b2
and then
c 2 = a 3 b1 a 1 b3 .
 25

Substitute these expressions for c1 and c2 back into equation

(1) and solve for c3 :

a1 (a2 b3 a3 b2 ) + a2 (a3 b1 a1 b3 ) + a3 c3 = 0,

or
c 3 = a 1 b2 a 2 b1 .
To summarize:
2 3
a 2 b3 a 3 b2
6 7
c = a ⇥ b = 4 a 3 b1 a 1 b3 5 .
a 1 b2 a 2 b1

This provides a formula for a vector that is perpendicular to

both a and b. We still do not know whether it satisfies the
right hand rule and we do not know what its length is.
The length of a ⇥ b:
Consider kck2 :

kck2 = (a2 b3 a3 b2 )2 + (a3 b1 a1 b3 )2 + (a1 b2 a 2 b1 ) 2

= a22 b23 2a2 a3 b2 b3 + a23 b22
+a23 b21 2a1 a3 b1 b3 + a21 b23
+a21 b22 2a1 a2 b1 b2 + a22 b21
(1.14)
Collecting terms gives

kck2 = a21 (b22 +b23 )+a22 (b21 +b23 )+a23 (b21 +b22 ) 2(a2 a3 b2 b3 +a1 a3 b1 b3 +a1 a2 b1 b2 )

The last term may not be well known, but it may be worth
noting that it also appears in the square of the dot product:

(a · b)2 = (a1 b1 + a2 b2 + a3 b3 )2
= a21 b21 + a22 b22 + a23 b23 + 2(a2 a3 b2 b3 + a1 a3 b1 b3 + a1 a2 b1 b2 )

therefore

2(a2 a3 b2 b3 +a1 a3 b1 b3 +a1 a2 b1 b2 ) = a21 b21 +a22 b22 +a23 b23 (a·b)2 .

Substituting this expression into (1.14) gives

kck2 = a21 (b22 + b23 ) + a22 (b21 + b23 ) + a23 (b21 + b22 ) + a21 b21 + a22 b22 + a23 b23 (a · b)2
= a21 (b21 + b22 + b23 ) + a22 (b21 + b22 + b23 ) + a23 (b21 + b22 + b23 ) (a · b)2
= a2 b2 (a · b)2
= a2 b2 (ab cos ✓)2
= a2 b2 (1 cos2 ✓)
= a2 b2 sin2 ✓

and therefore
kck = ab sin ✓.
 26

This result is often also stated as follows:

a ⇥ b = ab sin ✓ ĉ

where ĉ is a unit vector perpendicular to both a and b.

Example 1.5

Consider the following two vectors. Find the angle between

them twice: (a) by using the dot product, and (b) by using
the cross product.
2 3 2 3
1 4
6 7 6 7
a4 2 5, b = 4 2 5.
2 4

SOLUTION:
p p
a= 12 + 22 + 22 = 3, b= 42 + 22 + 42 = 6,

(a)
2 3 2 3
1 4
6 7 6 7
4 2 5 · 4 2 5 = 4 + 4 + 8 = 16 = 3 ⇥ 6 ⇥ cos(✓)
2 4

therefore
16
✓ = arccos( ) = 27.27 .
18
(b) 2 3 2 3 2 3
1 4 4
6 7 6 7 6 7
c=4 2 5⇥4 2 5=4 4 5
2 4 6
and q p
kck = 42 + 42 + ( 6)2 = 68
then p
68 = 3 ⇥ 6 ⇥ sin ✓
and therefore
p
68
✓ = arcsin( ) = 27.27 .
18
Chapter 2

Lines and Planes

2.1 Introduction

In this chapter we discuss how to find di↵erent types of equa-

tions for straight lines and planes in 3D since, for example,
the well known expression y = mx + c that represents a
straight line in the xy plane is not appropriate for three di-
mensions. It will be shown how equations appropriate for
three dimensions can be developed using vectors.

2.2 Lines

Vector equation:
Similar to the situation for 2D, only one line can pass through
any two distinct points in 3D. Consider two distinct points,
A(xA , yA , zA ) and B(xB , yB , zB ), and let point P (x, y, z)
represent any point on the line through A and B. To determ-
ine the equation for the line through A and B, it is necessary
to find the position vector, r, of point P (x, y, z) on the line
since by varying the coordinates (x, y, z) all points on the
line will be traced out by the end point of the position vec-
tor r. The position vector between the two known points
on the line is, a = rB rA . Another vector that is also
on the line and includes the general point is r rB . Here
r rA could also have been selected. Vectors a and r rB
are parallel to each other and, therefore, they only di↵er by
a scalar multiple such that

r rB = ta

27
 28

where t is the scalar multiple. The equation can be restated

as
r = rB + ta (2.1)
which is the vector equation of the line through points A
and B. Vector a is referred to as the direction vector and r-rB P
z B
vector rB is the anchor vector. The vector equation of a a r y
A rB
line is not unique since r = rA + ta, r = rA + t( a) and
r = rA + t(k(a)) with k a non-zero scalar, are all also valid rA

equations for the line.

x
Parametric equations:
Equation (2.1) can be expressed in terms of Cartesian co- Figure 2.1. Definition sketch for a line through
ordinates as points A and B.

xi + yj + zj = (xB + tax )i + (yB + tay )j + (zB + taz )k

and by equating components we obtain the parametric equa-

tions of the line through points A(xA , yA , zA ) and B(xB , yB , zB ),

x = xB + ax t, y = yB + ay t, z = zB + az t. (2.2)

The variable t is a parameter and as t increases from 1 to

1 the point P (x, y, z) traces out the entire line. If the para-
meter t is restricted to a closed interval then P traces out
a line segment that corresponds to the range of t. Another
point to note is that the direction vector, a, does not have to
be on the line, but only parallel to it since a only indicates
the direction in 3D (which is similar to the gradient of the
line in 2D).
Symmetric equations:
The parameter, t, is the same in all the parametric equa-
tions (2.2). The symmetric equations for the line under con-
sideration are obtained by solving for the parameter in the
di↵erent equations and by equating,
x xB y yB z zB
t= = =
ax ay az
provided that the three numbers ax , ay and az are non-zero.
It is customary to write the symmetric equations as
x xB y yB z zB
= = . (2.3)
ax ay az
If any of the numbers ax , ay or az is zero in (2.3), then we
use the remaining two equations to eliminate the parameter
t. For example, if ax = 0, ay 6= 0 and az 6= 0, then
y yB z zB
x = xB , =
ay az
 29

are the symmetric equations for the line.

Example 2.1

Find the vector equation for the line through points

A(3, 1, 2) and B( 2, 7, 4).
Then find the parametric equations and the symmetric equa-
tion for the line.

Vector equation:
As direction vector choose the vector from A to B,

a = ( 2 3)i + (7 1)j + ( 4 ( 2))k

= 5i + 6j 2k.

If the position vector of point B is selected as the anchor

vector, then the vector equation of the line is given by

r = rB + ta
= 2i + 7j 4k 5ti + 6tj 2tk

which can be simplified to

r=( 2 5t)i + (7 + 6t)j + ( 4 2t)k. (2.4)

Parametric and symmetric equations:

The parametric equations of the line are obtained by writing
out each component from (2.4) and noting that r is the
position vector of a general point on the line,

x= 2 5t y = 7 + 6t z= 4 2t.

The symmetric equations are obtained by eliminating the

parameter t,
x+2 y 7 z+4
= = .
5 6 2

2.3 Planes

Vector equation:
Vectors can also be used to determine equations of planes in
3D. An infinite number of planes can pass through a point
in space. However, the vector that is perpendicular to a par-
 30

ticular plane uniquely defines the orientation of the plane.

Therefore, if a point, A(xA , yA , zA ) and a vector n that is
normal to the plane through A(xA , yA , zA ) are given, then
there is only one plane that satisfies these conditions. If
P (x, y, z) represents any point on the plane and r is the posi-
tion vector of point P (x, y, z), then the vector r rA is in the
plane where rA is the position vector of point A(xA , yA , zA )
and both vectors have their end points on the plane. If it
is given that n is a normal vector on the plane and since n
and r rA are perpendicular, n

n · (r rA ) = 0 (2.5) A r-rA

is the vector equation of the plane. The vector r represents P

all points on the plane.
Cartesian equation:
Figure 2.2. Vector n is perpendicular to the
If the normal vector is specified by n = ai + bj + ck, then by
plane through the fixed point A(xA , yA , zA ). Point
expanding the dot product, (2.5) can be expressed as
P (x, y, z) represents all points on the plane.

a(x xA ) + b(y yA ) + c(z zA ) = 0 (2.6)

which is the Cartesian equation of the plane containing point

A(xA , yA , zA ) with normal vector n. By simplifying, (2.6)
can be written as

ax + by + cz + d = 0

where d = axA byA czA is a known scalar since all the

Checkpoint 2.1:
values are known. Any linear equation of the form ax + by +
cz + d = 0 with a, b, c and d not all zero, represents a plane.
Find a vector that is normal to the plane
A plane is also determined by three points that are not colin- x + 2y + 3z 13 = 0.
ear, which means the three points do not lie in a straight line.
To obtain an equation of such a plane plane, it is necessary
to form two vectors between two pairs of points. These two
vectors lie in the plane and are known. The cross product of
these two vectors is a vector normal to the plane and can be
used as the normal vector, n, in (2.5). The position vector
of any of the points can be used for rA and (2.5) can be used
to find the vector equation of the plane through the three
noncolinear points. This is illustrated in the next example.
 31

(rB -rA ) x (rC-rA )

Example 2.2

A C
rC -rA
Find the equation of the plane determined by the points r-rA rB-rA B
A(4, 3, 1), B(6, 4, 7) and C(1, 2, 2). P

Construct two vectors in the plane by finding the position

vectors between two groups of point, Figure 2.3. Vectors rB rA and rC rA are
in the plane and their cross product, vector n, is
rAB = 2i j + 6k and rAC = 3i + 5j + k.
perpendicular to the plane. Vector r rA is in
The cross product of these two vectors yields a vector per- the plane and normal to n. Point P (x, y, z), with
pendicular to the plane position vector r, represents a general point on
the plane.
n = rAB ⇥ rAC = 31i 20j + 7k.

If point A, with the position vector rA = 4i 3j + k, is

selected as the known point on the plane, then the equation
of the plane follows from (2.5),

n · (r rA ) = 0

31(x 4) 20(y + 3) + 7(z 1) = 0

and the equation of the plane is

31x 20y + 7z + 57 = 0.
Chapter 3

Kinematics with vectors

3.1 Mechanics: an overview

Many students may already be familiar with some aspects

of mechanics, while others may not have encountered any
mechanics at all. This section is then included either as a
review or as a short introduction to the topic.
Mechanics is the mathematical formulation of motion, de-
formation and forces. The laws of classiscal mechanics were
developed during the second half of the seventeenth century,
most notably by Sir Isaac Newton. His famous Philosophiae
Naturalis Principia Mathematica (Mathematical principles
of natural philosophy) is a classical masterpiece where the
principles of motion and its causes have been formulated sys-
tematically. It was published in five volumes of which the
first appeared in 1687. It may be considered as an excellent
application where mathematics was used to model natural
phenomena.
The broad subject of mechanics has since then undergone
various refinements (continuum mechanics, statistical mech-
anics, quantum mechanics, relativistic mechanics). How-
ever, for physicists, applied mathematicians and engineers
knowledge of basic mechanics is still very important and it
is a prerequisite for understanding other theories and more
advanced applications of mechanics.
Mechanics is divided into three categories:

— Kinematics, the geometrical description of motion.

— Dynamics, the study of the influence of forces on free

32
 33

bodies.

— Statics, the study of the influence of of forces on fixed

bodies.

In these notes we shall mainly be concerned with kinematics

and dynamics.
In mechanics ideal objects moving under ideal circumstances
are assumed. The bodies as well as the conditions under
which they move are represented by idealized models. Some
examples will serve to highlight this idea.

— A very small object is represented by a mathematical

point, called a particle.

— A long thin rod carrying load much larger than its own
weight, is represented as a massless line segment. It is
called a light rod.

— A plane on which a particle may move freely, is repres-

ented as a perfect plane with no friction. It is called a
smooth plane.

— A string connecting two bodies, where the mass of the

string is significantly less than the masses of the two
bodies, is represented as a massless line segment join-
ing the two bodies. It is called a light string.

In order to describe the position, orientation, shape, mass

and other properties of a body mathematically, it is neces-
sary to associate a number or numbers to each of these prop-
erties. Viewed di↵erently, each property must be measured.
These properties are called quantities and they are measured
in units. For example, time is a quantity and it is measured
in seconds (or hours, etc.). Mass is a quantity and it is
measures in kilograms (or pounds, for example).
In developing elementary mechanics, we are not concerned
with how these quantities are measured. It is implicitly as-
sumed that these quantities can be measured accurately at
any moment of time. Mechanics only provides the mathem-
atics (i.e. the formulae) by which changes in these quantities
can be described. It therefore forecasts the position, orienta-
tion or form of a body, given the initial position, orientation
or form of the body.
 34

3.2 Position, velocity and acceleration

Position
Consider a particle that is free to move in a straight line.
It is convenient to associate a number line to this physical
line. The x-coordinate where the particle is located is called
the position of the particle. Note that x may be positive or
negative, depending on whether the particle is to the right or
to the left of the origin. The di↵erence between two positions
at two di↵erent times is called the displacement.
As the particle moves along the x-axis, x becomes a function
of time. We shall denote the time by t, and we shall write
the position as
x(t).
The time dependence of x(t) is always assumed, although
we sometimes suppress the dependence on t and write only
x.
Velocity
Velocity is an indication of how fast or slowly the particle
moves. Intuitively we feel that a particle moves faster if it
covers a larger distance in the same time span. Alternat-
ively, we may also feel that it moves faster if it covers the
same distance in a shorter time span. This idea forces us
to consider velocity as distance traveled divided by the time
taken.
If the particle is at position x0 at time t0 , and it is at position
x1 at time t1 , then the particle has undergone a displacement
of x1 x0 in a time span of t1 t0 . The velocity v is then
given by
x
v=
t
where x = x1 x0 and t = t1 t0 . It ought to be clear
that v is positive if the particle moves to the right, and v is
negative if the particle moves to the left.
If the particle moves with constant velocity, then the same
value of v will be obtained independent of the size of t.
However, if the velocity changes, then t ought to be taken
as small as possible in order to obtain an accurate value
of the velocity at time t0 . The instantaneous velocity is
therefore the derivative of the position with respect to time:
dx
v(t) = . (3.1)
dt

Expressed in words, the velocity is defined as:

 35

Velocity is the rate at which position increases.

In what follows, the word ‘velocity’ will always refer to the

instantaneous velocity. It is also customary to write ẋ for
velocity.
The overdot notation was introduced by Isaac Newton and
it always denotes a derivative with respect to time. A double
dot denotes the second derivative with respect to time, for
example
d2 x
ẍ = 2 .
dt
If the position is known as a function of time, then the velo-
city is obtained by di↵erentiating it. If the velocity is known
as a function of time, then the position is obtained by integ-
rating it. Z Z
dx = v(t)dt
Z
x(t) = v(t)dt

The integration constant that is required is obtained from

the initial position (that is the position at time t = 0.) If
the velocity is constant, then we find that
Z
x(t) = vdt

= vt + K

By substituting x(0) = x0 into this expression, the following

is obtained:
x(t) = vt + x0 . (3.2)
(The student may also recognize the following notation which
is often used in school syllabuses: x(t) is written as s and
x0 is taken as zero, then the equation becomes s = vt.)
Acceleration
If the velocity also changes, then it is sensible to measure
how fast it changes. The concept of instantaneous accelera-
tion is then introduced and we define:

Acceleration is the rate at which velocity increases.

The symbol a is used to denote acceleration. In symbols the

definition is
dv
a(t) = . (3.3)
dt
Using the overdot notation for time derivatives, the acceler-
ation can be expressed as v̇ or ẍ.
 36

If the acceleration is known, then the velocity is obtained by

integrating, and the integration constant is obtained from
the initial velocity.
Standard units and symbols:
The following table gives the variables used in the formulae
and it also gives the standard SI units in which each quantity
is measured:
Quantity Symbol Measurement unit
time t second (s)
position x meter (m)
velocity v m/s
acceleration a m/s2
The initial time, position or velocity is denoted by the cor-
responding symbol with a zero subscript (for example t0 , x0 ,
v0 ).
When doing a problem in mechanics, it is advisable to con-
vert all quantities to the same unites of time and space, e.g.
all velocities in km/h, all lengths in km and all times in h.
Especially when presented with mixed units, e.g. t1 = 1 12
hours and t2 = 45 minutes. Convert this to t1 = 1.5h and
t2 = 0.75h. (Or alternatively t1 = 5400s and t2 = 2700s.)
The kinematic formulae are derived by integration and fur-
ther manipulation from only two basic definitions:
Quantity Definition
dx
velocity v=
dt
dv
acceleration a=
dt

If the acceleration is not constant, then the kinematic equa-

tions are still obtained by integration of these definitions,
however, the formulae derived are somewhat more involved.
The derivative of acceleration:
We do not need to stop with a definition of acceleration, but
we may continue to define the rate of increase of acceleration
as
da
dt
and we may give it a name and denote it by some sym-
bol. However, experimental evidence from nature indicates
that the interaction of bodies on each other is described ad-
equately in terms of acceleration and any further definitions
 37

seem to be superfluous. We shall not give it a name nor

assign a symbol to it.

3.3 Rectilinear motion at constant ac-

celeration

Equations for constant acceleration

If we have constant acceleration, i.e. a is not a function of
time, then we can integrate as follows
Z Z
dv = a dt

v(t) = at + K.

Let v(0) = v0 then the latter equation becomes

v(t) = at + v0 or v = u + at (3.4)

(On the right hand side we have rewritten the equation in

’school notation’, where v0 is replaced by u. The ‘school
notation’ also uses s in the place of x x0 .)
By substituting v = dx/dt and by di↵erentiating a second
time, the position may be found as a function of time
Z Z
dx = (at + v0 )dt
1 2
x(t) = 2 at + v0 t + K

Taking x(0) = x0 this equation becomes

x(t) = 12 at2 + v0 t + x0 , or s = ut + 12 at2 . (3.5)

By eliminating certain variables in (3.4) and (3.5), other use-

ful formulae may be obtained. For example, by eliminating
t we obtain

v 2 = v02 + 2a(x x0 ), or v 2 = u2 + 2as. (3.6)

By eliminating a we obtain

x = x0 + 12 (v0 + v) or s = 12 (u + v)t. (3.7)

These equations are called kinematic equations for constant

acceleration.
 38

Example 3.1

A particle moves from rest at constant acceleration in a

straight line and during the fifth second of its motion, it
covers 45m. Calculate

(a) the acceleration,

(b) the time needed to cover 245m,

(c) the velocity of the particle at that moment.

Solution:
(a) Let the particle cover a distance of s1 from the position
of rest up to the beginning of the fifth second of motion.
(That will take four seconds.) Let s2 be the distance covered
from the position of rest up to the end of the fifth second of
motion. (That will take five seconds.) Using s = ut + 12 at2 ,
we find
1
s1 = a(4)2 .......................(1)
2
en
1
s2 = a(5)2 .......................(2)
2
But s2 s1 = 45, therefore we can subtract equations (1)
from equation (2),

s2 s1 = 45 = 12.5a 8a

and then a = 10m/s2 .

(b) Using s = ut + 12 at2 , we find

245 = 0 + 12 (10)t2

and therefore the time to cover 245 m, is 49 s.

(c) Using v 2 = u2 + 2as, we have

v 2 = 02 + 2(10)(245)

and therefore the velocity is 70 m/s.

 39

Gravitational acceleration:
We shall use the theory of motion at constant acceleration
to model the motion of free falling bodies. Objects falling
freely near the surface of the earth experience a constant
acceleration of about 9.81 m/s2 . This acceleration is called
gravitational acceleration and is usually denoted by the sym-
bol g.
The value of g = 9.81m/s2 is approximate. It changes
slightly from place to place and it decreases as one goes fur-
ther away from the centre of the earth (i.e. going higher).
The four equations (3.4), (3.5), (3.6) and (3.7) can be ap-
plied to solve problems for a body that falls straight down or
is thrown straight upwards, or that moves horizontally with
no friction. We shall illustrate this with another example.
 40

Example 3.2

A particle is shot upwards at time t = 0. On the upward

track it reaches a height h at time t = 4s and on the down-
ward track it reaches the same height h at time t = 7s. Take
g = 9.81 m/s2 , and calculate

(a) the launch velocity of the particle,

(b) the height h,

(c) the maximum height attained.

Solution:
(a) and (b) Let s be the displacement of the particle meas-
ured from the ground (i.e. form where it was launched). By
using s = ut + 12 at2 we can write down two expression for
the height h:

h = u(4) + 12 ( 9.81)42 = 4u 80 ...................... (1)

h = u(7) + 12 ( 9.81)72 = 7u 245 ....................... (2)

Simultaneous solution of u and h from (1) and (2) gives:
u = 53.96m/s and h = 137.34m.
(c) When the particle reaches the highest point, its velocity
is momentarily zero, i.e. v = 0. Using v 2 = u2 + 2as we find
that

0 = 53.962 + 2( 9.81)s, or s = 148.4m.

3.4 Integration techniques

Constant acceleration is easy to integrate. However, in many

application the acceleration is a function of time, for example
in a car where the driver controls the acceleration by means
of the accelerator pedal. The acceleration may also be a
function of position. For example, a particle at the end
of a spring experiences such acceleration — the further the
spring is extended, the greater its acceleration. It is then
necessary to integrate a function of t or a function of x.
Expressions for a:
 41

The acceleration in terms of t is

dv
a=
dt
which means that if a is a function of t, then v can be ob-
tained by this form.
An alternative expression for a in terms of x may be derived
as follows:
By the chain rule, we may write
dv dv dx
a= = ,
dt dx dt
dx
but = v and therefore the acceleration may also be ex-
dt
pressed as
dv
a=v . (3.8)
dx
Separation of variables:
In the models we are going to explore, the acceleration my be
given as a function of time. In other cases the acceleration
may be supplied as a function of position or as a function of
velocity. We therefore have equations of the form

(I) a = f (t), or
(II) a = f (x), or
(III) a = f (v).
An equation where a is expressed in terms of other variables
is called an equation of motion.
dv
In case (I), one must replace a with , so that an equation
dt
is obtained that depends only on v and t:
dv
= f (t). (3.9)
dt
dv
In case (II), one must replace a with v , so that an equa-
dx
tion is obtained that depends only on x and v:
dv
v = f (x). (3.10)
dx
dv
In case (III), one must replace a with , so that an equation
dt
is obtained that depends only on x and v:

dv
= f (v). (3.11)
dt
 42

In all three cases an ordinary di↵erential equation is ob-

tained depending on only two variables, and this equation
can be solved with a technique known as separation of the
variables, followed by integration. The integration constants
that emerges after integration, obtain their values by sub-
stitution of the initial conditions.
After integration of (3.9), (3.10) or (3.11), v is obtained as
a function of either x or as a function of t:

(IV ) v = g(t), or
(V ) v = g(x).

dx
In both cases we may express the velocity as v = , and
dt
then separation of the variables may be applied again to ob-
tain x as a function of t. Afterwards other useful kinematic
equations may be obtained by eliminating some variables
from some equations.
A number of examples will be given now.
 43

Example 3.3

If
t
a=e
find v as a function of t and integrate again to find x as a
function of t. Use v0 and x0 as initial position and velocity
respectively.
SOLUTION:
dv
Use a = and separate variables:
dt
dv t
=e
dt
Z Z
dv = e t dt.

Integration gives
t
v= e + K.
With initial condition v = v0 when t = 0, we have K =
v0 + 1, and the equation is
t
v = v0 + 1 e

This equation may be integrated further to yield x as a func-

tion of the time:
dx t
= v0 + 1 e
dt
Z Z
dx = (v0 + 1 e t )dt.

After integration and substitution of x = x0 when t = 0, we

obtain
x = x0 + (v0 + 1)t + e t 1.
 44

Example 3.4

Let the acceleration be

a= v2.

Find v as a function of x and integrate again to find x as a

function of t. Use v0 and x0 as initial position and velocity
respectively.
SOLUTION:
dv
Set a = v and separate variables:
dx
dv
v = v2
dx
Z Z
dv
= dx.
v
Integration gives

ln v = x + K, of v = eK e x
.

Using v = v0 when x = x0 as initial velocity, we find eK =

v0 ex0 , and the equation is

v = v 0 e x0 x
.

This equation may be integrated further to yield x as a func-

tion of time:
dx
= v0 ex0 x
dt
Z Z
x x0
e dx = v0 dt.

After integration we find

x = x0 + ln(v0 t + K)

and by using x = x0 when t = 0, we obtain ln K = 0, there-

fore K = 1 and the equation for the position as a function
of time is
x = x0 + ln(v0 t + 1).
 45

Example 3.5

Let
a = 6x2 .
Find only the velocity as a function of x here.
SOLUTION:
dv
Use a = v and separate variables:
dx
dv
v = 6x2
dx
Z Z
vdv = 6x2 dx.

Integration yields
1 2
v = 2x3 + K
2
or q
v = ±2 x3 + K/4.
K must be obtained by the velocity (say v0 ) when the
particle is at x0 .
dx
Note: By substituting v = and separation of variables
dt
and integration, x may be found in principle as a function
of t. However, here the integration part is difficult.

3.5 Time graphs

It is quite instructive to look at the graphs of various kin-

ematic quantities as functions of time. We shall, for ex-
ample, call the graph of v (on the vertical axis) versus t (on
the horizontal axis), the v-t graph.
Velocity as an area under the a-t graph:
Since the integral of acceleration is velocity, the change in
velocity over the interval [t0 , t1 ] is the area under the v-t
graph between t0 and t1 .
 46

Velocity as an area: a
Change in velocity = area under the a-t graph

Z t1
v(t1 ) v(t0 ) = a(t)dt
t0

Position as an area under the v-t graph:

t1 t
Similarly, since the integral of velocity is position, the change t0
in position (i.e. the displacement) over the time interval
[t0 , t1 ] is the area under the v-t graph between t0 and t1 .
Figure 3.1. The change in velocity is the area
under the a-t graph.
Displacement as an area:

Change in position = area under the v-t graph

Z t1
x(t1 ) x(t0 ) = v(t)dt
t0

Piecewise functions:
The motion of a particle may consist of various parts, each
having a di↵erent acceleration. This means that the accel-
eration graph may be discontinuous. Consider the following
example.
 47

Example 3.6

A car accelerates from rest at constant acceleration 6m/s2

for 10s. Then it decelerates at 2m/s2 until it comes to a stop
again. Find its velocity as a function of time and find the
time at which it stops again.
SOLUTION:
For the first part of the motion its velocity is

v = 6t. a
After 10s its velocity is 60m/s.
6
For the second part of the motion its velocity is

v= 2t + C,
10 40 t
but at t = 10s (time still measured from the start) its velo- -2
city was 60m/s. We can then find C as follows

60 = 2(10) + C, C = 80. v
Therefore its velocity for the second part is 60

v= 2t + 80,

and is stops when v = 0 therefore

0= 2t + 80, t = 40s. 40
t
10

The acceleration may be expressed as

(
Figure 3.2.
6, t 2 [0, 10]
a(t) =
2, t 2 [10, 40]

Its velocity may be expressed as

(
6t, t 2 [0, 10]
v(t) =
2t + 80, t 2 [10, 40]

Figure 3.2 shows a and v as functions of t.

Continuity conditions at jumps in the acceleration:

We shall now discuss some aspects of the motion described
by such piecewise functions.
The acceleration as a function of time, may therefore appear
 48

as is in the a-t graph of Figure 3.3.

The velocity however cannot jump from one value to another
in no time, since this would mean that the acceleration at
that moment was infinite (and infinite quantities do not ap-
pear in our finite universe). The v-t graph is therefore always
continuous, but it may have a kink. Physically, this means
that the end velocity of the first part of the motion, is equal
to the initial velocity of the second part of the motion. This a
is also shown in Figure 3.3.
The displacement is the area under the v-t graph. Just like
velocity, the displacement is continuous, since a particle can-
not jump from one position to another in no time. But is
it even more continuous — since the velocity is continuous
at the transition point, the first derivative of the displace- t
ment (which is the velocity) is also continuous there. When t0 t1 t2
integrating velocity to obtain the displacement, one should v
ensure that the end position of the first part is the begin-
value left = value right
ning position of the second part. If the velocity was obtained
correctly, then the slope at the end of the first part will auto-
matically be the same as the slope at the start of the second
part. This is also shown in Figure 3.3
When finding the change in velocity in a case as in Figure t
3.3, it should be integrated as follows: t0 t1 t2
Let the acceleration be given by x
(
f (t), t 2 [t0 , t1 ],
a(t) = value left = value right
g(t), t 2 [t1 , t2 ]. and
slope left = slope right

then Z t1 Z t2
v(t2 ) v(t0 ) = f (t)dt + g(t)dt. t
t0 t1 t0 t1 t2

Example 3.7 Figure 3.3. Illustration of the continuity

conditions for v and x where a has a discontinuity.

..... to be inserted later ......

3.6 Paths in 2D

Paths: When a particle moves in 2D, it traces out a path in

2D. This path is parameterized by t, and the position vector
 49

then is a function of t, i.e.

" #
x(t)
r(t) = , t 2 [t0 , t1 ]
y(t)

One may view this as a vector r(t), anchored at the origin,

that changes its length and orientation as time goes by. As
it moves about, its arrow head traces out the path on the
xy plane.
It is often a good idea to find the equation of the path by
eliminating t, if possible, and to draw it on the xy-plane. It
is also advisable to put some arrows on the path to indicate
the direction in which the particle goes.
Consider the following example.

Example 3.8

Draw the path of the particle having the following position

vector:
" # y
t
r(t) = , t 2 [0, 2]
t2
r(2) t=2
Solution:
If possible, one may eliminate t from both components and
find the path as a function of x and y only. Eliminating t
from x = t and y = t2 , gives

y = x2 . t=1

The path is a parabola in the xy-plane. At time t = 0 r(1)

x
the particle starts at the origin and follows the parabola
traveling upwards along the right branch. Note that the x- t=0
coordinate of the position, x(t), moves at constant speed,
while the y(t) is motion with constant acceleration. Figure Figure 3.4.
3.4 shows the path. Two instances of the position vector
(at t = 1 and at t = 2) are also shown, though it is not
necessary to shows these vectors.
 50

Example 3.9

Draw the path of the particle that moves with the following
position vector:
" #
2 cos t y
r(t) = , t 2 [0, 4]
2 sin t

Solution: We have the following

t=0
x = 2 cos t
x
y = 2 sin t.

The variable t can be eliminated by squaring both sides and

adding. Then
t=4
x2 + y 2 = (2 cos t)2 + (2 sin t)2 = 4.

This is a circle with radius 2. Note that the particle starts on Figure 3.5.
the x-axis at x = 2, at time t = 0, and travels at constant
speed counter-clockwise. The path ends when t = 4 i.e.
it ends at the point (cos 4, sin 4) which is ( 0.65, 0.76).
[Do Checkpoint 3.1].
 51

Checkpoint 3.1:
Describe the path of a particle that moves
3.7 The derivative of a vector according to
" #
As mentioned before, a position vector r(t) that depends on 3 sin t
r(t) = , t 2 [0, 2]
t draws out a path with its arrowhead as t advances. 3 cos t
The vector derivative:
Make sure that you determine the direc-
The derivative of such a vector to t, is defined as follows: tion of travel correctly, and also put ar-
dr r(t + t) r(t) rows along the path to show this direction.
= lim
dt t!0 t

Note that the derivative is also a vector that depends on t.

dr
It is customary to write ṙ for .
dt
When r(t) is expressed in terms of its components, it should
be clear that each component is separately di↵erentiated
with respect to t, i.e. if
r(t) = x(t)i + y(t)j
then
ṙ(t) = ẋ(t)i + ẏ(t)j.

How should we interpret this derivative vector? Suppose

we put a number of points regularly spaced on the path.
The path itself may be approximated by a number of small
vectors drawn from one point to the next point on the path.
Each of these vectors may be expressed as
r = r(t + t) r(t)
If we choose t smaller, then the vector r comes closer to Figure 3.6.
the direction of the path at the point r(t). The derivative is
given by
r
ṙ = lim
t!0 t
and therefore the derivative vector is tangent to the path at
the point where it is calculated. If t denotes time then ṙ
denotes the instantaneous velocity of the tip of the arrow.
The vector integral:
Similarly integration of a vector is done component-wise:
Z Z Z Z
r(t)dt = (x(t)i + y(t)j)dt = i x(t)dt + j y(t)dt.

The integral of a vector is therefore again a vector. As this

stage we shall not supply a geometric interpretation of the
integral vector.
 52

3.8 Planar Kinematics: Velocity

When the motion of a particle occurs in a plane, the posi-

tion, velocity as well as the acceleration are all vectors. The
definition of velocity is still the derivative of position with
respect to time, however the derivative is now the vector
derivative.
Position, displacement and distance covered:
The position of a particle that describes a path in a plane is
given by the time-dependent vector:

r(t) = x(t)i + y(t)j.

It is called the position vector.

The magnitude of the position vector is the direct distance
between the particle and the origin, and is denoted by r or
|r|. q
r = |r| = x2 + y 2 .
As the position vector changes in time, it describes a path
or orbit in the xy-plane.
The displacement experienced by a particle from a
time t up to a time t + t, is

r = r(t + t) r(t).

The magnitude of this displacement is the called the distance

travelled in the time t.
Velocity and speed:
The (instantaneous) velocity of a particle is defined as

dr
v(t) = .
dt
Component-wise it is given by
dx dy
v(t) = i+ j
dt dt

Using the overdot notation to denote a time derivative we

may write this as

v = ṙ = iẋ + jẏ.

The velocity vector is tangent to the path of a particle.

When drawing the velocity vector on a figure it is advisable
Figure 3.7. The position vector (depending on
not to put the velocity vector at the origin, but rather to
time) and the path described by it.
 53

anchor it where the particle is at that moment. It will then

be evident that it is a tangent vector to the path.
The magnitude of the velocity is called the speed and it is
denoted by v: q
v = |v| = ẋ2 + ẏ 2 .

Example 3.10

The position of a particle is given by

r = ti + t2 j

Find its instantaneous velocity and instantaneous speed.

SOLUTION:
The velocity is
v = ṙ = i + 2tj
and the speed is
p
v = |ṙ| = 1 + 4t2 .

If the velocity (as a function of time) and the initial posi-

tion (call it r0 ) is known, then the position is obtained by
integration:
dr
=v
dt
and therefore Z Z
dr = vdt
Z
r = vdt.

The integration constant is also a vector and is determined

from the initial position.
 54

Example 3.11

A particle moves with velocity

v = i + 2tj

and at time t = 0 it is at position r0 = i + j. Find the

vector form for its position as a function of time. Also find
its acceleration.
SOLUTION:
The position is given by
Z
r = (i + 2tj)dt
Z Z
= i 1dt + j 2tdt

= ti + t2 j + r0 .

But
r(0) = i + j = r0
therefore

r = ti + t2 j + i + j = (t + 1)i + (t2 + 1)j.

3.9 Planar Kinematics: Acceleration

The (instantaneous) acceleration of a particle is defined as

dv
a(t) = .
dt
Component-wise it is given by
dv1 dv2
a(t) = i+ j
dt dt
where the velocity is given by v(t) = v1 (t) i + v2 (t) j.
Using the overdot notation to denote a time derivative we
may write this as

a = v̇ = r̈ = ẍ i + ÿ j.

The geometric interpretation of the acceleration vector is

beyond the scope of this course. However, we shall only
 55

briefly mention the following: if a particle moves at constant

speed (though it may change direction) the acceleration vec-
tor remains perpendicular to the path. On the other hand,
if the path is a straight line (though it may change speed)
the acceleration vector remains tangent to the path. But in
general, when both speed and direction change, the accel-
eration vector is neither perpendicular nor tangent to the
path.
In sketches it is also advisable to anchor the acceleration
vector to the point where the particle is at that moment,
rather than at the origin.
Chapter 4

Projectiles

4.1 Constant acceleration

The simplest acceleration is constant acceleration. Free fall-

ing bodies near the surface of the earth experience this kind
of acceleration when air resistance is ignored.
If the acceleration is constant, it means that we write ac and
consider it to be independent of either t, r or v. For such ac-
celeration the expressions for v and r assume a particularly
simple form. Let us briefly rederive it here.
For the definition of acceleration we have
dv
= ac .
dt
After separation and adding the integrals, we obtain
Z Z
dv = av dt + C

which integrates to

v = ac t + v 0 (4.1)

where v0 is the initial velocity (at t = 0).

By using the definition of the velocity, we obtain:
dr
= ac t + v 0
dt
and after separation and adding integrals, we obtain
Z Z
dr = (ac t + v0 )dt.

56
 57

After integration the following is obtained,

r = 12 ac t2 + v0 t + r0 (4.2)

where r0 is the initial position.

If we write out all vectors fully, they are

r = xi + yj, r 0 = x 0 i + y0 j

v = ẋi + ẏj = vx i + vy j v0 = (v0 )x i + (v0 )y j

and
ac = ax i + ay j.
Remember that x, y, vx and vy are functions of time while
ax and ay are constant.
Then (4.1) becomes

vx = (v0 )x + ax t
vy = (v0 )y + ay t

and (4.2) becomes

x = x0 + (v0 )x t + 12 ax t2
y = y0 + (v0 )y t + 12 ay t2 .

In summary: the components of velocity are linear functions

of time, while the components of position are quadratic func-
tions of time (if ax 6= 0 and ay 6= 0.)

4.2 Gravitational acceleration

Free falling bodies near the surface of the earth all experience
an almost constant downward acceleration in vacuum (i.e. if
the motion only covers a relatively small vertical distance).
This acceleration is called gravitational acceleration and its
magnitude is usually denoted by the symbol g. The value
of g di↵ers slightly from one place to another on the surface
of the earth but it is constant for every specific place on the
earth. The 1980 International Gravitation Formula gives g
at sea level as follows:

g0 = 9.780327(1 + 0.005279 sin2 + 0.000023 sin4 ) m/s2

where denotes the latitude of the place on the surface of

the earth. This formula contains the outward component of
the centrifugal force due to the rotation of the earth as well
 58

as the fact that the earth is an approximate oblate spheroid

(a flattened ellipsoid).
This gravitational constant must be corrected further for
height above sea level, as follows:
2
g = g0 (0.000003086 s )h
where h is the height above sea level in meters.
In most practical problems, it is custom to take g = 9.81 m/s2
or g = 9.8 m/s2 , and sometimes it is rounded to g = 10 m/s2 .
Gravitational acceleration (as a vector) points towards the
centre of the earth. However, when the motion of free falling
bodies close to the surface of the earth is considered for
motion that spans only a small portion of space, it is usually
good enough to take the direction directly downward, and
therefore we shall write

ac = gj.
If the motion occurs on a larger scale (such as satellites that
move around the earth) then the correct direction (toward
the centre of the earth) must be taken.

4.3 Projectile motion

The term projectile is used to describe an object that receives

an initial velocity and thereafter it is allowed to fall freely
under the influence of gravity. A missile on the other hand,
is an object that has its own propulsion while it moves about
under the influence of gravity.
The equation of motion:
With the assumption that gravity acts only downward and
with air resistance neglected, the acceleration of a projectile
is constant and the equations (4.1) and (4.2) apply, where
a= gj.
We shall assume that the projectile is launched from the
origin (r0 = 0) with an initial velocity with magnitude u
making an angle ↵ with respect to the positive x-axis. That u
is
v0 = u cos ↵ i + u sin ↵ j.

The equations of the velocity and the position are then

v = u cos ↵ i + u sin ↵ j gtj (4.3)
Figure 4.1.
 59

1 2
r = u cos ↵ it + u sin ↵ tj 2 gt j. (4.4)

‘Unpacked’ as components, we can express it as

vx = u cos ↵ (4.5)
vy = u sin ↵ gt (4.6)

x = x0 + u cos ↵ t (4.7)
1 2
y = y0 + u sin ↵ t 2 gt . (4.8)

From this it is clear that the projectile maintains a constant

velocity with magnitude u cos ↵ in the x direction, while the
y-component executes the same motion as a particle that
falls vertically.
The trajectory:
Choose the origin of the axes system at the start of the
trajectory, i.e. x0 = 0 and y0 = 0. The trajectory (or path)
is obtained by eliminating t from (4.7) and (4.8). First make
t the subject of the formula in (4.7),
x
t=
u cos ↵
and then substitute it into (4.8),
✓ ◆2
x 1 x
y = u sin ↵ 2g
u cos ↵ u cos ↵
g
= x tan ↵ x2 (4.9)
2u2 cos ↵ 2

1
By using the identity = 1 + tan2 ↵ the trajectory
cos ↵ 2
may be expressed as
g
y = x tan ↵ (1 + tan2 ↵)x2 (4.10)
2u2

This expression where the only trigonometric function em-

ployed is tan ↵, is useful when x, y and u is known and ↵ is
to be found.
The range:
u

}
Figure 4.2 shows a typical trajectory together with two im-
portant lengths associated with it. This is the range R and H
the greatest (maximum) height H.
The range R is the horizontal distance covered over a ho-
}
rizontal surface before the projectile hits the ground. It is R
clear that R is a root of the equation for the trajectory.
Figure 4.2.
 60

Let us make use of the equation (4.9) for the path to find
R. This means, set y = 0 in this equation and solve for x.
✓ ◆
g
0 = x tan ↵ x
2u2 cos ↵ 2
from which we find
2u2 sin ↵ cos ↵
x=0 or x= .
g

Clearly R is the second root. By using the identity sin(2↵) =

2 sin ↵ cos ↵ , the range can be expressed as

u2 sin(2↵)
R= . (4.11)
g

The time of flight T to cover the range may be obtained by

substituting y = 0 in (4.8):
⇣ ⌘
1
0 = t u sin ↵ 2 gt

and therefore
2u sin ↵
T = . (4.12)
g

(Of course, R could also have been found by first obtaining

the expression for T and then substituting it into (4.8).)
The maximum range:
There are a number of sensible questions that can be posed
with regard to R.
For a constant given launch speed u, at which launch angle
↵ will the range be a maximum?
In order to find the maximum R as al is varied, take the
formula for R, di↵erentiate it to ↵, set the result to zero,
and solve for ↵.

@R 2u2 cos(2↵)
= =0 (4.13)
@↵ g

We have to solve cos(2↵) = 0. Therefore the maximum

range is attained when

↵ = 45

and substituting this into (4.13), gives

u2
Rmaksimum = .
g
 61

Two angles for the same range:

For any range less than Rmaksimum there are two launch angles
that can achieve the given range. Make sin(2↵) the subject
of the formula (4.11):

Rg
sin(2↵) =
u2

For every positive right hand side, there are two positive
values of ↵ that satisfies the equation. We shall illustrate it
with a small example.

Example 4.1

A projectile is launched with u = 70 m/s and must reach

R = 250 m on a horizontal plane. Find two launch angles
that will achieve this. Also find the time of flight for each
trajectory.
SOLUTION:
Rg 250 ⇥ 9.81
sin(2↵) = 2
= = 0.5
u 702
Then
2↵ = 30 or 150 .
Therefore
↵ = 15 or 75
Figure 4.3 shows the two trajectories.
The time for each trajectory is obtained from
2u sin ↵
T =
g
and in this case the lower orbit with ↵ = 15 takes 3.70 s
while the higher orbit takes 18.80 s. u2
u1

As illustrated by the previous example, the time of flight 75 15

may di↵er considerably between the two available trajector-
}
ies for a given range. In ball-and-net games where air res-
R
istance is small enough to be neglected (this excludes bad-
minton, for example) the player will always use the lower
Figure 4.3.
trajectory so that his opponent will have the smallest pos-
sible time to react.
The maximum height:
 62

In order to find the maximum height, denoted by H (see

Figure 4.2), one can find the extremum of (4.9) by di↵eren-
tiation. However, from the symmetry of the trajectory it is
clear that the x coordinate of the turning point is 12 R is. Set
x = 12 R and set y = H in (4.9), then
! !2
u2 cos ↵ sin ↵ g u2 cos ↵ sin ↵
H= tan ↵
g 2u cos ↵ 2
2 g
and therefore
u2 sin ↵ 2
H= . (4.14)
2g

Here is yet another method to find H. Use (4.6). The max-

imum height is achieved when ẏ = 0, therefore take
0 = u sin ↵ gt
and then the time to reach the maximum height is
u sin ↵
t= .
g
Substitute this into (4.8) with y = H.

4.4 Examples of projectile motion

Some examples will serve to illustrate how these formulae

are used in practical applications.

Example 4.2

Two particles A and B are launched simultaneously: A is

shot from a point on the ground and B is shot from a point
40 m directly above this point. B is launched horizontally
with speed 28 m/s. The two particles hit the ground simul- B
taneously at the same spot on the ground. Calculate

(a) the time it takes for B to hit the ground,

(b) the magnitude and direction of the initial velocity of A

A.

(c) Show that at the point where A and B hit the

ground, their velocities make an angle of approxim- Figure 4.4.
ately 18.4 with respect to each other.
 63

SOLUTION:
(a) The vertical displacement of B is given by
1 2
y = 40 + u(sin 0 )t 2 gt

and therefore
40 = 12 (9.81)t2
and therefore t = 2.856 s.
(b) The horizontal displacement of B is given by

x = uB (cos 0 )t = (28)(2.856) = 79.959.

Let A have initial speed u with launch angle ↵. The motion

of A is then given by
1 2
x = u cos ↵ t y = u sin ↵ t 2 gt

Substitute t = 2.856, x = 79.959 and y = 0 into this equa-

tion

79.959 = u cos ↵ (2.856), ; dus u cos ↵ = 28.0

and
1 2
0 = u sin ↵ (2.856) 2 (9.81)(2.856) , dus u sin ↵ = 14.007

The magnitude of the initial velocity of A is

u2 sin ↵ 2 + u2 cos ↵ 2 = 28.02 + 14.0072 = 980.2

or u = 31.31 m/s. Its direction is

u sin ↵ 14.007
= tan ↵ =
u cos ↵ 28.0
or
↵ = 26.577 .
(c) A hits the ground at the same angle (its launch angle)
because of symmetry, therefore ↵ = 26.577 . The velocity
of B
vB = 28.0 i (9.81)t j
and when B hits the ground, we have

vB = 28 i (9.81)(2.856) = 28.0 i 28.014 j

Let the angle between vB and the ground be ✓, then

28.014
tan ✓ =
28.0
and therefore
✓ = 45.015
The di↵erence between these angles then is 45.015 - 26.577
= 18.44 .
 64

In many problems involving projectile motion, the prob-

lem reduces to finding the coefficients of a parabola where
enough information is given. Such information may be: (a)
three points through which the parabola goes, (b) two points
through which the parabola goes together with a slope at
some point.
In such cases the formula for the trajectory may be used.
Here it is often convenient to collect some constants into
one symbol to save writing e↵ort, for example, it may be
beneficial to set the following
g g
C= or B= (1 + t2↵ )
2u2 2u2
etcetera. Much of the work that follows then involves solving
two or three equations simultaneously in order to eliminate
some variables and solve for others.
The following example may illustrate this.

Example 4.3

A particle is shot from a point on the ground so that it just

clears a wall of height b at a distance a from the launch
point. It lands at a distance c from the wall on the other
side of the wall on the same horizontal plane as the launch
point. Find the launch speed and launch angle.
SOLUTION:
The trajectory of the particle is
b
y = x tan ↵ Bx2 (4.15)
a c
where
g
B= (1 + tan ↵2 ). (4.16)
2u2 Figure 4.5.
Apart from the origin (which already satisfies equation
(4.15), there are two other points on the trajectory, viz.
(a, b) and (a + c, 0). Figure 4.5 shows the setup. Let us sub-
stitute these two points. We therefore obtain two equations.
 65

Set x = a and y = b, then

b = a tan ↵ Ba2 (4.17)

Set x = a + c and y = 0, then

0 = (a + c) tan ↵ B(a + c)2

which may be simplified to

0 = tan ↵ B(a + c) (4.18)

since (a + c) 6= 0.
In these equations we consider a, b, c and g as known, while
↵ and u are to be found.
Note that B contains both u and ↵, while tan ↵ contains
only ↵. It will be beneficial to eliminate B, so that tan ↵ is
the only unknown.
Eliminating B from equations (4.17) and (4.18) gives

a2 tan ↵
b = a tan ↵
a+c
or
b(a + c)
tan ↵ = . (4.19)
ac
Note that resubstitution of this expression into (4.18) gives

b
B= . (4.20)
ac
Finally replace B from (4.16) into (4.20), then

g b
2
(1 + tan ↵2 ) =
2u ac
and we resubstirute tan ↵ from (4.19) into this, then
✓ ◆2 !
g b(a + c) b
1+ =
2u2 ac ac

Solving for u2 gives

!
2 g a 2 b2 + b 2 c 2 + c 2 a 2
u = 2b + . (4.21)
2 abc
 66

4.5 Summary

We shall briefly summarize the main formulae that were de-

rived.

Kinematic equations: The Path:

Position: Trajectory: ✓ ◆
x = x0 + u cos ↵t g 1
1 2
y = x tan ↵ x2
y = y0 + u sin ↵ 2 gt
2u2 cos ↵2
g
Velocity: or y = x tan ↵ 2u2
(1 + tan ↵2 )x2
ẋ = vx = u cos ↵
ẏ = vy = u sin ↵ gt Range:
u2 sin(2↵)
R=
Acceleration: g
ẍ = ax = 0 2u2 sin ↵ cos ↵
or R=
ÿ = ay = g g

Maximum height:
u2 sin ↵ 2
H=
2g
Chapter 5

Relative motion

5.1 Relative position

We have up to now always expressed positions and velocit-

ies with respect to a fixed system of axes. In applications
of navigation it is often necessary to find positions and ve-
locities with respect to a moving coordinate system.
Suppose particle A is at position rA , and particle B is at
position rB . These positions are with respect to the origin
O, i.e it is how an observer at O sees the two particles.
In a coordinate system where A is the origin, O will be at
position rA . We refer to this position as the position vector
of O relative to A and we write

rO/A = 0 rA = rA .

Notice that we have subtracted the position of A relative to

O from the position of O relative to O (which is 0).
Similarly the position of B relative to A is given by

rB/A = rB rA .

Of course, it is instructive to note that the position of A

relative to A is rA rA = 0, in other words sees itself as at
its own origin (as expected).
The general rule is as follows:

67
 68

Relative position: B
C
If the position vectors of particles A, B, C, ... relat-
ive to O are given by rA , rB , rC ... respectively, then the
position vectors of A, B, C, ... relative to A are given by C

rA/A = rA rA = 0 A
rB/A = rB rA
rC/A = rC rA
... etc. O

The position of all particles relative to a specific particle is B

obtained by subtracting the position of that particle from C
the positions of all the other particles. B/A

The distance between particles A and B is obtained by the

C
length of the vector rBA .
A
distance between A and B = krB/A k = krB rA k
O/A

Wind directions: O
The direction of the positions (or velocities) of vehicles, ships
and aeroplanes are often given in terms of the wind direc- Figure 5.1.
tions. It is then convenient to convert these directions to
angles taken counter clockwise from the positive x-axis. Re-
member that on the xy axis, East is at 0 , North is at
90 , West is at 180 and South is at 270 . For example,
20 West of North is ✓ = 20 + 90 = 110 . Another ex-
ample is, 10 West of South is 270 10 = 260 .
An example may help to illustrate how velocities and posi-
tions are used simultaneously in the same problem:
 69

Example 5.1

Ship A sees a lighthouse at 120 West of North at a distance

of 6 km. Because of fog, Ship A cannot see Ship B, but the
lighthouse sta↵ informs it that Ship B is 40 West of North
from the lighthouse at at a distance of 14 km. (Ship B is
stationary and remains so.) In which direction must A sail
to go directly to B and how fast must it sail in order to
arrive there in 2 hours 30 minutes?
SOLUTION:
It is best to convert all directions to standard directions
(counter clockwise from the positive x-axis). Let the light-
house be at O.
" # " #
6 cos 210 5.20
rO/A = =
6 sin 210 3.00

and therefore
" #
5.20
rA = rO/A = .
3.00
" # " #
14 cos 130 9.00
rB = =
14 sin 130 10.72
Then
" # " # " #
9.00 5.20 14.20
rB/A = rB rA = =
10.72 3.00 7.72

The direction of rB/A is found as follows:

7.72
tan ✓ = = 0.5437, then ✓ = 151.5 .
14.20
The distance from A to B is
q
krB/A k = ( 14.20)2 + (7.72)2 = 16.16 km.

Using v = s/t, the speed at which A must sail is

16.16 km
v= = 6.464 km/h.
2.5 h
 70

5.2 Relative velocity

Relative velocity is a similar concept. Let particle A have

velocity vA relative to the fixed observer at O. Let particle
B have velocity vB relative to O. Once again notice that the
velocity of O relative to itself is 0.
Relative velocity:
We require to know how A would see the various velocities.
Firstly A would see itself as stationary, i.e. the relative
velocity of A with respect to A is

vA/A = vA vA = 0.

The velocity of O relative to A is

vO/A = 0 vA = vA .

The velocity of B relative to A is

vB/A = vB vA .

The general rule is similar to position:

Relative velocity:

If the velocity vectors of particles A, B, C, ... relat-

ive to O are given by vA , vB , vC ... respectively, then the
velocity vectors of A, B, C, ... relative to A are given by

vA/A = vA vA = 0
vB/A = vB vA B
V

vC/A = vC vA V

V
... etc. A
A

Position with constant velocity: O O

In problems where all particles move at constant velocity, is
is often necessary to relate their various positions to each Figure 5.2.
other. A crucial observation is, that if A starts at rA (0) (at
time t = 0) and travels with constant velocity vA , then its
position is given by

rA = vA t + rA (0) (5.1)

It is often more convenient to convert all velocities accord-

ing to (5.1) to positions and then to consider the relative
positions of particles.
 71

Also take caution that you do not put position vectors and
velocity vectors in the same diagram — one cannot add ve-
locities and position together. Rather make two separate
diagrams, one showing the position vectors, and another
showing the velocity vectors.
Speed:
In many problems the speed of a ship or aeroplane is known,
because the captain or pilot can read this speed o↵ from
his instrument panel in the cabin. This speed is always
a speed relative to the medium (water or air) in which the
craft moves. In many problems this medium itself moves (as
sea currents or winds) and therefore such speeds are always
considered to be the magnitude of the relative velocity.
You are reminded that if vA/M is the relative velocity of
particle A with respect to the medium M in which it travels,
then its speed is kvA/M k. It is also often more convenient
to work with the square of the magnitude (to avoid having
a square root to keep track of all the time).
We shall illustrate the concept with some examples:

Example 5.2

A ship sails at constant velocity at 15 km/h in a steady

current that flows at 5 km/h directly west.
The ship sails along a straight line from port P to port Q
which is 100 km directly south west of P . Find the true
speed of the ship (relative to the earth) and also find the
direction in which it must be steered to reach port Q.
SOLUTION:
Let the velocity of the water (current) be vW and the velo-
city of the ship be vS . Let the velocity of the ship relative
to the water be vS/W .
From the information given,

vW = 5i, vS/W = 15 and the direction of vS is 225 .

We do not know the absolute speed of the ship, therefore

just denote it by vS .

vS = vS cos(225 )i + vS sin(225 )
vS vS
= p i p j.
2 2
 72

Therefore
✓ ◆ ✓ ◆
v v
vS/W = vS vW = pS + 5) i pS j
2 2
and kvS/W k = 15, therefore
✓ ◆2 ✓ ◆2
v v
pS + 5) + pS . = (15)2 .
2 2
From this we obtain
10
vS2 p vS 200 = 0,
2
or
vS = 18.113 or 11.041.
Speed is the magnitude of velocity and as such, it is always
positive. We therefore take the positive value here. The true
speed of the ship relative to the earth is 18.113 km/h.
The direction in which the ship must sail is given by the
direction of the velocity of the ship relative to the water:
✓ ◆ ✓ ◆
18.113 18.113
vS/W = p + 5) i p j
2 2
= 7.808i 12.808j

12.808
tan ✓ = = 1.640
7.808
✓ = 238.6 .
Expressed as a wind direction, Ship A must sail 31.4 West
of South.
 73

Example 5.3

Ship A sails with constant speed 15 km/h directly East and

ship B sails with constant speed 20 km/h directly North.
The extended courses of the ships cross at C. By 12:00 B is
at C and A is 7.5 km due West of C. Find

(a) the velocity of ship B relative to ship A,

(b) the shortest distance between the ships, and at which

time this happens.

SOLUTION:
The following information is given:

vA = 15i, vB = 20j, rA (0) = 7.5i rB (0) = 0.

(a)
vB/A = vB vA = 15i + 20j
(b) We first find the positions of A and B as functions of
time:
rA (t) = vA t + rA (0) = (15t 7.5)i
rB (t) = 20tj
Then
rA/B (t) = (15t 7.5)i 20tj
|rA/B (t)|2 = (15t 7.5)2 + ( 20t)2 = 625t2 225t + 56.25
This distance must be a minimum, therefore

d|rA/B (t)|2
= 1250t 225 = 0,
dt
and therefore the minimum distance is reached at time t =
0.18, and the minimum distance is
q
625(0.18)2 225(0.18) + 56.25 = 6 km.

Trigonometric equations:
In some problems involving relative motion, the direction in
which a craft must travel (let us call it ✓) in order to arrive at
some position (usually relative to some other moving craft)
may be unknown. In such cases the equations in which the
 74

unknown ✓ appears may contain terms in both sin ✓ as well

as cos ✓.
As an example consider

15 cos ✓ + 10 sin ✓ = 6.

How can one solve such an equation?

The technique for solving this, consists of taking one of
the trigonometric terms to the right hand side (for example
10 sin ✓) and then squaring the equation on both sides:

15 cos ✓ = 6 10 sin ✓,

then
225 cos ✓2 = (6 10 sin ✓)2 .
Then replace cos ✓2 with 1 sin ✓2 , and multiply out.

325 sin ✓2 120 sin ✓ 189 = 0

This is a quadratic equation in sin ✓. When solved we find

3 63
sin ✓ = , or sin ✓ = ,
5 65
so that
✓= 36.87 , or ✓ = 75.75 .

One should then return to the physical situation of the prob-

lem to determine which of the angles is the correct one.
 75

Example 5.4

An enemy car travels on a road due North at speed v. A

canon is positioned a distance a due West of the road. The
canon shoots a bullet at speed u (Assume that the speed
of the bullet remains constant.) If the canon shoots at the
moment the vehicle is due east, in which direction must it
aim in order to hit the car?
SOLUTION:
Let vC and vB be the velocities of the car and the bullet
respectively.
The velocities are

vC = v j, and vB = u cos ✓ i + u sin ✓ j.

Set the time equal to zero when the bullet leaves the canon.
The positions are therefore

rC = vt j
rB = u cos ✓t i + u sin ✓t j a i.

The position of the bullet relative tot the car is

rB/C = rB rC
= u cos ✓t i + u sin ✓t j a i vt j
= (u cos ✓t a) i + (u sin ✓t vt) j.

When the bullet hits the car, rB/C = 0, therefore

u cos ✓t a =0
u sin ✓t vt = 0.

Dividing the second equation by t and solving ✓ gives

✓ ◆
v
✓ = arcsin .
u
The time when the bullet strikes the car is found by substi-
tuting ✓ into the first equation and solving for t:
a
t= p .
v2 u2
Chapter 6

Dynamics

6.1 Forces and Newton’s laws

Up to this stage we were doing kinematics, which is the

description of motion without referring to the causes of the
motion. The topic introduced now, is dynamics, which is
the study of the influence of forces on free bodies.
A force is an abstract concept. It refers to an “influence”
(sometimes acting over a distance) that may change either
the position or the shape of a body. Experimental evidence
indicates that a force influences a body di↵erently if it acts
from di↵erent directions — it is therefore a vector quantity.
A force may influence a body primarily in two ways:

— It will accelerate bodies which are free to move — it

therefore has an e↵ect on the position of the body.

— It will deform bodies that are fixed — it therefore has

an e↵ect on the shape of a body.

In this chapter we shall consider only the former influence.

Newton’s three laws:
In his famous volume, Principia, Sir Isaac Newton has for-
mulated three well-known laws of motion which explains the
connection between motion and forces:

[I] Every particle maintains its state of either rest or mo-

tion in a straight line at constant speed, if no force
acts on it.

76
 77

[II] When a single force acts on a free particle it will ac-

celerate in the direction of the force. The magnitude
of the acceleration is directly proportional to the mag-
nitude of the force.

[III] When a force acts on a body in a certain direction, but

the body does not move in that direction, it is because
there is another force opposing the applied force. This
law is often stated as “For every action there is an
equal but opposite reaction”.

In what follows, we shall simply refer to these laws as New-

ton I, Newton II and Newton III. Newton I and II were actu-
ally already formulated in principle almost 50 years earlier
by Galileo Galilei.
In vector form Newton’s second law is:

F = ma (6.1)

where F is the force acting on a free particle, a is the ac-

celeration experienced by the particle due to the force and
m is a proportionality constant. Physical experiments show
that the proportionality constant m di↵ers between various
particles. It may be viewed as an indication of the “slug-
gishness” the particle to accelerate under the influence of a
force and is called the inertial mass of the particle.
In SI units mass is measured in kilogram (kg). The unit for
force is kg.m/s2 . However, a shorter name has been assigned
to this unit, viz. the newton with abbreviation N. Therefore

1 N = 1 kg.m/s2 .

Newton’s first law is already contained in the equation F =

ma — if F is zero then a is also zero. You are reminded
that zero acceleration does not necessarily imply that the
particle is at rest. Zero acceleration means that the velocity
is constant (i.e. it does not change its speed or direction).
Free and constrained particles:
Note that Newton’s second law is only applicable to particles
that are completely free to move. When a force acts on a
particle that is fixed, then it cannot accelerate in the dir-
ection of the force, but it remains at rest. In order not to
cause this behaviour to clash with Newton II, Newton has
formulated his third law. Let us restate the third law in
other words:
 78

If a force acts on a fixed particle then the medium constrain-

ing the particle applies an equal but opposite force to the
particle. The resultant force on the particle is therefore zero
and it will not accelerate. This force which the medium
exerts on the particle is called the reaction force.
The particle may be constrained only in a certain direc-
tion. Consider, for example, a particle resting on a hori-
zontal surface under gravity, see the Figure 6.1 on the left.
The particle is free to move upwards as well as horizontally
along the surface. However, it is constrained by the surface
to move downwards. If any forces are applied to the particle,
the downward component of the resultant force, will be can- N N
celed by the reaction force. The reaction force is always a
directed normal to the surface but opposite in direction to
the component of the applied force into the surface. That
is, the reaction force always points out of the surface. Its
magnitude is equal to the downward component of the total
resultant applied force.
F F
It therefore makes sense to decompose the forces on a particle
that slides along a surface in two components: a component
Figure 6.1. LEFT: The acceleration in the
tangential to the surface and a component normal to the sur-
vertical direction is zero, therefore N = F.
face. It is the normal component of the applied force that
RIGHT: The acceleration in the vertical direction
solicits the reaction force, while the tangential component
is zero, therefore N = F cos ✓ j.
has nothing that opposes it, and it causes the particle to
accelerate along the surface.
In the Figure 6.1:RIGHT, the applied force is decomposed
into a vertical component (F cos ✓) and a horizontal com-
ponent (F sin ✓). In the vertical direction the acceleration
is zero, therefore the reaction force simply balances F cos ✓.
The horizontal component, however, causes acceleration along
the surface.
Types of forces:
There are various physical forces in nature. We list some of
them here:

(1) gravitation,

(2) elastic deformation force (a compressed spring exerts

this type of force),

(3) thrust (force due to change in momentum of a stream

of particles. Wind, air resistance and rocket motors
all exert this type of force),

(4) electrostatic force (a charged particle exerts this force

on another charged particle)
 79

(5) magnetic force (the force between two magnets),

(6) friction (the resistance force when a particle is dragged
along a rough surface).

We shall study the e↵ects of only (1), (2), (3) and (6) here.

6.2 Friction

If a particle is set in motion on a frictionless horizontal plane,

it must keep its velocity forever. A surface where the e↵ects
of friction can be neglected, is called a smooth surface.
However, from everyday experience we know that most sur-
faces do not behave like that. When a particle moves on
a rough horizontal surface, it is decelerated (slowed down)
until it comes to rest. There must therefore be some force
that decelerate the particle until it comes to rest. This force
is called a friction force and is due to the roughness of both
the surface and the underside of the particle.
Friction comes in two types: static friction and kinetic fric-
tion. Friction forces can be complicated, but we shall employ
here only a simplified model of how the two types of friction
work. N
Static friction is similar to the normal reaction force, in that
is opposes the applied force, but it acts tangentially to the
surface (and not normal to it). It is, however, di↵erent in
that it can only oppose a force up to a certain maximum
magnitude and then it ’gives in’. Problems involving static
friction are usually of the type where the system is still at R F
rest, but is ’on the verge of moving’.
Let us firstly consider a particle at rest. It has mass m and
is lying on a rough horizontal surface. There is a down-
ward force on the particle due to its weight mg, and in the
upward direction the normal reaction force N acts on the -mg j
particle. Because the particle experiences no acceleration in
the upward direction, we must have that N = mg.
Figure 6.3.
We shall now apply a horizontal force F to the particle. Be-
cause of friction caused by the interaction between the un-
derside of the particle and the horizontal surface, the particle
does not accelerate. The friction force therefore ought to
cancel the applied force, and we have that
R= F.

As F is increased in magnitude, so R increases in magnitude.

 80

However, at some stage R reaches a maximum value and

thereafter the particle starts to move and then it acceler-
R
ates due to the ‘excess’ force F Rmax . Once the particle
is in motion, the model for kinetic friction is employed to
After this point the static friction
determine the friction force. force ceases, the particle comes
R max into motion and the kinetic friction
force takes over.
Figure 6.3 shows R as a function of the magnitude of the
applied force F .
Coulomb’s law: F
R max
Experimental evidence shows that the maximum friction
force (or limiting friction) depends on how strongly the particle
is pressed against the rough surface. Coulombs law of fric- Figure 6.3.
tion states that for most surfaces that has a uniform level
of roughness, the magnitude of the maximum friction is dir-
ectly proportional to the normal force. The proportionality
constant is called the static coefficient of friction and is usu-
ally denoted by µs . Therefore

Rmax = µs N.

Measuring static friction:

The static coefficient of friction becomes important only
when a particle is ‘on the verge of’ moving. One of the
easiest ways to measure it for a particular surface, is to put
a particle on a plane and to gradually increase the inclina-
tion of the plane until the particle starts to move. At that N
moment, R has just over-reached it maximum value, and it
s
N
is slightly larger than µs N . We usually assume that R is
close enough to Rmax so that we may set it equal to µs N .
Let the inclination angle be ↵ at this stage. Figure 6.4 shows
the setup.
Newton II for this particle gives

(µs N mg sin ↵)i + (N mg cos ↵)j = 0

mg
and from this follows
Figure 6.4.
µs N = mg sin ↵ ................. (1)

N = mg cos ↵ ................. (2)

(1) divided by (2) gives

µs = tan ↵.

This furnishes us with the easiest practical method for de-

termining µs .
 81

Kinetic friction:
The kinetic friction force appears whenever a particle is
already in motion along a rough surface and it always acts
in the opposite direction to that of the local velocity.
Similar to the maximum static friction, the kinetic friction is
assumed to be directly proportional to the magnitude of the
normal reaction force, and a similar Coulomb’s law applies:

Rkin = µk N,

where µk is the kinetic coefficient of friction. For most sur-

faces, µk is slightly less than µs .

6.3 Moment of a force

Definition of a moment
In this chapter we will briefly introduce the concept of the
moment of a force. A detailed discussion on moments will
be given elsewhere.
The moment of a force is a measure of its tendency to cause
a (rigid) body to rotate about a specific point or axis. This
is di↵erent from the tendency for a body to move, or trans-
late, in the direction of the force. In order for a moment to
Figure 6.5. Scalar formulation
develop, the force must act upon the body in such a manner
that the body would tend to twist. This occurs every time a
force is applied so that it does not pass through the centroid
of the body. A moment is due to a force not having an equal
and opposite force directly along it’s line of action.
Moment of a Force - Scalar formulation
The magnitude of the moment of a force acting about a point
or axis is directly proportional to the distance of the force
from the point or axis. It is defined as the product of the Figure 6.6.
force, F , and the moment arm d. The moment arm or lever
arm is the perpendicular distance between the line of action
of the force and the center of moments. The direction of the
moment is obtained using the right-hand rule as illustrated
in Figure 6.5.
Resultant moment, ⌃MO , is moments of all the forces (Fig-
ure 6.6):
⌃MO = ⌃F d.
 82

Moment of a Force - Vector formulation

Moment of force F about point O can be expressed using
cross product (figure 6.7):

⌃MO = r ⇥ F.

Direction and sense of MO are determined by right-hand

rule as illustrated in Figure 6.8.
Resultant moment of a system of forces about point O, Fig- Figure 6.7. Vector formulation
ure 6.9, can be determined by:

⌃MO = ⌃r ⇥ F.

The principle of moments

The principle of moments state that if a rigid object is in
equilibrium, the sum of the moments about any point must Figure 6.8.
be equal to zero, that is for an object in equilibrium:

⌃MO = 0.

6.4 Pulleys and strings

In a number of problems you may encounter, there will be Figure 6.9.

particles joined by strings that run over pulleys. When such
a string is taut (not slack) there exists a tension force in
the string. The direction of this force is always along the
string (in contrast to rigid beams where the forces exerted
by a beam may not necessarily be along the direction of the
beam).
In the models we are going to use, we shall always assume
that the string is massless (a light string) and that it does not
stretch (an inelastic string). The first assumption implies
that the force in the string is constant everywhere along the
string, and the second assumption implies that the length of
the string remains constant, therefore the motion of its left
end is mimicked by its right end (same displacement, same
speed, same acceleration).
The direction in which the string acts, may however be
changed by running the string over a pulley. In the models
we shall study, we shall assume that the pulley is massless
(a light pulley) and that is experiences no friction on its axel
(a smooth pulley). These two assumptions also ensure that
the force in the string has the same magnitude everywhere.
 83

To summarize, a pulley may be employed to change the point 2

of application of a force from one particle to another and
it may simultaneously change the direction of the applied
force, but the magnitude remains the same.
We shall illustrate these principles by means of examples.
Consider a smooth light pulley over which a light inelastic
s
string is run and on both sides the string goes downward.
Let T be the magnitude of the tension force in the string. s

Note that where a string passes over a pulley it pulls with a

force T on both sides. If the pulley remains at rest (does not
also accelerate) then the axel of the pulley applies an upward
Figure 6.10.
force of magnitude 2T on the pulley in order to keep it at
rest.
The next point requiring attention is the motion of the
string. If the left end moves up by a distance s, then the
right end must move down by a distance s. This is a dir-
ect consequence of the inelasticity of the string. The speeds
of the ends of the string are therefore given by ṡ and their
accelerations are s̈.
The moment a string becomes slack, the force in it becomes
zero.

Example 6.1

Two particles A and B with masses mA and mB are joined

by a light inelastic string that passes over a smooth light
pulley. The particles hang vertically down on both sides.
(a) Find the acceleration of the system in terms of mA , mB A
and g. Also find an expression for the tension in the string.
(b) If mA = 3 kg and mB = 5 kg, find the acceleration of
the system. B

Figure 6.11.
 84

SOLUTION: It is important to establish a system of axes for j

each particle. We shall choose j upwards for both particles. T
The next step is to draw all the forces acting on each particle.
The forces on particle A are: its weight ( mA gj) and the i j
tension in the string. Similarly the forces acting on particle T
B are: its weight ( mB gj) and the tension in the string.
a
The next step is to put the acceleration vectors in the dia- i
gram, but do not put them with the forces. Assume that
B is the heavier particle, then B will accelerate downwards mA g a
and A will accelerate up. If this assumption turns out to be
wrong then the acceleration will come out with a negative
sign.

mB g

(a) Figure 6.12.

For particle A:

"j: T mA g = mA a .............. (1).

For particle B:

"j: T mB g = mB a .............. (2).

Eliminating T from (1) and (2), gives

(mA + mB )a = (mB mA )g

or
mB mA
a= g.
mA + mB
(b) If mA = 3 kg and mB = 5 kg, then
5 3
a= g = 2.45m/s2 .
5+3
 85

Example 6.2

A particle A with mass mA lies on a smooth inclined plane

with inclination ↵. It is tied to a particle B with mass mB by
means of a light inelastic string going over a smooth pulley,
see Figure 6.13.

(a) Derive an expression for the acceleration of the system A

in terms of mA , mB , g, and ↵.
B
(b) Find the condition in terms of mA , mB , and ↵ such
that the system has no acceleration.

(c) If mA = 1 kg, and mB = 2 kg, for which inclination Figure 6.13.

angle will A move up 0.35m along the plane in 0.4s,
starting from rest?

SOLUTION:
We establish a convenient system of axes for each particle
(For A it is better to have one direction tangential to the
plane and the other normal to the plane.) Then all the forces
acting on each particle is drawn in on the particle. We also
N j
add arrows for the accelerations. Note that the magnitude
i
of the acceleration of A is equal to the magnitude of the T
a T
acceleration of B, because they are attached to the ends of j
an inelastic string.
We then write down Newton II in each direction for each i
particle: a
Particle A:
mA g
mB g
!i: T mA gs↵ = mA a .............(1)
" j: N mA g cos ↵ = 0 .............(2)
Figure 6.14.
Particle B:

" j: T mB g = mB ( a) .............(3)

(a) We do not really need equation (2) (but it is needed if

the magnitude of the normal force is required).
 86

Eliminate T from equations (1) and (3), and make a the

subject of the formula:
mB mA sin ↵
a= g.
mA + mB

(b) If a = 0, then

mB = mA sin ↵.

(c) If A moves up 0.9m along the plane in 0.6s, then (using

x x0 = v0 t + 12 at2 ), we find

1
0.35 = a(0.4)2 , or a = 4.375.
2
Substituting this into the formula derived in (a) gives

4.375 2 1 sin ↵
= , or sin ↵ = 0.6607
9.8 1+2
and then ↵ = 41.4 .

Example 6.3 s

A
Particle A with mass mA lies on a rough inclined plane with
B
inclination angle ↵ and its static coefficient of friction is µ,
where tan ↵ > µ. It is connected by means of a light inelastic
string going over a smooth light pulley to another particle mAg
B with m,ass mB hanging freely. (a) Find a formula for mBg
the smallest mass of B such that A does not slide down the
plane. (b) Find a formula for the largest mass of B such A
does not start sliding up the plane.
SOLUTION:
The fact that tan ↵ > µ ensures that A will definitely slide
A
down the plane if the mass of B is not large enough to ‘hold s

it back’. B
If A is on the verge of moving down the plane then the
friction force is upward and it has reached its maximum
mAg
value, i.e. R = ⌫s N . However, a is still zero. The forces
mBg
acting on the particles are shown in Figure 6.15:TOP.

Figure 6.15. TOP: Particle A on the verge of

sliding down. BOTTOM: Particle A on the verge
of sliding up.
 87

For particle A:

(T + µs N mA g sin ↵)i + (N mA g cos ↵)j = 0

and from this follows

N = mA g cos ↵ .................... (1)

and

T + µs N mA g sin ↵ = 0 .................... (2)

Eliminating N gives

T = mA g( µs cos ↵ + sin ↵) .................... (3)

For particle B

(T mB g)j = 0 and therefore T = mB g .................... (4)

 88

Eliminating T from (3) and (4) gives

mB g = mA g( µs cos ↵ + sin ↵)

and therefore the smallest mass of B that will just prevent

A from sliding down is

mB = mA ( µs cos ↵ + sin ↵)

(b) As the mass of B is increased the static friction force de-

k
creases in magnitude. At some stage it is zero and thereafter
it changes direction (facing down the slope) to oppose the
resultant tangential force which is now up along the slope. A
stage it reached where this downward friction force reaches 9.8
its maximum magnitude. Figure 6.15:BOTTOM shows this
setup.
Equations (1) and (4) remain the same, but equation (3)
becomes 9.8

T = mA g(µs cos ↵ + sin ↵) .................... (30 )

Figure 6.16.

Eliminating T from (30 )

and (4), gives the maximum mass
of B for which the system will be still just in equilibrium as

mB = mA (µs cos ↵ + sin ↵)

 89

Example 6.4

A particle A with mass 1 kg lies on a rough horizontal table

and is connected to another particle B with mass 1 kg by
means of a light inelastic string going over a light frictionless
pulley. B hangs freely. The system is released from rest and
after 2s A has covered 4m. Assuming that A had constant
acceleration, calculate the kinetic coefficient of friction.
SOLUTION:
We first need to do some kinematics at constant acceleration
in order to determine a. From x x0 = v0 t + 12 at2 , we find
that a = 2m/s2 .
The forces acting on A and B are shown in the figure. Since
A is moving, the opposing friction is a kinetic friction force,
and its magnitude is µk N .
For particle A:

(N 9.8)i + (T µN )j = 2i

and from this follows that

N = 9.8

and therefore

T 9.8µk = 2 .................... (1)

B has the same acceleration as A (but it is downward). Then

(T 9.8)j = 2j

and therefore T = 7.8. Substituting this into (1) gives µk =

0.592.
Chapter 7

Polar Coordinates

7.1 Review of Cartesian coordinates

Cartesian coordinates in the 2D plane, that is, x along the

horizontal axis and y along the vertical, are well known. We
shall now introduce another coordinate system in 2D which
is quite often more convenient than Cartesian coordinates
for the mathematical description of phenomena in certain
geometries. For example, for the description of motion in a
central force field, polar coordinates are more suitable than
Cartesian. y
Review of Cartesian coordinates:
Every point in the 2D plane is uniquely described by two
coordinates (x, y). A curve in the 2D plane may be expressed
as all coordinate pairs satisfying some relationship, such as (3,2)

y = f (x),

and it can be plotted in the xy-plane by choosing values for

x
x and calculating the corresponding value for y.
The grid (think of it as a printed graph paper) for Cartesian
coordinates consists of a grid of squares. In fact, the ver-
tical lines all correspond to the curves such as x = 0, x = 1,
x = 2 etc. (Of course these curves are lines here.) Similarly
the horizontal lines all correspond to another set of curves
of the form y = 0, y = 1, y = 2, etc. This is a fundamental
property of any coordinate system: It’s graph paper consists Figure 7.1. A Cartesian grid showing the point
of two sets of intersecting curves, and each curve in the first (3, 2). This is at the intersection of the lines x = 3
set is described by ‘first coordinate = constant’, and sim- and y = 2.
ilarly each curve in the second set is described by ‘second

90
 91

coordinate = constant’. A Cartesian graph paper is shown

in Figure 7.1.
Each point (x0 , y0 ) is the intersection of one of the curves
for x =constant from the first set with one of the curves for
y =constant from the second set.

7.2 Polar coordinates

Polar coordinates:
We now introduce a coordinate system, called polar coordin-
ates. In polar coordinates every point in the 2D plane is
described by a set of two coordinates (r, ✓). Let r be the po-
sition vector of the point. The first coordinate, r, is called
the radial coordinate and describes the distance between the
point and the origin. The second coordinate, ✓, is called the
angular or tranverse coordinate, which is the counterclock-
wise angle between a fixed reference line called the polar
axis, and the r-axis. Usually ✓ is measured in radians.
Note that the same point may now be described by two

}
di↵erent coordinate systems — in Cartesian it is at (x, y),
and in polar coordinates it is at (r, ✓). If the polar axis is r
collinear with the x-axis, the relationship between (x, y) and }} y
(r, ✓) is given by

p
x = r cos ✓, r= x2 + y 2 x
✓ ◆ (7.1)
y Figure 7.2.
y = r sin ✓, ✓ = atan
x

The polar grid:

The polar grid consists of r =constant (concentric circles)
and ✓ =constant (radial lines). Figure 7.3 shows an example
of this grid.
Any point (r, ✓) may now be plotted on the 2D plane. Fig-
ure 7.3 shows the points A:(2, 0), B:(3, ⇡/4), C:(4, ⇡/2) and 1 2 3 4
D:(3, 45 ⇡).

7.3 Graphs in polar coordinates

Plotting functions in polar coordinates:

We may now plot functions such as
r = g(✓) Figure 7.3. A polar grid, showing the points
A:(2, 0), B:(3, ⇡/4), C:(4, ⇡/2) and D:(3, 54 ⇡).
 92

on the polar grid.

The simplest example is probably r = 1. This is circle with
radius=1. Note that we may convert
p this function
p back to
Cartesian by substituting r = x2 + y 2 , i.e. x2 + y 2 = 1,
or if we square both sides, we obtain x2 + y 2 = 1, which is
the familiar expression for a circle with radius 1.
Familiar Cartesian curves in polar form:
Note that any expression in terms of x and y can be con-
verted to an expression in terms of r and ✓, by utilizing the
transformation formulae:

x = r cos ✓, and y = r sin ✓.

Example 7.1

Convert the line x + y = 2, to polar coordinates.

SOLUTION:

r cos ✓ + r sin ✓ = 2

or
2
r= .
cos ✓ + sin ✓

Example 7.2

Convert the parabola y = x2 , to polar coordinates.

SOLUTION:

r sin ✓ = (r cos ✓)2

or
sin ✓
r= .
cos2 ✓

Some well-known polar curves:

There are a number of ‘beautiful’ curves discovered in the
course of history, which are easier to describe in polar co-
ordinates than in Cartesian. Many of them have specific
 93

geometric properties that make them useful in some applic-

ations. We shall only mention some here.
The spiral of Archimedes:
A curve of the form
r = a✓
is called the Spiral of Archimedes. It is an example of many
spirals that exist. Figure 7.4 shows an example.
The Limaçon:
Figure 7.4. The spiral of Archimedes r = a✓,
A curve of the form
with a = 0.5.
r = a + b cos ✓

is called the Limaçon of Pascal.

It is obtained by fixing a circle at the origin and then rolling
another circle (with another radius) on the outer boundary
of the fixed circle, while following the locus of an o↵-center
point on the second circle. It is shown in Figure 7.5.
The Lemniscate
A curve of the form
Figure 7.5. The Limacon of Pascal r = a +
r2 = 2a2 cos(2✓) b cos ✓, with a = 1, b = 2.

is called a Lemniscate.
Converting it to Cartesian coordinates gives

(x2 + y 2 )2 = 2a2 (x2 y 2 ).

There exist a number of curves that are shaped like a fig-

ure ‘8’, and they mostly go by the name of a Lemnicate of Figure 7.6. The Lemniscate of Bernoulli r2 =
‘someone’. The one we have given here is called the Lem- 2a2 cos(2✓).
niscate of Bernoulli, and an example is shown in Figure 7.6.
The N-leafed rose
A curve of the form

r = a cos(N ✓)

is called an N-leafed rose. Figure 7.7 shows an example with

N = 5.

7.4 The polar unit vectors Figure 7.7. The Five-leafed rose r = a cos(5✓).

In Cartesian coordinates, vectors are expressed in terms of

the unit vectors i and j. Polar coordinates have their own
unit vectors, denoted by r̂ and ✓ˆ.
 94

Let us first review the Cartesian unit vectors. i is associated y

with x, it has unit length and it points in the direction in
j
which x increases. j is associated with y, it has unit length j
and it points in the direction in which y increases.
i j
The polar unit vectors: j i

In a similar fashion, the unit vector r̂ is associated with r, i x

it has unit length and it points in the direction in which r i
j
increases, while keeping ✓ fixed. It therefore always points j
radially outwards from the origin. Note that its direction i
depends on the position where it is employed. i
Similarly, the unit vector ✓ˆ is associated with ✓, it has unit
length and it points in the direction in which a point will
move (instantaneously) when ✓ is increased, while keeping r
fixed. Its direction also depends on the position where it is Figure 7.8.

r^
employed.

^
Figure 7.9 shows a polar grid with some sets of unit vectors
placed at various positions on the grid. r
^
^
^
r
It is important to establish a relationship between i and j
^ ^
on the one hand, and r̂ and ✓ˆ on the other.
Consider Figure 7.10. It should be clear that the components ^
r
of r̂ are cos ✓ in the x-direction and sin ✓ in the y-direction. ^
We may therefore express r̂ as

^
r
^

r̂ = i cos ✓ + j sin ✓ (1). ^

r^

^
r
Similarly
✓ˆ = i sin ✓ + j cos ✓ (2).

These equations express the polar unit vectors in terms of i

and j. We sometimes need the inverse relationship. It may Figure 7.9.
be derived geometrically, however, we shall simply use (1)
and (2) and eliminate either i or j, in order to make the
other Cartesian unit vector the subject of the formula.
Taking sin ✓⇥equation (1) + cos ✓⇥equation (2), gives ^

j = r̂ sin ✓ + ✓ˆ cos ✓.
r
^
One may similarly take cos ✓(1) - sin ✓(2), and obtain i, as
follows
i = r̂ cos ✓ ✓ˆ sin ✓.

Let us summarize these relationships:

Figure 7.10.
 95

Unit vectors in Cartesian and in polar:

r̂ = cos ✓ i + sin ✓ j i = cos ✓ r̂ sin ✓ ✓ˆ.

✓ˆ = sin ✓ i + cos ✓ j j = sin ✓ r̂ + cos ✓ ✓ˆ.

Adding vectors in polar coordinates:

Note that the polar unit vectors depend on ✓. When working
with polar unit vectors, the position where the vector is
anchored, determines which angle ✓ is used to describe the
polar unit vectors at that point.
Any vector anchored at the origin (i.e. a position vector),
may be described by
a = ar̂
where a is the magnitude of the vector and its direction is
contained in r̂. In fact we may sometimes require that the
dependence of either r̂ or ✓ˆ is explicitly shown, i.e. we may
write r̂(✓) and ✓ˆ(✓).
Consider a = ar̂(✓1 ). It is anchored at the origin and makes
an angle ✓1 with respect to the polar axis. See Figure 7.11.
If we add another vector

b = b1 r̂(✓1 ) + b2 ✓ˆ(✓1 )

to a, and call the result c, then

c =a+b
d
= ar̂(✓1 ) + b1 r̂(✓1 ) + b2 ✓ˆ(✓1 )
= (a + b1 )r̂(✓1 ) + b2 ✓ˆ(✓1 )
y

This result is correct and may be expanded as

c
"
cos ✓1
# "
sin ✓1
# ^ b
c = (a + b1 ) + b2 . ^
r
sin ✓1 cos ✓1

2 a
However, if a third vector d = dr̂(✓2 ) is to be added to c, 1 x
where ✓2 depends of the anchor of c, one cannot simply add
d here, without first converting c to the form

c = cr̂(✓2 ).

Vectors are normally not added in polar coordinates, since

the procedure is cumbersome. Figure 7.11.
The position vector:
 96

The position vector (which is anchored at the origin) has a

particular simple form in polar coordinates. It is
r = rr̂.

It has great use in central force fields (such as for a planet

moving around the sun) where the centre of the force field
(i.e. the sun) is at the origin and the force on the planet
always points towards the sun.
Derivatives of polar unit vectors:
Since the polar unit vectors depend on ✓, any derivative
which involves ✓ will necessitate taking the derivative of the
polar unit vector as well.
Consider
r̂ = i cos ✓ + j sin ✓
Its derivative with respect to ✓ is
dr̂
= i( sin ✓) + j cos ✓
d✓
but the right hand side is simply ✓ˆ.
Similarly, consider
✓ˆ = i sin ✓ + j cos ✓
Its derivative with respect to ✓ is
d✓ˆ
= i( cos ✓) + j( sin ✓)
d✓
and here the right hand side is r̂.
Let us summarize:

✓-derivatives of polar unit vectors:

dr̂ ˆ d✓ˆ
= ✓, = r̂
d✓ d✓

We shall discuss velocity and acceleration in polar coordin-

ates in the next section. However, we would like to mention
here that when taking a time derivative, of any vector in
polar coordinates where ✓ depends on time, one must take
the time derivative of r̂ and ✓ˆ as well.
The chain rule helps to convert derivatives to t to derivatives
to ✓. For example
dr̂ dr̂ d✓
= = ✓˙✓ˆ
dt d✓ dt
 97

and
d✓ˆ d✓ˆ d✓ ˙
= = ✓r̂.
dt d✓ dt

7.5 Velocity

We shall now consider the curvilinear motion of a particle

that is described in polar coordinates by r(✓). The angle is
✓(t), i.e. it is a function of time.
We shall describe all motion with respect to the origin and
make use of the polar unit vectors r̂ and ✓ˆ instead of i and
j.
Review of polar unit vectors:
You are reminded again that

dr̂ ˆ d✓ˆ
= ✓, and = r̂.
d✓ d✓
This means that (using the chain rule of di↵erentiation)

dr̂ dr̂ d✓
= = ✓˙✓ˆ
dt d✓ dt
and
d✓ˆ d✓ˆ d✓ ˙
= = ✓r̂.
dt d✓ dt

The variable ✓˙ is called the angular velocity, since it indicates

the time rate of change of the angle ✓. Often the symbol !
is used for the angular velocity, especially in applications to
rigid body dynamics.

Angular velocity:

Angular velocity is defined as

d✓
✓˙ = ! = .
dt

Angular velocity is measured in radians per second, often

expressed as rad/s. (However, the correct SI unit is just
s 1 , since the radian is actually unitless).
Position:
The position of a particle is given by

r = rr̂.
 98

Velocity:
Using the definition of velocity, we have that

dr dr̂
v= = ṙr̂ + r
dt dt
˙
= ṙr̂ + r✓.

The velocity now has two components: one in the ✓ˆ direction

that depends on the angular velocity (as well as the radius),
and another component in the r̂ direction that depends on
the rate at which the radius grows, i.e. ṙ.
It is often best to obtain the path of the particle in polar
coordinates if it is not given in that form. Once r(✓) is
known, ṙ can be found by di↵erentiation, and the velocity
may then be expressed in terms of ✓˙ and the polar unit
vectors.

7.6 Acceleration

Acceleration is the time derivative of velocity, therefore

dv
a =
dt
d ⇣ ⌘
= ṙr̂ + r✓˙✓ˆ
dt
dr̂ d✓ˆ
= r̈r̂ + ṙ + ṙ✓˙✓ˆ + r✓¨✓ˆ + ṙ✓˙
dt dt

= r̈r̂ + ṙ✓˙✓ˆ + ṙ✓˙✓ˆ + r✓¨✓ˆ + ṙ✓˙✓ˆ ˙

r✓r̂
= (r̈ r✓˙2 )r̂ + (2ṙ✓˙ + r✓) ¨ ✓ˆ

The variable ✓¨ is called the angular acceleration and may

also be denoted by the symbol ↵.

Angular acceleration:

Angular acceleration is defined as

d✓˙
✓¨ = ↵ = .
dt

Angular acceleration is measured in radians per second-squared,

expressed as rad/s2 .
 99

The acceleration in polar coordinates is therefore:

a = (r̈ r✓˙2 )r̂ + (2ṙ✓˙ + r✓)

¨ ✓ˆ.

Note that this general acceleration has four components:

In the r̂ direction:
r̈ acceleration due to accelerating radius
r✓ ˙ 2 centripetal acceleration
In the ✓ˆ direction:
2ṙ✓˙ Coriolis acceleration
r✓¨ acceleration due to angular acceleration
The component (r̈ r✓˙2 )r̂ is called the radial acceleration
and the component (2ṙ✓˙ + r✓)
¨ ✓ˆ is called the transverse ac-
celeration.
A note on terminology: a vector that points along the path
(in the direction of the velocity) is called tangential, and a
vector that points perpendicular to the path (in the direc-
tion of the radius of curvature) is called normal. In general
motion the normal and the radial acceleration are not neces-
sarily co-linear. Similarly in general motion the tangential
and the transverse acceleration are not necessarily co-linear.
However, during circular motion (i.e. ṙ = 0 and r̈ = 0),
radial acceleration and centripetal acceleration point in op-
posite directions and transverse acceleration and tangential
acceleration are the same.

7.7 Circular motion in polar coordin-

ates

Considering a particle traveling in a circle with constant

radius r. This implies that ṙ = 0.
Angular velocity:
You are reminded that

dr̂ ˆ d✓ˆ
= ✓, and = r̂.
d✓ d✓

The position of a particle moving in a circle with constant

radius r, is
r = rr̂.
 100

The velocity is
dr dr̂
v= = ṙr̂ + r
dt dt
dr̂ d✓
= 0r̂ + r
d✓ dt
= r✓˙✓ˆ.

dr̂
Note that we have employed the chain rule for .
dt
If ✓˙ is positive, the velocity is clearly in the direction of ✓ˆ.
If ✓˙ is negative (turning clockwise), the velocity is in the
direction of ✓ˆ.
Constant angular velocity:
Consider a particle traveling in a circular path with radius r
and constant angular velocity !c . In order to find its angular
position, we separate variables and integrate to t.

d✓
= !c
dt
or
Z Z
d✓ = !c dt

✓ = !c t + C

If the particle started at ✓0 at time t = 0, the integration

constant C is ✓0 , and the angular position of the particle is
✓ = ✓0 + !c t. (7.2)

The speed of the particle is also constant and is given by

v = r!c .

The period and revolutions:

The time it takes for the particle to complete one revolution
is called the period and it is given by

2⇡r 2⇡
T = = .
v !
The angular velocity is often given in revolutions per minute
or rpm. It is then necessary to convert it to rad/s before
using it in the formulae above. The conversion is as follows
1 rev 2⇡ rad 2⇡
1 rpm = = = rad/s.
1 min 60 s 60
 101

Acceleration:
For a particle moving in a circle, the acceleration is obtained
by di↵erentiating the velocity to time:
dv d ⇣ ˙ ˆ⌘
a= = r✓✓
dt dt
d✓ˆ d✓
= r✓¨✓ˆ + r✓˙
d✓ dt
= r✓¨✓ˆ + r✓(
˙ r̂)✓˙
= r↵✓¨ r✓˙2 r̂

Once again note the two types of acceleration: tangential

acceleration with magnitude r✓¨ that points in the direction
of the velocity, and centripetal acceleration with magnitude
r✓˙2 that points towards the centre of the circle.
 102

Example 7.3

A wheel with radius 40 cm spins at constant angular velocity

120 rpm. A point P on the rim of the wheel is initially at
✓0 = 30 . What is the angular position of P after 20 s Also
find the vector form of its position, velocity and acceleration
in terms of i and j at that moment. Take the origin at the
axle of the wheel.
SOLUTION:
The radius at which point P travels is 0.4 m. At t = 0,
✓0 = ⇡6 . The angular velocity of the wheel is

120 ⇥ 2⇡
!= = 4⇡ rad/s.
60
Note that ↵ = 0 since the angular velocity remains constant.
After 20 s,
✓ = ✓0 + !t
= ⇡6 + (4⇡)(20)
= 251.9 rad
We shall use this angle further in radians, therefore it is
necessary that the mode of the calculator is set to radians
here.
The position of point P is

r = rr̂
= (0.4)(cos ✓i + sin ✓j)
= (0.4) cos(251.9)i + (0.4) sin(251.9)
= 0.336i + 0.217j m

The velocity of point P is

v = r! ✓ˆ
= (0.4)(4⇡)( sin(251.9)i + cos(251.9)j)
= 2.72i + 4.22j m/s

The acceleration of point P is

a = r↵✓ˆ r! 2 r̂
= 0 (0.4)(4⇡)2 (cos(251.9)i + sin(251.9)j)
= 53.1i 34.2j m/s2 )
 103

7.8 Circular motion at constant angu-

lar acceleration

We shall now consider a particle that executes circular mo-

tion with constant angular acceleration — that means it
spins faster and faster. Though ✓¨ = ↵c is constant, ✓˙ = !
and ✓ change with time.
From the definition of angular acceleration, we have
d!
= ↵c .
dt
Separating variables and integrating yields
Z Z
d! = ↵c dt

✓ = ↵c t + C

Let the particle start with angular velocity !0 at time t = 0,

then C = !0 , and have

! = !0 + ↵c t. (7.3)

We can use the definition of ! here, separate variables and

integrate to yield
d✓
= !0 + ↵ c t
dt
or Z Z
d✓ = (!0 + ↵c t)dt

✓ = !0 t + 12 ↵c t2 + C
At time t = 0 the particle starts at position ✓0 . Then C = ✓0 ,
and

✓ = ✓0 + !0 t + 12 ↵c t2 . (7.4)

Eliminating t from (7.3) and (7.4) yields another useful equa-

tion
! 2 = (!0 )2 + 2↵c (✓ ✓0 ). (7.5)

One may easily notice the similarity between these equa-

tions for rotation and the similar equations for translation
at constant linear acceleration.
We shall illustrate the use of these equations with an ex-
ample.
 104

Example 7.4

A motor drives a pulley A with diameter 18 cm at constant

counterclockwise angular acceleration ↵ = 0.4 rad/s, start-
ing from rest (Figure 7.12). Pulley A is connected with a
thin massless belt to pulley B with diameter 32 cm. A point
P on the rim of pulley B is initially right at the top. (a) Find
the speed of the belt at time t = 3s. (b) Then express the
position rP and velocity vP of point P at time t = 3 s in
terms of i and j in a system of axes where the origin is on
the axle of pulley B. (c) Also express the acceleration, aP
of point P at time t = 3 s in terms of i and j.
SOLUTION: Pulley A has radius 0.09 m and pulley B has
radius 0.16 m.

(a) Consider pulley A:

P
!A = !A0 + ↵A t
B
B
At t = 3s, A
A
!A = 0 + (0.4)(3) = 1.2 rad/s.
A point on the rim of pulley A, then has speed

v = rA !A = (0.09)(1.2) = 0.108 m/s.

Figure 7.12.
This is also the speed of the belt at this moment.

(b) Consider pulley A again. A point on the rim of pulley

A has tangential acceleration (only the magnitude)

at = rA ↵A = (0.09)(0.4) = 0.036 m/s2 .

This is also the acceleration of the belt, and it is therefore

also the tangential acceleration of the point P on the rim of
pulley B. Since
a t = rB ↵ B
we have that
at 0.036
↵B = = = 0.225 rad/s2 .
rB 0.16
Figure 7.13. The position of P (blue arrow),
its velocity (red arrow) and its acceleration (green
arrow) at time t = 3 s.
 105

Now, point P starts at initial angular position ✓0 = ⇡/2,

from rest (i.e. !0 = 0) and experiences constant angular
acceleration 0.225 rad/s2 . Therefore, at time t = 3 s its
angular position and angular velocity are respectively

✓ = ✓0 + !0 t + 12 ↵t2
= ⇡2 + 0 + 12 (0.225)(32 ) = 2.583 rad

! = !0 + ↵t
= 0 + (0.225)(3) = 0.675 rad/s

Its position (with respect to a system of axes centered on

the axle of pulley B) is

rP = rr̂
= r(cos ✓i + sin ✓j)
= (0.16)(cos(2.583) i + sin(2.583)j)
= 0.136 i + 0.085 j m

Note that the angle is given in radians and therefore you

need to set your calculator to radians before calculating tri-
gonometric functions here. You may also keep your calcu-
lator in degrees, but then you must convert the angle to
degrees (2.583 rad = 148.0 ).
The velocity of point P is

vP = r! ✓ˆ
= r!( sin ✓i + cos ✓j)
= (0.16)(0.675)( sin(2.583)i + cos(2.583)j)
= 0.057 i 0.092 j m/s

(c) The magnitude of the tangential acceleration of point P

and its angular velocity are

↵ = 0.225 rad/s2 , ! = 0.675 rad/s.

The acceleration of P is

aP = r↵✓ˆ r! 2 r̂
= r↵( sin ✓i + cos ✓j) r! 2 (cos ✓i + sin ✓j)
= (0.16)(0.225)( sin(2.583)i + cos(2.583)j) (0.16)(0.6752 )(cos(2.583) i + sin(2.583)j)
= 0.019 i 0.031 j + 0.062 i 0.039 j
= 0.043 i 0.070 j m/s2
 106

7.9 Kinetics in polar coordinates

From Chapter 5 it follows that Newton’s second law in vector

form is:
F = ma (7.6)
where F is the force acting on a free particle, a is the accel-
eration experienced by the particle due to the force and m
is the mass of the particle.
Cartesian coordinates:
In 3-D Cartesian coordinates,the three components of equa-
tion (7.6) are:

Fx = max
Fy = may (7.7)
Fz = maz ,

where the acceleration and resultant force are given by:

a = ax i + ay j + az k
q
a= a2x + a2y + a2z
F = Fx i + Fy j + Fz k
q
|F| = Fx2 + Fy2 + Fz2 .

Polar coordinates:
In polar coordinates,the two components of equation (7.6)
are:

Fr = mar
(7.8)
F✓ = ma✓

where ar = r̈ r✓˙2 and a✓ = 2ṙ✓˙ + r✓.

¨
 107

Example 7.5

The smooth 0.5-kg double-collar in Figure (7.14) can freely

slide on arm AB and the circular guide rod. If the arm
rotates with a constant angular velocity ✓˙ = 3 rad/s,
determine the force the arm exerts on the collar at the
instant ✓ = 45 . Motion is in the horizontal plane.

SOLUTION:
Free-body diagram: The normal reaction Nc of the circular
guide rod and the force F of the arm AB act on the collar
in the plane of motion, Figure (7.15). Note that F acts
perpendicular to the axis of arm AB, i.e. in the direction
of the circular path at ✓ = 45 . The four unknowns are NC ,
F , ar , a✓ .

Equations of motion:

%)Fr = mar : NC cos 45 = (0.5)ar (1)

-)F✓ = ma✓ : F NC sin 45 = (0.5)a✓ (2)

Kinematics: Using the chain rule, the first and second de-
rivatives of r when ✓ = 45 , ✓˙ = 3 rad/s, ✓¨ = 0, are

r = 0.8 cos ✓ = 0.8 cos 45 = 0.5657 m

ṙ = 0.8 sin ✓✓˙ = 0.8 sin 45 (3) = 1.6971 ms

r̈ = 0.8[sin ✓ ✓¨ + cos ✓ ✓˙2 ]
= 0.8[sin 45 ; 0 + cos 45 ; (3)2 ] = 5.091 m/s2
Figure 7.14.
It then follows that

ar = r̈ r ✓˙2 = 0.5657 (0.5657)(3)2 = 10.18 ms2

a✓ = r✓¨ + 2ṙ✓˙ = (0.5657)(0) + 2( 1.6971)(3)

= 10.18 m/s2
Substituting these results in equations (1) and (2), yields:

NC = 7.20 N

F =0

Figure 7.15.
Chapter 8

Normal and Tangential Coordinates

8.1 Introduction

In the previous chapters we have described the position, ve-

locity and acceleration of a particle in fixed Cartesian and
polar coordinate systems. We have consequently expressed
these vectors in a polar coordinate system in 2D, and also
extended it to cylindrical coordinates in 3D. Another way to
describe the velocity and acceleration is to use base (unit)
vectors which is defined by the motion, or more specifically,
the coordinate system is aligned with the velocity vector of
the particle, which is tangent to the particle’s trajectory,
and the normal to this trajectory. The two orthogonal unit
vectors aligned with the particular directions, also define a
third direction, called the binormal, which is normal to both
the velocity vector and the vector normal to the velocity vec-
tor. Such a coordinate system is called Path coordinates or
Normal and Tangential coordinates (in 2D) or an Intrinsic
coordinate system.

8.2 Normal and tangential coordinates

Consider the motion of a particle of which its position vector Figure 8.1. Position.
r(t) traces a known curved path that is given by the scalar
function s(t), where s(t) is the arc length along the curve
measured from point O, as shown in Figure 8.1. The base
vectors are defined as:
(1) the unit vector tangent to the path êt , and

91
 92

(2) the unit vector normal to the path ên .

Each circle segment ds has a radius of curvature ⇢ and a
center of curvature O0 . The t axis (tangential axis) is tangent
to the curve at the point and is positive in the direction
of increasing s. The n axis (normal axis) is perpendicular
to the t axis and its direction is towards the center of the
curvature O0 .
Velocity
The path s(t) of the the particle is a function of time. The
Figure 8.2. The t axis is in the direction of the
particle’s velocity v has a direction that is always tangent
increasing s; the n axis is on the concave side of s
to the path (as shown in Figure 8.3) , and its magnitude is
and perpendicular to the t axis.
ds/dt. Therefore

v = vêt

where

ds
v= = ṡ
dt

Time derivative of êt

Similar to polar coordinates, the base vectors êt and ên are
not constant. As the particle moves along s the direction
of êt may change as illustrated in Figure 8.4, but its length
remains unity. From Figure 8.4 it follows that ê0t = êt + dêt Figure 8.3. Velocity.
where dêt is the change in direction of êt , with magnitude
det = (1)d✓ and pointing in the direction of ên . Since ds =
⇢d✓, it follows that ✓˙ = ṡ/⇢, and therefore

dêt ˙ n = ṡ ên = v ên

= ✓ê
dt ⇢ ⇢

Acceleration
The acceleration of the particle is the time derivative of the
(a)
velocity, yielding:

dêt v2
a = v̇ = v̇êt + v = v̇êt + ên
dt ⇢

The acceleration a is then

a = at êt + an ên ,

where the tangential and normal components of the acceler- (b)

ation are, respectively:
Figure 8.4. Change in direction of êt .
 93

at = v̇

and
v2
an = .
⇢

The magnitude of the acceleration is:

q
a= a2t + a2n .

Three-dimensional motion
If a particle moves along a curved path in space, the t and
n axes are defined as described above. At any point, the t
axis is tangential to the path and the n axis points towards
the center of the curve segment. The vectors êt and ên and
their respective coordinates t and n define two orthogonal
directions. The plane defined by these two directions, is
called the osculating plane. This plane changes from point
to point, and can be thought of as the plane that locally
contains the trajectory. In order to define a right-handed
set of axes, an additional unit vector which is orthogonal
to êt and ên is introduced. This vector is called the binor-
mal and is defined as êb = êt ⇥ ên (where ⇥ is called the
cross product which will be defined later). The three path
coordinate directions are shown in Figure 8.5.
The radius of curvature
In some case the trajectory of the particle will be known as
a curve of the form y = f (x). The radius of curvature, ⇢,
can be determined from the equation:

[1 + (dy/dx)2 ]3/2 Figure 8.5. The three orthogonal path

⇢= coordinate directions.
|d2 y/dx2 |
 94

Example 8.1

When the skier shown in Figure 8.6 reaches point A along

the parabolic path y = (1/20)x2 , she has a speed of 6 m/s
which is increasing at 2 m/s2 . Neglect the size of the skier
in the calculations.
(a) Determine the magnitude and direction of her velocity,
and also give her velocity in normal and tangential coordin-
ates.
(b) Give her acceleration in normal and tangential coordin-
ates.

SOLUTION:
The velocity is always directed tangent to the path and
therefore the velocity in n t coordinates is:

v = vêt .

Since
1 2
y= x
20
then
dy 1
= x
dx 10

At x = 10 m, dy/dx = 1, and at A, v makes an angle of

✓ = tan 1 (1) = 45 with the x-axis. Therefore vA = 6 m/s
at an angle of 45 (.).
Figure 8.6.
The acceleration is determined from a = v̇êt + (v 2 /⇢)ên . In
order to be able to calculate the acceleration we require the
radius of curvature, ⇢, at point A, i.e. at (10, 5)). Since
d2 y 1
dx2
= 10 , then
" #
1
[1 + (dy/dx)2 ]3/2 [1 + ( 10 x)2 ]3/2
⇢= = 1 = 28.28 m
|d2 y/dx2 | | 10 | x=10 m

The acceleration is then

v2
a = v̇êt + ⇢ ên

62
= 2êt + 28.28 ên

= (2êt + 1.273ên )m/s2

Chapter 9

Work and Energy

9.1 Introduction

So far we have used Newtons second law ΣF = ma to estab-

lish the instantaneous relation between the sum of the forces
acting on a particle and the acceleration of that particle.
Once the acceleration is known, the velocity (or position) is
obtained by integrating the expression of the acceleration (or
velocity). Newtons law and the conservation of momentum
give us a vector description of the motion of a particle in
three dimensions There are two situations in which the cu-
mulative effects of unbalanced forces acting on a particle are
of interest to us. These involve:

• Forces acting along the trajectory. In this case, in-

tegration of the forces with respect to the displacement
leads to the principle of work and energy.

• Forces acting over a time interval. In this case, in-

tegration of the forces with respect to the time leads
to the principle of impulse and momentum as will be
discussed later.

9.2 Mechanical Work

Consider a force F acting on a particle that moves along a

path as shown in Figure 9.??. Let r be the position of the
particle measured relative to the origin O. The work done by
the force F when the particle moves an infinitesimal amount

Figure 9.??.
94
 95

dr is defined as

dU = F · dr.

That is, the work done by the force F, over an infinitesimal

displacement dr, is the scalar product of F and dr, and it
then follows that the work is a scalar quantity. Using the
definition of the scalar product, we have that dU = F · dr =
F ds cos θ, where ds is the magnitude of dr, and θ is the angle
between F and dr. Since dr is parallel to the tangent vector
to the path, êt , (i.e. dr = ds êt ), we have that F · êt = Ft .
Thus

dU = Ft ds,

which implies that only the tangential component of the

force does work.
In the international system, SI, the unit of work is the Joule
(J). We have that 1 J = 1 N · m. In the English system the
unit of work is the ft-lb.

Work of a Variable Force

During a finite increment in which the particle moves from
position r1 to position r2 , Figure 9.??, the total work done
by F is

! r2 ! s2 Figure 9.??.
U1−2 = F · dr = F cos θds. (9.1)
r1 s1

Here, s1 and s2 are the path coordinates corresponding to

r1 and r2 .

Work of a Constant Force Working Along a Straight Line

Consider the constant force Fc that acts at a constant angle
θ from its straight-line path, Figure 9.??. Then the compon- Figure 9.??.
ent of Fc in the direction of displacement is always Fc cos θ.
The work done by Fc when the particle is displaced from s1
to s2 is
! s2
U1−2 = Fc cos θ ds,
s1

U1−2 = Fc cos θ(s2 − s1 ). (9.2)

 96

Work of Weight
Consider a particle of weight W, which moves along the
path, shown in Figure 9.??, from position s1 to s2 . From
equation (??) it follows that

" "
U1−2 = rr12 F · dr = ss12 (−W j) · (dxi + dyj + dzk)
" y2
= y1 −W dy = −W (y2 − y1 ),

or

Figure 9.??.
U1−2 = −W ∆y. (9.3)

The work done by weight is independent of the path and

only depends on the weight of the particle and its vertical
displacement.

Work of a Spring Force

If an elastic spring of stiffness k is stretched a distance ds,
Figure 9.??, the force exerted by the spring on the particle
is F = −ksi. From the definition of work, if the particle is
displaced from position s1 to s2 , the work done by the spring
is:
" "
U1−2 = rr12 F · dr = rr12 (−ksi) · dsi Figure 9.??.
" s2
= s1 (−ks)ds,

1 1
U1−2 = −( ks22 − ks21 ). (9.4)
2 2

9.3 Principles of Work and Energy

Consider the particle in Figure 9.?? which is located on the

path defined relative to an inertial coordinate system. If the
particle has mass m and is subjected to the external forces
FR = ΣF, then the equation of motion in the tangential Figure 9.??.
direction is ΣFt = mat . Assume that the initial conditions
are s = s1 and v = v1 , and the final conditions are s = s2
and v = v2 . Applying the kinematic equation at = vdv/ds
and integrating both sides, yield
 97

! s2 ! s2 ! v2
Σ Ft ds = Σ F cos θds = mvdv,
s1 s1 v1

and therefore from equation (??) it follows that

1 1
ΣU1−2 = mv22 − mv12 . (9.5)
2 2
The two terms on the right had sign of equation (??) are of
the form T = 12 mv 2 and is defined as the kinetic energy of
the particle. Kinetic energy, with units joules (J) or ft·lb, is a
scalar and is always positive. Work, however, can be positive
or negative. For convenience, equation (??), is often written
as:

T1 + ΣU1−2 = T2 , (9.6)

which states the particle’s final kinetic energy is equal to its

initial kinetic energy plus all the work done on the particle.

9.4 Power and Efficiency

Power
In many situations it is useful to consider the rate at which
a device can deliver work. The work per unit time is called
the power, P . Thus,

dU dr
P = =F· = F · v. (9.7)
dt dt
The unit of power in the SI system is Watt (W). We have
that 1 W = 1 J /s. In the English system the unit of power
is the ft-lb/s. A common unit of power is also the horse
power (hp), which is equivalent to 550 ft-lb/s, or 746 W.

Efficiency
The ratio of the power delivered out of a system, Pout , to the
power delivered into the system, Pin , is called the mechanical
efficiency, ε, of the system, i.e.:

Pout
ε= . (9.8)
Pin
This definition assumes that the energy into and out of the
system flows continuously and is not retained within the
 98

system. The efficiency of any actual machine is always less

than unity since there is always some mechanical energy
dissipated as heat due to friction forces.

9.5 Potential Energy

Conservative forces
If the work of a force is independent of the path and only
depends on the force’s initial and final positions, the force
is classified as a conservative force. Examples include the
weight of a particle and work done by spring forces.
A non-conservative force is when the work done by the par-
ticular force is dependent on the path along which the particle
travels. An example of a non-conservative force is kinetic
friction as a body slips over a surface.

Energy
Energy is defined as the capacity to do work. The kinetic
energy is a measure of a particle’s capacity to do work which
is associated with motion. When energy is the result of the
position of a particle, measured from a fixed datum or ref-
erence plane, it is called potential energy. Thus potential
energy is a measure of the amount of work done by conser-
vative forces to move the particle from its current position
to the datum.

Gravitational potential energy

If a particle is located a distance y above an arbitrary selected
datum, Figure 9.??, the particle’s weight W has positive
gravitational potential energy, Vg , since W has the capacity
to do positive work when the particle is moved back to the
datum, and Figure 9.??. Gravitational po-

tential energy
Vg = W y. (9.9)

If the particle is place below the datum, then the work done
by gravity to move the datum will be negative, then

Vg = −W y. (9.10)
 99

Elastic potential energy

Elastic potential energy occurs in the deformation of an
elastic body, such as a spring. For small displacement s,
tensile or compressive, the force supported by a spring is
F = ks. The elastic potential energy of the spring is the
work done on it to deform it an amount s. Thus, we have
! s 1
Ve = (ks)ds − ks2 . (9.11)
0 2

Here Ve is always positive since, in the deformed position,

the force of the spring has the capacity to do positive work,
as illustrated in Figure 9.??. Figure 9.??. Elastic potential
If the deformation, either tensile or compressive, increases
from s1 to s2 during the motion, then the change in potential energy
energy of the spring is the difference between its final and
initial values, or,

1
∆Ve = k(s22 − s21 ).
2

9.6 Conservation of Energy

If a particle is subjected to both gravitational and elastic

forces, the particle’s potential energy can be expressed in
term of a potential function as

V = Vg + Ve . (9.12)

The work done by conservative forces is:

U1−2 = V1 − V2
= −(V2 − V1 )
= −∆V.

From Equation (??) it follows that

T1 + ΣU1−2 = T2 ,

T1 + (ΣU1−2 )cons + (ΣU1−2 )non−cons = T2 ,

T1 + [(V1 )g − (V2 )g ] + [(V1 )e − (V2 )e ] + (ΣU1−2 )non−cons = T2 ,

 100

T1 + V1 + (ΣU1−2 )non−cons = T2 + V2 , (9.13)

where V1 and V2 include both gravitational and elastic po-

tential energy. If there are only conservative forces, then
equation (??) reduces to

T1 + V1 = T2 + V2 , (9.14)

which is called the conservation of mechanical energy or

simply conservation of energy.

If an isolated system of particles is subjected only to con-

servative forces, equation (??) for a system of particles be-
comes:

ΣT1 + ΣV1 = ΣT2 + ΣV2 . (9.15)

Chapter 10

Linear Impulse and Momentum

10.1 Introduction

In this chapter we will consider the equations that result

from integrating Newtons second law, F = ma, with re-
spect to time. This approach leads to the equations of linear
impulse and momentum. These equations are useful when
solving problems in which we are interested in determining
the global e↵ect of a force acting on a particle over a time
interval.

10.2 Linear impulse and linear momentum

Consider the curvilinear motion of a particle of mass, m,

under the influence of a force F as shown in Figure 10.1.
The resultant force F of all the forces on the particle is in
the direction of the acceleration v̇. Assuming that the mass
does not change, we have from Newtons second law,

d
F = mv̇ = dt (mv).

The linear momentum vector, L, is defined as

L = mv.

Thus, an alternative form of Newtons second law is

Figure 10.1. Velocity.

F = L̇, (10.1)

101
 102

which states that the total force acting on a particle is equal

to the time rate of change of its linear momentum.
The Linear Impulse-Momentum Principle
In the previous section we have rewritten Newton’s second
law in an alternative form in terms of momentum. Consider
now a force acting on the particle between time t1 and time
t2 . Equation 10.1 can then be integrated in time to obtain:

Z t1 Z t1
F(t)dt = L̇dt = L2 L1 = L. (10.2)
t0 t0

Here L1 = L(t1 ) and L2 = L(t2 ). The term

Z t1
I= F(t)dt = L = (mv)2 (mv)1 ,
t0

is called the linear impulse. Thus, the linear impulse on a

particle is equal to the linear momentum change L. In
many applications, certain forces are very large and of short
duration. Such forces are called impulsive forces and need to
be taken into consideration when applying equation (10.2).
Non-impulsive forces are often ignored in the calculations.
An example of a non-impulsive force is the weight of a ball
when struck by a bat.
Impulsive forces
Typically the impulsive forces are forces of very large mag-
nitude that act over a very small interval of time, but cause
a significant change in the momentum. Examples of impuls-
ive forces are those generated when a ball is hit by a tennis
racquet or a baseball bat, or when a steel ball bounces on a
steel plate. The table below lists typical time intervals over
which some of these impulses occur.
Time interval t [s]
Racquet hitting a tennis ball 0.005-0.05
Bat hitting a baseball 0.01-0.02
Golf club hitting a golf ball 0.001
Shotgun 0.001
Steel ball bouncing on steel plate 0.0002
For calculation purposes it is often convenient to write equa-
tion (10.2) in terms of its components:
 103

R t1
m(v1 )x + t0 Fx dt = m(v2 )x
R t1
m(v1 )y + t0 Fy dt = m(v2 )y (10.3)
R t1
m(v1 )z + t0 Fz dt = m(v2 )z

NOTE: Linear impulse and momentum have the same units.

In the SI system they are N·s or kg·m/s, whereas in the Eng-
lish system they are lb·s, or slug ft/s.

Example 10.1

A tennis player strikes the tennis ball with her racket when
the ball is at the uppermost point of trajectory as shown.
The horizontal velocity of the ball just before impact with
the racket is v1 = 15 m/s and just after impact its velocity
is v2 = 21 m/s directed at the 15 angle as shown in Figure
10.2. If the 60-g ball is in contact with the racket for 0.02
s, determine the average force R exerted by the racket on
the ball. Also determine the angle made by R with the Figure 10.2.
horizontal.
 104

SOLUTION:
From equation (10.3) it follows:

Z t1
m(v1 )x + Fx dt = m(v2 )x : 0.060(15)+Rx (0.02) = 0.060(21cos 15 )
t0
Z t1
m(v1 )y + Fy dt = m(v2 )y : 0.060(0)+Ry (0.02) (0.060)(9.81) = 0.060(21sin 15 )
t0

Solving for the impact forces, Rx and Ry , yield:

Rx = 105.9 N

Ry = 16.89 N
The average impact force is therefore:

R = {105.9i + 16.89j} N

The angle is:

1 16.89
= tan ( ) = 9.06 %
105.9

10.3 Conservation of Linear Momentum

We see from equation (10.2) that if the resultant force on a

particle is zero during an interval of time, then its linear mo-
mentum L must remain constant. Since equation (10.2) is a
vector quantity, we can have situations in which only some
components of the resultant force are zero. For instance, in
Cartesian coordinates, if the resultant force has a non-zero
component in the y direction only, then the x and z com-
ponents of the linear momentum will be conserved since the
force components in x and z are zero. Consider now two
particles, m1 and m2 , which interact during an interval of
time. Assume that interaction forces between them are the
only unbalanced forces on the particles. Let F be the inter-
action force that particle m2 exerts on particle m1 . Then,
according to Newton’s third law, the interaction force that
particle m1 exerts on particle m2 will be F. Using (10.2),
we will have that L1 = L1 , or L = L1 + L2 = 0.
That is, the changes of momentum of particles m1 and m2
are equal in magnitude and opposite in sign, and the total
momentum change equals zero. Recall that this is true if
 105

the only unbalanced forces on the particles are the interac-

tion forces. We note that the above argument is also valid
in a componentwise sense. That is, when two particles in-
teract and there are no external unbalanced forces along a
given direction, then the total momentum change along that
direction must be zero.

Example 10.2

The 80-kg man can throw the 20-kg box horizontally at

4 m/s when standing on the ground. If instead he firmly
stands on the 120-kg boat and throws the box, as shown in
Figure 10.3, determine how far the boat will move in the
three seconds. Neglect water resistance.

SOLUTION:
Applying the conservation of linear momentum to the man,
boat and box system:

!) ⌃(mv)1 = ⌃(mv)2 : 0 + 0 + 0 = (mm + mb )vb mbox vbox

0 = (80 + 120)vb (20)vbox

vbox = 10vb (1)
Since the velocity of the box relative to the man (and the
boat), vbox/b , is known, then vb can be related to vbox using
the relative velocity equation.

!) vbox = vb + vbox/b

vbox = vb 4 (2)
Solving for equations (1) and (2): Figure 10.3.

vbox = 3.64 m/s

vb = 0.3636 m/s !
The displacement of the boat in three seconds is therefore

sb = vb t = (0.3636)(3) = 1.09 m

Figure 10.4.
Chapter 11

Angular Impulse and Momentum for a

Particle

11.1 Introduction

Similar to the equations for linear impulse and momentum, a

set of equations that relate angular impulse and momentum
can be derived.

11.2 Angular momentum

Consider a particle of mass m, with velocity v, moving under

the influence of a force F . The angular momentum about
point O is defined as the moment of the particles linear
momentum, L = mv, about O (Figure 11.1(a))
Therefore,the particles angular momentum is given by:

HO = r ⇥ mv = r ⇥ L. (11.1)

The units for the angular momentum are kg·m2 /s in the SI

system, and slug·ft2 /s in the English system.

From its definition, it follows that the angular momentum is

a vector which is perpendicular to the plane defined by r and
v. Therefore it may in some situations be more convenient
to determine the direction of HO from the right-hand rule,
and its modulus directly from the definition of the cross Figure 11.1. Angular impulse.

106
 107

product,
HO = mvr sin ✓. (11.2)
where ✓ is the angle between r and mv (Figure 11.1(b)).

In other situations, it may be convenient to directly calculate

the angular momentum in component form. For instance,
using a right handed Cartesian coordinate system, the com-
ponents of the angular momentum are calculated as

i j k
H0 = Hx i+Hy j+Hz k x y z = m(vz y vy z)i+m(vx z vz x)j+m(vy x vx y)k.
mvx mvy mvz

Similarly, in cylindrical coordinates we have

r̂ ✓ˆ k
H0 = Hr r̂+H✓ ✓ˆ+Hz k = r 0 z = mv✓ zr̂+m(vr z vz r)✓ˆ+mv✓ rk.
mvr mv✓ mvz

Rate of Change of Angular Momentum

We now want to examine how the angular momentum changes
with time. We examine this in two di↵erent coordinate sys-
tems: system 11.2 is about a fixed point O; system 11.3 is
about the center of mass of the particle. Of course system
11.3 is rather trivial for a point mass, but its later exten-
sions to rigid bodies will be extremely important. Even at
this trivial level, we will obtain an important result.
About a Fixed Point O
The angular momentum about the fixed point O is
Figure 11.2.

HO = r ⇥ mv = r ⇥ L.

Taking a time derivative of this expression , we have

ḢO = ṙ ⇥ mv + r ⇥ mv̇.

Here, we have assumed that m is constant. If O is a fixed

point, then ṙ = v and ṙ ⇥ mv = 0. Thus, we end up with,
Figure 11.3.
ḢO = r ⇥ mv̇. (11.3)
 108

Applying Newtons second law to the right hand side of the

above equation, we have that r ⇥ ma = r ⇥ F = MO , where
MO is the moment of the force F about point O. The equa-
tion expressing the rate of change of angular momentum is
then written as

MO = ḢO . (11.4)

We note that this expression is valid whenever point O is

fixed. The above equation is analogous to the equation de-
rived for the rate of change of linear momentum. It states
that the rate of change of linear momentum about a fixed
point O is equal to the moment about O due to the resultant
force acting on the particle. Since this is a vector equation, it
must be satisfied for each component independently. Thus,
if the force acting on a particle is such that the component
of its moment along a given direction is zero, then the com-
ponent of the angular momentum along this direction will
remain constant. This equation is a direct consequence of
Newtons law. It will not give us more information about the
momentum of a particle, but a clever choice of coordinates
may make angular momentum easier to apply in any given
case.
About the Center of Mass
For a particle, the angular momentum is zero. We examine
carefully the expression for the rate of change of angular
momentum, equation (11.3).
Since the coordinate system moves with the particle, both ṙ
and v̇ are zero. This is true even if the coordinate system is
not inertial, in contrast to the application of Newtons Law
for linear momentum. Therefore, for a mass point, the rate
of change of angular momentum is zero in a coordinate sys-
tem moving with the particle. This implies that no moments
can be applied to a mass point in a coordinate system mov-
ing with the point. This result will later be extended to
bodies of finite size; we can apply equation (11.4) to a rigid
body (body of finite size) even in an accelerating coordinate
system if the origin of our coordinate system is the center of
mass.
Principle of Angular Impulse and Momentum
Equation (11.4) gives us the instantaneous relation between
the moment and the time rate of change of angular mo-
mentum. Imagine now that the force considered acts on a
particle between time t1 and time t2 . Equation (11.4) can
then be integrated in time to obtain
 109

Z t2 Z t2
MO dt = ḢO dt = (HO )2 (HO )1 = HO . (11.5)
t1 t1

Here, (HO )1 = HO (t1 ) and ((HO )2 = HO (t2 ). The term

Z t2
MO dt,
t1

is called the angular impulse. Thus, the angular impulse

on a particle is equal to the angular momentum change.
Equation (11.5), called the principle of angular impulse and
momentum, is often written as:

Z t2
(HO )1 + MO dt = (HO )2 , (11.6)
t1

is particularly useful when we are dealing with impulsive

forces. In such cases, it is often possible to calculate the
integrated e↵ect of a force on a particle without knowing in
detail the actual value of the force as a function of time.
Conservation of Angular Momentum
We see from equation (11.1) that if the moment of the res-
ultant force on a particle is zero during an interval of time,
then its angular momentum HO must remain constant. Con-
sider now two particles m1 and m2 which interact during an
interval of time. Assume that interaction forces between
them are the only unbalanced forces on the particles that
have a non-zero moment about a fixed point O. Let F be
the interaction force that particle m2 exerts on particle m1 .
Then, according to Newtons third law, the interaction force
that particle m1 exerts on particle m2 will be F. Using
expression (11.5), we will have that:

(HO )1 = (HO )2 . (11.7)

 110

Example 11.1

A small boy has the position and velocity indicated in

Figure 11.4 and is acted upon by the force F. Determine
the angular momentum HO about point O and the time
derivative ḢO .

SOLUTION:
The angular momentum is:

HO = r ⇥ mv
= (3i + 6j + 4k) ⇥ 2(5j)
= 40i + 30j N · m/s

From 11.4,

ḢO = MO
=r⇥F
Figure 11.4.
= (3i + 6j + 4k) ⇥ 10k
= 60i 30j N · m

As with the moments of forces, the position vector must run

from the reference point (O in this case) to the line of action
of the linear momentum mv. Here r runs directly to the
particle.