Example: (Not in lectures, still a useful illustration.

) The distance between the points A

and B is d. An aeroplane flies on a straight course from A to B and back, with constant speed
v relative to the air. There is a wind of speed w < v blowing in the direction inclined at angle
θ to AB opposing the motion. Find the total time taken to get from A to B and back.
Solution Let v(1) represent the velocity of the aeroplane relative to the ground, and let v(2)
be the velocity of the wind relative to the ground. Hence v(12) is the velocity of the aeroplane
relative to the air. Introducing an appropriate coordinate system, it follows that

v(1) = u j, where u ∈ R is unknown, and v(2) = w = w(− sin θi − cos θj),

and

v(12) = v e = v(1) − v(2) == w sin θi + (u + w cos θ)j, where e is unknown unit vector, kek = 1.

Hence, taking magnitudes gives

p
v 2 = w2 sin2 θ + (u + w cos θ)2 =⇒ u = − w cos θ ± v 2 − w2 sin2 θ.

Thus, the time taken for the forward journey is

d d
t1 = =p ,
u v 2 − w2 sin2 θ − w cos θ
taking the positive root (otherwise u < 0). For the return journey, the negative root must be
taken, so the time t2 taken on the return journey is

−d
t2 = p .
v 2 − w2 sin2 θ + w cos θ
Hence the total journey time, t, is
p
2d v 2 − w2 sin2 θ
t = t1 + t2 = .
v 2 − w2
Notice that this time t = t(θ) is minimal when θ = π/2, which physically represents a crosswind.
2

5 Laws of Motion

Up till now the equations of motion (namely, vector differential equations) have just been written
down and then solved. In this section we’ll show how they can be derived from physical “laws”.
In §5.1 we’ll look at Kepler’s laws and derive an inverse square law for the acceleration of a
planet. In §5.2 we’ll consider Newton’s laws which lead to the universal law of gravitation.

51
5.1 Kepler’s laws of planetary motion (see Anton §11.6 pp. 779–, §13.7)

In the 16th century, the astronomer Tycho Brahe made detailed observations of the motion of
the planets. His student, Johannes Kepler, analysed these and in 1609 (i.e. before the invention
of calculus) published his three laws which are valid for all planets in the solar system:

K1: Each planet moves in an ellipse with the sun at a focus.

y

A2
A1

S x

K2: Areas swept out by the radius vector from the sun to a planet in equal times are equal.

K3: The square of the period of revolution T of a planet is proportional to the cube of the
semi-major axis,
T 2 = ca3 , (5.1)

the constant of proportionality c being independent of the planet in the solar system.

Implications of the Kepler’s laws: Take the polar axes in the plane of the orbit with the
sun at the origin.
K1: In polar coordinates an ellipse is given by (§4.4)
l
r = , (0 < e < 1), (5.2)
1 + e cos θ
where the constants l and e take different values for different planets. Recall from §4.4, example
2), that the semi-major axis, a, and the semi-minor axis, b, are given by
l l
a= , b= √ , (5.3)
1 − e2 1 − e2
and that |OA| (see diagram) is the shortest distance from the sun to the orbit.
K2: Suppose in a small time increment δt, the angle θ increases by an amount δθ. Let A(t) be
the area swept by the radius vector from time t0 to t. The area swept out by the radius vector
in time δt is
1 2
r δθ + O (δθ)2


δA =
2

52
 y

l δθ
−
1−e
A
x
O l
1+e

1 2
(i.e., to first order with respect to small δθ, δA is approximated by r δθ ). Divide by δt and
2
let δt → 0, to give
dA δA 1 δθ 1 2
= lim = lim r 2 = r θ̇.
dt δt→0 δt δt→0 2 δt 2
dA
K2 implies that the rate of change of the swept out area is constant, so dropping the factor
dt
1
2 we have

r 2 θ̇ = h, a constant . (5.4)

Note that the constant h takes different values for different planets.
K3: First, note that the period of revolution T is the time taken to go from θ = 0 to θ = 2π,
i.e. for the swept out area to coincide with the area of the ellipse. Since, from the above,

dA 1
= h
dt 2

is constant, by the fundamental theorem of calculus,

1
A(T ) − A(0) = h T = area of the ellipse.
2

The area of an ellipse with semi-axes a and b is know to equal πab. [Exercise: show this in
Cartesian coordinates, cf. MA10208A.]1
1
One can also evaluate the area A of an ellipse in polar coordinates, as follows:
Zπ Zπ
1 2 l2 1
A = r dθ = dθ
2 2 (1 + e cos θ)2
−π −π

2ds 1 − s2
Now use the substitution s = tan θ/2, so dθ = , and cos θ = , to obtain
1+s 2 1 + s2
„ «
π 1+e 2
+∞ 2 1 + tan φ 1
(1 + s2 )ds 1−e
Z Z
(1 + e) 2
A = l2 = l 2 2
1 sec φ dφ
((1 + e) + (1 − e)s2 )2 (1 + e)2 sec4 φ (1 − e) 2
−∞ −π
2

53
 Hence
2πab
T = .
h
Finally, using (5.3),
4π 2 a2 b2 4π 2 l4 4π 2 l
 
2
T = = = a3 . (5.5)
h2 h2 (1 − e2 )3 h2
K3 states that T 2 = c a3 , with the constant of proportionality c being independent of the planet.
4π 2 l
Thus 2 is constant, same for all the planets. Therefore K3 implies that
h
h2
µ := , (5.6)
l

is independent of the particular planet under consideration and so is a ‘solar’ constant.

The acceleration of a planet

We now use the results we have just derived along with the theory in §4.3 on velocity and
acceleration in polar coordinates to show that K1-K3 provide a vector differential equation for
the motion of a planet. The following is the classics in mechanics, discovered by Isaac Newton:

Theorem 5.1 The acceleration of any planet is always directed towards the sun and has mag-
nitude inversely proportional to the square of the distance from the sun, with the constant of
proportionality µ (the solar constant given in (5.6)) being independent of the planet. Mathemat-
ically:
µ µ
ẍ = − 2
er = 3 x, where r = kxk, er = x/r. (5.7)
r r

Proof: In polar coordinates the acceleration of x is given by, (4.9),

ẍ = (r̈ − r θ̇ 2 )er + (r θ̈ + 2ṙ θ̇)eθ .

1 d  2  1 d
Now r θ̈ + 2ṙ θ̇ = r θ̇ = (h) = 0, since r 2 θ̇ = h is constant from (5.4). So the angular
r dt r dt
component of acceleration is zero. Next, we need to find an expression for r̈. To do this we use
1 1
where we have used the further substitution (1−e) 2 s = (1+e) 2 tan φ to get the second integral. After simplifying
the integrand,
π
Z2
l2 πl2
(1 − e) cos2 φ + (1 + e) sin2 φ dφ =
` ´
A = 3 3 = π a b.
(1 − e2 ) 2 (1 − e2 ) 2
−π
2

(Exercise: Check the details in the above calculations.) 2

54
(5.2), re-written as

l
= 1 + e cos θ.
r
l
Differentiating in t, − ṙ = − e sin θ θ̇.
r2
eh
∴ ṙ = sin θ, having used (5.4);
l
eh
∴ r̈ = cos θ θ̇
l
h2 h2

h l h
= −1 = − , having again used (5.2).
l r r2 r3 lr 2

h2
So, since (5.4) implies r θ̇ 2 = ,
r3
 2
h2 h2


2
 h
ẍ = r̈ − r θ̇ er = − 2 − 3 er
r3 lr r
h 2 h 2 µ
= − 2 er = − 3 x = − 3 x, having used (5.6), (5.8)
lr lr r
µ
and kẍk = . 2
r2
We’ll see in the next section that this theorem can be used to deduce Newton’s law of gravitation.
Newton deduced this theorem but it took him 20 years and he had to invent calculus en route!
He then studied the motion of the moons of Jupiter and Saturn and obtained the same result
but with different values for the constant µ.

5.2 Forces and Newton’s Laws

We start with some definitions and physical concepts.

(a) An inertial frame of reference (or coordinate system) is one which is fixed or is moving
with uniform velocity compared to a “universal” fixed frame of reference. We shall assume
that the sun (and sometimes the earth too) forms an inertial system. (This is a reasonable
assumption as long as speeds aren’t too large.) Newton’s Laws hold under the assumption
that all measurements or observations are made with respect to an inertial system.

(b) Any body (lump of matter) has associated with it a numerical value, called its mass. It
is an intrinsic property of the body, and is a measure of its resistance to change in its
motion (or a measure of its inertia). We’ll refer to all bodies whose size does not matter
in the considered context and which can therefore be regarded as points as particles, for
example, tennis balls, apples, satellites, planets....

55
 (c) The linear momentum of a particle is defined to be the vector p = mv = mẋ, where m
is the mass. The linear momentum is a measure of how hard it is to stop the body.

(d) Forces cause particles to move (as experience suggests). Experimental evidence sug-
gests that there are fundamental and non-fundamental forces (see Feynmann, ‘Lectures on
Physics I’ for a physical discussion). Fundamental forces are gravitational (long range),
electromagnetic (long range), and nuclear (short range). Non-fundamental forces in-
clude molecular, elastic, and frictional forces. We will be mostly considering in this course
the gravitational and frictional forces.

Newton’s laws of motion:

In an inertial frame (which is thereby postulated to exist):

N1: Any freely moving particle moves with uniform velocity.

N2: A particle of mass m subject to a total force F undergoes an acceleration a given by

F = m ẍ = m a.

N3: Action and reaction are equal in magnitude and opposite in direction. So if F12 denotes
the force exerted by particle 2 on particle 1, and F21 denotes the force exerted by particle
1 on particle 2, then
F12 = − F21 .

5.3 Newton’s law of gravitation

We now use Theorem 5.1 (derived from K1-K3) and N2-N3 to obtain Newton’s Law of Gravita-
tion (for planets). We assume the existence of an attractive force between two particles called
gravitation.
From (5.7), the acceleration of a planet in orbit round the sun is given by

µ
ẍ = − x. (5.9)
r3

Let the mass of the planet be m1 and the mass of the sun be m2 . (We’ll later use the fact that
m2 is “much bigger” (>>) than m1 , m2 >> m1 .) Now from N2, F12 , the gravitational force
exerted by the sun on the planet is

m1 µ
F12 = m1 ẍ = − x.
r3

56
 m1 F12 m2
F21
P1 P2

x1
x2

But by N3, F12 = −F21 , where F21 is the gravitational force exerted by the planet on the sun.
Now, F12 is proportional to m1 , and by ‘symmetry’, F21 must be proportional to m2 . So Newton
postulated:

Gm1 m2
F12 = (x2 − x1 ), (5.10)
kx2 − x1 k3
where G is the (universal) gravitational constant. By N2:

Gm1 m2 (x2 − x1 )
m1 ẍ1 = F12 =
kx1 − x2 k3
Gm2 (x2 − x1 )
ẍ1 =
kx2 − x1 k3
Gm2
∴ kẍ1 k = .
kx2 − x1 k2

So we can state Newton’s Law of Gravitation: “Any two particles of matter of masses
m1 and m2 respectively attract each other with a force proportional to m1 m2 and inversely
proportional to the square of the distance between them, acting along the line joining them.”

Now from N3 and N2, m1 ẍ1 = −m2 ẍ2 , and within the solar system, m2 (sun) >> m1 (any
other planet), so
m1
kẍ2 k = − ẍ1 << kẍ1 k.
m2
Therefore, to a good approximation ẍ2 = 0, and we can regard mass m2 (the sun) to be fixed
at x2 = 0, and set x = x1 − x2 to give

Gm2
ẍ = − x, r := kxk,
r3

Thus we recover (5.7) with

µ = Gm2 .

Here µ is a constant for our solar system, G is a universal constant, and m2 is the mass of the
sun.

57
5.4 Particle motion in the neighbourhood of the Earth
m

F
M
O

Consider a two-particle system with M (Earth) >> m (particle, for example, a satellite). Then
we may regard the earth as fixed and the equation of motion of the particle is

GM m
mẍ = F = − x, r = kxk.
r3
GM GM
∴ ẍ = − x, kẍk = .
r3 r2
Particle motion very close to the Earth
Assume now that the earth is a sphere of radius R, and that the particle is sufficiently close to
the surface of the Earth so we may take kxk = R. So, to a good approximation,

GM m GM m
F = mẍ = − 3
x = − x̂,
R R2
x x
where x̂ = = , is the unit vector pointing ‘away’ from the centre of the earth. Introduce
kxk R
GM
the scalar g := 2 , and the vector g := − gx̂. Then, for a particle very close to the earth, the
R
acceleration is given by
ẍ = g (‘free fall’ acceleration).

Definition 5.2 The weight of a particle close to the surface of the earth is defined to be F =
mg.

Important Note: The above discussion regards the earth as a particle. This seems reasonable
for planetary motion (that is, when kxk >> R). However, Newton showed that even near the
surface, the gravitational attraction is the same as if all Earth’s mass were concentrated at its
centre, assuming the Earth has a spherically symmetric mass distribution.

6 Projectiles

In this section we will use the Newton’s second law (N2) to derive a vector ODE and then solve it
and analyse the solution, to solve projectile problems. We first analyse the motion of projectiles

58