Book A
Book A
Book A
Cover image: This shows arrows representing a vector field and a corresponding contour map. The
vector field shows the direction of change for a model of two competing populations of animals. You
will meet this model in Unit 12.
This publication forms part of an Open University module. Details of this and other
Open University modules can be obtained from the Student Registration and Enquiry Service, The
Open University, PO Box 197, Milton Keynes MK7 6BJ, United Kingdom (tel. +44 (0)845 300 6090;
email [email protected]).
Alternatively, you may visit the Open University website at www.open.ac.uk where you can learn
more about the wide range of modules and packs offered at all levels by The Open University.
To purchase a selection of Open University materials visit www.ouw.co.uk, or contact Open
University Worldwide, Walton Hall, Milton Keynes MK7 6AA, United Kingdom for a brochure
(tel. +44 (0)1908 858779; fax +44 (0)1908 858787; email [email protected]).
Contents
Unit 1 First- and second-order differential equations 1
Introduction 3
Solutions to exercises 67
Acknowledgements 91
Introduction 95
Index 255
Unit 1
Introduction
An important class of equations that arise in mathematics consists of those
that feature the rates of change of one or more variables with respect to
one or more others. These rates of change are expressed mathematically by
derivatives, and the corresponding equations are called differential
equations. Equations of this type crop up in a wide variety of situations.
They are found, for example, in models of physical, electronic, economic,
demographic and biological phenomena.
First-order differential equations, which are the particular topic of
Section 1, feature derivatives of order one only; that is, if the rate of
change of variable y with respect to variable x is involved, then the
equations feature dy/dx but not d2 y/dx2 , d3 y/dx3 , etc.
When a differential equation arises, it is usually an important aim to solve
the equation. For an equation that features the derivative dy/dx, this
entails expressing the dependent variable y directly in terms of the
independent variable x. You will see four possible approaches to finding a
solution in Section 1.
This unit also considers second-order differential equations, that is,
differential equations that involve a second (but no higher) derivative.
Examples of second-order differential equations are A second-order differential
equation may or may not
d2 y dy d2 y include a first derivative.
−3 + 2y = 4ex and 3 + y 2 sin x = x2 .
dx2 dx dx2
3
Unit 1 First- and second-order differential equations
This principle applies to any quantity whose change, over a given time
interval, is due solely to the specified input and output.
The accumulation δP of population over the time interval δt is the
population at the end of the interval minus the population at the start of
the interval, that is,
δP = P (t + δt) − P (t).
The input is the number of births (equation (1)), and the output is the
number of deaths (equation (2)). The input–output principle now enables
us to express the accumulation δP of the population over the time
4
1 First-order differential equations
interval δt as
δP ( b P (t) δt − c P (t) δt = (b − c) P (t) δt.
Dividing through by δt, we obtain
δP
( (b − c) P (t).
δt
The approximations involved in deriving this equation become
progressively more accurate for shorter time intervals. So, finally, by
letting δt tend to zero, we obtain This is the step that requires P
to be a continuous (rather than
dP
= (b − c) P (t). discrete) function of t.
dt
(This follows because
dP P (t + δt) − P (t)
= lim
dt δt→0 δt
is the definition of the derivative of P .)
This is a differential equation because it describes dP/dt rather than the
eventual object of our interest (which is P itself). The purpose of this unit
is to enable you to solve a wide variety of such equations.
We can simplify the above equation slightly by using the proportionate
growth rate r, which is the difference between the proportionate birth
and death rates: r = b − c. Then our model becomes
dP
= rP. (3)
dt P
For very simple population models, r is taken to be a constant. As we will
see, this leads to a prediction of exponential growth (or, if r < 0, decay) in
population size with time, as illustrated in Figure 1. This may be a very
good approximation for certain populations, but it cannot be sustained r>0
indefinitely if r > 0.
In practice, both the proportionate birth rate and the proportionate death t
rate will vary, and so therefore will the proportionate growth rate. It turns
out to be convenient to model these changes as being dependent on the
population size, so the proportionate growth rate r becomes a function Figure 1 Population growth
of P . The justification for this is as follows. When the population is low,
one may assume that there is potential for it to grow (assuming a
reasonable environment). The proportionate growth rate should therefore
be high. However, as the population grows, there will be competition for
resources. Thus the proportionate growth rate will decline, and in this way
unlimited (exponential) growth does not occur.
A particularly useful model arises from taking r(P ) to be a decreasing
linear function of P . We write this as
' -
P
r(P ) = k 1 − , (4) You will see later why this
M particular form is chosen.
where k and M are positive constants. Looking at this formula, you can
see that the proportionate growth rate r decreases linearly with P , from
the value k (when P = 0) to the value 0 (when P = M ).
5
Unit 1 First- and second-order differential equations
Exercise 1
Suppose that a population obeys the logistic model (with the
proportionate growth rate given by equation (4)), and that you are given
the following information. When P = 10 the proportionate growth rate
is 1, and when P = 10 000 the proportionate growth rate is 0.
Find the corresponding values of k and M .
6
1 First-order differential equations
and f denotes the function that relates them), we could write y = y(t), Strictly speaking, this is an
which is read as ‘y is a function of t’. abuse of notation, since there is
ambiguity as to exactly what the
The following definitions explain just what are meant by a differential symbol y represents: it is a
equation, by the order of such an equation, and by a solution of it. variable on the left-hand side of
y = y(t), but a function on the
right-hand side. However, it is a
A differential equation for y = y(x) is an equation that relates the very convenient abuse.
independent variable x, the dependent variable y, and one or more
derivatives of y.
The order of such a differential equation is the order of the highest
derivative that appears in the equation. Thus a first-order
differential equation for y = y(x) features only the first derivative,
dy/dx.
A solution of such a differential equation is a function y = y(x) that
satisfies the differential equation.
Exercise 2
Verify that each of the functions given below is a solution of the
corresponding differential equation.
dy
(a) y = 2ex − (x2 + 2x + 2); = y + x2 .
dx
dy
(b) y = tan x + sec x; = y tan x + 1.
dx
(c) y = t + Ce−t ; ẏ = −y + t + 1. (Here C is an arbitrary constant.)
7
Unit 1 First- and second-order differential equations
Exercise 3
(a) Verify that for any value of the constant C, the function
y = C − 13 e−3x is a solution of the differential equation
dy
= e−3x .
dx
(b) Verify that for any value of the constant C, the function
CM ekt
P =
1 + Cekt
See equation (5). is a solution of the logistic equation
' -
dP P
= kP 1 − .
dt M
8
1 First-order differential equations
The word ‘initial’ in these definitions arises from those (frequent) cases in
which the independent variable represents time. In such cases, the
differential equation describes how the system being modelled behaves
once started, while the initial condition specifies the configuration with
which the system is started off. In fact, if the initial condition is
y(x0 ) = y0 , then we are often interested in solving the corresponding
initial-value problem for x > x0 . If x represents time, then x > x0
is ‘the future’ after the system
We usually require that an initial-value problem should have a unique has been started off.
solution, since then the outcome is completely determined by how the
system behaves and its configuration at the start. Almost all the
differential equations in this module do have unique solutions, and we will
assume that all the initial-value problems in this unit have unique
solutions.
Example 1
Using the result given in Exercise 3(a), solve the initial-value problem
dy
= e−3x , y(0) = 1.
dx
Solution
From Exercise 3(a), a solution of the differential equation is In fact, as will be shown in
1 −3x
Example 2, this is the general
y=C− 3e , solution.
where C is a constant.
The initial condition says that y = 1 when x = 0, and on feeding these
values into the above solution we find that
1 = C − 13 .
9
Unit 1 First- and second-order differential equations
Exercise 4
The size of a population (measured in units of hundreds of thousands) is
modelled by the logistic equation
' -
dP P
= kP 1 − , P (0) = 1,
dt M
where k = 0.15 and M = 10.
(a) Use your answer to Exercise 3(b) to solve this initial-value problem.
(b) Can you predict the long-term behaviour of the population size from
your answer?
Finally in this subsection, note that one needs to keep an eye on the
domain of the function defining the differential equation. ‘Gaps’ in the
domain usually show up as some form of restriction on the nature of a
solution curve. For example, consider the differential equation
dy 1
= . (6)
dx x
It turns out that there are two distinct families of solutions of this
Since |x| = −x if x < 0, we can equation, given by y = ln x + C (if x > 0) and y = ln(−x) + C (if x < 0).
write
( These two families of solutions are illustrated in Figure 2. Notice that the
1
dx = ln |x|. right-hand side of equation (6) is not defined at x = 0, and that there is no
x solution that crosses the y-axis.
y
4
2
−10 −5 −2 5 10 x
−4
−6
10
1 First-order differential equations
Example 2
(a) Find the general solution of the differential equation
dy
= e−3x .
dx
(b) Find the particular solution of this differential equation that satisfies
the initial condition y(0) = 53 .
Solution
(a) On applying direct integration, we obtain the general solution
(
y = e−3x dx = − 13 e−3x + C,
Once the general solution has been found, it is possible to single out a
particular solution by specifying a value for the constant C. This value
may be found by applying an initial condition.
11
Unit 1 First- and second-order differential equations
Exercise 5
Solve each of the following initial-value problems.
dy
(a) = 6x, y(1) = 5.
dx
dv
(b) = e4u , v(0) = 2.
du
t
(c) ẏ = , y(0) = 2.
1 + t2
(Hint: For the integral, try the substitution u = 1 + t2 .)
The proof of this result involves differentiating the solution and showing
that y satisfies the differential equation.
In Exercise 5(c) we had f (t) = 1 + t2 and f ! (t) = 2t, with k = 21 . The
initial right-hand side t/(1 + t2 ) had to be manipulated slightly to get it
into the right form. Spotting integrands of this form (or of this form apart
from a constant multiple) can allow you to solve some quite tricky-looking
problems.
Exercise 6
Find the general solution of each of the following differential equations,
where a is a non-zero constant.
dy 1
(a) = (u =
) a)
du u−a
dy 1
(b) = (x )= 0, x )= 1/a)
dx x(1 − ax)
+ $
1 1 a
Hint: First verify that = + .
x(1 − ax) x 1 − ax
12
1 First-order differential equations
Example 3
(a) Find the general solution of the differential equation
dy 3x2 y
= (y > 0).
dx 4 + x3
(b) Find the general solution of the differential equation
dy x
=− (y > 0),
dx 3y
and the particular solution that satisfies the initial condition y(0) = 3.
Solution
(a) The equation is of the form
dy
= g(x) h(y),
dx
where the obvious choices for g and h are
3x2
g(x) = and h(y) = y.
4 + x3
Dividing both sides of the given differential equation by y gives
1 dy 3x2
= .
y dx 4 + x3
Integrating both sides with respect to x gives
( ( (
1 dy 1 3x2
dx = dy = dx,
y dx y 4 + x3
where the second integral comes from applying the rule for integration
by substitution.
13
Unit 1 First- and second-order differential equations
Spotting that the right-hand integrand is of the form f ! (x)/f (x) with
f (x) = 4 + x3 , we have
ln |y| = ln |4 + x3 | + C,
where the arbitrary constants associated with the two indefinite
integrals have been lumped into a single arbitrary constant C.
The solution is not yet in the explicit form y = F (x) for some
function F . If we write C = ln A (A > 0), then the right-hand side
becomes ln |4 + x3 | + ln A = ln(A|4 + x3 |), and we can take
exponentials of both sides to give the explicit form of the solution as
If necessary, A can be negative y = A(4 + x3 ).
to take into account changes in
sign for y or (4 + x3 ). (b) The equation is of the form
dy
= g(x) h(y),
dx
where
Notice that since y > 0, h(y) is g(x) = −x and h(y) = 1/(3y).
never zero.
On dividing through by h(y) = 1/(3y) (i.e. multiplying through by 3y)
and integrating with respect to x, the differential equation becomes
( (
3y dy = −x dx.
14
1 First-order differential equations
Exercise 7
A mass m(t) of a uranium isotope, which is present in an object at time t,
declines over time due to radioactive decay. Its behaviour is modelled by This model can be applied to
the differential equation other radioactive substances by
selecting the appropriate value
dm of the parameter λ.
= −λm (m > 0),
dt Note the condition m > 0. You
where the decay constant λ is a positive constant characteristic of the can see that m = 0 also satisfies
uranium isotope. the differential equation.
(a) Find the general solution of this differential equation.
(b) Find the particular solution for which the initial amount of uranium
present (at time t = 0) is m0 .
The condition m > 0 in Exercise 7 arose from the modelling context. This
condition enabled us to find the general solution without needing to worry
about dividing by zero at Step 1 of the separation of variables method
(and hence without needing to restrict the image set further).
15
Unit 1 First- and second-order differential equations
Exercise 8
Find the general solution of each of the following differential equations.
dy y−1 dy 2y
(a) = (x > 0) (b) = 2
dx x dx x +1
Exercise 9
The differential equation here is (a) Solve the initial-value problem
the logistic equation (5), which, ' -
as was pointed out earlier, may dP P
= kP 1 − , P (0) = P0 (where P0 > 0),
be used as a model for the size dt M
P (t) of a population at time t.
where k and M are positive constants.
(Hint: For the integral involving P , the solution to Exercise 6(b)
should be useful.)
(b) Describe what happens to the solution P (t) as t becomes large.
Exercise 10
Decide whether or not each of the following first-order differential
equations is linear.
dy
(a) = x sin x (b) ẏ + y 2 = t
dx
dy dy
(c) x + y = y2 (d) (1 + x2 ) + 2xy = 3x2
dx dx
Theorem 1
If the functions g(x) and h(x) are continuous throughout an interval
(a, b) and x0 belongs to this interval, then the initial-value problem This includes the possibility that
either a = −∞ or b = ∞, so the
dy interval might be all of the real
+ g(x) y = h(x), y(x0 ) = y0 ,
dx line.
has a unique solution throughout the interval.
This is a very powerful result, since it means that once you have found a
solution in a particular interval, that solution will be the only one.
The integrating factor method is a technique for solving linear differential
equations. It derives from the rule for integration by parts or, equivalently,
from the product rule for derivatives. To introduce the topic, consider the
differential equation
dy
(1 + x2 ) + 2xy = 3x2 . (10) As you saw in Exercise 10(d),
dx this differential equation is
Note first that 2x (the coefficient of y) is the derivative of 1 + x2 (the linear; but it is not soluble by
coefficient of dy/dx). It follows from the product rule that direct integration or by
separation of variables.
d% # dy
(1 + x2 )y = (1 + x2 ) + 2xy.
dx dx
The right-hand side of this equation is the same as the left-hand side of
equation (10), so we can rewrite the latter as
d% #
(1 + x2 )y = 3x2 . (11)
dx
17
Unit 1 First- and second-order differential equations
Now the left-hand side here is just the derivative of (1 + x2 )y, so we can
apply direct integration to equation (11) to obtain
(
(1 + x )y = 3x2 dx = x3 + C,
2
18
1 First-order differential equations
so
'( -
p = exp g(x) dx ,
2. Multiply equation (16) by p(x) to recast the differential equation You can, if you wish, check that
as you have found p correctly by
checking that
dy
p(x) + p(x) g(x) y = p(x) h(x). dy
dx p(x) + p(x) g(x) y
dx
3. Rewrite the differential equation as d
= (p(x) y) ,
d dx
(p(x) y) = p(x) h(x). i.e. by checking that
dx dp/dx = p(x) g(x).
4. Integrate this last equation to obtain
( The integral in Step 4 will
p(x) y = p(x) h(x) dx. involve an arbitrary constant C.
5. Divide through by p(x) to obtain the general solution in explicit It is a good idea to check, by
form. substitution into the original
equation, that the function
obtained is indeed a solution.
19
Unit 1 First- and second-order differential equations
Exercise 11
Find the general solution of each of the following differential equations.
dy dy y−1
(a) − y = ex sin x (b) = (x > 0)
dx dx x
Exercise 12
Which method would you use to try to solve each of the following linear
first-order differential equations?
dy dy
(a) + x3 y = x5 (b) = x sin x
dx dx
dv dy
(c) + 5v = 0 (d) (1 + x2 ) + 2xy = 1 + x2
du dx
20
1 First-order differential equations
Example 4
(a) Part of the direction field for the logistic equation
' -
dP P
=P 1− This is equation (18) with k = 1
dt 1000 and M = 1000.
is sketched in Figure 4. Using this diagram, sketch the solution curves
that pass through the points
(0, 1500), (0, 1000), (0, 100), (0, 0), (0, −100). We do not normally need to
consider P < 0 since populations
From your results, describe the graphs of particular solutions of the must be non-negative.
differential equation.
21
Unit 1 First- and second-order differential equations
1500
1000
500
0
2 4 6 8 10 t
−500
(b) What does your answer to part (a) tell you about the predicted
behaviour of a population whose size P (t) at time t is modelled by this
logistic equation?
Solution
(a) The slope is shown to be zero at all points on the horizontal lines
P = 0 and P = 1000, so these correspond to constant solutions of the
differential equation. (As pointed out earlier in the text, these two
solutions can also be spotted directly from the form of the differential
equation.)
The graphs of solutions through starting points above the line
P = 1000 appear to decrease, but at a slower and slower rate, tending
from above towards the limit P = 1000 as t increases.
The graphs of solutions through starting points in the region
0 < P < 1000 are increasing, with slope growing before the level
P = 500 is reached and declining thereafter. For large values of t,
these graphs tend from below towards the limit P = 1000.
For starting points in the region P < 0, the graphs decrease without
limit and with steeper and steeper slope.
These various cases are illustrated by typical graphs in Figure 5.
22
1 First-order differential equations
1500
1000
500
0
2 4 6 8 10 t
−500
23
Unit 1 First- and second-order differential equations
slope = f (x0 , y0 )
P1 (x1 , Y1 )
Y1 − y 0
P0
(x0 , y0 )
h
0 x0 x1 x
The idea is that the point P1 , whose coordinates have been denoted by
The reason for using Y1 here, (x1 , Y1 ), provides an approximate value Y1 of the solution function y(x) at
rather than y1 , will be explained x = x1 . Now, unless the solution function follows a straight line as x moves
shortly. from x0 to x1 , Y1 is unlikely to give the exact value of y(x1 ). However, the
hope is that because we headed off from x0 along the correct slope, as
given by the direction field, Y1 will be reasonably close to the exact value.
Before worrying about accuracy, let us continue with the construction of
the points in our sequence.
The next thing that we need to do, before proceeding to the second step in
the construction process, is determine formulas for x1 and Y1 in terms of
x0 , y0 , h and f (x0 , y0 ). By the construction described, as the point P1 is
reached from P0 by taking a step to the right of horizontal length h, we
have
x1 = x0 + h. (21)
24
1 First-order differential equations
Y2 − Y1
(x1 , Y1 ) P1
P0 h
0 x0 x1 x2 x
25
Unit 1 First- and second-order differential equations
Yi+1 − Yi
(xi , Yi ) Pi
h
0 xi xi+1 x
P3
y = y(x)
P2
P0 P1
0 x0 x1 x2 x3 x4 x
Nevertheless, the formulas (25) and (26) provide a method for finding
approximate solutions to the initial-value problem (20), in terms of The accuracy of such
numerical estimates Y1 , Y2 , Y3 , . . . at the respective domain values approximate solutions, and ways
x1 , x2 , x3 , . . .. This is called Euler’s method, after Leonhard Euler of improving accuracy, will be
considered shortly.
(Figure 10). It is summarised below.
27
Unit 1 First- and second-order differential equations
Example 5
Consider the initial-value problem
dy
= x + y, y(0) = 1.
dx
Use Euler’s method, with step size h = 0.2, to obtain an approximation
to y(1).
Solution
We have x0 = 0, Y0 = y0 = 1, and f (xi , Yi ) = xi + Yi . The step size is
given as h = 0.2. Equation (25) with i = 0 gives
x1 = x0 + h = 0 + 0.2 = 0.2,
and equation (26) with i = 0 gives
Y1 = Y0 + h f (x0 , Y0 ) = 1 + 0.2 × (0 + 1) = 1.2.
For the second step, we have (from equation (25) with i = 1)
x2 = x1 + h = 0.2 + 0.2 = 0.4,
and (from equation (26) with i = 1)
Y2 = Y1 + h f (x1 , Y1 ) = 1.2 + 0.2 × (0.2 + 1.2) = 1.48.
If more than a couple of steps of such a calculation have to be computed
by hand, then it is a good idea to lay out the calculation as a table. In this
case, by continuing as above and putting i in turn equal to 2, 3 and 4, we
obtain Table 1.
Table 1
i xi Yi f (xi , Yi ) = xi + Yi Yi+1 = Yi + h f (xi , Yi )
0 0 1 1 1.2
1 0.2 1.2 1.4 1.48
2 0.4 1.48 1.88 1.856
3 0.6 1.856 2.456 2.347 2
4 0.8 2.347 2 3.147 2 2.976 64
5 1.0 2.976 64
(After each value of Yi+1 has been calculated from the formula and entered
in the last column, it is transferred to the Yi column in the next row. Once
the value at x = 1.0 has been found, no further calculations are necessary.)
So at x = 1, Euler’s method with step size h = 0.2 gives the approximation
y(1) ( 2.976 64.
28
1 First-order differential equations
Exercise 13
Use Euler’s method, with step size h = 0.2, to obtain an approximation to
y(1) for the initial-value problem
dy
= y, y(0) = 1.
dx
29
Unit 1 First- and second-order differential equations
y
estimate
with h = 0.4
with h = 0.2
y = y(x)
with h = 0.1
exact solution
at x = 0.4
In fact, it can be shown that the accuracy of Euler’s method does indeed
usually improve when we take a smaller step size.
To demonstrate this, consider the initial-value problem from Exercise 13.
This has the exact solution y = ex (as you can verify), and its value at
x = 1 is y(1) = e = 2.718 282, to six decimal places. In Exercise 13 you
showed that with a step size h = 0.2, Euler’s method gives the
approximation 2.488 32 for y(1). Table 2 shows the corresponding results
(to six decimal places) obtained when we apply Euler’s method to this
In Exercise 13, where h = 0.2, same initial-value problem but with successively smaller step sizes h.
the value of y(1) was
approximated by Y5 . From the Table 2
column for ‘Number of steps’ in
Table 2, you can see that y(1) is h Approximation Absolute Number of
approximated by: to y(1) error steps
Y10 when h = 0.1;
Y100 when h = 0.01; 0.1 2.593 742 0.124 539 10
Y1000 when h = 0.001; 0.01 2.704 814 0.013 468 100
Y10000 when h = 0.0001. 0.001 2.716 924 0.001 358 1000
0.0001 2.718 146 0.000 136 10000
As expected, the absolute errors in the third column of the table become
progressively smaller as h is reduced.
Looking more carefully at these absolute errors, we notice that they seem
to tend towards a sequence in which each number is a tenth of the previous
one. Since each value of h in the table is a tenth of the previous one, this
suggests that
absolute error is approximately proportional to step size h.
This turns out to be a general property of Euler’s method, for sufficiently
small values of the step size. So not only do we know that accuracy can be
30
1 First-order differential equations
improved by decreasing the step size h, but this general property also tells
us that by making h small enough, the absolute error in an approximation
can be made as small as desired. In other words, the absolute error
approaches the limit zero as h approaches zero (as you might have
expected from the intuitive argument preceding Figure 11).
Exercise 14
Suppose that when Euler’s method is applied to the problem in
Exercise 13, the absolute error in approximating y(1) is proportional to the
step size h, for sufficiently small h.
Use the last row of Table 2 to estimate the constant of proportionality,
k say, and hence estimate the step size required to compute y(1) correct to
five decimal places (i.e. so that the absolute error is less than 5 × 10−6 ).
A few words of caution are necessary at this point. Although the absolute
error can be made as small as we please by making the step size h
sufficiently small, this is valid only if the arithmetic is performed using
sufficient decimal places. Where a calculator or computer is involved, the
number of decimal places that can be used is limited, and as a result
rounding errors may be introduced into the calculations. After a certain
point, any increase in accuracy brought about by reducing the size of h
may be swamped by these rounding errors.
Moreover, rounding errors are not the only problem. Before concluding
that h should always be chosen to be very small, we must also consider the
cost of this additional accuracy. Now, by cost is meant the effort involved,
which can be measured in a variety of ways; commonly for iterative
methods (such as Euler’s method) it is measured by the number of steps
taken. In general for numerical methods, the greater the accuracy
required, the greater the cost. To illustrate this, look back at Table 2. The
last column of the table shows how the number of steps required for the
calculation goes up in inverse proportion to the step size: to move from In general, to move from a to b
x = 0 to x = 1, it takes 10 steps with step size h = 0.1, 100 steps with step (where b > a) with step size h
size h = 0.01, and so on. Since, for sufficiently small h, the error in Euler’s takes (b − a)/h steps.
method is approximately proportional to the step size, it follows that for
this method a tenfold improvement in accuracy is paid for by a tenfold
increase in the number of steps required.
So for Euler’s method and similar methods, the choice of step size has to
be based on a compromise between the two opposing requirements of
accuracy and cost. There are better numerical methods for solving
initial-value problems that are considerably more efficient than Euler’s Greater efficiency means that
method. In fact, Euler’s method is not suitable for high-accuracy work. Its the same or better numerical
virtue lies rather in its simplicity and its clear illustration of the basic accuracy is achieved with fewer
numerical computations.
principles of how differential equations may be solved numerically.
In any practical case, calculations of the type described in this subsection
are ideally suited to being performed on a computer.
31
Unit 1 First- and second-order differential equations
Exercise 15
(a) Without plotting the direction field, say what you can about the
slopes defined by the differential equation
dy
= f (x, y) = y + x2 .
dx
(b) Verify that your conclusions are consistent with the direction field
diagram in Figure 12.
y
2
−2 −1 0 1 2 x
−1
−2
(c) On the basis of the direction field, what can be said about the graphs
of solutions of the differential equation?
(d) Write down the formulas required in order to apply Euler’s method to
the initial-value problem
dy
= y + x2 , y(−1) = −0.2,
dx
using a step size h = 0.1.
32
2 Homogeneous second-order differential equations
2 Homogeneous second-order
differential equations
You will recall that a particular solution of a first-order differential
equation is obtained by applying a single condition (known as an initial
condition) to the general solution in order to find a particular value of the
single arbitrary constant. In the case of a second-order differential
equation, a particular solution is obtained by applying two conditions to
the general solution in order to find particular values of the two arbitrary
constants. The following example illustrates this.
Example 6
Suppose that a car is travelling with constant acceleration a along a
straight road. If, at time t, its distance from a fixed point is s, then its
velocity is given by ds/dt, its acceleration is given by d2 s/dt2 , and its
motion is modelled by
d2 s
= a. (27)
dt2
If the car is initially stationary at position s = 0 and thereafter has a
constant acceleration of 2 m s−2 , how long does it take for the car to attain
a velocity of 30 m s−1 , and what distance has it travelled in that time?
Solution
Integrating equation (27) leads to
ds
= at + C and s = 12 at2 + Ct + D,
dt
where C and D are arbitrary constants. To find these constants (and
hence answer the questions asked), we need to make use of the conditions
given. These are that the car is initially stationary (i.e. ds/dt = 0 when
t = 0) at position s = 0 (i.e. s = 0 when t = 0). The first of these
conditions together with the equation ds/dt = at + C tells us that C = 0.
With C = 0, the second equation becomes s = 21 at2 + D, and this together
with the second condition tells us that D = 0.
Therefore when a = 2, we have In mathematical modelling, the
parameters of the model, in this
d2 s ds case a, are retained for as long
= 2, = 2t, s = t2 .
dt2 dt as possible, in order to generate
So the velocity is ds/dt = 30 when 2t = 30, that is, after 15 seconds, and in results that could be used in
other situations, with different
this time the car has travelled a distance s = 152 , that is, 225 metres. values.
Compare the definitions for A second-order differential equation for y = y(x) is linear if it can be
first-order equations in expressed in the form
Subsection 1.4. The important
feature is the linear combination d2 y dy
of y and its derivatives on the
a(x) 2
+ b(x) + c(x) y = f (x), (28)
dx dx
left-hand side.
where a(x), b(x), c(x) and f (x) are given continuous functions.
A linear second-order differential equation is constant-coefficient if
the functions a(x), b(x) and c(x) are all constant, so that the
equation is of the form
d2 y dy
a +b + cy = f (x), (29)
dx2 dx
If a = 0, then the equation is where a )= 0.
first-order.
A linear constant-coefficient second-order differential equation is said
to be homogeneous if f (x) = 0 for all x, and inhomogeneous (or
non-homogeneous) otherwise.
34
2 Homogeneous second-order differential equations
Exercise 16
Consider the following second-order differential equations.
d2 y d2 y dy d2 y dy
(i) = x2 (ii) 3 + 4 + y = x 2
(iii) 3 +4 +y =0
dx2 dx2 dx dx2 dx
d2 y dy
(iv) xy !! + x2 y = 0 (v) 2y 2 + 4y = 3
dx dx
d2 t dt
(vi) 2 2 + 3 + 4t = sin θ (vii) ẍ = −4t (viii) ẍ = −4x
dθ dθ
(a) Which of the equations are both linear and constant-coefficient?
(b) Which of the linear constant-coefficient equations are homogeneous?
(c) For each equation, identify the dependent and independent variables.
Principle of superposition
A key theoretical result will turn out to be extremely useful throughout
this module. This is known as the principle of superposition, and it is a
fundamental property of linear differential equations.
Suppose that we have a solution y1 (x) of
d2 y dy
a 2
+b + cy = f1 (x), Here a, b and c can be functions
dx dx of x.
and a solution y2 (x) of
d2 y dy
a 2
+b + cy = f2 (x).
dx dx
Then we claim that the linear combination k1 y1 + k2 y2 , where k1 and k2
are constants, is a solution of
d2 y dy
a +b + cy = k1 f1 (x) + k2 f2 (x). (30)
dx2 dx
In fact, this is easy to see, for if we substitute k1 y1 + k2 y2 directly into
equation (30), we obtain
d2 d
a 2
(k1 y1 + k2 y2 ) + b (k1 y1 + k2 y2 ) + c (k1 y1 + k2 y2 )
dx ' dx- ' -
d2 y1 d2 y2 dy1 dy2
= a k1 2 + k2 2 + b k1 + k2 + c (k1 y1 + k2 y2 )
dx dx dx dx
' 2 - ' 2 -
d y1 dy1 d y2 dy2
= k1 a 2 + b + cy1 + k2 a 2 + b + cy2
dx dx dx dx
= k1 f1 (x) + k2 f2 (x),
as required.
35
Unit 1 First- and second-order differential equations
36
2 Homogeneous second-order differential equations
dy/dx = λAeλx and d2 y/dx2 = λ2 Aeλx , and substituting into the left-hand
side of equation (31) gives
d2 y dy
a 2
+b + cy = aλ2 Aeλx + bλAeλx + cAeλx
dx dx
= (aλ2 + bλ + c)Aeλx .
Hence y = Aeλx is indeed a solution of equation (31), for any value of A,
provided that λ satisfies Note that the discussion here
2 applies irrespective of whether λ
aλ + bλ + c = 0. is real or complex. The
This equation plays such an important role in solving linear consequences of λ being complex
are explained later.
constant-coefficient second-order differential equations that it is given a
special name.
Example 7
Write down the auxiliary equation of the differential equation
d2 y dy
3 2
−2 + 4y = 0.
dx dx
Solution
The auxiliary equation is
3λ2 − 2λ + 4 = 0.
Exercise 17
Write down the auxiliary equation of each of the following differential
equations.
d2 y dy
(a) 2
−5 + 6y = 0 (b) y !! − 9y = 0 (c) ẍ + 2ẋ = 0
dx dx
37
Unit 1 First- and second-order differential equations
Example 8
(a) Write down the auxiliary equation of the differential equation
d2 y dy
2
−3 + 2y = 0,
dx dx
and find its roots λ1 and λ2 .
(b) Deduce that y1 = Cex and y2 = De2x are both solutions of the
differential equation, for any values of the two constants C and D.
(c) Show that y = Cex + De2x is also a solution of the differential
equation, for any values of the two constants C and D.
Solution
(a) The auxiliary equation is
λ2 − 3λ + 2 = 0.
Using the formula This equation may be solved, for example, by factorising in the form
√
−b ± b2 − 4ac (λ − 1)(λ − 2) = 0, to give the two roots λ1 = 1 and λ2 = 2.
λ1 , λ2 =
2a (b) Since λ1 = 1 and λ2 = 2 are the roots of the auxiliary equation,
produces the same answer.
y1 = Cex and y2 = De2x are solutions of the differential equation, for
It does not matter which of the
roots is called λ1 and which is any values of C and D.
called λ2 . (c) To show that y = Cex + De2x is a solution of the differential equation,
we differentiate and substitute into the differential equation.
Differentiating to obtain the first and second derivatives of y gives
dy d2 y
= Cex + 2De2x and = Cex + 4De2x .
dx dx2
Substituting these into the left-hand side of the differential equation
gives
d2 y dy
−3 + 2y
dx2 % dx # % # % #
= Cex + 4De2x − 3 Cex + 2De2x + 2 Cex + De2x
= C(1 − 3 + 2)ex + D(4 − 6 + 2)e2x
= 0.
Hence y = Cex + De2x is a solution of the differential equation, for any
values of C and D.
38
2 Homogeneous second-order differential equations
It can be seen that if λ1 and λ2 are distinct roots of the auxiliary equation
of a homogeneous linear constant-coefficient second-order differential
equation, then any solution of the form
y = Ceλ1 x + Deλ2 x , (34)
for some choice of constants C and D, is also a solution. Furthermore, it
can be shown that a solution of this form, for distinct roots of the auxiliary
equation, and involving two arbitrary constants, is the general solution of
the homogeneous linear second-order differential equation.
Exercise 18
Use the auxiliary equation to find the general solution of each of the
following differential equations.
d2 y dy d2 y dy d2 z
(a) 2
+5 + 6y = 0 (b) 2 2
+3 =0 (c) − 4z = 0
dx dx dx dx du2
We now consider an example where the two roots of the auxiliary equation
are equal, in which case the above method does not work! Indeed, in light
of the earlier discussion, you might expect the solution always to be of the
form y = Aeλ1 x + Beλ2 x , where A and B are arbitrary constants. But if
λ1 = λ2 , this reduces to y = (A + B)eλ1 x = Ceλ1 x , where C = A + B is a
single arbitrary constant, so this cannot be the general solution of a
second -order differential equation, which requires two arbitrary constants.
Example 9
(a) Write down the auxiliary equation of the differential equation
d2 y dy
+6 + 9y = 0,
dx2 dx
and find its roots λ1 and λ2 .
(b) Deduce that y1 = Ce−3x is a solution of the differential equation, for
any value of the constant C.
(c) Show that y2 = Dxe−3x is also a solution, for any value of the
constant D.
(d) Deduce that y = (C + Dx)e−3x is also a solution of the differential
equation, for any values of the two constants C and D.
Solution
(a) The auxiliary equation is
λ2 + 6λ + 9 = 0.
The left-hand side is the perfect square (λ + 3)2 , so the auxiliary
equation has equal roots λ1 = λ2 = −3.
39
Unit 1 First- and second-order differential equations
Note that the ‘other’ root (b) Since λ1 = −3 is a root of the auxiliary equation, y1 = Ce−3x is a
λ2 = −3 gives the same solution. solution of the differential equation, for any value of C.
(c) To show that y2 = Dxe−3x is a solution of the differential equation, we
differentiate and substitute into the differential equation.
Differentiating to obtain the first and second derivatives of y2 gives
dy2 % #
Here we are using the product = De−3x + Dx −3e−3x = D(1 − 3x)e−3x ,
rule for differentiation. dx
d2 y2
= −3De−3x + D(1 − 3x)(−3e−3x ) = D(−6 + 9x)e−3x .
dx2
Substituting these into the left-hand side of the differential equation
gives
d2 y2 dy2
2
+6 + 9y2 = D(−6 + 9x)e−3x + 6D(1 − 3x)e−3x + 9Dxe−3x
dx dx
= D(−6 + 6)e−3x + D(9 − 18 + 9)xe−3x
= 0.
Hence y2 = Dxe−3x is a solution of the differential equation, for any
value of D.
(d) Since y1 = Ce−3x and y2 = Dxe−3x are both solutions of the
differential equation, the principle of superposition tells us that so is
y = Ce−3x + Dxe−3x = (C + Dx)e−3x , for any values of C and D.
40
2 Homogeneous second-order differential equations
Exercise 19
Use the auxiliary equation to find the general solution of the following
differential equations.
d2 y dy
(a) 2
+2 +y =0 (b) s̈ − 4ṡ + 4s = 0
dx dx
Equations (34) and (37) give us the general solution of equation (35) for
the cases where the roots λ1 and λ2 of the auxiliary equation are distinct
and equal, respectively. However, the distinct roots of a quadratic equation
may not be real – they could consist of a pair of complex conjugate roots Recall that the complex
λ1 = α + βi and λ2 = α − βi. If the auxiliary equation has such a pair of conjugate of the complex
roots, we can still write the general solution in the form number α + βi is α − βi.
y = Aeλ1 x + Beλ2 x = Ae(α+βi)x + Be(α−βi)x , You will soon see why we use A
and B for the arbitrary
but we now have a complex-valued solution. constants (rather than our usual
choice of C and D).
Since equation (35) has real coefficients, we would like a real-valued
solution. In order to achieve this, we need to allow A and B to be
complex. Then we can use Euler’s formula, which tells us that Euler’s formula is given in the
Handbook.
eiβx = cos βx + i sin βx and e−iβx = cos βx − i sin βx.
Now
y = Aeλ1 x + Beλ2 x
= Ae(α+βi)x + Be(α−βi)x
= Aeαx eiβx + Beαx e−iβx
= Aeαx (cos βx + i sin βx) + Beαx (cos βx − i sin βx)
% #
= eαx (A + B) cos βx + (Ai − Bi) sin βx
= eαx (C cos βx + D sin βx), The constants in the final
expression are now C and D, in
where C = A + B and D = (A − B)i. Provided that any initial conditions keeping with our previous
are real-valued, C and D are real, and this is the required real-valued solutions.
solution containing two arbitrary constants.
41
Unit 1 First- and second-order differential equations
Example 10
(a) Write down the auxiliary equation of the differential equation
d2 y dy
−6 + 13y = 0,
dx2 dx
and show that its roots are λ1 = 3 + 2i and λ2 = 3 − 2i.
(b) Confirm that y1 = e3x cos 2x and y2 = e3x sin 2x are both solutions of
the differential equation.
(c) Deduce that y = e3x (C cos 2x + D sin 2x) is also a solution of the
differential equation, for any values of the two constants C and D.
Solution
(a) The characteristic equation is
λ2 − 6λ + 13 = 0.
The formula method gives
√ √
6 ± 36 − 4 × 1 × 13 6 ± −16
λ= = = 3 ± 2i,
2 2
With the previous notation we so the two complex conjugate roots are λ1 = 3 + 2i and λ2 = 3 − 2i.
have α = 3 and β = 2.
(b) To confirm that y1 = e3x cos 2x is a solution of the differential
equation, we differentiate and substitute into the differential equation.
Differentiating to obtain the first and second derivatives of y1 gives
dy1
= 3e3x cos 2x + e3x (−2 sin 2x) = e3x (3 cos 2x − 2 sin 2x),
dx
d2 y1
= 3e3x (3 cos 2x − 2 sin 2x) + e3x (−6 sin 2x − 4 cos 2x)
dx2
= e3x (5 cos 2x − 12 sin 2x).
Substituting these into the left-hand side of the differential equation
gives
d2 y1 dy1
−6 + 13y1 = e3x (5 cos 2x − 12 sin 2x)
dx2 dx
− 6e3x (3 cos 2x − 2 sin 2x) + 13e3x cos 2x
= e3x [(5 − 18 + 13) cos 2x + (−12 + 12) sin 2x]
= 0.
Hence y1 = e3x cos 2x is a solution.
Similarly, for y2 = e3x sin 2x we have
dy2 d2 y2
= e3x (2 cos 2x + 3 sin 2x), = e3x (12 cos 2x + 5 sin 2x),
dx dx2
and substituting into the left-hand side of the differential equation
gives
d2 y2 dy2
2
−6 + 13y2 = e3x [(12 − 12) cos 2x + (5 − 18 + 13) sin 2x]
dx dx
= 0.
42
2 Homogeneous second-order differential equations
Exercise 20
Use the auxiliary equation to find the general solution of each of the
following differential equations.
d2 y dy d2 θ
(a) 2
+4 + 8y = 0 (b) + 9θ = 0
dx dx dt2
43
Unit 1 First- and second-order differential equations
Exercise 21
Find the general solution of each of the following differential equations.
d2 y
(a) + 4y = 0 (b) u!! (x) − 6u! (x) + 8u(x) = 0
dx2
d2 y dy d2 y dy
(c) 2
+ 2 = 0 (d) 2
−2 +y =0
dx dx dx dx
d2 y
(e) − ω2 y = 0, where ω is a real constant
dx2
d2 y dy
(f) 2
+4 + 29y = 0
dx dx
Exercise 22
Small oscillations of the pendulum of a clock can be modelled by the
θ differential equation
l g
θ̈ = − θ,
l
where g is the magnitude of the acceleration due to gravity, l is the length
of the pendulum, and θ is the angle that the pendulum makes with the
vertical (see Figure 13).
Figure 13 A pendulum
Solve the differential equation to obtain an expression for θ in terms of g
and l.
3 Inhomogeneous second-order
differential equations
Section 2 was concerned with finding the general solution of homogeneous
linear constant-coefficient second-order differential equations. This section
concerns inhomogeneous linear constant-coefficient second-order
differential equations, that is, equations of the form
d2 y dy
a +b + cy = f (x), (38)
dx2 dx
where a, b, c are real constants, a )= 0, and f (x) is a given continuous
real-valued function of x.
44
3 Inhomogeneous second-order differential equations
Example 11
Show that y = Ce−2x + De−3x + 2 is a solution of the inhomogeneous
differential equation
d2 y dy
2
+5 + 6y = 12, (39)
dx dx
for any values of the constants C and D.
Solution
We know from Exercise 18(a) that the homogeneous differential equation
d2 y dy
2
+5 + 6y = 0 (40)
dx dx
has a general solution yc = Ce−2x + De−3x , where C and D are arbitrary The notation yc and yp will be
constants. explained shortly.
45
Unit 1 First- and second-order differential equations
Let
d2 y dy
a 2
+b + cy = f (x) (41)
dx dx
be an inhomogeneous linear constant-coefficient second-order
differential equation.
Its associated homogeneous equation is
d2 y dy
a 2
+b + cy = 0.
dx dx
The general solution yc of the associated homogeneous equation is
known as the complementary function for the original
inhomogeneous equation (41).
Any particular solution yp of the original inhomogeneous
equation (41) is referred to as a particular integral for that
equation.
Later in this section we will show how to find particular integrals for a
wide variety of equations. Before we do that, it is important that you
realise the full significance of finding just one particular integral.
Exercise 23
Suppose that we have found two different particular integrals yp1 , yp2 for
equation (41). Use the principle of superposition to show that the function
yp1 − yp2 is then a solution of the associated homogeneous equation.
Theorem 3
If yc is the complementary function for an inhomogeneous linear
constant-coefficient second-order differential equation, and yp is a
particular integral for that equation, then yc + yp is the general
solution of that equation.
46
3 Inhomogeneous second-order differential equations
Example 12
Find the general solution of the differential equation
d2 y
+ 9y = 9x + 9. (42)
dx2
Solution
The associated homogeneous equation is
d2 y
+ 9y = 0,
dx2
which has the general solution yc = C cos 3x + D sin 3x, where C and D are See Exercise 20(b), although
arbitrary constants. This is the complementary function for equation (42). there different symbols were
used for the variables.
A particular integral for equation (42) is yp = x + 1.
You will see in the next
This may be verified by differentiation and substitution: yp! = 1 and subsection how to find such a
yp!! = 0, and substituting into the left-hand side of equation (42) gives particular integral.
47
Unit 1 First- and second-order differential equations
Exercise 24
For each of the following equations:
• write down its associated homogeneous equation and its
complementary function yc
• find a particular integral of the form yp = p, where p is a constant
• write down the general solution.
d2 y d2 y dy
See Exercise 21(a) and (a) + 4y = 8 (b) −3 + 2y = 6
Example 8. dx2 dx 2 dx
48
3 Inhomogeneous second-order differential equations
when f (x) is a polynomial, exponential or sinusoidal function, or a sum of There exist procedures for
such functions. finding a particular integral for
fairly general types of continuous
You saw an example of the approach in Exercise 24. There the functions function f (x), but these are not
f (x) were constants and you tried a constant function y = p as a considered in this module.
particular integral, substituting into the differential equation to find a
suitable value for p. In general, we try a function of the same form as f (x)
as a particular integral, and substitute into the differential equation to find
suitable values for its unknown coefficients. The function that we try is
known as a trial solution, and the method is known as the method of
undetermined coefficients.
The following examples illustrate the method. Bear in mind, though, that
the method (and hence each example) finds only a particular integral for
the differential equation; to find the general solution you would need to
find the complementary function and combine it with the particular
integral, according to Procedure 6.
A polynomial function
We consider f (x) = mn xn + mn−1 xn−1 + · · · + m1 x + m0 .
Let us first consider a case where f (x) is a linear function (i.e. a
polynomial of degree 1).
Example 13
Find a particular integral for
d2 y dy
3 −2 + y = 4x + 2.
dx2 dx
Solution
We try a solution of the form
y = p1 x + p0 ,
where p1 and p0 are coefficients to be determined so that the differential
equation is satisfied. To try this solution, we need the first and second
derivatives of y:
dy d2 y
= p1 , = 0.
dx dx2
Substituting these into the left-hand side of the differential equation gives
d2 y dy
3 2
−2 + y = 3 × 0 − 2p1 + (p1 x + p0 ) = p1 x + (p0 − 2p1 ).
dx dx
Therefore for y = p1 x + p0 to be a solution of the differential equation, we
require that
p1 x + (p0 − 2p1 ) = 4x + 2 for all x. (43)
To find the two unknown coefficients p1 and p0 , we compare the Comparing coefficients works
coefficients on both sides of equation (43). Comparing the terms in x gives because two polynomials are
p1 = 4. Comparing the constant terms gives p0 − 2p1 = 2, so equal if and only if all their
corresponding coefficients are
p0 = 2 + 2p1 = 2 + 2 × 4 = 10. the same.
49
Unit 1 First- and second-order differential equations
Exercise 25
Find particular integrals of the form y = p1 x + p0 for each of the following
differential equations.
d2 y dy
(a) −2 + 2y = 2x + 3
dx2 dx
d2 y dy
(b) 2
+2 + y = 2x
dx dx
50
3 Inhomogeneous second-order differential equations
Exercise 26
Find a particular integral for
ÿ − y = t2 .
An exponential function
We consider f (x) = mekx .
Example 14
Find a particular integral for
d2 y
+ 9y = 2e3x .
dx2
Solution
We try a solution of the form
y = pe3x , Since the derivative of e3x is
3e3x , the exponent (3x)
where p is a coefficient to be determined so that the differential equation is appearing in y(x) should be the
satisfied. Differentiating y = pe3x gives same as that appearing in f (x),
and only the coefficient p is to
dy d2 y
= 3pe3x , = 9pe3x . be determined.
dx dx2
Substituting these into the left-hand side of the differential equation gives
d2 y
+ 9y = 9pe3x + 9pe3x = 18pe3x .
dx2
Therefore for y = pe3x to be a solution of the differential equation, we
require that 18pe3x = 2e3x for all x. Hence p = 91 , and
yp = 19 e3x
is a particular integral for the given differential equation.
Exercise 27
Find a particular integral for
d2 y dy
2 2
−2 + y = 2e−x .
dx dx
51
Unit 1 First- and second-order differential equations
A sinusoidal function
This type of function is We consider f (x) = m cos Ωx + n sin Ωx.
particularly important in many
practical applications. Following on from earlier ideas, the trial solution must contain terms like
those in f (x) and all its derivatives; so even if f (x) contains only a sine or
only a cosine, the trial solution y(x) must contain both a sine and a cosine.
However, the value of the parameter Ω appearing in y(x) should be the
same as that appearing in f (x).
Example 15
Find a particular integral for
d2 y dy
2
+2 + 2y = 10 sin 2x.
dx dx
Solution
We try a solution of the form
y = p cos 2x + q sin 2x,
where p and q are coefficients to be determined so that the differential
equation is satisfied. Differentiating y gives
dy d2 y
= −2p sin 2x + 2q cos 2x, = −4p cos 2x − 4q sin 2x.
dx dx2
Substituting these into the left-hand side of the differential equation gives
d2 y dy
2
+2 + 2y = (−4p cos 2x − 4q sin 2x)
dx dx
+ 2 (−2p sin 2x + 2q cos 2x)
+ 2(p cos 2x + q sin 2x)
= (−2p + 4q) cos 2x + (−4p − 2q) sin 2x.
Therefore for y = p cos 2x + q sin 2x to be a solution of the differential
equation, we require that
(−2p + 4q) cos 2x + (−4p − 2q) sin 2x = 10 sin 2x for all x. (44)
Comparing coefficients works To find p and q, we compare the coefficients of cos 2x and sin 2x on both
because the cosine and sine sides of equation (44). Comparing cos 2x terms gives −2p + 4q = 0, and
functions are linearly comparing sin 2x terms gives −4p − 2q = 10. Solving these simultaneous
independent:
if a sin rx + b cos rx = 0 for all x,
equations gives p = −2, q = −1. Hence
then a = b = 0. yp = −2 cos 2x − sin 2x
is a particular integral for the given differential equation.
Exercise 28
Find a particular integral for
d2 y dy
− = cos 3t + sin 3t.
dt2 dt
52
3 Inhomogeneous second-order differential equations
Exercise 29
What form of trial solution should you use in order to find a particular
integral for each of the following differential equations?
d2 y
(a) − y = e2x
dx2
d2 y dy
(b) 2
+2 − 4y = sin 4x
dx dx
Exercise 30
Use Procedures 6 and 7 to find the general solution of each of the following
differential equations. The roots of the auxiliary
equation are −1 ± i for part (a),
d2 y dy and 0 and −3 for part (b).
(a) +2+ 2y = 4
dx2 dx
d2 θ dθ
(b) 2
+ 3 = 13 cos 2t
dt dt
53
Unit 1 First- and second-order differential equations
Exercise 31
A long horizontal rectangular beam of length l rests on rigid supports at
l each end. It is important in civil engineering to determine how much such
a beam ‘sags’. A simple model of this ‘sag’, or vertical displacement y, is
the differential equation
y Ry !! − Sy + 12 Q(l − x)x = 0,
x displacement
where R, S and Q are constants related to the structure of the beam, and
Figure 14 A horizontal x is the distance from one end of the beam (as illustrated in Figure 14).
beam resting on supports
Find the general solution of the differential equation in the case where R,
S and Q are all equal to 1.
In Subsection 3.4 you will see how the principle of superposition can be
used in combination with Procedure 7 to solve differential equations whose
right-hand-side functions f (x) are sums of polynomial, exponential and
sinusoidal functions. But first let us look at some exceptional cases for
which Procedure 7 does not work and needs to be adapted.
Example 16
Find a particular integral for
d2 y
− 4y = 2e2x .
dx2
Solution
Using Procedure 7, let us try y = pe2x . Differentiating this gives
dy d2 y
= 2pe2x , = 4pe2x .
dx dx2
Substituting these into the left-hand side of the differential equation gives
d2 y
− 4y = 4pe2x − 4pe2x = 0.
dx2
So there is no value of p that gives a particular integral of the form
y = pe2x .
The trouble is that the complementary function, that is, the general
solution of the associated homogeneous equation
d2 y
− 4y = 0,
dx2
See Exercise 18(c). is y = Ce−2x + De2x , where C and D are arbitrary constants, thus the
trial solution is a solution of the associated homogeneous equation (with
54
3 Inhomogeneous second-order differential equations
The problem with the trial solution being a solution of the associated
homogeneous equation can occur irrespective of the form of the trial
solution (i.e. polynomial, exponential or sinusoidal), but in most cases it
can be overcome by multiplying the trial solution suggested in Procedure 7
by x. When using Procedure 7, you should check whether the proposed
trial solution is a solution of the associated homogeneous equation, and if
so try multiplying it by x. (This is why it is important to find yc before yp
in Procedure 6.)
Exercise 32
Find a particular integral for each of the following differential equations. The complementary functions
are given in the solutions to
d2 y dy d2 y dy Example 8 and Exercise 18(b).
(a) −3 + 2y = 4ex (b) 2 +3 =1
dx2 dx dx2 dx
Exercise 33
The motion of a marble dropped from the Clifton Suspension Bridge into
the River Avon can be modelled by the differential equation
mẍ + rẋ − mg = 0, x
55
Unit 1 First- and second-order differential equations
We have seen that Procedure 7 fails if the trial solution is a solution of the
associated homogeneous differential equation; in such cases we multiply the
suggested trial solution by the independent variable and use this as the
trial solution. Another situation in which it is necessary to multiply the
suggested trial solution by the independent variable is illustrated in the
following example.
Example 17
Find a particular integral for
d2 y dy
2
+2 = 2x + 2.
dx dx
Solution
Using Procedure 7, let us try y = p1 x + p0 . Differentiating this gives
dy d2 y
= p1 , = 0.
dx dx2
Substituting these into the left-hand side of the differential equation gives
d2 y dy
2
+2 = 2p1 .
dx dx
But there is no value of p1 that satisfies 2p1 = 2x + 2 for all x.
The problem this time is that (from Exercise 21(c)) the complementary
This example again illustrates function is y = C + De−2x , where C and D are arbitrary constants, so the
why it is better to find the p0 part of the trial solution is a solution of the associated homogeneous
complementary function before equation (with C = p0 , D = 0). Hence on substituting the trial solution
looking for a particular integral.
y = p1 x + p0 into the inhomogeneous equation, the p0 part disappears, and
the trial solution effectively reduces to y = p1 x. The result in this case is
that after substituting the trial solution and its derivatives into the
left-hand side of the equation, there are not enough terms on the left-hand
side to compare with the terms in the right-hand-side function.
Again, the difficulty can be overcome by multiplying the trial solution
suggested by Procedure 7 by x, to give y = p1 x2 + p0 x. Differentiating this
gives
dy d2 y
= 2p1 x + p0 , = 2p1 .
dx dx2
Substituting these into the left-hand side of the differential equation gives
d2 y dy
2
+2 = 2p1 + 2(2p1 x + p0 ) = 4p1 x + (2p1 + 2p0 ).
dx dx
Therefore y = p1 x2 + p0 x is a solution of the differential equation provided
that 4p1 x + (2p1 + 2p0 ) = 2x + 2 for all x. This gives p1 = 21 , p0 = 12 , so
yp = 21 (x2 + x)
is a particular integral for the given differential equation.
56
3 Inhomogeneous second-order differential equations
Exercise 34
Find a particular integral for You found the complementary
function in Exercise 21(d).
d2 y dy
−2 + y = ex .
dx2 dx
Example 18
Find a particular integral for
d2 y
+ 9y = 2e3x + 18x + 18. (45)
dx2
57
Unit 1 First- and second-order differential equations
Solution
In Example 14 you saw that yp = 19 e3x is a particular integral for
d2 y
+ 9y = 2e3x ,
dx2
and in Example 12 you saw that yp = x + 1 is a particular integral for
d2 y
+ 9y = 9x + 9.
dx2
Therefore, by the principle of superposition, a particular integral for
equation (45) is
yp = 91 e3x + 2 × (x + 1) = 19 e3x + 2x + 2.
Exercise 35
Find particular integrals for each of the following differential equations.
d2 y dy
See Exercise 34. (a) −2 + y = 4ex − 3e2x
dx2 dx
d2 x dx
(b) 2 2 + 3 + 2x = 12 cos 2t + 10
dt dt
Exercise 36
You will find some help for Find the general solution of each of the following differential equations.
parts (a), (d), (e) and (f) in
Exercises 18 and 21, and
d2 θ
(a) + 4θ = 2t
Example 8. dt2
(b) u!! (t) + 4u! (t) + 5u(t) = 5
d2 Y dY
(c) 3 2
−2 − Y = e2x + 3
dx dx
d2 y
(d) − 4y = e−2x
dx2
d2 y
(e) + 4y = sin 2x + 3x
dx2
d2 y dy
(f) −3 + 2y = 2ex − 5e2x
dx2 dx
58
4 Initial conditions and boundary conditions
59
Unit 1 First- and second-order differential equations
Let us now see how initial conditions can be used to determine values for
the two arbitrary constants and hence find a particular solution.
Example 19
Find the particular solution of the differential equation
d2 y dy
2
−3 + 2y = 0
dx dx
that satisfies the initial conditions y = 0 and dy/dx = 1 when x = 0.
Solution
From Example 8 we know that the general solution is
y = Cex + De2x , (46)
where C and D are arbitrary constants. One of the initial conditions
involves the derivative of the solution, so we need to obtain that derivative:
dy
= Cex + 2De2x . (47)
dx
The initial conditions state that y(0) = 0, y ! (0) = 1. Substituting x = 0,
y = 0 into equation (46) gives
0 = Ce0 + De0 = C + D,
while substituting x = 0, dy/dx = 1 into equation (47) gives
1 = Ce0 + 2De0 = C + 2D.
Note that when you check a Solving these equations gives C = −1, D = 1, so the required particular
particular solution, you should solution is
check that it satisfies the initial
conditions as well as the y = −ex + e2x .
differential equation.
60
4 Initial conditions and boundary conditions
Exercise 37
Find solutions to the following initial-value problems.
% # % #
(a) u!! (t) + 9u(t) = 0, u π2 = 0, u! π2 = 1. See Exercise 20(b).
d2 y dy
(b) −3+ 2y = 4ex , y(0) = 4, y ! (0) = 2. See Exercise 32(a).
dx2 dx
d2 y dy
(c) 2
−2 + y = 4ex − 3e2x , y(0) = 4, y ! (0) = −1. See Exercises 21(d) and 35(a).
dx dx
Theorem 4
The initial-value problem
d2 y dy
a 2
+b + cy = f (x), y(x0 ) = y0 , y ! (x0 ) = z0 ,
dx dx
where a, b, c are real constants with a )= 0, and f (x) is a given
continuous real-valued function on an interval (r, s), with x0 ∈ (r, s),
has a unique solution on that interval.
61
Unit 1 First- and second-order differential equations
Example 20
Find the particular solution of the differential equation
d2 y
+ 9y = 0
dx2
that satisfies the boundary conditions y = 0 when x = 0 and dy/dx = 1
when x = π3 .
Solution
From Exercise 20(b), the general solution is
y = C cos 3x + D sin 3x, (48)
where C and D are arbitrary constants.
One of the boundary conditions involves the derivative of the solution, so
we need to obtain that derivative:
dy
= −3C sin 3x + 3D cos 3x. (49)
dx
% #
The boundary conditions state that y(0) = 0, y ! π3 = 1. Substituting
x = 0, y = 0 into equation (48) gives
0 = C cos 0 + D sin 0 = C,
so C = 0. Substituting x = π3 , y ! = 1 and C = 0 into equation (49) gives
1 = 3D cos π = −3D.
Therefore C = 0, D = − 31 , so the required particular solution is
y = − 31 sin(3x).
62
4 Initial conditions and boundary conditions
Exercise 38
Solve the boundary-value problem
d2 y dy
−3 + 2y = 4ex , y ! (0) = 2, y(1) = 0. See Exercise 32(a).
dx2 dx
Exercise 39
Use the differential equation of Exercise 31, with R = S = Q = 1, namely
y !! − y + 12 (l − x)x = 0,
to determine the vertical displacement at the centre of a beam of length
2 metres resting on rigid supports at its ends.
Example 21
Try to solve the boundary-value problem
d2 y %π#
+ 4y = 0, y(0) = 0, y 2 = 1.
dx2
Solution
From Exercise 21(a), the general solution is
y = C cos 2x + D sin 2x,
where C and D are arbitrary constants.
%π#
The boundary conditions state that y(0) = 0, y 2 = 1. Substituting each
of these into the general solution in turn gives
0 = C cos 0 + D sin 0 = C,
1 = C cos π + D sin π = −C.
There is no solution for which C = 0 and C = −1, so there is no solution of
the differential equation that satisfies the given boundary conditions.
63
Unit 1 First- and second-order differential equations
Example 22
Solve the boundary-value problem
d2 y dy
+4 + 5y = 5, y(0) = 1, y(π) = 1.
dx2 dx
Solution
From Exercise 36(b), the general solution is
y = e−2x (C cos x + D sin x) + 1,
where C and D are arbitrary constants.
The boundary conditions state that y(0) = 1, y(π) = 1. Substituting each
of these into the general solution in turn gives
1 = e0 (C cos 0 + D sin 0) + 1 = C + 1,
1 = e−2π (C cos π + D sin π) + 1 = −Ce−2π + 1.
Both of these equations reduce to C = 0, but D can take any value, so any
solution of the form
y = De−2x sin x + 1
satisfies the differential equation and the boundary conditions.
Exercise 40
You found the general solution For each of the following problems, identify the conditions as either initial
of the differential equation in conditions or boundary conditions, and solve the problem.
Exercise 21(a).
(a) u!! (x) + 4u(x) = 0, u(0) = 1, u! (0) = 0.
% #
(b) u!! (x) + 4u(x) = 0, u(0) = 0, u π2 = 0.
% # % #
(c) u!! (x) + 4u(x) = 0, u π2 = 0, u! π2 = 0.
% #
(d) u!! (x) + 4u(x) = 0, u(−π) = 1, u π4 = 2.
% #
(e) u!! (x) + 4u(x) = 0, u! (0) = 0, u! π4 = 1.
64
Learning outcomes
Exercise 41
Find the solution (if any) to each of the following problems.
(a) u!! (t) + 4u! (t) + 5u(t) = 0, u(0) = 0, u! (0) = 2. The roots of the auxiliary
equation are −2 ± i.
d2 y dy dy
(b) +2 + 2y = 0, where y = 0 and = 0 when x = 0. Roots are −1 ± i.
dx2 dx dx
% #
(c) ẍ + 9x = 3(1 − πt), x(0) = 31 , ẋ π3 = 0. Roots are ±3i.
Learning outcomes
After studying this unit, you should be able to:
• understand and use the basic terminology relating to differential
equations and their solutions
• check by substitution whether a given function is a solution of a given
first-order differential equation or initial-value problem
• find from the general solution of a first-order differential equation the
particular solution that satisfies a given initial condition
• appreciate the difficulties with domains and image sets for the solution
of some differential equations
• deduce the qualitative behaviour of solutions from consideration of a
first-order differential equation itself, as visualised from its direction
field
• set up the formulas required by Euler’s method for solving an
initial-value problem, and carry out a few steps of the method
• recognise when a first-order differential equation is soluble by direct
integration, and carry out that integration when appropriate, in
simple cases
• recognise when a first-order differential equation is separable, and
apply the method of separation of variables in simple cases
• recognise when a first-order differential equation is linear, and solve
such an equation by the integrating factor method in simple cases
• understand and use the terminology relating to linear
constant-coefficient second-order differential equations
• understand the key role of the principle of superposition in the solution
of linear constant-coefficient second-order differential equations
• obtain the general solution of a homogeneous linear
constant-coefficient second-order differential equation using the
solutions of its auxiliary equation
65
Unit 1 First- and second-order differential equations
66
Solutions to exercises
Solutions to exercises
Solution to Exercise 1
We have
' -
P
r(P ) = k 1 − ,
M
so we need to solve the pair of simultaneous equations
' - ' -
10 10 000
k 1− = 1, k 1 − = 0.
M M
From the second equation, since k > 0, we see immediately that
M = 10 000. Substituting in the first equation leads to
999 1000
k = 1, so k = .
1000 999
Solution to Exercise 2
(a) In each case we need to show that the given function satisfies the
differential equation, that is, it gives the same expression for either
side of the equation.
If y = 2ex − (x2 + 2x + 2), then differentiating y gives
dy
= 2ex − 2x − 2,
dx
and substituting the expression for y into the right-hand side of the
given differential equation gives
y + x2 = 2ex − (x2 + 2x + 2) + x2 = 2ex − 2x − 2,
as required.
(b) If y = tan x + sec x, then
dy
= sec2 x + tan x sec x,
dx
and substituting the expression for y into the right-hand side of the
given differential equation gives
y tan x + 1 = (tan x + sec x) tan x + 1
= (tan2 x + 1) + sec x tan x
= sec2 x + tan x sec x,
as required.
(c) If y = t + Ce−t , then
dy
ẏ = = 1 − Ce−t ,
dt
and substituting the expression for y into the right-hand side of the
given differential equation gives
−y + t + 1 = −(t + Ce−t ) + t + 1 = 1 − Ce−t ,
as required.
67
Unit 1 First- and second-order differential equations
Solution to Exercise 3
(a) In each case we need to show that the given function satisfies the
differential equation, that is, it gives the same expression for either
side of the equation.
If y = C − 31 e−3x , then
dy
= e−3x ,
dx
as required.
(b) If
CM ekt
P = ,
1 + Cekt
Rules such as the quotient rule then, using the quotient rule for differentiation,
can be found in the Handbook.
dP (1 + Cekt )(CM kekt ) − (CM ekt )(Ckekt )
=
dt (1 + Cekt )2
' - ' -
CM ekt 1 + Cekt − Cekt
=k
1 + Cekt 1 + Cekt
' - ' -
CM ekt Cekt
=k 1−
1 + Cekt 1 + Cekt
' -
P
= kP 1 − ,
M
as required.
Solution to Exercise 4
(a) From Exercise 3(b) and using k = 0.15 and M = 10, we know that
CM ekt 10Ce0.15t
P = =
1 + Cekt 1 + Ce0.15t
is a solution of the differential equation, where C is a constant. The
initial condition P (0) = 1 then implies
10C
1= , so C = 19 .
1+C
A particular solution is therefore
10 0.15t
9 e 10e0.15t
P = = .
1 + 19 e0.15t 9 + e0.15t
(b) Dividing top and bottom by e0.15t , we see that
10
P = .
9e−0.15t
+1
For large values of t, the exponential term on the bottom will be very
small. The result is that P will approach the value 10 in the long term.
68
Solutions to exercises
Solution to Exercise 5
(a) In each case, we apply direct integration to find the general solution,
and C is an arbitrary constant.
The differential equation dy/dx = 6x has general solution
(
y = 6x dx = 3x2 + C.
Solution to Exercise 6
(a) Each of the differential equations is soluble by direct integration.
The general solution of dy/du = 1/(u − a), where u )= a, is given by
(
1
y= du.
u−a
Using equation (8), integration produces the general solution
y = ln |u − a| + C,
where C is an arbitrary constant.
69
Unit 1 First- and second-order differential equations
Solution to Exercise 7
(a) The differential equation is dm/dt = −λm, where m > 0. Following
Procedure 2, we obtain
( (
1
dm = (−λ) dt,
m
and since m > 0, integration produces
ln m = −λt + C,
where C is an arbitrary constant. On solving this equation for m, by
taking the exponential of both sides, we obtain
m = e−λt+C = eC e−λt = Be−λt ,
where B = eC is a positive (since eC > 0 for all C) but otherwise
arbitrary constant. The general solution is therefore
m = Be−λt ,
where B is a positive but otherwise arbitrary constant.
(b) The initial condition is m(0) = m0 , from which we have m0 = Be0 , so
B = m0 . The required particular solution is therefore
m = m0 e−λt .
Solution to Exercise 8
(a) The differential equation is
dy y−1
= , where x > 0.
dx x
In order to apply the separation of variables method, we need to
exclude the case where y = 1.
70
Solutions to exercises
71
Unit 1 First- and second-order differential equations
Solution to Exercise 9
(a) The given equation is dP/dt = kP (1 − P/M ). First, note that the
constant functions P = 0 and P = M are both solutions. Assuming
that we are considering neither of these possibilities (we are certainly
not interested in P = 0 since we know that P0 > 0), we can use the
separation of variables method to obtain
( (
1 1
dP = 1 dt.
k P (1 − P/M )
The integral on the left-hand side is of the form evaluated in
Exercise 6(b), with 1/M in place of a. Hence we have
) )
1 )) 1 1 ))
− ln ) − = t + C,
k P M)
where C is an arbitrary constant. On solving for P , we find first that
1 1
− = ±e−k(t+C) = ±e−kC e−kt = Be−kt ,
P M
where B = ±e−kC is a non-zero but otherwise arbitrary constant.
However, note that B = 0 corresponds to the constant solution
P = M already noted, so the restriction B )= 0 may be dropped.
Hence we obtain
' -−1
1 −kt
P = + Be (ekt )= −M B),
M
where B is an arbitrary constant.
From the initial condition P (0) = P0 , we have
' -−1
1 0 1 1
P0 = + Be , so B = − .
M P0 M
The solution to the initial-value problem is therefore
' ' - -−1
1 1 1
P = + − e−kt ,
M P0 M
which yields
M
P = .
1 + (M/P0 − 1)e−kt
(b) As t → ∞, we have e−kt → 0, and consequently the value of P (t)
approaches M . This is true whether the starting value P0 is greater
than or less than M .
Solution to Exercise 10
(a) The equation dy/dx = x sin x is linear, with g(x) = 0 (for all x) and
h(x) = x sin x.
(b) The equation ẏ + y 2 = t is not linear (because of the y 2 term).
(c) The equation x(dy/dx) + y = y 2 is not linear (because of the y 2 term).
72
Solutions to exercises
Solution to Exercise 11
(a) The given equation is dy/dx − y = ex sin x. Comparison with
equations (16) and (17) shows that the integrating factor is
'( -
p(x) = exp (−1) dx = exp(−x) = e−x .
73
Unit 1 First- and second-order differential equations
Solution to Exercise 12
(a) This requires the integrating factor method.
(b) This is best solved by direct integration.
(c) This can be solved by separation of variables or the integrating factor
method.
(d) This requires the integrating factor method.
Solution to Exercise 13
For the initial-value problem
dy
= y, y(0) = 1,
dx
we have x0 = 0, Y0 = y0 = 1 and f (xi , Yi ) = Yi . The step size is given as
h = 0.2. Equation (25) with i = 0 gives
x1 = x0 + h = 0 + 0.2 = 0.2,
and equation (26) with i = 0 gives
Y1 = Y0 + h f (x0 , Y0 ) = 1 + 0.2 × 1 = 1.2.
Applying equations (25) and (26) in turn for i = 1, 2, 3, 4, we obtain the
following table.
Solution to Exercise 14
Since we are told that for a sufficiently small step size h, the absolute error
is proportional to h, we can deduce from the last row of Table 2 that there
exists a constant k such that
0.000 136 = 0.0001k,
so k = 1.36. In order to determine y(1) correct to five decimal places, h
must be such that
1.36h < 5 × 10−6
74
Solutions to exercises
or
5 × 10−6
h< ( 3.7 × 10−6 .
1.36
So a suitable choice of h would be 10−6 = 0.000 001.
(In fact, using this value of h gives an approximation to y(1) of 2.718 280,
which is correct to five decimal places.)
Solution to Exercise 15
(a) The slope defined by the direction field f (x, y) = y + x2 is zero when
y = −x2 , which is a parabola in the lower half-plane with vertex at
the origin. Below this parabola we have y < −x2 and f (x, y) < 0,
while above the parabola we have y > −x2 and f (x, y) > 0. Thus all
slopes for points of the plane below the parabola y = −x2 are
negative, and all slopes for points above it are positive.
Also, if x is fixed, then f (x, y) = y + x2 is an increasing function as y
increases. If instead y is fixed, then for x > 0, f (x, y) increases as x
increases, and for x < 0, f (x, y) increases as x becomes more negative.
These observations indicate that the slope given by the direction field
increases as we move from bottom to top along any vertical line,
whereas on moving along any horizontal line, the slope increases with
distance from the y-axis.
(b) The features described in the solution to part (a) are all apparent on
the direction field diagram. This direction field diagram is repeated
below, with the parabola y = −x2 superimposed on it. (Note that this
parabola does not represent a solution of the differential equation.)
y
2
−2 −1 0 1 2 x
−1
−2
75
Unit 1 First- and second-order differential equations
(c) It appears from the direction field that there are several types of
solution. Any solution whose graph cuts the y-axis above the origin
has positive slope at all points. The solution graph that passes
through the origin has zero slope there, but positive slope everywhere
else. Any solution graph that cuts the y-axis below the origin has a
maximum (where it meets y = −x2 for x < 0). Some of these graphs
also have a minimum (where they meet y = −x2 for x > 0). Others
have no minimum (though this is not clear from the diagram given).
A solution graph of each type is sketched below.
y
2
−2 −1 0 1 2 x
−1
−2
76
Solutions to exercises
Solution to Exercise 16
(a) Equations (i), (ii), (iii), (vi), (vii) and (viii) are linear and
constant-coefficient. (Equation (v) is non-linear; (iv) is linear but not
constant-coefficient.)
(b) Of the linear constant-coefficient equations, only (iii) and (viii) are
homogeneous.
(c) In equations (i)–(v) the (dependent, independent) variable pairs are
all (y, x). In equations (vi), (vii) and (viii) they are (t, θ), (x, t) and
(x, t), respectively.
Solution to Exercise 17
(a) λ2 − 5λ + 6 = 0
(b) λ2 − 9 = 0
(c) λ2 + 2λ = 0
Solution to Exercise 18
(a) The auxiliary equation is λ2 + 5λ + 6 = 0. Solving this by
factorisation as (λ + 2)(λ + 3) = 0 gives the roots λ1 = −2 and
λ2 = −3. The general solution is therefore
y = Ce−2x + De−3x ,
where C and D are arbitrary constants.
(b) The auxiliary equation is 2λ2 + 3λ = 0. This can be factorised as
λ(2λ + 3) = 0, so its roots are λ1 = 0 and λ2 = − 32 . The general
solution is therefore
y = Ce0 + De−3x/2 = C + De−3x/2 ,
where C and D are arbitrary constants.
(c) The auxiliary equation is λ2 − 4 = 0, that is, λ2 = 4, so its roots are
λ1 = −2 and λ2 = 2. The general solution is therefore
z = Ce−2u + De2u ,
where C and D are arbitrary constants.
Solution to Exercise 19
(a) The auxiliary equation is λ2 + 2λ + 1 = 0, which can be factorised as
(λ + 1)2 = 0, giving equal roots λ1 = λ2 = −1. The general solution is
therefore
y = (C + Dx)e−x ,
where C and D are arbitrary constants.
(b) The auxiliary equation λ2 − 4λ + 4 = 0 factorises as (λ − 2)2 = 0,
which has equal roots λ1 = λ2 = 2. The general solution is therefore
s = (C + Dt)e2t ,
where C and D are arbitrary constants.
77
Unit 1 First- and second-order differential equations
Solution to Exercise 20
(a) The auxiliary equation is λ2 + 4λ + 8 = 0, which has solutions
√
−4 ± 16 − 32
λ= = −2 ± 2i.
2
The general solution is therefore
y = e−2x (C cos 2x + D sin 2x),
where C and D are arbitrary constants.
(b) The auxiliary equation is λ2 + 9 = 0, which has solutions
λ = ±3i.
The general solution is therefore
θ = e0 (C cos 3t + D sin 3t) = C cos 3t + D sin 3t,
where C and D are arbitrary constants.
Solution to Exercise 21
(a) The auxiliary equation is λ2 + 4 = 0, which has solutions λ = ±2i.
The general solution is therefore
y = C cos 2x + D sin 2x,
where C and D are arbitrary constants.
(b) The auxiliary equation is λ2 − 6λ + 8 = 0, which has solutions λ1 = 4
and λ2 = 2. The general solution is therefore
u = Ce4x + De2x ,
where C and D are arbitrary constants.
(c) The auxiliary equation is λ2 + 2λ = 0, which has solutions λ1 = 0 and
λ2 = −2. The general solution is therefore
y = C + De−2x ,
where C and D are arbitrary constants.
(d) The auxiliary equation is λ2 − 2λ + 1 = 0, which has solutions
λ1 = λ2 = 1. The general solution is therefore
y = (C + Dx)ex ,
where C and D are arbitrary constants.
(e) The auxiliary equation is λ2 − ω2 = 0, which has solutions λ = ±ω.
The general solution is therefore
y = Ceωx + De−ωx ,
where C and D are arbitrary constants.
78
Solutions to exercises
Solution to Exercise 22
&
The auxiliary equation is λ2 + g/l = 0, which has solutions λ = ±i g/l.
The general solution is therefore
'" - '" -
g g
θ = C cos t + D sin t .
l l
This problem will be discussed again in Unit 10.
Solution to Exercise 23
We could check this directly, by substituting y = yp1 − yp2 into the
associated homogeneous equation. However, it is easier to appeal to the
principle of superposition. Since yp1 and yp2 both satisfy
d2 y dy
a 2
+b + cy = f (x),
dx dx
Theorem 2 shows that the combination y = yp1 − yp2 indeed satisfies
d2 y dy
a 2
+b + cy = f (x) − f (x) = 0,
dx dx
as required.
Solution to Exercise 24
(a) The associated homogeneous equation is d2 y/dx2 + 4y = 0. The
complementary function (see Exercise 21(a)) is
yc = C cos 2x + D sin 2x, where C and D are arbitrary constants.
Trying a solution of the form yp = p, where p is a constant, in the
original equation d2 y/dx2 + 4y = 8 gives 0 + 4p = 8, so p = 2. Thus a
particular integral is yp = 2.
By Procedure 6, the general solution is
y = C cos 2x + D sin 2x + 2.
(b) The associated homogeneous equation is d2 y/dx2 − 3 dy/dx + 2y = 0.
The complementary function (see Example 8) is yc = Cex + De2x ,
where C and D are arbitrary constants.
Trying a solution of the form yp = p in the original equation
d2 y/dx2 − 3 dy/dx + 2y = 6 gives 0 − 0 + 2p = 6, so p = 3. Thus a
particular integral is yp = 3.
By Procedure 6, the general solution is
y = Cex + De2x + 3.
79
Unit 1 First- and second-order differential equations
Solution to Exercise 25
(a) Substituting y = p1 x + p0 and its derivatives into the differential
equation gives
0 − 2p1 + 2(p1 x + p0 ) = 2p1 x + (2p0 − 2p1 ) = 2x + 3.
Equating coefficients gives p1 = 1, p0 = 25 . Therefore a particular
integral is
yp = x + 25 .
(b) Substituting y = p1 x + p0 and its derivatives into the differential
equation gives
0 + 2p1 + (p1 x + p0 ) = p1 x + (2p1 + p0 ) = 2x.
Hence p1 = 2, p0 = −4, and a particular integral is
yp = 2x − 4.
Solution to Exercise 26
We try y = p2 t2 + p1 t + p0 , which has derivatives ẏ = 2p2 t + p1 , ÿ = 2p2 .
Substituting these into the differential equation gives
2p2 − (p2 t2 + p1 t + p0 ) = −p2 t2 − p1 t + (2p2 − p0 ) = t2 .
Hence p2 = −1, p1 = 0, p0 = −2, and a particular integral is
yp = −t2 − 2.
Solution to Exercise 27
We try a solution of the form y = pe−x , which has derivatives
dy/dx = −pe−x , d2 y/dx2 = pe−x . Substituting these into the differential
equation gives
2pe−x + 2pe−x + pe−x = 5pe−x = 2e−x .
Hence p = 52 , and a particular integral is
yp = 25 e−x .
Solution to Exercise 28
We try y = p cos 3t + q sin 3t, which has derivatives
dy d2 y
= −3p sin 3t + 3q cos 3t, = −9p cos 3t − 9q sin 3t.
dt dt2
Substituting into the differential equation gives
(−9p cos 3t − 9q sin 3t) − (−3p sin 3t + 3q cos 3t)
= −(9p + 3q) cos 3t + (3p − 9q) sin 3t
= cos 3t + sin 3t.
Hence we have a pair of simultaneous equations
−9p − 3q = 1,
3p − 9q = 1.
80
Solutions to exercises
Adding three times the second equation to the first, to eliminate p, gives
2 1
−30q = 4, so q = − 15 , whence p = − 15 . A particular integral is thus
1 2
yp = − 15 cos 3t − 15 sin 3t.
Solution to Exercise 29
(a) Try y = pe2x .
(b) Try y = p cos 4x + q sin 4x.
Solution to Exercise 30
(a) The complementary function is yc = e−x (C cos x + D sin x), where C
and D are arbitrary constants.
To find a particular integral, try y = p0 . Substituting into the
differential equation gives
0 + 0 + 2p0 = 2p0 = 4.
Hence p0 = 2, and a particular integral is yp = 2.
Therefore the general solution is
y = e−x (C cos x + D sin x) + 2.
(b) The complementary function is θc = C + De−3t , where C and D are
arbitrary constants.
To find a particular integral, try θ = p cos 2t + q sin 2t. Differentiating
gives
dθ d2 θ
= −2p sin 2t + 2q cos 2t, = −4p cos 2t − 4q sin 2t.
dt dt2
Substituting into the differential equation gives
(−4p cos 2t − 4q sin 2t) + 3(−2p sin 2t + 2q cos 2t)
= (6q − 4p) cos 2t − (4q + 6p) sin 2t
= 13 cos 2t.
Comparing the coefficients of cos 2t and sin 2t gives a pair of
simultaneous equations to solve:
−4p + 6q = 13,
−6p − 4q = 0.
Hence p = −1, q = 32 , and a particular integral is
θp = − cos 2t + 32 sin 2t.
Therefore the general solution is
θ = C + De−3t − cos 2t + 3
2 sin 2t.
81
Unit 1 First- and second-order differential equations
Solution to Exercise 31
Putting the equation into the usual form and using R = S = Q = 1 gives
y !! − y = − 21 (l − x)x = − 12 lx + 12 x2 .
The associated homogeneous equation is y !! − y = 0, which has auxiliary
equation λ2 − 1 = 0. This has roots λ = ±1, so the complementary
function is
yc = Cex + De−x ,
where C and D are arbitrary constants.
To obtain a particular integral, we try a function of the form
y = p2 x2 + p1 x + p0 . Its derivatives are y ! = 2p2 x + p1 , y !! = 2p2 .
Substituting into the differential equation gives
2p2 − (p2 x2 + p1 x + p0 ) = −p2 x2 − p1 x + (2p2 − p0 )
= 12 x2 − 12 lx.
Hence p2 = − 21 , p1 = 21 l, p0 = −1, and a particular integral is
yp = − 12 x2 + 21 lx − 1.
Therefore the general solution is
y = Cex + De−x − 21 x2 + 21 lx − 1.
Solution to Exercise 32
(a) From Example 8, the associated homogeneous equation has general
solution y = Cex + De2x , where C and D are arbitrary constants, and
the trial solution y = pex suggested by Procedure 7 is a solution of
this equation (with C = p, D = 0). So we try y = pxex instead.
Differentiating twice gives
dy
= pex + pxex = p(1 + x)ex ,
dx
d2 y
= pex + p(1 + x)ex = p(2 + x)ex .
dx2
Substituting into the differential equation gives
p(2 + x)ex − 3p(1 + x)ex + 2pxex = −pex = 4ex .
Hence p = −4, and a particular integral is
yp = −4xex .
(b) From Exercise 18(b), the associated homogeneous equation has
general solution y = C + De−3x/2 , where C and D are arbitrary
constants, and the trial solution y = p0 suggested by Procedure 7 is a
solution of this equation (with C = p0 , D = 0). So we try y = p0 x.
Differentiating twice gives
dy d2 y
= p0 , = 0.
dx dx2
82
Solutions to exercises
Solution to Exercise 33
The auxiliary equation of the differential equation mẍ + rẋ = mg is
mλ2 + rλ = 0,
with solutions λ = 0 and λ = −r/m. The complementary function is
therefore
xc = C + De−rt/m ,
where C and D are arbitrary constants.
The inhomogeneous term is mg, so Procedure 7 suggests a trial solution
x = p0 . However, this is a solution of the associated homogeneous
equation (with C = p0 , D = 0). Hence we try x = p0 t instead, where t is
the independent variable in this problem. Differentiating and substituting
gives
rp0 = mg,
so
mg
p0 = .
r
Hence a particular integral is
mgt
xp = ,
r
and the general solution is
mgt
x = C + De−rt/m + .
r
Solution to Exercise 34
From Exercise 21(d), the associated homogeneous equation has general
solution y = (C + Dx)ex , where C and D are arbitrary constants. So not
only is the trial solution y = pex suggested by Procedure 7 a solution of
the associated homogeneous differential equation (with C = p, D = 0), but
so is y = pxex (with C = 0, D = p). So we try y = px2 ex . Differentiating
twice gives
dy
= 2pxex + px2 ex = p(2x + x2 )ex ,
dx
d2 y
= p(2 + 2x)ex + p(2x + x2 )ex = p(2 + 4x + x2 )ex .
dx2
Substituting into the differential equation gives
p(2 + 4x + x2 )ex − 2p(2x + x2 )ex + px2 ex = 2pex = ex .
Hence p = 21 , and a particular integral is
yp = 12 x2 ex .
83
Unit 1 First- and second-order differential equations
Solution to Exercise 35
(a) From Exercise 34, yp = 12 x2 ex is a particular integral for
d2 y dy
−2 + y = ex .
dx2 dx
So, using the principle of superposition, we can find a particular
integral for the given differential equation if we can find one for
d2 y dy
−2 + y = −3e2x .
dx dx
We try y = pe2x , which has derivatives
dy d2 y
= 2pe2x , = 4pe2x .
dx dx2
Substituting into the differential equation gives
4pe2x − 4pe2x + pe2x = pe2x = −3e2x .
Hence p = −3, and yp = −3e2x is a particular integral for the
differential equation with right-hand side −3e2x .
Thus, using the principle of superposition, a particular integral for the
given differential equation is
% #
yp = 4 21 x2 ex − 3e2x = 2x2 ex − 3e2x .
(b) This time we do not have a particular integral for any part of the
right-hand-side function, so we need to start from scratch.
First consider the 12 cos 2t term on the right-hand side, and try
x = p cos 2t + q sin 2t as a trial solution. This has derivatives
dx d2 x
= −2p sin 2t + 2q cos 2t, = −4p cos 2t − 4q sin 2t.
dt dt2
Substituting into the differential equation gives
2(−4p cos 2t − 4q sin 2t) + 3(−2p sin 2t + 2q cos 2t)
+ 2(p cos 2t + q sin 2t)
= 6(q − p) cos 2t − 6(p + q) sin 2t
= 12 cos 2t.
So p + q = 0, q − p = 2, hence p = −1, q = 1, and a particular integral
is
xp = − cos 2t + sin 2t.
Now consider the 10 term, and try x = p0 . Substituting into the
differential equation gives 2p0 = 10, so p0 = 5, and a particular
integral is
xp = 5.
Therefore, using the principle of superposition, a particular integral
for the differential equation with f (t) = 12 cos 2t + 10 is
xp = − cos 2t + sin 2t + 5.
84
Solutions to exercises
Solution to Exercise 36
(a) From Exercise 21(a), the complementary function is
θc = C cos 2t + D sin 2t,
where C and D are arbitrary constants.
To find a particular integral, try θ = p1 t + p0 . Substituting this and
its derivatives into the differential equation gives
4(p1 t + p0 ) = 2t.
Hence p1 = 21 , p0 = 0, and a particular integral is
θp = 21 t.
Therefore the general solution is
θ = C cos 2t + D sin 2t + 12 t.
(b) The auxiliary equation is λ2 + 4λ + 5 = 0, which has solutions
λ = −2 ± i. So (using Procedure 5) the complementary function is
uc = e−2t (C cos t + D sin t),
where C and D are arbitrary constants.
To find a particular integral, try u = p0 . Substituting gives 5p0 = 5.
Hence p0 = 1, and a particular integral is
up = 1.
Therefore the general solution is
u = e−2t (C cos t + D sin t) + 1.
(c) The auxiliary equation is 3λ2 − 2λ − 1 = 0, which has solutions λ1 = 1
and λ2 = − 13 . So the complementary function is
Yc = Cex + De−x/3 ,
where C and D are arbitrary constants.
Consider first the e2x term on the right-hand side of the equation. To
find a particular integral, try Y = pe2x . The derivatives are
dY /dx = 2pe2x and d2 Y /dx2 = 4pe2x . Substituting gives
3(4pe2x ) − 2(2pe2x ) − pe2x = 7pe2x = e2x .
Hence p = 17 , and a particular integral is
Yp = 71 e2x .
Now consider the 3 term on the right-hand side of the equation, and
try Y = p0 . Substituting gives −p0 = 3, so p0 = −3, and a particular
integral is
Yp = −3.
Therefore, using the principle of superposition, a particular integral
for the differential equation with f (x) = e2x + 3 is
Yp = 17 e2x − 3.
85
Unit 1 First- and second-order differential equations
86
Solutions to exercises
Solution to Exercise 37
(a) From Exercise 20(b), the general solution is
u = C cos 3t + D sin 3t,
where C and D are arbitrary constants. Its derivative is
u! = −3C sin 3t + 3D cos 3t.
Substituting the initial condition t = π2 , u = 0 into the general solution
gives D = 0. Substituting the initial condition t = π2 , u! = 1 into the
derivative gives C = 13 . Hence the required particular solution is
1
u= 3 cos 3t.
87
Unit 1 First- and second-order differential equations
Solution to Exercise 38
From Exercise 32(a), the general solution is
y = Cex + De2x − 4xex ,
where C and D are arbitrary constants, and its derivative is
y ! = Cex + 2De2x − 4(1 + x)ex .
Substituting the boundary condition x = 0, y ! = 2 into the derivative gives
C + 2D − 4 = 2. Substituting x = 1, y = 0 into the general solution gives
Ce + De2 − 4e = 0, which can be rearranged to give C + eD = 4. Hence
C = (8 − 6e)/(2 − e), D = 2/(2 − e), and the required particular solution is
8 − 6e x 2 2x
y= e + e − 4xex .
2−e 2−e
Solution to Exercise 39
From Exercise 31, the general solution of the differential equation is
y = Cex + De−x − 21 x2 + 12 lx − 1, which for l = 2 becomes
y = Cex + De−x − 21 x2 + x − 1,
where C and D are arbitrary constants.
The boundary conditions, resulting from the beam resting on supports at
its two ends, are y(0) = 0, y(2) = 0.
88
Solutions to exercises
Solution to Exercise 40
(a) The differential equation is the same in each case, and from
Exercise 21(a) its general solution is
u = C cos 2x + D sin 2x,
where C and D are arbitrary constants, and the derivative is
u! = −2C sin 2x + 2D cos 2x.
In this case, we have an initial-value problem.
The condition u(0) = 1 gives C = 1. The condition u! (0) = 0 gives
D = 0. The required solution is therefore
u = cos 2x.
(b) This is a boundary-value problem.
% #
The condition u(0) = 0 gives C = 0. The condition u π2 = 0 gives
C = 0 also. D therefore remains arbitrary, so there is an infinite
number of solutions, of the form
u = D sin 2x.
(c) This is an initial-value problem.
% # % #
The condition u π2 = 0 gives C = 0. The condition u! π2 = 0 gives
D = 0. The required solution is therefore the zero function
u = 0.
(Alternatively, since the differential equation is homogeneous and the
initial values are both equal to zero, by the remarks after Theorem 4,
the solution is the zero function u = 0.)
(d) This is a boundary-value problem.
%π#
The condition u(−π) = 1 gives C = 1. The condition u 4 = 2 gives
D = 2. The required solution is therefore
u = cos 2x + 2 sin 2x.
89
Unit 1 First- and second-order differential equations
Solution to Exercise 41
(a) This is an initial-value problem, therefore by Theorem 4 it has a
unique solution.
The general solution is
u = e−2t (C cos t + D sin t),
where C and D are arbitrary constants. Its derivative is
% #
u! = e−2t (−2C + D) cos t − (C + 2D) sin t .
The condition u(0) = 0 gives C = 0. The condition u! (0) = 2 gives
D = 2. The solution is therefore
u = 2e−2t sin t.
(b) This is an initial-value problem, therefore by Theorem 4 it has a
unique solution.
The differential equation is homogeneous and the initial values are
both equal to zero. Hence the solution is the zero function y = 0.
(c) This is a boundary-value problem, which may have no solution, a
unique solution, or an infinite number of solutions.
The complementary function is
xc = C cos 3t + D sin 3t,
where C and D are arbitrary constants.
To find a particular integral, try x = p1 t + p0 . Substituting into the
differential equation gives
9(p1 t + p0 ) = 3(1 − πt).
Hence p1 = − π3 , p0 = 31 , and a particular integral is
xp = − π3 t + 13 .
Therefore the general solution is
x = C cos 3t + D sin 3t − π3 t + 31 ,
and its derivative is
ẋ = −3C sin 3t + 3D cos 3t − π3 .
%π#
The condition x(0) = 13 gives C = 0. The condition ẋ 3 = 0 gives
D = − π9 . The solution is therefore
x = − π9 sin 3t − π3 t + 13 .
90
Acknowledgements
Acknowledgements
Grateful acknowledgement is made to the following source:
Figure 10: Uploaded by ‘Soerfm’, from
http://upload.wikimedia.org/wikipedia/commons/6/60/Leonhard Euler 2.jpg.
Every effort has been made to contact copyright holders. If any have been
inadvertently overlooked, the publishers will be pleased to make the
necessary arrangements at the first opportunity.
91
Unit 2
Introduction
This module discusses Newtonian mechanics. Isaac Newton (Figure 1) was
the great English mathematician whose name is given to this subject. His
Philosophiae Naturalis Principia Mathematica of 1687 (Mathematical
Principles of Natural Philosophy, known as the Principia for short)
incorporates one of the most celebrated examples of mathematical
modelling. It was in the Principia that Newton laid down the foundations
of Newtonian mechanics. This great book, which showed for the first time
how earthly and heavenly movements obey the same laws, is cast in the
form of a set of propositions all deriving from three axioms, or laws of
motion. It is these, here translated into modern English from the original
Latin, that still provide the basis for Newtonian mechanics.
Law I Every body continues in a state of rest, or moves with constant
velocity in a straight line, unless a force is applied to it.
Figure 1 Sir Isaac Newton
Law II The rate of change of velocity of a body is proportional to the (1642–1727)
resultant force applied to the body, and happens in the direction of the
resultant force.
Law III To every action (i.e. force) by one body on another there is
always opposed an equal reaction (i.e. force) – that is, the actions of two
bodies on each other are always equal in magnitude and opposite in
direction.
One of the central concepts in Newtonian mechanics is that of a force. The
word ‘force’ is used in everyday conversation in a variety of ways: he forced
his way in; the force of destiny; to put into force; the labour force. In
mathematics and science, the word ‘force’ has a precise definition.
However, this definition relies on the movement of objects thus is deferred
until the next unit. Essentially, though, this definition says that a force
either changes the shape of the object on which it acts, or causes
movement of the object. When we experience a force, in the mathematical
sense of the word, we feel it through contact: pulling on a rope, lifting a
shopping bag, pushing against a car, holding a child aloft. In each case,
the force that we experience has a magnitude and a direction, so we model
a force as a vector quantity.
We often need to represent physical quantities – such as mass, force,
velocity, acceleration, time – mathematically. Most of the physical
quantities that we need can be classified into two types: scalars and
vectors. Scalar quantities are quantities, like mass, temperature, energy,
volume and time, that can be represented by a single real number. Other
quantities, like force, velocity and acceleration, possess magnitude and
direction in space, and cannot be represented by a single real number; they
are called vector quantities.
The definitive vector quantity is displacement. The displacement of a point
specifies the position of the point in space relative to some reference point.
We use the concept of displacement whenever we want to describe spatial
relationships.
95
Unit 2 Vector algebra and statics
96
1 Describing, representing and combining vectors
We read |v| as ‘the modulus The modulus or magnitude of the vector v is denoted by |v|, or sometimes,
of v’ or ‘the magnitude of v’, or where there is no possibility of ambiguity, by v; |v| is a non-negative scalar.
simply ‘mod v’.
Exercise 1
Represent the following two vectors on a diagram by arrows:
• vector a has magnitude 3 units and points in the positive y-direction
• vector b has magnitude 4 units and points in the direction at
π/3 radians (60◦) to the positive x-direction.
What about a vector whose magnitude is zero? Clearly its length is zero,
but what is its direction? The answer is that it does not have one! We
make the following definition.
Zero vector
Be particularly careful to The zero vector is the unique vector with magnitude zero and no
underline the zero vector direction. It is denoted by 0.
(0 or 0 ) in your written work,
∼
and be aware that the vector 0
is different from the scalar 0! A constant velocity is defined by a magnitude and a direction. For
instance, in a weather forecast, a typical wind velocity might be 35 mph
from the north-west. It is not sufficient to say that ‘the wind velocity is
35 mph’; the obvious question about such a statement would be ‘from
which direction?’. The vector v representing this velocity has
magnitude 35 and direction from the north-west and towards the
south-east (since the air is travelling in the south-easterly direction).
98
1 Describing, representing and combining vectors
Two vectors are said to be equal if they have the same magnitude
and the same direction. y
3
This definition tells us that the two features needed to define a vector b
uniquely are its magnitude and direction. This means that any two arrows 2
that are drawn at different places on the page but are equal in length, are
parallel and have the same sense, can be used to represent the same vector. 1
For instance, the two arrows in Figure 5 are each of length 2 units and d
point in the positive x-direction. They represent two equal vectors, and we x
0 1 2 3
write b = d. In other words, the arrow representing a vector does not have
to be drawn so that its tail is at any particular point. Figure 5 Two equal vectors
0 0
v x 3v x
99
Unit 2 Vector algebra and statics
Exercise 3
(a) If v represents the velocity of a wind of 35 mph from the north-east,
what vector represents a wind of 35 mph from the south-west?
(b) Relate the direction and magnitude of −1.5v to those of v, where v is
any given non-zero vector. Do the same for −kv, where k is an
arbitrary positive number.
(c) If v is any non-zero vector, what are the magnitude and direction of
1
the vector v?
|v|
1
The vector v in Exercise 3(c) is a vector that has magnitude 1 and
|v|
points in the direction of v. It is called the unit vector in the direction of v
and is often denoted by the symbol v * (read as ‘v hat’).
Unit vectors are often used to denote directions in the plane or in space.
A particular example is provided by the unit vectors in the positive
directions of the x- and y-axes in the plane Cartesian coordinate system.
(We will develop the Cartesian representation of vectors in Section 2.)
These unit vectors are denoted by i and j, respectively, and are called
Cartesian unit vectors. Note that we do not denote these unit vectors
as *i and *j as they are widely used without the ‘hat’ in mathematics.
100
1 Describing, representing and combining vectors
Figure 8 shows these Cartesian unit vectors and two other vectors, a y
and b. The vector a has magnitude 2 and points in the positive
3.5 b
x-direction; b has magnitude 3.5 and points in the positive y-direction. 3
The unit vector i has magnitude 1 and points in the same direction as a.
Thus we can write a in terms of i by a scaling: 2
a = 2 i. a
1 j
Similarly, we can write b in terms of j: i
0 1 2 3 4 x
b = 3.5 j.
Any vector parallel to the x- or y-axis can be written as a scaling of i or j. Figure 8 Vectors a and b,
and two unit vectors, i and j
Note that although i and j are shown in Figure 8 with their tails at the
origin, this is not necessary. They can be drawn at any convenient
position, provided only that they are of unit magnitude and point in the
positive x- and y-directions, respectively.
Exercise 4
Let the unit vectors i and j denote the directions of east and north,
respectively. Specify the following vectors as scalings of i and j.
(a) A wind velocity of 35 km per hour due south
(b) The displacement of Bristol from London (112 miles due west)
Let us consider what is meant by the addition of vectors. Suppose that we Leeds
make a journey from Bristol to Leeds, and then another journey from d1
Leeds to Norwich. The first journey produces a displacement of d1 and the
second a displacement of d2 . The net result of the two journeys is a
d2
displacement of d3 from Bristol to Norwich. This is illustrated by the
Norwich
triangle of displacements shown in Figure 9. Displacements are said to add d3
by the triangle rule, and we write d3 = d1 + d2 . The vector d3 is called Bristol
the resultant of d1 and d2 . Figure 9 A triangle of
Velocities also add by the triangle rule, and so do forces, accelerations and displacements
all other vector quantities. Thus the triangle rule is also called the vector
addition rule.
To add any two vectors a and b: choose an origin O; draw the line b
OP in the direction of a and with length equal to the magnitude of a;
and draw the line P Q in the direction of b and with length equal to
O P
the magnitude of b (as in Figure 10). Then a + b is the vector with a
magnitude equal to the length of OQ and with direction from O to Q.
Figure 10 The triangle rule
The vector a + b is called the sum or resultant of a and b. or vector addition rule
101
Unit 2 Vector algebra and statics
Exercise 6
A vector a has magnitude 3 units and points in the positive x-direction.
A vector b has magnitude 4 units and points in the positive y-direction.
Draw a diagram showing the vectors a + b and a − b.
a Vector addition is commutative, that is, the order in which we add two
P2 Q vectors does not matter. This can be illustrated by reference to vectors a
and c of Exercise 5 (see Figure 12). The triangle OP1 Q illustrates the
a addition a + c, while triangle OP2 Q illustrates c + a. The same resultant
−−→
c c ++ c c OQ is obtained in both cases. Thus
a
a + c = c + a.
O P1 An alternative geometric construction for adding two vectors can be seen
a
from Figure 12. It is called the parallelogram rule. Draw the two
Figure 12 The vectors a and b with the same beginning point O. Complete the
parallelogram rule parallelogram OP1 QP2 . Then the resultant vector c = a + b is the vector
on the diagonal of the parallelogram. The parallelogram rule gives the
same resultant as the triangle rule.
The algebraic rules are summarised below.
102
1 Describing, representing and combining vectors
Example 1
Simplify the expression 2(a + b) + 3(b + c) − 5(a + b − c).
Solution
We have
2(a + b) + 3(b + c) − 5(a + b − c)
= 2a + 2b + 3b + 3c − 5a − 5b + 5c
= 8c − 3a.
Exercise 7
Simplify the expression 4(a − c) + 3(c − b) + 2(2a − b − 3c).
103
Unit 2 Vector algebra and statics
Figure 13 A cube
A three-dimensional Cartesian coordinate system
Consider a two-dimensional Cartesian coordinate system Oxy. Draw a
third axis, the z-axis, through the origin O, perpendicular to both the x-
and y-axes of the two-dimensional system. This produces a coordinate
system with three mutually perpendicular axes, the x-, y- and z-axes (see
The z-axes shown in Figure 14 Figure 14), intersecting at O.
are meant to point out of the
plane of the page.
104
2 Cartesian components and products of vectors
e
-p lan y
)
(y ,z B Q
3
S P 2
p2 P
1 (2, 3, 4)
O A 0
p1 x-axis x
1 2
p3 1
C 2
R 3
z4
(x, z)-plane
z-axis
(a) (b)
x
Figure 15 Choosing the
direction of positive z
105
Unit 2 Vector algebra and statics
positive y-direction
positive x-direction
positive z-direction
z
Figure 17 The right-hand grip rule
106
2 Cartesian components and products of vectors
Exercise 8
Decide which of the sets of perpendicular axes in Figure 18 define
right-handed coordinate systems.
y y x y
z
z z y x
x x z
(a) (b) (c) (d)
Figure 18
(The x-axis points out of the plane of the page in (a) and (b). The z-axis
points into and out of the plane of the page in (c) and (d), respectively.)
The components of vectors not based at the origin are defined similarly, as
follows.
107
Unit 2 Vector algebra and statics
−−→
A vector a = P Q in three-dimensional space, where P is the point
(p1 , p2 , p3 ) and Q is the point (q1 , q2 , q3 ), has component form
Note that this component form a = a1 i + a2 j + a3 k, (3)
may also be written as in (2).
where a1 = q1 − p1 , a2 = q2 − p2 , a3 = q3 − p3 , and i, j, k are the
Cartesian unit vectors. The numbers a1 , a2 , a3 are the (Cartesian)
components of a.
−−→
For example, in Figure 14(b), we have OP = 2 i + 3 j + 4 k.
As in two dimensions, the operations of vector algebra can be expressed in
terms of components.
N
a3
a2 j
a3 k
A
a2
a1 i
O a1 x
z
Figure 20 Finding the magnitude of a vector
108
2 Cartesian components and products of vectors
Exercise 9
Consider the vectors a = i + j + k, b = 2i − 3j − k and c = 3i + k.
(a) Express d = 2a − 3b and e = a − 2b + 4c in component form.
(b) Find the magnitudes of the vectors d and e.
(c) Evaluate |a|, and write down a unit vector in the direction of a.
(d) Find the components of a vector x such that a + x = b.
Exercise 10
Write the vectors 0, i, j and k as column vectors in three dimensions.
Example 2
Find the position vector of a point P lying on a straight-line segment AB
in terms of the position vectors of A and B.
Solution
−−→ y
Let P be any point on AB (see Figure 21). The position vector OP of P A
relative to the origin can also be written, using the triangle rule, as P
a
−−→ −→ −→
OP = OA + AP . r
B
−→ −−
→
Now AP = s AB, for some number s, and the point P traces out the line b
segment AB as s varies from 0 to 1. Thus the straight-line segment AB is
described by the vector equation
−−→ −→ −−→
OP = OA + s AB (0 ≤ s ≤ 1). O x
Figure 21 Finding a general
point on a straight line
109
Unit 2 Vector algebra and statics
−→ −−→ −−→
Writing a = OA, b = OB, r = OP , and noting (using the triangle rule)
−−
→ −−→ −→
that AB = OB − OA = b − a, this equation can also be written as
r = a + s(b − a) = (1 − s)a + sb (0 ≤ s ≤ 1).
Note that if the parameter s in Example 2 is allowed to range over all the
real numbers (−∞ < s < ∞), then the point P traces out the entire
straight line of which AB is a segment. Also note that the ideas in
Example 2 are easily extended to three dimensions.
Exercise 11
(a) Write down, in component form, the vector equation of the straight
line on which lie the points with Cartesian coordinates (1, 1, 2) and
(2, 3, 1).
(b) Find the coordinates of the point where the line cuts the (x, z)-plane.
Exercise 12
Let a = 2i − j, b = i + 3j + 5k and c = i + j − 2k.
(a) Find the magnitudes of a and b, and describe the direction of a.
(b) Find the vectors a + b, 2a − b and c + 2b − 3a in component form.
(c) What is the endpoint Q of the displacement represented by the vector
2a − b if (0, 2, 3) is its beginning point P ?
110
2 Cartesian components and products of vectors
Exercise 13
(a) A straight line in Cartesian form is expressed as
x+1 y−8
= = 1 − z.
3 −2
Express this equation in parametric and vector form. If A is the point
corresponding to s = 0, and B is the point corresponding to s = 1,
−−
→
determine the component form of AB and the distance between A
and B.
(b) (i) Determine the vector position of M , the point midway between A
and B.
(ii) Determine the vector position of N , the point that divides the
line AB in the ratio 3 : 2.
(c) Find the vector equation of the straight line through the points
(−1, −2, 0) and (2, 1, −1).
(d) Determine whether the lines in parts (a) and (c) intersect. If so,
determine the coordinates of the point of intersection.
θ
The dot product of two vectors is a scalar quantity, that is, it is a real
number: a · b is the product of the three scalars |a|, |b| and cos θ. So the
a
operation of the dot product combines two vectors to define a scalar, and
for this reason the dot product is also called the scalar product. The Figure 22 The angle
angle θ lies in the range 0 ≤ θ ≤ π, and the value of a · b is: between two vectors
• positive for 0 ≤ θ < π2 , i.e. when θ is an acute angle
π
• negative for 2 <θ ≤ π, i.e. when θ is obtuse
π
• zero for θ = 2 , i.e. when θ is a right angle.
111
Unit 2 Vector algebra and statics
b
Exercise 14
Three vectors a, b and c of magnitudes 2, 4 and 1 units, respectively, lying
4 in the same plane, are represented by arrows as shown in Figure 23. The
angle between the vectors a and b is π3 radians, and that between the
c π vectors b and c is π6 radians. Use the definition of the dot product to find
6
1 π the values of a · b, b · c, a · c and b · b.
3
2 a
This exercise demonstrates two important properties of the dot product.
Figure 23
• If two vectors a and b are perpendicular to each other (i.e. the angle
between them is π2 radians), then since cos π2 = 0,
a · b = |a| |b| cos π2 = 0.
• The dot product of a vector with itself gives the square of the
magnitude of the vector, that is,
a · a = |a| |a| cos 0 = |a|2 .
The converse of the first property also holds: if a and b are two non-zero
vectors such that a · b = 0, then the definition of the dot product tells us
that cos θ = 0; therefore θ = π2 and the vectors are perpendicular.
In a product of real numbers, xy = 0 implies that either x or y (or both) is
zero. In contrast, for the dot product, a · b = 0 gives an extra possibility:
either a or b (or both) is the zero vector, or the angle between a and b is
π
2 radians.
The following example shows how these properties can be used to simplify
expressions.
112
2 Cartesian components and products of vectors
Example 3
Expand the expression x · y, given that x = 2u + v and y = u − 5v.
Calculate its value when u and v are perpendicular unit vectors.
Solution
x · y = (2u + v) · (u − 5v)
= 2(u · u) − 10(u · v) + v · u − 5(v · v)
= 2(u · u) − 9(u · v) − 5(v · v).
Now u · u = |u|2 = 1 and v · v = |v|2 = 1 when u and v are unit vectors.
Furthermore, u · v = 0 when u and v are perpendicular vectors. So when
u and v are perpendicular unit vectors, we have
x · y = 2 − 0 − 5 = −3.
Exercise 15
(a) Expand the expression (a + b) · (a − b).
(b) Expand the expression |a + b|2 . Recall that |v|2 = v · v.
Exercise 16
Suppose that a and b are perpendicular unit vectors.
(a) Find the value of m such that the two vectors 2a + 3b and ma + b are
perpendicular.
(b) Find the value of |c| if c = 3a + 5b.
A word of caution: (a · b)c is not in general the same as a(b · c). The In general, if m is a scalar and a
vector (a · b)c is a scaling of c by the number a · b, whereas a(b · c) is a is a vector, we can write ma or
scaling of a by the number b · c. Clearly these two vectors are not generally am as is convenient, although
ma is more usual. Thus a(b · c)
even parallel, let alone equal. For example, if a = b = i and c = j, then means the same as (b · c)a.
(a · b)c = (i · i)j = j but a(b · c) = i(i · j) = 0.
j First, observe that by definition, i, j and k are unit vectors and are
perpendicular to one another (see Figure 24). Thus
i · j = j · i = 0, i · k = k · i = 0, j · k = k · j = 0,
i
k i · i = 1, j · j = 1, k · k = 1.
Figure 24 The Cartesian If two vectors a and b have component forms a = a1 i + a2 j + a3 k and
unit vectors (k points out of b = b1 i + b2 j + b3 k, then the dot product of a and b may be written as
the plane of the page)
(a1 i + a2 j + a3 k) · (b1 i + b2 j + b3 k).
We can now apply properties of the dot product and the above rules for
combining i, j and k to this expression to obtain a very simple formula for
the dot product of vectors in component form. Specifically, we have
(a1 i + a2 j + a3 k) · (b1 i + b2 j + b3 k)
= a1 b1 (i · i) + a1 b2 (i · j) + a1 b3 (i · k)
+ a2 b1 (j · i) + a2 b2 (j · j) + a2 b3 (j · k)
+ a3 b1 (k · i) + a3 b2 (k · j) + a3 b3 (k · k)
= a1 b1 + a2 b2 + a3 b3 .
This extremely important formula is worth remembering.
Exercise 17
If a = 4i + j − 5k and b = i − 3j + k, show that a · b = −4. What does the
negative sign tell us?
114
2 Cartesian components and products of vectors
Example 4
√ √
Let a = i + j + 2k, b = 2i + j + k and c = i + j − 2k.
(a) Find the angle between the vector a and the x-axis.
(b) Find the angle between the vectors a and b.
(c) Show that c is perpendicular to a.
Solution
(a) The direction of the x-axis is the same as the direction of i, and the
angle θ between a and i is given by
a·i a1 1 1
cos θ = = =√ = .
|a| |i| |a| 1+1+2 2
Thus the angle between a and the x-axis is π/3 radians.
√ √ √
(b) We have |a| = 1 + 1 + 2 = 2, |b| = 4 + 1 + 1 = 6 and
√ √
a · b = (1 × 2) + (1 × 1) + ( 2 × 1) = 3 + 2.
Therefore the angle θ between a and b is given by
√
3+ 2
cos θ = √ = 0.901 05,
2× 6
so θ = 0.4486 (radians).
(c) To test whether a and c are perpendicular, we calculate their dot
product:
√ √
a · c = (i + j + 2k) · (i + j − 2k)
√ √
= (1 × 1) + (1 × 1) + ( 2 × 2)
= 0.
Since a · c = 0 and a and c are non-zero vectors, c is perpendicular
to a.
Exercise 18
Consider the vectors a = 2i − 3j + k and b = −i + 2j + 4k.
Find the magnitudes of a and b, and the angle between them.
115
Unit 2 Vector algebra and statics
Exercise 19
If a = a1 i + a2 j + a3 k, find the values of a · i, a · j and a · k.
A
a The solution to Exercise 19 shows the important fact that the i-component
of any vector a may be found by taking the dot product a · i. The j- and
k-components can be found similarly (by taking dot products with j
θ P and k, respectively).
*
u We can also find the components of a vector in other directions. Suppose
O −→
Figure 25 Finding the *
that a vector a, represented by OA, makes an angle θ with a unit vector u
component of a vector in an *.
(see Figure 25). Draw the line AP perpendicular to the direction of u
arbitrary direction Then the distance OP is seen from simple trigonometry to be |a| cos θ.
* is
Now observe that the dot product of a and u
Note that a · u * will be negative * = |a| |*
a·u u| cos θ = ±|OP | (since |*
u| = 1).
if θ > π2 , i.e. if P and u* lie on
opposite sides of O. The signed distance ±|OP | represents the component of a in the direction
*.
of u
Exercise 20
Consider the vectors a = 2i − 3j + k and b = −i + 2j + 4k.
(a) Which of the vectors
c = −i + j + 3k, d = −2i + k, e = −i − j − k,
is perpendicular to a?
(b) Find the component of the vector a + 2b in the direction of the line
joining the origin to the point (1, 1, 1).
Example 5
Figure 26 shows a line inclined at an angle α to the x-axis, and unit
vectors i and j aligned along and perpendicular to the line. The vectors
a, b, c and d have magnitudes 1, 1.5, 1.5 and 2, respectively.
y
c
b
d
j a
i α
x
Figure 26
α
j π
−α a
i α 2
x
Figure 27
117
Unit 2 Vector algebra and statics
118
2 Cartesian components and products of vectors
b *
d b
a a
*
n
(a) (b)
Figure 29 Using the right-hand grip rule for the cross product:
(a) a × b; (b) b × a
Exercise 22
Three vectors u, v and w lie in the (x, y)-plane. Their magnitudes are 2, 3
and 4 units, respectively, their directions make angles π6 , π3 and π6 radians,
respectively, with the positive x-axis, and they have positive j-components.
Use the definition of the cross product to find the vectors u × v, u × w
and v × w.
119
Unit 2 Vector algebra and statics
These properties can all be derived from the definition of the cross
product, but the derivations are not given here.
Exercise 23
(a) Show that i × j = k, j × k = i and k × i = j.
(b) Calculate j × i, k × j and i × k.
(c) Calculate i × i, j × j and k × k.
(d) Expand and simplify
(i + k) × (i + j + k) and (i × (i + k)) − ((i + j) × k).
(e) For two non-zero non-parallel vectors a and b, simplify
(a + b) × (a + 2b) and (a + b) × (a − b).
120
2 Cartesian components and products of vectors
This formula is not easy to remember or use in this form. Another quick
way to evaluate cross products is to use determinants. This method will be The determinant form of the
introduced in Unit 4 when we discuss determinants. If you already know cross-product a × b, where
this method, then we suggest that you continue to use it. a = a1 i + a2 j + a3 k and
b = b1 i + b2 j + b3 k, can be
written as + +
Exercise 24 +i j k+
+ +
a × b = + a1 a2 a3 +
If +b b2 b3 +
1
= (a2 b3 − a3 b2 )i
a = 2i − 3j + k, b = −i + 2j + 4k and c = −4i + 6j − 2k,
+ (a3 b1 − a1 b3 )j
find a × b, a × c and b × c. From your results, what can you say about a + (a1 b2 − a2 b1 )k.
and c?
Exercise 25
If
a = 2i + 2j + k and b = 4i + 4j − 7k,
find a unit vector whose direction is perpendicular to the directions of both
a and b.
Geometric applications
We close this subsection with some useful geometric applications of the
cross product. The following example is the first step. b
Example 6 θ
Any two non-zero and non-parallel vectors a and b define a parallelogram, a
as shown in Figure 31. Express the area of the parallelogram in terms of Figure 31 A parallelogram
a × b. defined by two vectors
121
Unit 2 Vector algebra and statics
Solution
|a|
The area A of the parallelogram defined by the two vectors a and b is the
b
same as the area of the rectangle of height |b| sin θ and width |a| (see
|b| sin θ Figure 32). Thus A = |a| |b| sin θ, and this is the magnitude of a × b. So
θ A = |a × b|.
a
Figure 32 Finding the area
of a parallelogram
Area of a parallelogram
The area of a parallelogram with sides defined by vectors a and b is
|a × b|.
This idea is easily extended for the area of a triangle. Any two non-zero
b non-parallel vectors a and b define a triangle (see Figure 33). The area of
this triangle is half that of the corresponding parallelogram, so it is
1
2 |a × b|.
a
Area of a triangle
Figure 33 Finding the area
of a triangle The area of a triangle with sides defined by vectors a and b is
1
2 |a × b|.
Using the formula for the area of a parallelogram, we can find the volume
A parallelepiped is like a of a parallelepiped (see Figure 34).
distorted brick. All of its faces
are parallelograms. The base is a parallelogram (assumed to be in the (x, z)-plane) defined by
the vectors a and b. The base therefore has an area equal to the
magnitude of a × b. Now the vertical height h is the component of the
vector c in the direction of the Cartesian unit vector j pointing vertically
c upwards, that is, it is the y-component of c, given by c · j = j · c. So the
j
h volume of the parallelepiped is |a × b|(j · c). But the vector product a × b
b points vertically upwards and can therefore be expressed as |a × b| j.
Hence we have
a
|a × b|(j · c) = (|a × b| j) · c = (a × b) · c.
Figure 34 Finding the
volume of a parallelepiped Of course, the scalar (a × b) · c can be negative if one of the defining
vectors a or b is chosen to be in the direction opposite to the one chosen in
Figure 34, or if the order of the cross product is reversed. We use modulus
signs to ensure that the volume comes out positive:
volume of parallelepiped = base area × vertical height h
= |(a × b) · c|. (15)
The scalar quantity (a × b) · c is an example of a scalar triple product.
122
3 Modelling forces
Exercise 26
Suppose that the vectors r and s are directed towards north and
north-east, respectively, and define r × s = t.
(a) What is the direction of t?
(b) In what direction is s × r?
(c) In what direction is t × r?
(d) If |r| = |s| = 1, what is |t|?
(e) Calculate the vector t × (r × s).
(f) If |r| = |s| = 1, what is the value of r · s?
(g) If |r| = |s| = 1, what is the value of s · (t × r)?
Exercise 27
Find the area of the triangle ABC, where the coordinates of A, B and C
are (2, 1, −3), (1, 0, 2) and (4, −2, −1).
3 Modelling forces
This section shows how four common types of force can be modelled: the
force of gravity, the force exerted by a surface on an object in contact with
it, the tension force due to a string, and the friction force between two
surfaces. These forces and the situations in which they occur are modelled
and analysed in Subsections 3.2–3.5. In Subsection 3.6 we look at the
forces on two particle systems. First, however, we look at one way of
modelling the objects on which forces act.
3.1 Particles
When we create a mathematical model, the aim is to simplify the real
situation being modelled so that only the essential features are included.
This enables us to analyse the situation mathematically. In mechanics, the
most important things to model are the forces acting on objects, and
throughout this unit and the other mechanics units you will see how to do
this. However, we also need to model the objects on which the forces act.
The simplest model for an object is a particle.
123
Unit 2 Vector algebra and statics
3.2 Weight
When you hold a shoe, your fingers experience a force. The shoe, like all
objects, has a force associated with it, and if you do not provide opposition
to this force in holding the shoe, it will fall to the ground. But what is the
source of the force exerted by the shoe?
This force is due to the attraction of the shoe to the Earth. The force of
attraction of objects to the Earth is called the force of gravity or the
gravitational force. The gravitational force acting on a particular object
is not constant, but depends on the position of the object relative to the
Earth: there is a small variation of this force with height above ground (or
depth below ground), and there is an even smaller variation with latitude
and longitude. When applied to a particular object, this force is called the
weight of the object. In this module, we assume that the weight of a
particular object is constant near the Earth’s surface.
In everyday speech, the words mass and weight are interchangeable.
Mathematically, however, they are different. The mass of an object is the
amount of matter in the object and is independent of the object’s position
in the Universe; it is a scalar quantity, measured in kilograms (kg) in the
SI system. The weight of an object is the gravitational force on the
object, and is dependent on where the object is situated; it is a vector
quantity, whose magnitude is measured in newtons (N) in the SI system
and whose direction is downwards towards the centre of the Earth.
124
3 Modelling forces
Mass and weight are, however, related in that an object of mass m has The relationship between mass
weight of magnitude mg, where g is a constant known as the magnitude and weight is based on Newton’s
of the acceleration due to gravity. Near the Earth’s surface, g has the second law of motion, which is
discussed in Unit 3.
value approximately 9.81 m s−2 , and we assume this value for g throughout
this module. If the Cartesian unit vector j points vertically upwards from
the surface of the Earth, then the weight W of an object of mass m is
mg(−j) (where we need the negative sign because the force of gravity acts
vertically downwards, that is, the weight acts vertically downwards).
Exercise 28
What is the weight (in newtons) of a particle of mass 3 kg in a coordinate
system where the k-direction is vertically downwards?
i
Exercise 29 π/6
15 kg
Later in this unit we will find it convenient to use axes that are not
horizontal and vertical. Express the weight W of a particle of mass 15 kg j
W
in terms of the Cartesian unit vectors i and j, where i and j both lie in a
vertical plane and are oriented as shown in Figure 37. Figure 37 Using axes that
are not horizontal and
vertical
In Exercise 29, ‘nice’ angles (i.e. multiples of π6 (30◦)) were chosen in order
to help you to evaluate the sine and cosine involved without having to use a
calculator. The sines and cosines of some ‘convenient’ angles are as follows: These values are given in the
√ Handbook.
3
sin 0 = 0, sin π6 = 21 , sin π4 = √1 ,
2
sin π3 = 2 , sin π2 = 1,
√
3
cos 0 = 1, cos π6 = 2 , cos π4 = √1 ,
2
cos π3 = 1
2, cos π2 = 0.
Sometimes, obtuse angles are used; sines and cosines of such angles can be
derived from the addition formulas given in the Handbook. For example,
%π π,
cos 2π π π π π 1
3 = cos 2 + 6 = cos 2 cos 6 − sin 2 sin 6 = − 2 .
125
Unit 2 Vector algebra and statics
126
3 Modelling forces
the cable, so the cable must exert a force on the lamp. The force exerted
by the cable on the lamp is a vector quantity called the tension force.
Tension forces occur whenever objects are tautly joined, for example by
cables, ropes, strings or threads. These cables and ropes can be modelled
in different ways. For example, if we want to model the ceiling lamp and
are interested only in the force in the cable, then we can model the cable
as a model string, defined as an object possessing length, but no area,
volume or mass, and which does not stretch (i.e. it is inextensible). On the
other hand, if we are interested in how much the cable stretches under the
weight of the lamp, then we can model the cable as a model spring, which Springs are discussed in Unit 9.
has properties similar to those of a model string (i.e. it has no area, volume
or mass), but allows extension. In this unit we consider only strings.
The ceiling lamp example is illustrated in Figure 40. The tension force due
T
to the model string is denoted by T, and the weight of the lamp is denoted
by W. In a manner similar to the case of normal reaction forces, the j m
equilibrium condition for particles gives
W + T = 0. W
If the lamp has mass m, then W = mg (−j), hence Figure 40 Modelling a lamp
hanging from a ceiling
T = −W = mg j,
that is, tension is a force acting vertically upwards (along the length of the
model string) with the same magnitude as the weight of the lamp.
We assume that the tension force due to a model string acts along the
length of the string and away from the point of its attachment to an
object. As in the case of a normal reaction, the magnitude of this force (a
scalar quantity, often referred to as the tension in the string) depends on
the requirements necessary to maintain equilibrium, so it is potentially
unlimited. (In reality, a string can exert only a certain tension force before
it breaks, but a model string supports an unlimited tension force.)
As in the case of normal reaction forces, the tension force due to a string
need not be vertically upwards, as the following example illustrates.
Example 7 π
3 j
A hanging flower basket of mass 4 kg is suspended by one cord from a
porch and tied by another cord to the wall, as shown in Figure 41. Model
4 kg
the basket as a particle and the cords as model strings. What are the i
magnitudes of the tension forces due to the cords?
Figure 41 A flower basket
hanging in a porch
127
Unit 2 Vector algebra and statics
Solution
We choose axes as shown in Figure 41. Note that we need choose only two
axes because all the forces act in the same vertical plane. Denoting the
T1 tension forces by T1 and T2 , and the weight of the basket by W, we have
π the force diagram shown in Figure 42.
π 3
6 j In the diagram, the angle between the vector T1 and the unit vector j is
π
T2 3 calculated by imagining the right-angled triangle shown, and using the fact
i that the angles of a triangle sum to π radians. The angle between the
vectors T1 and i is π3 .
W
The equilibrium condition for particles tells us that
Figure 42 The force
diagram for the hanging T1 + T2 + W = 0. (17)
basket
To progress further, we need to express the three forces in terms of the
unit vectors i and j. To do this, we apply Procedure 1, the technique for
resolving vectors. Starting with the weight W, where |W| = 4g and W
acts in the direction −j, we have
W = −4g j.
Similarly, the tension force T1 has magnitude |T1 | in the direction
A vector u = cos α i + sin α j has * = cos π3 i + sin π3 j and can be expressed as
u
magnitude √
" 3
|u| = cos2 α + sin2 α = 1 T1 = |T1 | cos π3 i + |T1 | sin π3 j = 12 |T1 | i + 2 |T1 | j.
and is thus a unit vector. Finally, the tension force T2 has magnitude |T2 | in the direction −i and
can be written as
T2 = −|T2 | i.
Substituting in equation (17), the equilibrium condition is
√
1 3
2 |T1 | i + 2 |T1 | j − |T2 | i − 4g j = 0.
Either by separating out the i- and j-components, or equivalently, taking
the dot product with i and j in turn, gives the two scalar equations
1
2 |T1 | − |T2 | = 0, (18)
√
3
2 |T1 | − 4g = 0. (19)
Equation (19) gives
√
|T1 | = 8g/ 3 ) 45.31.
Substituting this into equation (18) gives
√
|T2 | = 4g/ 3 ) 22.66.
So the model predicts that the tension force due to the cord from the porch
has magnitude about 45.3 N, and the tension force due to the cord from
the wall has magnitude about 22.7 N, both correct to one decimal place.
The procedure that was used in Example 7 can be used to solve many
problems in statics, and may be summarised as follows.
128
3 Modelling forces
In this unit, the steps in this procedure will often be identified (using the
marginal abbreviations shown above) in the solutions to examples and
exercises. However, the procedure is intended to be a guide rather than a
rigid set of rules. For example, if it is not obvious which set of axes to
choose, then draw the force diagram first, and the best choice may become
more apparent. Try using the procedure in the following exercise.
Exercise 30
During December, a large plastic Christmas tree of mass 10 kg is
suspended by its apex using two ropes attached to buildings either side of
the high street of Treppendorf. The ropes make angles of π/6 and π/4
with the horizontal. Model the Christmas tree as a particle and the ropes
as model strings. What are the magnitudes of the tension forces due to the
two ropes?
3.5 Friction
sideways
Consider a book resting on a horizontal surface such as a table. There are push
two forces acting on the book: the weight downwards and the normal
reaction upwards. Suppose that you push the book gently sideways (see
Figure 43). If you do not push hard enough, the book will not move; it will
remain in equilibrium. The force opposing motion along the surface is Figure 43 Pushing a book
known as the friction force. It is considered to act parallel to the surface, resting on a table
that is, at right angles to the normal reaction, and in a direction that
opposes any (possible) motion along that surface. Modelling the book as a
N
particle, and denoting the pushing force by P, the friction force by F, the
weight by W and the normal reaction by N, the force diagram for this F P
example is shown in Figure 44.
W
Figure 44 The force
diagram for the book being
pushed
129
Unit 2 Vector algebra and statics
Friction forces are caused by the roughness of even seemingly very smooth
surfaces – a roughness that serves to inhibit the smooth movement of one
surface over another. So friction forces are present only where there is
movement or the possibility of movement. There is no friction force
present when an object is resting on a horizontal surface, where the only
two forces acting on the object are its weight and the normal reaction. But
when an object is being pushed or pulled, or is resting on a sloping surface,
In this unit we consider only then a friction force is present (see Figure 45).
cases where objects remain at
rest, so that there is only the N N
possibility of movement. Friction
in cases where there is
F
movement is considered in
Unit 3.
W W
no friction friction force
force present
130
3 Modelling forces
Example 8
A steel fork of mass 0.05 kg rests on a horizontal wooden table. Model the
fork as a particle. What is the maximum sideways force that can be
applied before the fork starts to move?
Solution
The situation is illustrated in Figure 46. Since all the forces act in a " Draw picture !
vertical plane, we can choose axes as shown. The force diagram is also " Choose axes !
shown in the figure, where F is the friction force, P is the sideways force,
" Draw force diagram(s) !
W is the weight, and N is the normal reaction. The obvious choice for the
directions of the unit vectors i and j is horizontal and vertical, respectively,
as all the forces act in these directions.
sideways j N
force 0.05 kg F P
µ = 0.55 i
W
131
Unit 2 Vector algebra and statics
Let us now apply these ideas to some examples, in which we will also apply
the steps of Procedure 2. In most of the situations that we investigate, we
will be concerned with limiting friction.
Exercise 31
A steel block of mass 0.5 kg rests on a horizontal dry steel surface (with
coefficient of static friction µ = 0.74) and is pulled by a horizontal force
of 2 N. Model the block as a particle. Use Procedure 2 to determine
whether the block will move. What is the magnitude of the friction force?
Exercise 32
(a) (b) A shallow box made of a uniform material and without a lid can be placed
on a horizontal table in two possible ways (as shown in Figure 47):
Figure 47 A box without a
lid: (a) with its base in (a) with its base in contact with the table surface
contact with a table; (b) with (b) with its open top in contact with the table surface.
its open top in contact with
the table Which of these two positions requires the smaller sideways force to start
the box slipping?
N
F
Inclined planes
W Consider now an object resting on a sloping plane surface, often referred to
as an inclined plane, such as the one shown in Figure 48. Provided that
Figure 48 An object on an the angle of inclination is not large, the object can remain at rest and does
inclined plane, with its force
not slide down the slope. The forces acting on the object are its weight,
diagram
132
3 Modelling forces
the normal reaction and friction. The weight W acts vertically downwards.
The normal reaction N acts normally to the surface between the object
and the slope. The friction force F is perpendicular to the normal reaction
and hence parallel to the slope, and it acts up the slope to counteract the
natural tendency of the object to move down the slope.
Example 9
A crate of empty bottles of total mass 30 kg is to be hauled by a rope up a
ramp. The rope is parallel to the ramp, and the ramp makes an angle of
π/6 radians with the horizontal. The coefficient of static friction between
the plastic crate and the wooden ramp is 0.2.
What is the tension force due to the rope when the crate is on the point of
moving upwards?
Solution
The situation is illustrated in Figure 49. " Draw picture !
j
µ = 0.2 30
k g
π
6 i
All the forces act in a vertical plane, so we need only two axes. We could " Choose axes !
choose i to be horizontal and j vertical as before, but it makes calculations
easier if we choose i to be parallel to the slope and j perpendicular to it, as
shown in Figure 49. This is because when we come to resolve the forces in
the i- and j-directions, three of the four forces (all except W) will then act
along one or other of the axes, making resolving them much simpler.
Modelling the crate as a particle and the rope as a model string, the force " Draw force diagram(s) !
diagram is as shown in Figure 50, where W is the weight, N is the normal
reaction, F is the friction force, and T is the tension force.
T N j
π π F i
6 3 π
6
W
Figure 50 The force diagram for the crate
133
Unit 2 Vector algebra and statics
Exercise 33
A full crate of bottles of mass 60 kg is at the top of the ramp described in
Example 9, ready to be lowered down it. What force needs to be applied
to the rope to keep the crate from sliding down the ramp?
134
3 Modelling forces
Exercise 34
(a) A box of mass m is resting on a surface inclined at an angle α to the
horizontal. If the box is on the point of slipping, what is the coefficient
of static friction?
(b) Two identical mugs are placed on a tray. One mug is half full of coffee,
while the other is empty. The tray is tilted slowly. Use your answer to
part (a) to determine which mug will start to move first.
Pulleys
The pulley is a common device with which you are probably familiar. You
may have seen pulleys in use on building sites, for example, as an aid to
raising or lowering heavy loads. The idea of a pulley is useful in modelling
mechanics problems, as it enables us to model a change in direction of a
tension force.
In diagrams, we use an idealised pulley as shown in Figure 51. In order to
keep the model simple, we make simplifying assumptions, which are
formally stated in the following definition.
The point to remember is that the result of these assumptions implies that
the tension forces due to the string on either side of the model pulley are
equal in magnitude, that is, the magnitude of the tension in the string is
the same on either side of the pulley (although the tension forces act in
different directions).
135
Unit 2 Vector algebra and statics
Example 10
A sack of flour of mass 50 kg is lying on the floor of a mill, ready to be
loaded into a cart. To help with the loading process, a light rope is
attached to the sack, passes over a pulley fixed to the ceiling immediately
j above the sack, and is attached at its other end to a stone of mass 15 kg
15 kg that hangs without touching the floor. The system is shown in Figure 52.
Model the sack and the stone as particles, the pulley as a model pulley,
50 kg and the rope as a model string, and consider the forces acting on each
particle separately.
(a) Calculate the normal reaction of the floor on the sack.
Figure 52 Raising a sack of
flour using a pulley (b) What force does the pulley exert on the ceiling?
Solution
" Choose axes ! (a) All the forces are vertical, so we need only one axis, as shown in
Figure 52.
" Draw force diagram(s) ! The force diagrams for the sack and the stone are shown in Figure 53,
where W1 and W2 represent the weights, T1 and T2 represent the
tension forces, and N is the normal reaction of the floor on the sack.
T1
N T2
j
50 kg 15 kg
W1 W2
sack stone
Figure 53 Force diagrams for the sack and the stone
136
3 Modelling forces
To solve the equations, the first step is to write the forces in " Solve equation(s) !
component form as
W1 = |W1 | (−j), N = |N| j, T1 = |T1 | j,
T2 = |T2 | j, W2 = |W2 | (−j).
Then resolving equations (24) and (25) in the j-direction gives
−|W1 | + |N| + |T1 | = 0,
−|W2 | + |T2 | = 0.
Therefore |T1 | = |T2 | = |W2 |, so
|N| = |W1 | − |T1 | = |W1 | − |W2 | = 50g − 15g = 35g ) 343.
So the normal reaction of the floor on the sack has magnitude about " Interpret solution !
343 N (35g) and is directed upwards. In the absence of the stone and
the pulley, the magnitude of the normal reaction would have been
equal to the magnitude of the weight of the sack (50g). The effect of
the stone, transmitted via the pulley, is as if the magnitude of the
sack’s weight were reduced by the magnitude of the weight of the
stone (15g).
Note that the sack remains in contact with the floor if |N| > 0, and is
on the point of losing contact if |N| approaches zero. The model
breaks down if the sack loses contact with the floor with |N| = 0.
(b) To answer this question, we need to model the forces on the pulley. " Draw force diagram(s) !
We can consider the pulley as a particle of no mass. Let T3 and T4
represent the tension forces in the left- and right-hand ropes,
respectively. Modelling the short piece of metal that attaches the
pulley to the ceiling as a model string, with tension force T5 , we have
the force diagram shown in Figure 54.
T5
j
T3
T4
pulley
137
Unit 2 Vector algebra and statics
Exercise 35
Suppose that the pulley in Example 10 is no longer immediately above the
sack, so that the rope attached to the sack makes an angle of π/4 to the
j
vertical, as shown in Figure 55.
(a) Set up the problem using Procedure 2.
π
i 15 kg
4 (b) Find the magnitude of the normal reaction of the floor on the sack.
(c) Find the magnitude of the friction force on the sack.
50 kg
(d) Find the smallest value of the coefficient of static friction that would
allow the system to remain in equilibrium.
Figure 55 (e) Find the weight of stone such that the sack just remains in contact
with the floor.
Slipping
You have already investigated slipping in the case of a one-particle system.
In this subsection we examine the phenomenon in systems of more than
one particle.
Example 11
Consider a scarf draped over the edge of a table. Model the scarf as two
particles, one of mass m1 hanging over the edge, and the other of mass m2
resting on the table, with the masses joined by a model string passing over
the edge of the table, which is modelled as a model pulley. Assume that the
scarf’s mass is uniformly distributed along its length, so that the masses of
the two particles are proportional to the corresponding lengths of scarf.
138
3 Modelling forces
If the coefficient of static friction between the scarf and the table surface
is µ, what proportion of the scarf’s length can hang over the edge of the
table before the scarf slips off the table?
Solution
We can answer this question if we can find the ratio of m1 (the mass of
scarf hanging over the edge) to m1 + m2 (the total mass of scarf) when the
scarf is on the verge of slipping.
The situation is illustrated in Figure 56, which also shows a suitable choice " Draw picture !
of axes. " Choose axes !
m2
j N T1
F m2 T2
m1 m1
i
W2 W1
Figure 56 Modelling a scarf draped over the edge of a table, with force
diagrams
There are two forces acting on the hanging particle: its weight W1 and the " Draw force diagram(s) !
tension force T1 . There are four forces acting on the particle on the table:
its weight W2 , the tension force T2 , the normal reaction force N, and the
friction force F. The force diagrams are shown in Figure 56.
While the scarf does not slip, we can apply the equilibrium condition for " Apply law(s) !
particles to each particle in turn. For the hanging particle, we have
T1 + W1 = 0. (27)
For the particle on the table, we have
F + N + T2 + W2 = 0. (28)
The assumption of a model pulley gives
|T1 | = |T2 |. (29)
When the particle is on the verge of slipping, we have
|F| = µ|N|. (30)
From Figure 56, the component forms of the force vectors can immediately " Solve equation(s) !
be written down:
T1 = |T1 | j, W1 = |W1 | (−j),
F = |F| (−i), N = |N| j, T2 = |T2 | i, W2 = |W2 | (−j).
Resolving equation (27) in the j-direction gives
|T1 | − |W1 | = 0,
so
|T1 | = |W1 | = m1 g.
139
Unit 2 Vector algebra and statics
Exercise 36
m A man of mass 80 kg is about to be lowered into a well from a rope that
80 kg
passes over a horizontal rotating axle above the well. The other end of the
µ = 0.45 rope is held by several men each of mass 80 kg, as shown in Figure 57, with
the rope between the men and the axle horizontal. Assume that n men can
j be represented by a single particle of mass 80n kg, the rope by a model
string and the axle by a model pulley.
i If the coefficient of static friction between the men’s boots and the ground
Figure 57 Lowering a man is 0.45, how many men are required to hold the man at the end of the rope
into a well before he is lowered into the well?
Exercise 37
An object of mass m1 , resting on a board inclined at an angle α to the
horizontal, is attached to an object of mass m2 by a string hanging over
m1 the edge of the board, as shown in Figure 58.
m2 Assuming that the objects can be modelled as particles, the string as a
α
model string and the edge of the board as a model pulley, find the
Figure 58 Two connected condition on the coefficient of static friction µ between the first object and
particles on an inclined plane the board for this system to remain in equilibrium.
(Hint: There are two ways in which the equilibrium can be disturbed.)
140
4 Torques
4 Torques
This section looks at solid bodies, and in particular at a phenomenon that
does not apply to particles: the turning effect of forces. We begin in
Subsection 4.1 by introducing ways of modelling objects when their size is
important, as it is when the turning effects of forces are considered.
Subsection 4.2 goes on to explain what is meant by the turning effect of a
force, and to provide a mathematical description of such an effect.
141
Unit 2 Vector algebra and statics
the ruler, if the two forces are not in line, then turning occurs. The weight
provides a turning effect if the vertical line of its action (through the
centre of mass) does not pass through your finger.
How do we measure the turning effect of a force in terms of what we
already know, namely the magnitude and direction of the force, and the
point on the object at which the force acts? In order to begin to answer
this question, try a simple experiment. Balance a 30-centimetre (12-inch)
ruler on a pencil rubber or a hexagonal pencil, or some other object that is
not too wide and so will act as a pivot. Place two identical small coins on
either side of the pivot so that each is 10 cm from the pivot (see Figure 59).
Figure 59 A ruler balanced
on a pencil, with coins at Then experiment with moving one of the coins in steps of 2 cm from its
each end initial position, and see how the other coin has to be moved in order to
re-establish balance. The conclusion from this experiment is not, perhaps,
a surprising one: coins of equal mass have to be placed at equal distances
on either side of the pivot for the ruler to remain balanced.
Next place the two coins together at a point on the ruler, say at 6 cm from
the pivot. Where does a third identical coin have to be placed to achieve
balance?
You should find that two identical coins placed together at 6 cm from the
pivot are balanced by another identical coin placed on the other side of the
ruler at 12 cm from the pivot. If you continue to experiment with varying
numbers of coins placed at various pairs of positions along the ruler, you
will find in each case that if the masses of the two sets of coins are
unequal, then in order to achieve balance, the greater mass has to be
placed nearer to the pivot than the smaller one. The turning effect due to
the weight of the coins acting at a point depends not only on the mass of
the coins, but also on the distance of the point of action from the pivot.
A long symmetrical object, such as the ruler in this experiment, can often
be modelled as a rigid body with length, but no breadth or depth. Such a
rigid body is known as a model rod and is often drawn as a straight line.
The pivot on which the object rests is often modelled as a model pivot,
which has a single point of contact with the rod and is often drawn as a
triangle. Using these notions, the above experiments should allow you to
believe the following result.
Balanced rod
A horizontal rod, pivoted at its centre, will remain horizontal under
F1 F2 the action of two forces F1 and F2 acting vertically downwards at
distances l1 and l2 , respectively, on either side of the pivot (see
l1 l2 Figure 60), provided that
In other words, the horizontal rod will remain in equilibrium provided that
the distances of the forces from the pivot are in inverse ratio to the
magnitudes of the forces.
142
4 Torques
Exercise 38
Jack and Jill are sitting on opposite ends of a seesaw. Jill is 1.2 m from the
pivot, and Jack is 1 m from the pivot. Jack’s mass is 60 kg. If the seesaw is
at rest and horizontal, what is Jill’s mass?
*,
r × F = (|r| |F| sin θ) n B
where n * is a unit vector perpendicular to both r and F, and with direction
F
out of the page (as given by the right-hand grip rule). This cross product
satisfies all the above requirements for the turning effect of a force. It R
includes a measure |F| of the force and the perpendicular distance |r| sin θ r
of its line of action AB from the fixed point O. The direction of the cross θ
product, represented by n * , corresponds to the direction of the turning
effect. In this example, the turning effect of F about O is anticlockwise,
which corresponds to the anticlockwise motion of the fingers in the A
right-hand grip rule. We refer to the cross product r × F as the torque of O
the force F relative to the origin O, and we use it as our measure of the
turning effect of the force. Figure 61 Calculating the
turning effect of force F
relative to O
The torque Γ of a force F about a fixed point O is the cross product
Γ = r × F,
where r is the position vector, relative to O, of a point on the line of
action of the force.
In SI units, the units of torque are newton metres, written as N m
(or kg m2 s−2 ).
143
Unit 2 Vector algebra and statics
Note that a torque is a vector quantity, but it does not have the same
y units as a force, which is measured in newtons.
F2 F1
3
F3 2
Exercise 39
1 Find the torque of each of the forces in Figure 62 relative to the origin,
where each force is of magnitude 3 N. Forces F1 and F2 each have their
−3 −2 −1 1 1 2 3 x point of action on the x-axis, and F3 has its point of action on the z-axis.
2
3 j
z Exercise 40
k i
(a) Show that if O is any point on the line of action of a force F, and r is
Figure 62 the position vector, relative to O, of any other point on the line of
action, then r × F = 0. Deduce that the torque of a force about a
point on its line of action is zero.
(b) Suppose now that O is not on the line of action of F, and let r1 and r2
be the position vectors, relative to O, of two points on the line of
action. Show that r1 − r2 is parallel to F, and hence that
r1 × F = r2 × F. Deduce that the torque of a force about a fixed
point O is independent of the choice of the position of the point on the
line of action.
Let us now convince ourselves that the definition of torque makes sense in
terms of the examples of turning forces that we have seen so far. To do
this, we need an equilibrium condition for rigid bodies that extends the
equilibrium condition for particles. You will not be surprised that it
requires not only that all the forces on a rigid body sum to zero, but also
that all the torques sum to zero.
144
5 Applying the principles
Example 12
A ladder of mass M and length l stands on rough horizontal ground, and
rests against a smooth vertical wall (see Figure 63). The ladder can be
modelled as a model rod. Find the minimum angle θ between the ladder
and the ground for which the ladder can remain static, if the coefficient of
static friction µ between the ladder and the ground is 0.5.
N2
j
l
M i
N1 k
W
θ
O µ = 0.5 F
Figure 63 A ladder resting against a wall
Solution
The forces acting on the rod are shown in the force diagram in Figure 63, " Draw force diagram(s) !
where W is the weight of the ladder, N1 is the normal reaction from the
ground, N2 is the normal reaction from the wall, and F is the friction force
at the bottom of the ladder. (The wall is smooth, so there is no friction
force at the top of the ladder.)
As two of the three unknown forces act at the bottom of the ladder, this is
a convenient point for the origin O. The axes are chosen as shown in " Choose axes !
Figure 63.
The equilibrium condition for rigid bodies gives " Apply law(s) !
N1 + N2 + F + W = 0, (33)
ΓN1 + ΓN2 + ΓF + ΓW = 0, (34)
where ΓN1 , ΓN2 , ΓF and ΓW are the torques with respect to O of N1 , N2 ,
F and W, respectively.
If the ladder is not going to slip, we must have
|F| ≤ µ|N1 |. (35)
145
Unit 2 Vector algebra and statics
" Solve equation(s) ! The position vectors rN1 and rF are both zero, hence the corresponding
torques ΓN1 and ΓF (relative to O) are also zero. To calculate the non-zero
torques, we need the position vectors of the points of application of the
forces. These are given by
rN2 = l cos θ i + l sin θ j,
rW = 12 l cos θ i + 21 l sin θ j.
All of the forces in this example are aligned with the coordinate axes, so
the components can be written down by inspection:
N1 = |N1 | j, N2 = |N2 | (−i), F = |F| i, W = M g (−j).
Now we can calculate the two non-zero torques:
ΓN2 = (l cos θ i + l sin θ j) × (−|N2 | i)
= −|N2 | l sin θ j × i
= |N2 | l sin θ k,
ΓW = ( 21 l cos θ i + 21 l sin θ j) × (−M g j)
= − 12 M gl cos θ i × j
= − 12 M gl cos θ k.
Substituting these torques into equation (34) gives
|N2 | l sin θ k − 12 M gl cos θ k = 0,
thus
|N2 | = 21 M g cot θ.
Resolving equation (33) in the i- and j-directions in turn gives
−|N2 | + |F| = 0, |N1 | − M g = 0.
Therefore
|F| = |N2 | = 12 M g cot θ, |N1 | = M g.
Substituting these into inequality (35) gives
1
Note that since M and g are 2 M g cot θ ≤ µM g,
positive, we can safely divide
through by them without which, on rearrangement and using µ = 0.5, gives
reversing the inequality.
cot θ ≤ 1.
" Interpret results ! Therefore the model predicts that the minimum angle that the ladder can
make with the ground before slipping is π/4 radians (45◦).
146
5 Applying the principles
Example 13
During the erection of a marquee, a heavy pole OA of mass m and length l A
must be held in place by a rope AB, as shown in Figure 64. The angle
between the pole and the ground is π/4, and the angle between the rope rope
and the ground is π/6. Model the pole as a model rod, and the rope as a pole
π π
model string. Assume that the pole is freely hinged at O, that is, the end 6 4
of the model rod is fixed at O and is free to pivot about O. B O
If the pole is in equilibrium, find the magnitude of the tension in the rope. Figure 64 A rope holding
Is the magnitude of the tension in the rope larger or smaller than it would up a pole
be if the pole were hanging freely on the end of the rope?
Solution
The best choice of axes is not obvious for this problem. So in this case we " Draw force diagram(s) !
proceed by drawing the force diagram first, as shown in Figure 65.
T
π j
3
π R π k i
6 4 W
O
Figure 65 The force diagram for the rope and pole
From Figure 65 we see that there are three forces acting on the rod: its " Choose axes !
weight W, the tension force in the string T, and the reaction force R at
the hinge. These forces act in three different directions, so we choose to
orient the axes as shown in the figure. In this problem, the choice of
origin O is vital, since we are going to take torques about the origin.
Choosing the origin at the base of the pole eliminates the reaction force
from the torque equation, which simplifies the calculations.
The equilibrium condition for rigid bodies gives " Apply law(s) !
R + W + T = 0, (36)
ΓR + ΓW + ΓT = 0, (37)
where ΓR , ΓW and ΓT are the torques with respect to O of R, W and T,
respectively.
In this case, equation (36) is not very useful since it contains two forces of " Solve equation(s) !
unknown magnitude, namely R and T. However, by taking torques
about O, one of the torques appearing in equation (37) is zero, namely ΓR .
Now we proceed by calculating the other two torques.
The position vectors of the points of application of the forces W and T (at
the centre and end of the pole, respectively) are
l
rW = 21 l cos π4 i + 12 l sin π4 j = √ (i + j),
2 2
l
rT = l cos π4 i + l sin π4 j = √ (i + j).
2
147
Unit 2 Vector algebra and statics
l |T| √
= √ ( 3 − 1)k.
2 2
Substituting these torques into equation (37) gives
lmg l |T| √
− √ k + √ ( 3 − 1)k = 0.
2 2 2 2
Rearranging gives
mg
|T| = √ .
3−1
√
" Interpret results ! So the magnitude of the tension in the rope is mg/( 3 − 1) ) 1.4mg,
which is greater than the magnitude of the weight of the pole, mg. So,
rather counter-intuitively, a stronger rope is needed to erect a pole in this
way than is needed to lift the pole.
Exercise 41
B A model rod OA of length l and mass m is fixed to a wall by a free hinge,
as shown in Figure 66. The rod is free to turn in a vertical plane about the
hinge, which is assumed to be smooth (i.e. there is no friction force
associated with it). The rod is supported in a horizontal position by a
O m θ string AB inclined at an angle θ to the horizontal.
A
l Find the reaction force at the hinge and the tension force acting on the rod
due to the string. Comment on the magnitudes and directions of these two
Figure 66 A rod fixed to a forces.
wall by a free hinge and
supported by a string
148
Learning outcomes
Exercise 42
A light ladder of length 3.9 m stands on a horizontal floor and rests against
a smooth vertical wall, as shown in Figure 67.
0.3 m
100 kg
3.9 m
θ m
0.3 m
µ = 0.25
1.5 m
Figure 67 A ladder with people standing on the top and bottom rungs
The base of the ladder is 1.5 m from the base of the wall. The coefficient of
static friction between the ladder and the floor is 0.25. The end rungs are
each 0.3 m from an end of the ladder. The ladder may be modelled as a
rod, and its mass may be neglected (as it is a light ladder, so its mass is
negligible compared with the masses of any people standing on it).
What is the minimum mass of a person standing on the bottom rung that
prevents the ladder from slipping when a person of mass 100 kg stands on
the top rung?
Exercise 43
At a building site, a plank OB of length 2l and mass m is resting against a
B
large smooth pipe of radius r, as shown in Figure 68. The pipe is fixed to
the ground, or else it would slide or roll to the left. The angle between the
A 2l
plank and the horizontal is π/3, so by symmetry the angle between the r
horizontal and the line between O and the centre of the pipe is π/6, as m
shown. In the figure, the distance OA is greater than the distance AB, so π
π6
the centre of mass of the plank is between O and A. 6
O
What is the least coefficient of static friction between the plank and the
ground that will ensure equilibrium? Figure 68 A plank resting
against a large smooth pipe
149
Unit 2 Vector algebra and statics
Learning outcomes
After studying this unit, you should be able to:
• understand the meaning of the terms scalar, vector, displacement
vector, unit vector and position vector, and know what it means to say
that two vectors are equal
• use vector notation and represent vectors as arrows on diagrams
• scale a vector by a number, and add two vectors geometrically using
the triangle rule (or the parallelogram rule)
• resolve a vector into its Cartesian components, and scale and add
vectors given in Cartesian component form
• calculate the dot product and cross product of two vectors
• determine whether or not two given vectors are perpendicular or
parallel to one another
• determine the magnitude of a vector and the angle between the
directions of two vectors
• write down the vector equation of a given straight line
• resolve a vector in a given direction
• manipulate vector expressions and equations involving the scaling,
addition, dot product and cross product of vectors
• use the cross product to determine the area of a parallelogram or
triangle
• appreciate the concept of a force
• understand and model forces of weight, normal reaction, tension and
friction
• recognise and model the forces that act on an object in equilibrium
• model objects as particles or as rigid bodies
• use model strings, model rods, model pulleys and model pivots in
modelling systems involving forces
• draw force diagrams, and choose appropriate axes and an origin
• use the equilibrium conditions for particles and for rigid bodies
• understand and use torques
• model and solve a variety of problems involving systems in equilibrium
and systems on the verge of leaving equilibrium.
150
Solutions to exercises
Solutions to exercises
Solution to Exercise 1
y
b
3 a
1
π/3
0 1 2 3 x
Solution to Exercise 2
N
70 mph
60◦
0 10 20 30 40 50 60 70
Scale (mph)
Solution to Exercise 3
(a) −v represents a wind of 35 mph from the south-west.
(b) The vector −1.5v has magnitude 1.5|v| and direction opposite to v.
The vector −kv (k positive) has magnitude k|v| and direction
opposite to v.
1 1
(c) v is a scaling of v by the positive scalar m = . The direction of
|v| |v|
1
v is thus the same as that of v, and its magnitude is
|v|
1
m|v| = |v| = 1.
|v|
Solution to Exercise 4
(a) −35 j (where |j| represents 1 km per hour).
(b) −112 i (where |i| represents 1 mile).
151
Unit 2 Vector algebra and statics
Solution to Exercise 5
(a) The figure below shows a + b and a + c.
y
a+c
c c
a
0 x
b
b a+b
(b) The figure below shows −b and the sum of a and −b.
y
a + (−b)
−b −b
0 a x
Solution to Exercise 6
y
b a+b b
a
0 x
−b a−b −b
152
Solutions to exercises
Solution to Exercise 7
4(a − c) + 3(c − b) + 2(2a − b − 3c)
= 4a − 4c + 3c − 3b + 4a − 2b − 6c
= 8a − 5b − 7c.
Solution to Exercise 8
Systems (b), (c) and (d) are right-handed.
Solution to Exercise 9
(a) d = 2(i + j + k) − 3(2i − 3j − k) = −4i + 11j + 5k,
e = (i + j + k) − 2(2i − 3j − k) + 4(3i + k) = 9i + 7j + 7k.
" √ √
(b) |d| = (−4)2 + 112 + 52 = 162 = 9 2,
√ √
|e| = 92 + 72 + 72 = 179.
√ √
(c) |a| = 12 + 12 + 12 = 3.
A unit vector in the direction of a is
1 1
a = √ (i + j + k).
|a| 3
(d) If a + x = b, then
x = b − a = (2i − 3j − k) − (i + j + k) = i − 4j − 2k.
Thus the components of x are 1, −4 and −2.
Solution to Exercise 10
0 = (0 0 0)T , i = (1 0 0)T , j = (0 1 0)T , k = (0 0 1)T .
Solution to Exercise 11
(a) Relative to the origin of the Cartesian coordinate system, the two
points have position vectors i + j + 2k and 2i + 3j + k. Thus the
vector equation of the line is
r = (1 − s)(i + j + 2k) + s(2i + 3j + k)
= (1 + s)i + (1 + 2s)j + (2 − s)k,
where −∞ < s < ∞.
(b) The line cuts the (x, z)-plane when y = 0, that is, 1 + 2s = 0, giving
s = − 12 . Hence r = 21 i + 52 k.
Solution to Exercise 12
" √
(a) |a| = 22 + (−1)2 =
5,
√ √ y
|b| = 12 + 32 + 52 = 35. φ
The vector a lies in the (x, y)-plane, and the angle φ that it makes √ 0 x
1 2
√ the figure in the margin) is given by cos φ = 2/ 5
with the x-axis (see
and sin φ = −1/ 5. Hence φ ) −0.4636 (radians). −1 a
153
Unit 2 Vector algebra and statics
(b) a + b = 3i + 2j + 5k,
2a − b = 3i − 5j − 5k,
c + 2b − 3a = −3i + 10j + 8k.
−−→
(c) The vector P Q is equal to 2a − b. The point Q is the end of the
−−→
vector OQ, which is given by
−−→ −−→ −−→
OQ = OP + P Q = (2j + 3k) + (3i − 5j − 5k) = 3i − 3j − 2k,
so Q is the point (3, −3, −2).
Solution to Exercise 13
(a) Writing
x+1 y−8
= =1−z =s
3 −2
gives
x = 3s − 1, y = 8 − 2s, z = 1 − s.
This gives the vector equation of the line as
r = (3s − 1)i + (8 − 2s)j + (1 − s)k
= (−i + 8j + k) + s(3i − 2j − k).
Hence A has position vector a = −i + 8j + k, B has position vector
−−→ √ √
b = 2i + 6j, and AB = 3i − 2j − k, with length 32 + 22 + 12 = 14.
(b) (i) The midpoint between A and B has s = 12 , so M has position
vector
m = 12 i + 7j + 12 k.
(ii) The point N has s = 0.6, so N has position vector
n = 0.8i + 6.8j + 0.4k.
(c) Let c = −i − 2j and d = 2i + j − k. Then d − c = 3i + 3j − k, so the
vector equation of the line (using a different parameter t) is
r = c + t(d − c) = (−i − 2j) + t(3i + 3j − k)
= (3t − 1)i + (3t − 2)j − tk.
(d) The lines intersect if there are values of s and t such that
(3t − 1)i + (3t − 2)j − tk = (3s − 1)i + (8 − 2s)j + (1 − s)k.
This gives rise to the three equations
3t − 1 = 3s − 1, 3t − 2 = 8 − 2s, −t = 1 − s.
Since there are no values of s and t that satisfy all three equations, the
lines do not intersect.
Solution to Exercise 14
a · b = |a| |b| cos θ = 2 × 4 × cos π3 = 4,
√
b · c = |b| |c| cos θ = 4 × 1 × cos π6 = 2 3,
154
Solutions to exercises
%π π
,
a · c = |a| |c| cos θ = 2 × 1 × cos 3 + 6 = 2 cos π2 = 0,
b · b = |b| |b| cos θ = 4 × 4 × cos 0 = 16.
Solution to Exercise 15
(a) (a + b) · (a − b) = a · a − a · b + b · a − b · b
= a · a − b · b.
(b) |a + b|2 = (a + b) · (a + b)
=a·a+a·b+b·a+b·b
= a · a + 2a · b + b · b.
Solution to Exercise 16
(a) If 2a + 3b and ma + b are perpendicular, then
(2a + 3b) · (ma + b) = 0.
Expanding this expression,
2ma · a + 2a · b + 3mb · a + 3b · b = 0.
Now a and b are perpendicular, so a · b = b · a = 0, and they are unit
vectors, so a · a = b · b = 1. Thus
2m + 3 = 0,
so m = −1.5.
(b) |c|2 = c · c
= (3a + 5b) · (3a + 5b)
= 9a · a + 15a · b + 15b · a + 25b · b.
Thus, since a and b are perpendicular unit vectors,
|c|2 = 9 + 25 = 34,
√
so |c| = 34 ) 5.831.
Solution to Exercise 17
a · b = (4 × 1) + (1 × −3) + (−5 × 1) = −4.
The negative sign tells us that the angle between a and b is between π/2
and π radians, that is, it is an obtuse angle.
Solution to Exercise 18
" √
|a| = 22 + (−3)2 + 12 = 14,
" √
|b| = (−1)2 + 22 + 42 = 21.
Also,
a · b = (2 × −1) + (−3 × 2) + (1 × 4) = −4,
so if θ is the angle between a and b, then
a·b −4 4
cos θ = =√ √ =− √ .
|a| |b| 14 × 21 7 6
The negative sign means that θ is obtuse; we have θ ) 1.806 (radians).
155
Unit 2 Vector algebra and statics
Solution to Exercise 19
a · i = (a1 i + a2 j + a3 k) · i
= a1 i · i + a2 j · i + a3 k · i
= a1 .
Similarly,
a · j = a2 and a · k = a3 .
(Notice that this means that the components of a vector are given by the
dot products of the vector with the Cartesian unit vectors i, j, k.)
Solution to Exercise 20
(a) a · c = −2, a · d = −3, a · e = 0.
Thus only e is perpendicular to a.
(b) First,
a + 2b = j + 9k.
Now a suitable vector along the line joining the origin to the point
(1, 1, 1) is i + j + k. The corresponding unit vector is
* = √13 (i + j + k). The component of a + 2b in the direction of this
u
line is
* = (j + 9k) ·
(a + 2b) · u √1 (i + j + k)
3
10
= √
3
.
Solution to Exercise 21
π
The angle between −i and p is α, and the angle between j and p is 2 − α.
Therefore the i-component of p is
−|p| cos(α) = −2.5 cos α,
and the j-component of p is
% ,
|p| cos π2 − α = 2.5 sin α.
Hence
p = −2.5 cos α i + 2.5 sin α j.
The angle between −i and q is β, and the angle between −j and q is
π
2 − β. Therefore the i-component of q is
−|q| cos(β) = −3 cos β,
and the j-component of q is
% ,
−|q| cos π2 − β = −3 sin β.
Hence
q = −3 cos β i − 3 sin β j.
156
Solutions to exercises
Finally, the angle between i and r is γ, and the angle between −j and r is
π
2 − γ. Therefore the i-component of r is
|r| cos γ = 2.5 cos γ,
and the j-component of r is
% ,
−|r| cos π2 − γ = −2.5 sin γ.
Hence
r = 2.5 cos γ i − 2.5 sin γ j.
Solution to Exercise 22
For the sake of clarity, here is a diagram showing u, v and w (where all
three vectors start at O) drawn in the (x, y)-plane. (The z-axis points out
of the page.)
y
v
π u
6
π
6
O x
Solution to Exercise 23
(a) i, j and k are unit vectors forming a right-handed system (see the
figure in the margin). j
Thus, using the definition of the cross product,
% ,
i × j = |i| |j| sin π2 k = k. i
k
Similarly,
j×k=i and k × i = j.
(b) Since (a × b) = −(b × a) for any vectors a and b, we have
j × i = −k, k × j = −i and i × k = −j.
157
Unit 2 Vector algebra and statics
Solution to Exercise 24
First, we list the components of the vectors:
a1 = 2, a2 = −3, a3 = 1,
b1 = −1, b2 = 2, b3 = 4,
c1 = −4, c2 = 6, c3 = −2.
To compute a × b, we need
a2 b3 − a3 b2 = −3 × 4 − 1 × 2 = −14,
a3 b1 − a1 b3 = 1 × (−1) − 2 × 4 = −9,
a1 b2 − a2 b1 = 2 × 2 − (−3) × (−1) = 1.
Hence a × b = −14i − 9j + k.
Similarly, for a × c we need
a2 c3 − a3 c2 = −3 × (−2) − 1 × 6 = 0,
a3 c1 − a1 c3 = 1 × (−4) − 2 × (−2) = 0,
a1 c2 − a2 c1 = 2 × 6 − (−3) × (−4) = 0.
Hence a × c = 0.
Finally, for b × c we need
b2 c3 − b3 c2 = 2 × (−2) − 4 × 6 = −28,
b3 c1 − b1 c3 = 4 × (−4) − (−1) × (−2) = −18,
b1 c2 − b2 c1 = (−1) × 6 − 2 × (−4) = 2.
Hence b × c = −28i − 18j + 2k.
Since a × c = 0, and neither vector is zero, the vectors a and c are
parallel. In fact, c = −2a.
158
Solutions to exercises
Solution to Exercise 25
A vector perpendicular to a and b is a × b. The components of a and b
are
a1 = 2, a2 = 2, a3 = 1,
b1 = 4, b2 = 4, b3 = −7.
So we have
a2 b3 − a3 b2 = 2 × (−7) − 1 × 4 = −18,
a3 b1 − a1 b3 = 1 × 4 − 2 × (−7) = 18,
a1 b2 − a2 b1 = 2 × 4 − 2 × 4 = 0.
Hence a × b = −18i + 18j.
We are asked for a unit vector, so the obvious choice is
1 −18i + 18j
(a × b) = √
|a × b| 18 2
= √12 (−i + j).
Solution to Exercise 26
(a) t is perpendicular to both r and s, and its sense is vertically down,
that is, into the ground.
(b) Conversely, the sense of s × r is vertically up.
(c) t × r is perpendicular to t (thus in the horizontal plane) and
perpendicular to r, and by the right-hand grip rule its sense is due
east.
(d) |t| = |r| |s| sin π4 = √1 .
2
(e) t × (r × s) = t × t = 0.
(f) r · s = |r| |s| cos π4 = √1 .
2
(g) s · (t × r) = |s| |t × r| cos π4 (by part (c))
% ,
= |s| |t| |r| sin π2 cos π4
=1× √1 ×1×1× √1 = 21 .
2 2
Solution to Exercise 27
−−→ −→
Two sides of the triangle are AB = −i − j + 5k and AC = 2i − 3j + 2k. So
the area of the triangle is
+ −→ −→+ 1 +
1 +− + +
2 AB × AC = 2 ((−1) × 2 − 5 × (−3))i + (5 × 2 − (−1) × 2)j
+
+ ((−1) × (−3) − (−1) × 2)k+
= 21 |13i + 12j + 5k|
√
= 21 169 + 144 + 25
√
= 13
2 2.
159
Unit 2 Vector algebra and statics
Solution to Exercise 28
The weight is 3g k.
Solution to Exercise 29
i Since the weight acts vertically downwards, its direction is in the vertical
π/6 plane defined by i and j, so W has no k-component. To find the i- and
j-components, the first step is to draw a diagram and work out the angles
j π/3 involved. In this case we use the two right angles marked in the diagram
W in the margin to work out the required angles.
π/6 * = − sin π6 i + cos π6 j.
A unit vector in the direction of W is u
Resolving the weight gives
% , % ,
W = |W| u * = −|W| sin π6 i + |W| cos π6 j
√
= −15g × − 21 i + 15g × 2
3
j
√
= − 15
2 gi+
15 3
2 g j.
Solution to Exercise 30
In this and other solutions, you may find that your diagrams and chosen
axes are different from those given. You should still be able to check the
validity of your solution against the given one, as the basic concepts are
unchanged by these differences. Any choice of axes should lead to the
" Draw picture ! same final answers as those given.
π/4
π/6
j
10 kg
i
" Choose axes ! The forces all lie in a vertical plane, so we need only two axes. The unit
vectors i and j are shown in the diagram.
" Draw force diagram(s) ! The force diagram, where the tension forces are denoted by T1 and T2 ,
and the weight of the tree is denoted by W, is as follows.
T2
T1 j
π/6 π/4
i
W
160
Solutions to exercises
Solution to Exercise 31
Draw the picture, choose axes and draw the force diagram. " Draw picture !
" Choose axes !
horizontal j N " Draw force diagram(s) !
force
5 kg F P
µ = 0.74 i
W
If the object does not move, the equilibrium condition for particles holds, " Apply law(s) !
so
F + N + P + W = 0.
If the block does not move,
|F| ≤ µ|N|,
where µ is the coefficient of static friction.
161
Unit 2 Vector algebra and statics
" Solve equation(s) ! Resolving the equilibrium equation in the i-direction gives
|F| = |P|.
Resolving in the j-direction gives
|N| = |W|.
Therefore if the block does not move,
|P| = |F| ≤ µ|N| = µ|W|.
Since |P| = 2 and |W| = 0.5g, the block does not move provided that
2
µ≥ ) 0.41.
0.5g
" Interpret solution ! The coefficient of static friction for steel on dry steel is µ = 0.74, so the
model predicts that the block will not move.
Since we have |F| = |P|, the magnitude of the friction force is 2 N.
Solution to Exercise 32
Since the friction force does not depend on the area of contact, the
sideways force required to start the box slipping is the same in both cases.
Solution to Exercise 33
" Choose axes ! Choose the same axes as in Example 9.
" Draw force diagram(s) ! Modelling the crate as a particle and the rope as a model string, the force
diagram is as shown below, where the notation is as in Example 9. Both F
and T act in the same direction (on the implied assumption that the crate
will slide down the ramp if nothing holds it back).
T
F N j
π π i
6 3 π
6
162
Solutions to exercises
Solution to Exercise 34
(a) The situation is illustrated below. " Draw picture !
" Draw force diagram(s) !
j F N
α i
α W
Choose axes so that i points down the slope and j is in the direction of " Choose axes !
the normal reaction.
The equilibrium condition for particles gives " Apply law(s) !
F + N + W = 0.
Since the box is on the point of slipping,
|F| = µ|N|,
where µ is the coefficient of static friction.
163
Unit 2 Vector algebra and statics
" Solve equation(s) ! Two of the forces are aligned with the axes and can be written down
immediately:
F = |F| (−i) and N = |N| j.
The third force is inclined to the axes, so resolving gives
% , % ,
W = |W| cos π2 − α i − |W| sin π2 − α j
= mg sin α i − mg cos α j.
Resolving the equilibrium equation in the i-direction gives
−|F| + 0 + mg sin α = 0,
so
|F| = mg sin α.
Resolving in the j-direction gives
0 + |N| − mg cos α = 0,
so
|N| = mg cos α.
Substituting |F| and |N| into the friction condition gives
|F|
µ= = tan α.
|N|
" Interpret solution ! (b) The result in part (a) tells us that for an object on an inclined plane,
the angle at which the object starts to slip depends only on the two
surfaces in contact (i.e. on the coefficient of static friction µ). The
mass of the object is irrelevant – so the half-full mug will start to slip
at the same angle as the empty one.
Solution to Exercise 35
(a) The sack, stone, pulley and rope are modelled as in Example 10.
" Choose axes ! This time we need two axes, as shown in Figure 55.
" Draw force diagram(s) ! The force diagrams for the sack and the stone, using the usual
notation, are as follows.
N π T2 j
4 T1
F
50 kg 15 kg
i
W1 W2
sack stone
164
Solutions to exercises
so the model predicts that the magnitude of the friction force is about
104 N.
(d) For equilibrium, we must have |F| ≤ µ|N|, that is,
µ ≥ |F|/|N| ) 104/386 ) 0.27.
So the model predicts that the smallest value of µ that allows the
system to remain in equilibrium is about 0.27.
(e) If |N| approaches zero, then |W1 | − √12 |T1 | = |W1 | − √12 |W2 |
√
approaches zero, so |W2 | tends to | 2 W1 |. Thus the stone would
have to weigh approximately 70.7 kg.
165
Unit 2 Vector algebra and statics
Solution to Exercise 36
Model the man in the well as a particle of mass 80 kg, and all the men on
the ground as a single particle of mass m = 80n, where n is the number of
men.
" Choose axes ! A suitable choice of axes is shown in Figure 57. A force diagram for each
" Draw force diagram(s) ! particle, using the usual notation, is shown below.
T1 N j
T2 F
80 kg m i
W1 W2
man in well men on ground
166
Solutions to exercises
Solution to Exercise 37
Equilibrium can be disturbed either by the object of mass m1 sliding down
the board and pulling up the object of mass m2 , or by the object of
mass m2 dropping down and pulling the object of mass m1 up the board.
Model the masses as particles joined by a model string hanging over a " Draw picture !
model pulley representing the edge of the board.
m1
m2
α
T2 j N j
T1
m2 m1
i
α F i
W2
W1
167
Unit 2 Vector algebra and statics
T2 j N j
T1
m2 F m1
i
i
W2 α
W1
" Apply law(s) ! The equilibrium condition and the use of a model pulley lead to the same
equilibrium equations as before.
" Solve equation(s) ! The only difference is that the friction force is in the opposite direction,
that is, F = −|F| i. This means that when the first equilibrium equation is
resolved in the i-direction, it now gives
−|T1 | + 0 − |F| + |W1 | sin α = 0,
leading to
|F| = |W1 | sin α − |T1 | = m1 g sin α − m2 g.
" Interpret solution ! For the friction condition to be satisfied, the model predicts that
|F| m1 sin α − m2
µ≥ = .
|N| m1 cos α
Both conditions for µ must hold for the system to be in equilibrium. Since
µ is non-negative and the right-hand sides are of opposite signs, we can say
the first condition applies when m2 > m1 sin α,
the second condition applies when m2 < m1 sin α.
If there is no friction, then µ = 0, and the conditions can both be satisfied
only if the right-hand sides are also zero, that is, m2 = m1 sin α.
168
Solutions to exercises
Solution to Exercise 38
We can model the seesaw as a model rod. Let Jill’s mass be m. Then,
assuming that neither Jack’s nor Jill’s feet are on the ground, we can use
equation (31) to obtain
mg × 1.2 = 60g × 1.
Therefore m = 50, so Jill’s mass is 50 kg.
Solution to Exercise 39
The simplest choice of position vector for a point on the line of action of
F1 is r1 = 2i, the position vector of the point (2, 0, 0) relative to the
origin. Similarly, suitably simple choices for F2 and F3 are r2 = −2i and
r3 = 2k. Hence the torques of the three forces relative to the origin are
r1 × F1 = 2i × 3j = 6k,
r2 × F2 = −2i × 3j = −6k,
r3 × F3 = 2k × 3j = −6i.
Solution to Exercise 40
(a) If O is on the line of action of F, and r is the position vector relative
to O of another point on that line of action, then r is parallel to F, so
r × F = 0.
Thus when O is on the line of action of a force F, the formula
Γ = r × F gives 0 whatever the choice of r.
(b) If r1 and r2 are position vectors, relative to O, of two points R1 and
R2 on the line of action of F (where O is not on this line of action),
then the triangle rule for the addition of vectors shows that r1 − r2 is R1
parallel to F (as shown in the diagram in the margin). F
−r2 r1 R2
Therefore (r1 − r2 ) × F = 0, and since r2
r1 − r 2
(r1 − r2 ) × F = r1 × F − r2 × F,
O
we have
r1 × F = r2 × F.
Thus the formula Γ = r × F gives the same vector, whatever the
choice of r.
Solution to Exercise 41
There are three forces acting on the rod: the weight W (at the centre of " Draw force diagram(s) !
the rod), the tension force T due to the string, and the reaction force R at
the hinge. The direction of R is not known, except that it is in the plane
of W and T. The force diagram is shown below.
R T j
θ
O k i
W
169
Unit 2 Vector algebra and statics
" Choose axes ! Choose the origin to be at O and axes as shown above.
" Apply law(s) ! The equilibrium condition for rigid bodies gives
R + T + W = 0,
ΓR + ΓT + ΓW = 0.
" Solve equation(s) ! Position vectors of the points of application are
rR = 0, rT = l i, rW = 21 l i.
From the force diagram,
W = mg(−j).
The force R is unknown except that it is in the plane of W and T, that is,
the plane defined by i and j. For convenience we write R in component
form as
R = Ri i + Rj j,
where Ri and Rj are the i- and j-components of R.
The remaining force can be resolved as
T = |T| cos θ (−i) + |T| sin θ j = −|T| cos θ i + |T| sin θ j.
The torques are given by
ΓR = 0,
ΓT = rT × T = l i × |T|(− cos θ i + sin θ j) = l |T| sin θ k,
ΓW = rW × W = 12 l i × mg(−j) = − 21 lmg k.
Resolving the second equilibrium equation in the k-direction gives
l |T| sin θ − 12 lmg = 0,
so
|T| = 12 mg cosec θ.
Resolving the first equilibrium equation in the i- and j-directions in turn
gives
Ri − |T| cos θ = 0,
Rj + |T| sin θ − mg = 0,
so
Ri = |T| cos θ = 21 mg cot θ,
Rj = mg − |T| sin θ = mg − 21 mg = 12 mg.
So the model predicts that the reaction force at the hinge is
R = 21 mg cot θ i + 21 mg j.
From the first equilibrium equation, the tension force due to the string is
T = −R − W = − 21 mg cot θ i + 12 mg j.
170
Solutions to exercises
The j-components of R and T are the same. The i-components are of " Interpret solution !
equal magnitude, but with opposite signs. Thus R and T have equal
magnitude and make the same angle θ with the horizontal, where θ is
measured anticlockwise for R and clockwise for T.
Solution to Exercise 42
Since the mass of the ladder may be neglected, the forces acting on the " Draw force diagram(s) !
ladder are the weights W1 and W2 of the two people, the normal reaction
forces N1 and N2 at the wall and at the floor, and the friction force F at
the floor. (There is no friction force at the wall, as it is smooth.) The force
diagram is shown below.
N1
W1 j
k i
N2
θ
O
F
W2
The origin O is best chosen to be at the foot of the ladder. The axes are " Choose axes !
shown above.
The equilibrium condition for rigid bodies gives " Apply law(s) !
N1 + N2 + W1 + W2 + F = 0,
ΓN1 + ΓN2 + ΓW1 + ΓW2 + ΓF = 0.
At the point of slipping,
|F| = µ|N2 |,
where µ = 0.25 is the coefficient of static friction.
Let l denote the length of the ladder (i.e. 3.9 m), and let d denote the " Solve equation(s) !
distance of the end rungs from the ends of the ladder (i.e. 0.3 m). Relative
to the origin O at the bottom of the ladder, two position vectors are zero,
namely rN2 and rF . The remaining position vectors are
rN1 = −l cos θ i + l sin θ j,
rW1 = −(l − d) cos θ i + (l − d) sin θ j,
rW2 = −d cos θ i + d sin θ j.
The forces are aligned with the axes, and can be put into component form
by inspection:
N1 = |N1 | i, N2 = |N2 | j,
W1 = M g (−j), W2 = mg (−j), F = |F| (−i),
where m is the mass of the person on the lower rung, and M is the mass of
the person on the upper rung (i.e. 100 kg).
171
Unit 2 Vector algebra and statics
Two of the torques are zero by choice of origin. The remaining three
torques can be calculated as
ΓN1 = (−l cos θ i + l sin θ j) × (|N1 | i) = −l |N1 | sin θ k,
ΓW1 = (−(l − d) cos θ i + (l − d) sin θ j) × (−M g j) = M (l − d)g cos θ k,
ΓW2 = (−d cos θ i + d sin θ j) × (−mg j) = mdg cos θ k.
Substituting these into the second equilibrium equation and resolving in
the k-direction gives
−l |N1 | sin θ + M (l − d)g cos θ + mdg cos θ = 0.
Now we need to find the magnitude of the normal reaction |N1 |, which we
do by resolving the first equilibrium equation in the i- and j-directions in
turn, to obtain
|N1 | − |F| = 0,
|N2 | − M g − mg = 0.
From these equations and the friction condition, we have
|N1 | = |F| = µ|N2 | = µ(M g + mg).
Substituting for |N1 | in the k-direction equation gives
−lµ(M g + mg) sin θ + M (l − d)g cos θ + mdg cos θ = 0,
thus
−M lµ tan θ − mlµ tan θ + M (l − d) + md = 0,
so
lµ tan θ − l + d
m=M .
d − lµ tan θ
" Interpret solution ! To use this
√ formula, we need the value of tan θ, which by trigonometry is
equal to 3.92 − 1.52 /1.5 = 2.4. Now we substitute in the known values,
namely µ = 0.25, d = 0.3, l = 3.9 and M = 100, to find
3.9 × 0.25 × 2.4 − 3.9 + 0.3
m = 100 × ) 61.8.
0.3 − 3.9 × 0.25 × 2.4
So to prevent the ladder from slipping, the model predicts that the person
on the bottom rung must have a mass of at least 61.8 kg.
Solution to Exercise 43
" Choose axes ! Take the origin at O at the base of the plank, and axes as shown below.
" Draw force diagram(s) ! The force diagram, using the usual notation, is also shown below.
B
B
N2
π π
6 A 6
A j r
x
W N1 k i
r π π
π 6 6
3
F O O
172
Solutions to exercises
174
Unit 3
Dynamics
Introduction
Introduction
Think of an object that is moving in some way. It might be a car
accelerating on a motorway, a tennis ball flying through the air, a comet
hurtling through space, or a pendulum swinging to and fro. Why does the
object move as it does? How will it move in the future? To what extent
can you influence its motion? Questions like these are very important from
a practical point of view. The control that the human race exerts over the
environment depends, to a large extent, on our ability to find the right
answers. For example, in Figure 1, can we predict whether the comet will
collide with the Earth in time for something to be done about it?
177
Unit 3 Dynamics
178
1 Describing motion
1 Describing motion
This section is devoted to describing the motion of objects modelled as The subject matter of this
particles. In Subsection 1.1 the motion is described by giving a position section, the description of
vector at each instant in time. The ideas in Subsection 1.1 apply whether motion, is often referred to by
the technical term kinematics.
the particle moves along a straight line or along a curve of some sort; in
Subsection 1.2, however, and for Sections 2 to 4, only motion along a
straight line is considered.
179
Unit 3 Dynamics
Example 1
A juggler throws a ball from one hand to the other, in a vertical plane, as
shown in Figure 4. The ball is modelled as a particle, and its motion, with
y respect to the horizontal and vertical axes shown in Figure 4, is described
by the two-dimensional vector function
O x
r(t) = 1.5ti + t(4 − 5t)j (0 ≤ t ≤ 1),
j where distances are measured in metres, and time t is measured in seconds
after the ball was thrown. The origin O is the juggler’s right hand just as
he throws the ball, which occurs at time t = 0.
i
Let x(t) be the component of r(t) in the i-direction, that is,
Figure 4 A juggler with a
ball x(t) = r(t) · i = 1.5t. Similarly, let y(t) be the component of r(t) in the
j-direction, that is, y(t) = r(t) · j = t(4 − 5t). These give the horizontal
distance travelled and the height of the ball, respectively, at time t.
(a) Calculate x(t) and y(t) at times t = 0, 0.2, 0.4, 0.6, 0.8, 1, and sketch
the graphs of y(t) against t and against x(t). Comment on what the
graphs represent.
(b) Using your graphs, or otherwise, answer the following questions.
(i) How high does the ball go, and what is its position at its highest
point?
(ii) Does the juggler catch the ball?
Solution
(a) The values are tabulated in Table 1, and the graphs are shown in
Figure 5.
Table 1
t 0 0.2 0.4 0.6 0.8 1
x(t) 0 0.3 0.6 0.9 1.2 1.5
y(t) 0 0.6 0.8 0.6 0 −1
y(t) y(t)
1 1
0.5 0.5
0 0
0.2 0.4 0.6 0.8 1 t 0.3 0.6 0.9 1.2 1.5 x(t)
−0.5 −0.5
−1 −1
(a) (b)
Figure 5 Graphs of (a) height against time, (b) height against horizontal
distance
180
1 Describing motion
Exercise 1
An ice hockey player aims to hit the puck towards the goal, as shown in
Figure 6, but misses. The puck hits the back wall and bounces straight
back, and is then hit by a second player who (incredibly) hits it along the
same path as the first player. Let the origin O be the point of impact of
the puck by the first player, which occurs at time t = 0. The motion of the
puck, with respect to the axis shown in Figure 6, is described by the x
i
one-dimensional vector function r(t) = x(t) i, where the function x(t) is O
defined by
5t, 0 ≤ t < 0.4,
Figure 6 The path of the
x(t) = 4 − 5t, 0.4 ≤ t < 1, puck
6t − 7, 1 ≤ t ≤ 1.5,
where x is measured in metres, and t is measured in seconds.
(a) Sketch the graph of the function x(t).
181
Unit 3 Dynamics
You saw in Unit 2 that some In Example 1 and Exercise 1, objects were modelled as particles. The
large static objects can be question arises as to when this is appropriate. It seems obvious that small
modelled as particles. objects can be modelled as particles, but what about large objects such as
the Earth? The answer is that it depends on the context. For example, to
calculate the Earth’s orbit around the Sun, it is permissible to model the
The Earth’s diameter (about Earth as a particle. One of Newton’s great achievements was to realise
13 000 km) is small compared that it is appropriate to model the Earth as a particle in this context.
with the Earth–Sun distance
(about 1.5 × 108 km). Whether a particle model will be satisfactory is not just a question of size.
For example, if a ball is placed on a rough sloping table, then it will roll
down the slope. A particle model could be used to describe the path of the
ball’s centre, but it would not be adequate to keep track of the rolling
motion that takes place about the centre. Note that this inadequacy of the
particle model occurs regardless of the ball’s size, since the same
consideration would apply to a football, a tennis ball, a marble or a
ball-bearing. However, if the ball is not rotating, or if it is assumed that
the rotations can be ignored, then the particle model is appropriate.
You might think from this last example that the particle model is of very
limited use for moving objects, but in fact the example hints at how this
model can be extended. You will see in Unit 19 that the motion of an
object can be described well by specifying:
You met the idea of centre of • the motion of the centre of mass of an object
mass in Unit 2. For the
symmetric objects considered in
• the motion of the whole object relative to its centre of mass.
that unit, the centre of mass is As you will see in Unit 19, the motion of the centre of mass may be
the geometric centre. predicted by considering a particle of the same mass as the object, placed
at the point defined by the centre of mass, and subjected to all of the
forces that act on the object. So even in this more complicated situation,
the concept of a particle is important. Alternatively, it may be appropriate
to think of an object as being composed of a number of elements, each of
which can be modelled individually as a particle.
182
1 Describing motion
The derivative of a vector function r(t) whose components are Compare this with the definition
smooth functions of t, is the vector function of the derivative of a function as
) 0 ) 0
dy y(x + h) − y(x)
dr(t) r(t + h) − r(t) = lim .
= lim . (1) dx h→0 h
dt h→0 h
The derivative dr(t)/dt is often written as r! (t) or, where t represents
time, as ṙ(t). Sometimes it is written more succinctly as dr/dt, r! Newton’s notation ṙ(t) is
or ṙ. commonly used in mechanics.
This definition makes use of the concept of the limit of a vector function.
As you might expect, the limit of a vector function f (h) as h → 0 is the
vector function whose components are the limits, as h → 0, of the
components of f (h).
Now, as you will recall, velocity is defined to be rate of change of position,
so the definition of the derivative of a vector function can be used to define
the velocity of a particle with position r(t) as
dr(t)
v(t) = . We sometimes consider velocity
dt as a variable rather than as a
The speed of the particle, which as you will recall is defined to be the function, and write v rather
magnitude of the velocity, is therefore given by |v(t)|. than v(t).
Theorem 1
If a vector function r(t) has the component form
r(t) = x(t) i + y(t) j + z(t) k,
where i, j, k are the (fixed-direction) Cartesian unit vectors, then its
derivative is given by
dr dx dy dz
= i + j + k. (2)
dt dt dt dt
Exercise 2
For each of the particles whose positions are given, calculate the velocity
and speed of the particle at t = 1, correct to two decimal places.
(a) r(t) = t2 i + 10tj (b) r(t) = (sin t)i + (cos t)j + tk
183
Unit 3 Dynamics
Exercise 3
Three particles, A, B and C, are moving along three different straight
lines. In each case, the straight line is chosen as the x-axis, so the vectors
defining the position, velocity and acceleration of the three particles have
only x-components. The graphs of the x-components x(t) of the position of
particle A, ẋ(t) of the velocity of particle B, and ẍ(t) of the acceleration of
particle C are shown in Figure 7. Use the graphs to answer the questions
that follow.
Example 2
A particle is moving in a straight line along the x-axis. The acceleration a
of the particle at time t is given by
a(t) = (12t2 + 2)i.
Note that the ‘initial’ condition After 1 second, the particle is 3 metres from the origin and has velocity
for a problem does not have to 2 m s−1 .
be at t = 0.
(a) Find the velocity v(t) and the position r(t) of the particle.
(b) Find the velocity and position of the particle at time t = 2.
Solution
(a) The acceleration is known and is a one-dimensional vector function.
Using the notation above, we have
a(t) = a(t) · i = 12t2 + 2.
In general, when trying to find Since a(t) = dv(t)/dt, we have the first-order differential equation
the position of a particle given
its acceleration, it is often easier dv
= 12t2 + 2.
to first find its velocity and then dt
find its position, and thus solve This equation can be solved by direct integration, giving
two first-order differential *
equations, rather than to find its v = (12t2 + 2) dt = 4t3 + 2t + A,
position directly by solving a
second-order differential
equation. where A is a constant.
To find the value of the constant A, we use the fact that v = 2 when
t = 1, so 2 = 4 + 2 + A, giving A = −4. This gives
v(t) = 4t3 + 2t − 4.
Since v(t) = dx(t)/dt, we have the first-order differential equation
dx
= 4t3 + 2t − 4.
dt
186
1 Describing motion
where B is a constant.
The initial condition x = 3 when t = 1 can be used to find the
constant B, so 3 = 1 + 1 − 4 + B, giving B = 5. This gives
x(t) = t4 + t2 − 4t + 5.
So the velocity and position of the particle are
v(t) = (4t3 + 2t − 4)i, r(t) = (t4 + t2 − 4t + 5)i.
(b) When t = 2, the velocity and position have values
v(2) = (4 × 23 + 2 × 2 − 4)i = 32i,
r(2) = (24 + 22 − 4 × 2 + 5)i = 17i.
Hence at time 2 seconds, the particle has position 17 metres along the
positive x-axis and velocity 32 metres per second in the direction of
the positive x-axis.
Exercise 4
A particle is moving in a straight line along the x-axis. At time t the
particle has an acceleration given by
a(t) = (18t − 20)i (t ≥ 0).
Initially, at t = 0, the particle has position r(0) = 7i and velocity v(0) = 3i.
Find the position, velocity and speed of the particle at time t = 10.
Exercise 5
A particle is moving in a straight line along the x-axis. At time t the
particle has an acceleration given by
a(t) = ge−kt i (t ≥ 0),
where g and k are positive constants. Initially, at t = 0, the particle is at
the origin (r(0) = 0) and is at rest (v(0) = 0).
Find the velocity and position of the particle as vector functions.
187
Unit 3 Dynamics
Example 3
A particle moves in a straight line along the x-axis with constant
The subscript 0 is used to acceleration a(t) = a0 i (a0 )= 0). The particle starts from x = x0 at time
distinguish the constants a0 and t = 0 with initial velocity v(0) = v0 i.
v0 from the variables a and v.
(a) Show that the velocity vector and position vector of the particle are
given by v(t) = v(t) i and r(t) = x(t) i, where
v(t) = v = v0 + a0 t,
x(t) = x = x0 + v0 t + 21 a0 t2 .
(b) By eliminating t between these two equations, show that
v 2 = v02 + 2a0 (x − x0 ).
Solution
(a) We have
dv
a(t) = = a0 ,
dt
which on integration yields
*
v = a0 dt = a0 t + A,
where A is a constant.
The initial condition v(0) = v0 gives A = v0 , so the velocity is given by
v(t) = v = v0 + a0 t. (3)
Hence
dx
= v0 + a0 t,
dt
from which
*
x = (v0 + a0 t) dt = v0 t + 12 a0 t2 + B,
where B is a constant.
Since the particle starts at x = x0 , we have B = x0 , so the position is
given by
x(t) = x = x0 + v0 t + 21 a0 t2 . (4)
(b) Rearranging equation (3) gives, since a0 )= 0,
v − v0
t= .
a0
Substituting this into equation (4) yields
) 0 ) 0
v − v0 1 v − v0 2
x = x0 + v0 + 2 a0 .
a0 a0
188
1 Describing motion
Constant acceleration
If a particle is moving in a straight line along the x-axis with constant
acceleration a(t) = a0 i, and at time t = 0 it has initial position
r(0) = x0 i and initial velocity v(0) = v0 i, then the components of
acceleration, velocity and position along the x-axis are given,
respectively, by
a = a0 , (5)
v = v0 + a0 t, (6)
1 2
x = x0 + v0 t + 2 a0 t . (7)
Furthermore,
v 2 = v02 + 2a0 (x − x0 ). (8)
189
Unit 3 Dynamics
Exercise 6
A particle moves in a straight line along the x-axis with constant
acceleration a(t) = a0 i. Initially, at time t = 0, the particle is at x = x0
and has velocity v(0) = v0 i.
Use the relationship a = v dv/dx to show that
v 2 = v02 + 2a0 (x − x0 ).
Exercise 7
How would you use the above formulas to substitute for a in the following
equations of motion in order to obtain the specified information? (You are
not expected to solve the resulting equations.)
(a) The x-component of the equation of motion is a = cos t; it is required
to find velocity and position in terms of time.
(b) The x-component of the equation of motion is a = −x; it is required to
find a relationship between velocity and position.
(c) The x-component of the equation of motion is a = −2x − 3v + cos t; it
is required to find position in terms of time.
The models developed in this section are used in practice as shown in the
next exercise.
Exercise 8
The data in Table 2, taken from the United Kingdom Highway Code, show
the shortest stopping distances of cars travelling along a straight road.
The thinking distance is defined to be the distance travelled by a car in the
maximum time that it takes for an alert driver to react to a hazardous
situation.
190
2 A theory of motion
Table 2
Speed Thinking Braking Overall
(mph) distance distance stopping
(metres) (metres) distance
(metres) The British Imperial unit for
20 6 6 12 speed is miles per hour (mph),
and 1 mph ( 0.447 m s−1 .
30 9 14 23
40 12 24 36
50 15 38 53
60 18 55 73
70 21 75 96
(a) The data in Table 2 are not from an experiment; they are the
predictions of models. Your task is to discover what models were used.
(i) What model (using SI units) was used to obtain the thinking
distance data?
(Hint: Think about the speed of the car before and after the
thinking phase, and use the constant acceleration formula
x = x0 + v0 t + 12 a0 t2 .)
(ii) What model (using SI units) was used to obtain the braking
distance data?
(Hint: Think about the speed of the car before and after the
braking phase, and use the constant acceleration formula
v 2 = v02 + 2a0 x.)
(b) Use your models from part (a) to predict the overall stopping distance
(in metres) for a speed of 45 mph.
2 A theory of motion
Section 1 introduced the basic concepts of position, velocity and
acceleration that are needed to describe motion. In this section, two
concepts introduced in Unit 2, force and mass, enable us to go beyond the
mere description of motion and formulate laws predicting what motions
take place.
At first sight it might seem that a different set of rules of motion would be
required for each type of object – one set for tennis balls, another set for
planets, and so on. Fortunately, there is a simple underlying pattern.
Newton was able to see beyond individual cases, and his three laws of Newton’s three laws of motion
motion form a framework, or theory, for predicting the motion of all were stated in the Introduction
objects. of Unit 2.
191
Unit 3 Dynamics
You met the idea of friction in Our instinctive ideas about motion are shaped by the presence of friction
Unit 2. It is discussed further in in almost all things in our everyday lives. Aristotle (384–322 bc)
Subsection 4.1 of this unit. constructed a theory out of this experience that turned out to be
completely wrong. This theory and the subsequent development of ideas of
motion led eventually to Newton’s laws. Consider the following thought
experiment to imagine motion without friction.
Consider a toboggan on a horizontal icy surface such as a frozen lake. Left
undisturbed, the toboggan remains static; it must be pushed or pulled in
some way if it is to be set in motion, that is, a force must act on the
Once the toboggan has been toboggan. However, if you give the toboggan a push and then release it,
released, the force of the push the toboggan will move across the ice at almost constant speed in the
ceases to act on it. direction in which it has been pushed.
This suggests that under ideal (i.e. frictionless) conditions, the following
applies:
in the absence of a force, the toboggan remains at rest or moves with
constant speed in a straight line.
Air resistance is discussed in In real life, the toboggan does eventually slow down, partly due to air
Subsection 4.2. resistance and partly due to friction between the toboggan runners and the
ice. In competitive tobogganing, the tobogganers go to great lengths to
reduce these resistive forces (i.e. forces resisting motion) by streamlining
the toboggan to reduce air resistance, and waxing the runners to reduce
friction.
Exercise 9
A car on a flat, straight road requires a motive force (supplied by its
engine) in order to maintain a constant speed of 70 miles per hour; if the
engine is switched off, then the car slows down. It might be thought from
the above example that if an object is moving with constant velocity, then
there is no force acting on it. Try to explain this apparent contradiction.
Exercise 10
A toboggan on an icy slope may accelerate even when it is not being
pushed. Try to identify the force that causes this acceleration.
192
2 A theory of motion
193
Unit 3 Dynamics
0.2 kg
10 N Exercise 11
If a mass of 200 grams on a smooth horizontal surface (see Figure 9) is
Figure 9 An object on a subjected to a horizontal force of magnitude 10 newtons, what is the
smooth horizontal surface magnitude of the acceleration produced?
Exercise 12
O An object of mass 10 kilograms is attached to a string hanging over the
x edge of a table, as shown in Figure 10. The other end of the string is
attached to another object, on top of the table. The object hanging over
10 kg the edge of the table is observed to be accelerating at 1 m s−2 downwards.
(a) What is the resultant force on the hanging object?
Figure 10 Two objects (b) Apply Newton’s second law to the hanging object, and hence calculate
connected by a string the tension force due to the string acting on the hanging object.
194
3 Predicting motion
Exercise 13
A fighter pilot can experience an acceleration of magnitude approximately
six times the magnitude of the acceleration due to gravity before being
rendered unconscious. If a fighter of mass 4000 kilograms is subjected to a
force 50 000 i + 60 000 j + 100 000 k (with magnitude in newtons) during an
aerobatic manoeuvre, will the pilot remain conscious?
3 Predicting motion
Newton’s second law of motion can be used to help to solve a huge variety
of mechanics problems. The example below considers the motion of an
object falling under gravity alone.
The steps involved in the solution are similar to those in Procedure 2 of
Unit 2. As in Unit 2, the steps are highlighted by labels in the margin.
Example 4
A marble, initially at rest, is dropped from the Clifton Suspension Bridge
and falls into the River Avon, 77 metres below. Assume that the bridge is
fixed and the only force acting on the marble is its weight due to gravity. In Section 4 this example will be
remodelled to include air
(a) Find the time taken before the marble hits the water. resistance.
(b) Find the speed of the marble just before it hits the water.
Solution
(a) The first step is always to draw a diagram that includes all the ! Draw picture #
relevant information given in the problem, as in Figure 11.
x
77 m i
Figure 11 Sketch of the Clifton Suspension Bridge with the data marked
on it
Choose an x-axis pointing vertically downwards, with its origin O at ! Choose axes #
the fixed point where the object is released, as shown in Figure 11.
(This makes the algebra simpler as all quantities are positive.)
The marble is modelled as a particle with its weight the only force ! State assumptions #
acting on it. This is represented by the force diagram in Figure 12, in ! Draw force diagram #
which m denotes the mass of the marble and W denotes its weight.
195
Unit 3 Dynamics
m
i
W
Figure 12 Force diagram for the marble
Try the following exercise, following the same steps as given in the margin
in the above example.
Exercise 14
A stone, dropped from rest, takes 3 seconds to reach the bottom of a well. Use the more general approach
Assume that the only force acting on the stone is gravity. of Example 4 rather than the
constant acceleration formulas of
(a) Estimate the depth of the well. Subsection 1.2.
(b) Estimate the speed of the stone when it reaches the bottom.
197
Unit 3 Dynamics
Exercise 15
A ball is thrown vertically upwards from ground level with an initial speed
of 10 metres per second. Assume that gravity is the only force acting on
the ball.
(a) Find the time taken for the ball to reach its maximum height.
(b) Find the maximum height attained.
198
3 Predicting motion
(c) Find the time taken for the ball to return to the ground.
(d) Find the speed of the ball as it reaches the ground on its return.
Exercise 16
A man leaning out from a window throws a ball vertically upwards from a
point 4.4 metres above the ground. The initial speed of the ball is
7.6 metres per second. It travels up and then down in a straight vertical
line, and eventually reaches the ground. Assume that its weight is the only
force acting on the ball.
(a) Estimate the time that elapses before the ball reaches the ground.
(b) Estimate the speed of the ball when it strikes the ground.
Exercise 17
A marble, initially at rest, is dropped from the Clifton Suspension Bridge
and falls into the River Avon, 77 metres below. Assume that the only force
acting on the marble is its weight due to gravity.
(a) By putting a = v dv/dx in Newton’s second law and solving the
resulting differential equation, find the marble’s speed v as a function
of the distance x through which it has fallen.
(b) By putting v = dx/dt in your answer to part (a), find the time t that
the object takes to fall a distance x.
(c) Hence find the time taken before the marble hits the water, and the
speed of the marble just before it hits the water.
In the previous examples and exercises, all of the forces acting on an object
were in the same direction; but this does not have to be so for the motion
to be in one dimension. This is illustrated in the following example.
Example 5
A crate of empty bottles of total mass 30 kilograms is being hauled by A similar problem was
rope up a smooth ramp from the cellar of a pub. The ramp makes an angle considered in Unit 2. The
of π/6 radians with the horizontal. When the crate has been hauled problem is reconsidered in the
presence of friction between the
2 metres up the ramp (i.e. 2 metres along the slope of the ramp), the rope crate and the ramp in Section 4.
suddenly breaks. It is estimated that if the crate hits the bottom of the
ramp at a speed of 5 metres per second or greater, then the bottles in the
crate will break.
199
Unit 3 Dynamics
30 2m
j k g y
O x
i π
6
! Choose axes # The crate moves down the slope, so we choose the x-axis to point down the
slope, with the origin O at the crate’s position when the rope breaks. (We
model the crate as a particle, so its position is at a point.) Choose a y-axis
perpendicular to this, as shown in Figure 13. Also shown in the picture are
the unit vectors i and j implied by this choice of axes.
! State assumptions # The only forces on the crate are the weight W and the normal reaction N
of the ramp (since we are neglecting friction and air resistance in this
example). So modelling the crate as a particle of mass m, the force
! Draw force diagram # diagram (showing also how the relevant angles are calculated) is as shown
in Figure 14.
N j
m
π π i
6 3 π
6
W
Figure 14 The force diagram
! Apply Newton’s 2nd law # Now, Newton’s second law for this system gives
As we have done elsewhere, we ma = W + N. (14)
use the letter m rather than the
value 30 kg for the mass since, as From the force diagram, N = |N| j. The weight can be resolved into
will be discussed in Unit 8, components using the formula from Unit 2, Example 9:
inserting actual data values too √
3
early in the modelling process W = |W| cos π3 i − |W| sin π3 j = 21 |W| i − 2 |W| j.
can obscure features of the
resulting model.
The motion is along the slope, so a = ai. (All we are saying here is that
there is no resultant force in the j-direction.) Now that we have all the
Since there is no motion in the vectors in component form, we can immediately resolve equation (14) in
j-direction, we do not need to the i-direction:
resolve in this direction. If we
did, we would have ma = 21 |W| + 0.
√
3
|N| − mg 2 = 0. Substituting |W| = mg gives ma = 12 mg, so
200
3 Predicting motion
a = 12 g.
Again m cancels out, so the results apply to a crate of any mass.
We want an equation for the velocity v in terms of the distance travelled x. ! Solve differential equation #
Since the acceleration is constant, we can use equation (8), that is,
v 2 = v02 + 2ax.
This gives the final velocity in terms of known quantities: the initial
velocity v0 is 0 m s−1 , the distance travelled x is 2 m, and the constant
acceleration is a = 21 g.
Putting the data into the equation gives ! Interpret solution #
2 2 1
v =0 +2× 2g × 2 ( 19.62 to 2 d.p.
So v ( 4.4 to one decimal place, which means that the crate is travelling at
about 4.4 metres per second when it hits the bottom of the ramp. This
speed is just below the estimated speed at which the bottles will break, so
– provided that the estimate of the speed at which the bottles will break is
a good one – this model predicts that the bottles will not break.
Exercise 18
A skier of mass 65 kilograms starts from rest at the top of a 120-metre ski
slope (i.e. 120 metres is the vertical distance from top to bottom), as
shown in Figure 15.
O
x j
65 kg
120 m
i
θ
A
The skier travels down the slope before leaving the ground and then
landing some distance further down the hill. Use the x-axis shown, which
points down the slope with the origin O at the top of the slope and point A
at the bottom of the slope. Let θ be the angle that the slope makes with
the horizontal. Assume that friction and air resistance can be neglected. This problem is reconsidered in
π the presence of friction in
(a) If θ = 4, what is the speed of the skier at point A at the bottom of the Section 4.
slope?
(b) If θ = π3 , what is the speed of the skier at point A at the bottom of the
slope?
201
Unit 3 Dynamics
4.1 Friction
F N
In Unit 2 you encountered a model for the friction force that acts on an
object at rest in contact with a surface. This force F has magnitude less
W than or equal to µ|N|, that is, |F| ≤ µ|N|, where N is the normal reaction
force and µ is the coefficient of static friction between the object and the
Figure 16 The forces acting surface. The direction of the friction force opposes any possible motion
on a block at rest on a slope (e.g. is up the slope in Figure 16).
The coefficient µ is called the coefficient of static friction to distinguish it
The names coefficient of from the coefficient of sliding friction, denoted by µ! , that is used when
dynamical friction or coefficient an object is moving along a surface. In this situation, experiments show
of kinetic friction are sometimes that the magnitude of the friction force F is equal to µ! |N|, that is,
used instead of the name
coefficient of sliding friction.
|F| = µ! |N|, where N is the normal reaction force and µ! is the coefficient
of sliding friction; the direction of the friction force is opposite to the
direction of motion.
Example 6
Consider again the problem of Example 5 concerning the crate of empty
bottles sliding down a cellar ramp, but this time assume that there is
friction between the crate and the ramp, with coefficient of sliding friction
µ! = 0.15.
202
4 Some more force models
(a) Estimate the speed of the crate when it reaches the bottom of the
ramp. Will the bottles break? In Example 5 it was assumed
the bottles would break if the
(b) Compare the answer to part (a) with the answer to Example 5, and speed at the bottom of the ramp
comment. exceeded 5 m s−1 .
Solution
(a) The situation is sketched in Figure 17. ! Draw picture #
µ! = 0.15
30 2m
j k g y
O x
i π
6
Choose the x-axis parallel to the ramp and the y-axis perpendicular to ! Choose axes #
it, as shown in Figure 17. Choose the origin O to be the crate’s
position when the rope breaks.
The crate is modelled as a particle. The forces acting on the crate are ! State assumptions #
its weight W, the normal reaction of the ramp on the crate N, and the
friction force between the ramp and the crate F, giving the force
diagram in Figure 18. ! Draw force diagram #
N j
F
m
π π i
6 3 π
6
W
Figure 18 The force diagram
Using m for the mass of the crate (as in Example 5), applying ! Apply Newton’s 2nd law #
Newton’s second law to this system gives
ma = W + N + F. (15)
From the force diagram, N = |N| j and F = −|F| i. It is also apparent
that the motion is one-dimensional parallel to the x-axis, that is,
a = ai. The weight is resolved into components as
W = |W| cos π3 i − |W| sin π3 j
√
3
= 21 mg i − 2 mg j.
Now we can immediately resolve equation (15) in the i-direction to
obtain
ma = 12 mg + 0 − |F|.
203
Unit 3 Dynamics
Table 3 shows some values of the coefficient of sliding friction that may be
useful in problems involving sliding objects.
204
4 Some more force models
Exercise 19
A tip-up truck is delivering a concrete block to a building site. The driver
increases the angle of tip of the carrier until the concrete block begins to
slide, then keeps the carrier at this constant angle. The coefficient of static
friction between a concrete block and metal is approximately µ = 0.4, and
the coefficient of sliding friction is approximately µ! = 0.3.
(a) Calculate the angle at which the concrete block begins to slide.
(b) If the concrete block has 3 metres to travel before leaving the carrier,
how long will it take to unload it?
Exercise 20
Repeat Exercise 18 under the new modelling assumption that friction
cannot be neglected; the coefficient of sliding friction between waxed skis
and dry snow is 0.03. Compare your answers here with those for
Exercise 18, and comment.
205
Unit 3 Dynamics
log10 (|R|)
5
Figure 19 The experimental data for air resistance together with broken
lines showing the two simplest models and their ranges of validity (shaded)
The graph has a complicated shape, so the air resistance force is modelled
by a complicated function of D|v|. However, there are two simple models
that fit the experimental data for wide ranges of situations, namely the
linear and quadratic models
|R| = c1 D|v| and |R| = c2 D2 |v|2 ,
where c1 and c2 are positive constants. The best fit of these models to the
experimental data (when using SI units) is given by c1 = 1.7 × 10−4 and
c2 = 0.20 over certain ranges:
The symbol " means ‘less than |R| ( 1.7 × 10−4 D|v| for D|v| " 10−5 ;
about’. |R| ( 0.2D2 |v|2 for 10−2 " D|v| " 1.
These approximations for air resistance are shown as broken lines in
Figure 19, together with their ranges of validity (shaded).
Consider a sphere of given diameter D. From Figure 19 it can be seen that
the linear model applies for low velocities, and the quadratic model applies
for an intermediate range of velocities. Notice also that there are large
ranges of velocities where neither of these simple models applies.
The air resistance force R always opposes the motion, that is, it is in the
Note that the vector and scalar direction of −v. So the vector equations for the air resistance models given
statements of the quadratic air above are
resistance model agree since
+ + R = −c1 Dv and R = −c2 D2 |v|v.
+−c2 D2 |v|v+ = c2 D2 |v| |v|
= c2 D2 |v|2 .
Air resistance
The air resistance force R on a smooth spherical object of diameter D
travelling with velocity v can be modelled as follows:
linear model R = −c1 Dv for D|v| " 10−5 , (17)
quadratic model R = −c2 D2 |v|v for 10−2 " D|v| " 1, (18)
where c1 ( 1.7 × 10−4 and c2 ( 0.20.
206
4 Some more force models
These models can be also used for other fluids, with different ranges
of applicability. For example, in water we have c1 = 9.4 × 10−3 for
D|v| " 10−6 and c2 = 156 for 10−3 " D|v| " 10−1 .
The use of these models is illustrated by examples, the first of which adds
air resistance to Example 4.
Example 7
In Example 4, the falling time for a marble dropped from the Clifton
Suspension Bridge into the River Avon was calculated to be 3.96 seconds.
The experimental value is 4.1 seconds. Can the discrepancy be accounted
for by a linear air resistance model?
The investigation is subdivided into the following steps.
(a) Find how, under a linear air resistance model, the distance from the
point of release varies with time for an arbitrary spherical object of
mass m and diameter D.
(b) Calculate the time, under a linear air resistance model, that a marble
of diameter 2 cm and mass 13 g takes to fall the 77 m from the Clifton
Suspension Bridge into the River Avon below.
(c) Comment on the validity of the linear air resistance model for this
problem.
Solution
(a) The picture is the same as in Example 4 – see Figure 11. ! Draw picture #
The x-axis points vertically downwards, with its origin O at the point ! Choose axes #
where the marble is released. The unit vector i acts downwards.
There are two forces acting on the marble, namely its weight due to ! State assumptions #
gravity downwards and air resistance upwards, as shown in the force
diagram in Figure 20. ! Draw force diagram #
W i
Figure 20 The force diagram
Applying Newton’s second law to the marble gives ! Apply Newton’s 2nd law #
ma = W + R. (19)
From the force diagram, W = |W| i = mgi. Using the linear air
resistance model, we have R = −c1 Dv = −c1 Dvi, since v = vi with
v ≥ 0. The acceleration is also downwards, so a = ai. Now we can
resolve equation (19) in the i-direction to obtain
ma = mg − c1 Dv.
207
Unit 3 Dynamics
! Solve differential equation # We want the distance x in terms of the time t. One way of obtaining
Notice that the acceleration is this is to first use the substitution a = dv/dt, and then substitute
not constant here, so we cannot dx/dt for v later. So we have
use the constant acceleration
dv
formulas. m = mg − c1 Dv.
dt
This assumption is checked later. We now make the assumption that mg − c1 Dv > 0 (i.e. the air
It is certainly true initially as resistance never becomes so strong as to overcome the marble’s
the marble is dropped from rest. weight). Under this assumption, the differential equation can be solved
The integrating factor method by the separation of variables method, to obtain
could also be used. * *
m
dv = 1 dt,
mg − c1 Dv
which, on dividing the numerator and denominator by m, and for
convenience writing k = C1 D/m, gives
* *
1
dv = 1 dt.
g − kv
Note that g − kv > 0 if Integrating, using the assumption that g − kv > 0, leads to
mg − c1 Dv > 0. 1
− ln(g − kv) = t + A, (20)
k
where A is an arbitrary constant. Rearranging gives
g − kv = e(−kt−kA) = Be−kt ,
where B = e−kA is another constant. The initial condition that the
marble is initially at rest (i.e. v = 0 when t = 0) gives B = g, so
g
Since c1 , D, t and m are all v = (1 − e−kt ). (21)
greater than or equal to zero, so k
that kt > 0, this equation Writing dx/dt for v gives
predicts that v < g/k = mg/c1 D
for all t, which is consistent with dx g
= (1 − e−kt ).
the assumption made in dt k
obtaining equation (20). Integrating directly gives
g g
x = t + 2 e−kt + C,
k k
where C is a constant.
The initial condition x = 0 when t = 0 gives C = −g/k 2 , so
g g
x = t − 2 (1 − e−kt ), (22)
k k
where k = c1 D/m.
Since e−kt → 0 as t → ∞, the speed v tends to g/k, and the position x
tends to gt/k − g/k 2 → gt/k.
! Interpret solution # (b) We want to find t given x = 77. To do this, we could try to rearrange
equation (22) to give t in terms of x; however, no such rearrangement
is possible. But we can use a numerical method. First we use the given
data to interpret equation (22) in the context of the current problem.
208
4 Some more force models
We have
c1 D 1.7 × 10−4 × 0.02
k= = ( 2.6 × 10−4 ,
m 0.013
so
9.81 9.81 . ,
−2.6×10−4 ×t
x( t − 1 − e . (23)
2.6 × 10−4 (2.6 × 10−4 )2
Now we already have two values for the time taken for the marble to
fall 77 metres: 3.96 seconds estimated in Example 4, and 4.1 seconds
given by experiment. Substituting these values for t into
equation (23), we obtain
x(3.96) ( 76.89 and x(4.1) ( 82.42.
So it looks like t = 3.96 is close to the solution of equation (23) for
x = 77.
So t = 3.96 (to two decimal places) when x = 77, that is, the marble More detailed calculation shows
takes about 3.96 seconds to reach the River Avon under the linear air that the linear air resistance
resistance model. This is exactly the same value, to two decimal model predicts that the marble
will hit the water approximately
places, as for the model without air resistance (Example 4). one thousandth of a second later
(c) A condition for the linear air resistance model to be valid is that the than it will in the case of the
product of the diameter of the marble and its speed is less than model without air resistance.
about 10−5 . Using equation (21) gives |v(3.96)| ( 38.8, so D|v| ( 0.78,
which is much greater than 10−5 m s−1 . So the linear air resistance In fact, D|v(t)| " 10−5 for
model is not appropriate. t " 5.1 × 10−5 , so the linear
model applies for only about the
first 51 microseconds of the
motion of the marble.
From Example 7 it seems that the linear air resistance model may apply
only to objects moving very slowly – not to the speeds experienced in
everyday life. So our attention turns to the quadratic air resistance model.
In general, the differential equations that arise from this model are harder
to solve than the ones that arise from the linear model. However, some of
the differential equations are easily soluble, as the following example shows.
Example 8
Revisit the Clifton Suspension Bridge problem in Example 7 using the
quadratic air resistance model.
(a) Derive an expression for the marble’s velocity in terms of its position.
(b) Use the expression derived in part (a) to estimate the speed of the
marble just before it hits the water.
(c) Is the quadratic air resistance model valid for this problem?
Solution
(a) Everything is the same as in Example 7 up to the point where we
apply Newton’s second law to the marble, to obtain ! Apply Newton’s 2nd law #
ma = W + R. (24)
209
Unit 3 Dynamics
210
4 Some more force models
(b) Substituting the values for the marble (see Example 7) gives ! Interpret solution #
c2 D2 0.2 × (0.02)2
k= = ( 6.2 × 10−3 ,
m 0.013
so from equation (25),
!
9.81 '
v= 1 − e−2×6.2×10−3 ×77 ( 31.2.
6.2 × 10−3
So the model predicts that the speed of the marble is about 31 metres
per second just before it hits the water.
(c) To test whether the quadratic air resistance model applies to the
motion, we must calculate D|v| and check that it is in the range 0.01
to 1. At the end of the motion, D|v(77)| ( 0.62, which is within the
limits of applicability of the model. The motion starts from rest, and
D|v(0)| = 0 is obviously outside the limits of the model. However,
D|v(0.01)| ( 0.01, so after the marble has dropped approximately one Admittedly, this calculation has
centimetre, the quadratic model applies. So the quadratic model been done using the quadratic
applies for almost all of the motion. model, but it can’t be far out!
The given air resistance models apply only to smooth spherical objects. To
apply these models to other objects, we have to model the objects as
smooth spheres. The diameter of the sphere used to model an object is
referred to as the effective diameter of the object. The following exercise The determination of an effective
makes use of this idea. diameter for an object is beyond
the scope of this module.
Exercise 21
In the discussion leading up to the statement of Newton’s second law in
Section 2, the motion of an empty toboggan sliding on ice was considered.
In this exercise, the effect of air resistance on the motion is examined.
Assume that the quadratic air resistance model applies, and that the
effective diameter of the toboggan is 5 centimetres. If the toboggan has
mass 2 kilograms and an initial speed of 2 metres per second, by what
percentage has its speed diminished after it has travelled 100 metres?
Terminal speed
The general equation for the velocity v = vi of an object falling from rest
under gravity, ignoring the effects of air resistance, was derived in
Example 4 as
v = gt.
This predicts that v increases indefinitely as t increases.
211
Unit 3 Dynamics
The general equation for the velocity v = vi of an object falling from rest
under gravity with linear air resistance was derived in Example 7 to be
g
v = (1 − e−kt ),
k
where k = c1 D/m. The exponential term decreases to zero as time
v increases, so
g mg
v→ = as t → ∞.
vT k c1 D
This behaviour, which is quite different from the case of the model that
neglects air resistance, is illustrated in Figure 21. The limiting value is
called the terminal speed of the object and is denoted by vT . So in the
t case of the linear air resistance model, the terminal speed (in SI units) is
Figure 21 Velocity graph g mg
vT = = .
k c1 D
The quadratic air resistance model predicts results that are qualitatively
similar to the linear case. The general equation for the speed of an object
falling from rest under gravity with quadratic air resistance was derived in
Example 8 to be
! '
g
v= 1 − e−2kx ,
k
where k = c2 D2 /m. The exponential term decreases to zero as x increases,
so the quadratic model predicts a terminal speed (in SI units) given by
! !
g mg
vT = = .
k c2 D2
There is another way of looking at the concept of terminal speed. The
equation of motion of an object falling from rest under gravity with linear
See Example 7. air resistance is ma = mg − c1 Dv; thus vT = mg/c1 D is exactly that value
of v for which a = 0. The equation of motion of an object falling from rest
See Example 8. under' gravity with quadratic air resistance is ma = mg − c2 D2 v 2 ; again
vT = mg/c2 D2 is such that a = 0. Thus in each case the terminal speed
is the speed at which the object can fall without accelerating, or in other
words at which the air resistance just balances the object’s weight. This
way of looking at terminal speed enables us to derive equations for the
terminal speed of any falling object, irrespective of whether or not the
object is falling from rest, as the following example illustrates.
Example 9
A small spider is blown by the wind into the high atmosphere and, once
the wind has died away, falls back to Earth. The spider may be modelled
as a sphere of effective diameter 2 millimetres.
If the spider has mass 0.004 grams, calculate the speed at which it will
land, assuming that the fall is long enough for the spider effectively to
reach terminal speed.
212
4 Some more force models
Solution
The first step is to draw a picture and mark the x-axis on it, then draw the
force diagram (see Figure 22).
x R
W i
Figure 22 The spider descending to Earth, and the force diagram
213
Unit 3 Dynamics
The results about the terminal speed of objects falling under the influence
of gravity and air resistance alone are summarised in the following box.
Exercise 22
A parachutist of mass 65 kilograms has a parachute of effective diameter
10 metres when fully opened. Estimate the landing speed of the
parachutist, assuming that the parachute jump is long enough for the
terminal speed effectively to be reached.
Exercise 23
The parachutist will start in The maximum speed at which a parachutist can land safely is about
freefall before pulling the rip 13 metres per second. Assuming that the parachute jump is long enough
cord, thus will be approaching for the terminal speed effectively to be reached, calculate the effective
the terminal speed from above.
It will be important that when diameter of a parachute that will enable a parachutist of mass
the rip cord is pulled, there is 70 kilograms to land safely.
sufficient time to slow down.
Exercise 24
Revisit the Clifton Suspension Bridge problem described in Example 8,
with different initial conditions. Assume now that instead of it being
dropped from rest, the marble is catapulted downwards with an initial
velocity of 50 m s−1 .
214
5 Projectiles
The initial stages of the solution will be exactly the same as in the
example, since the initial conditions are not used until the step solving the
differential equation. So the equation of motion of the marble is still
dv
v = g − kv 2 ,
dx
where k = c2 D2 /m is a positive constant.
(a) Derive an expression for the marble’s velocity in terms of its position.
(b) Describe the motion of the marble.
5 Projectiles
In Section 1 we discussed motion in one dimension. In this section we
discuss two-dimensional motion – more specifically, the motion of
projectiles modelled as particles. We will use the following terminology in Whenever we mention a
our discussion. During the period that the projectile is off the ground and projectile in this section, it will
subject only to the force of gravity (and possibly air resistance), it is said be modelled as a particle.
to be in flight. The start of the flight is the launch, and the initial velocity
is the launch velocity. The flight ends with an impact (often hitting the
ground again, but possibly hitting some other target). The time of flight is
the time between the moment of launch and the moment of impact. The
path of the projectile while in flight is its trajectory.
In this section, we consider only models in which the effect of air resistance In ignoring air resistance, we
is assumed to be negligible. also ignore effects such as the
swerve that may occur when a
In Subsection 5.1, we model the forces acting on a projectile and solve its football is kicked with spin, or
equation of motion. In Subsection 5.2, we look at the trajectory of a when a golf ball is sliced.
projectile and consider a variety of examples.
215
Unit 3 Dynamics
y j y j
B
u
i i
θ
O A x O x
(a) (b)
216
5 Projectiles
217
Unit 3 Dynamics
The sine function never exceeds 1 in value, and we have sin 2θ = 1 when
For a launch on horizontal 2θ = π2 , that is, when θ = π4 . Since the launch angle must be between 0
ground, we must have θ > 0. and π2 , other solutions can be ignored. So for a given launch speed u, the
(For a launch from above ground maximum range Rmax for a projectile launched from a horizontal surface
level, such as from a cliff or
bridge, we could have a launch
(ignoring air resistance) is obtained using a launch angle of π4 to the
angle below the horizontal, when horizontal, and this maximum range is
θ would be negative.) u2
Rmax = .
g
From equation (34), we see that y is a quadratic function of t, with a
negative coefficient of t2 . So the graph of y against t is part of a parabola
opening downwards. Such a parabola has a single stationary point where
The condition ẏ(t) = 0 is ẏ(t) = 0, and this will give the maximum value of y. Differentiating
equivalent to asserting that the equation (34) with respect to t gives
vertical component of the
velocity is zero when the ẏ(t) = u sin θ − gt,
projectile is at its maximum
height. and this is 0 when u sin θ − gt = 0, that is, when t = (u sin θ)/g.
Substituting t = (u sin θ)/g into the right-hand side of equation (34), the
corresponding maximum height is given by
) 0 ) 0
u sin θ g u sin θ 2
H=u sin θ −
g 2 g
2 2 2 2
u sin θ u sin θ
= −
g 2g
2 2
u sin θ
= . (36)
2g
Substituting t = (u sin θ)/g into the right-hand side of equation (33), the
x-component of the projectile’s position at the point of maximum height
is, using sin 2θ = 2 sin θ cos θ,
) 0
u sin θ
x=u cos θ
g
u2 sin θ cos θ
=
g
2
u sin 2θ
Thus the maximum height = .
occurs when x is half the 2g
range R, as you might expect. The results derived above apply to any projectile where the points of
launch and impact are in the same horizontal plane (and air resistance is
ignored). Some problems involving projectiles can conveniently be solved
by direct use of these results.
Example 10
During a particular downhill run, a short but sharp rise causes a skier to
leave the ground at 25 m s−1 at an angle of π6 above the horizontal. The
ground immediately beyond the rise is horizontal for 60 metres. After this,
the slope is again downhill. Will the skier land on the level ground or on
the downhill slope beyond it?
218
5 Projectiles
Solution
Model the skier as a particle launched from the end of the rise (see y
Figure 25). With u = 25 and θ = π6 , the expression in equation (35) for the
range on a horizontal surface gives 25 m s−1
252 sin π3 π
R= ( 55.17. 6
9.81 O x
60 m
Since this is less than 60 metres, it would seem that the skier will land on
the flat part of the run.
(This conclusion is expressed cautiously because of the underlying Figure 25 The skier at
modelling assumptions. Drag forces may reduce the range of a projectile, take-off
but a skier may also experience aerodynamic lift forces that would increase
the range. Also, the skier may change the position of her skis relative to
the position of her centre of mass – for example, by bending or
straightening her legs – which would affect the validity of the model of the
skier as a particle.)
Exercise 25
A ball kicked from a flat piece of ground at an angle of π6 above the
horizontal lands 40 metres away from where it was kicked. What is the
greatest height above the ground that the ball will have reached?
y u j
Exercise 26 θ
Consider a projectile launched at an angle θ above the horizontal, from a i
point at a height h above the origin O, with speed u. Take t = 0 to be the h
moment of launch, and use the coordinate system (and associated unit
vectors) shown in Figure 26. Find the solution of r̈(t) = −gj satisfying O x
these initial conditions. Figure 26 A projectile at
launch
219
Unit 3 Dynamics
Exercise 27
After a road accident, a crashed car is found on a sandy beach at the base
of a cliff. The cliff is vertical and is 18 metres high. The investigating
police officer finds that the marks in the sand resulting from the car’s
impact on the beach start 8 metres from the base of the cliff, and that the
point of impact is at roughly the same horizontal level as the cliff base.
The car appears to have been travelling at right angles to the cliff when it
went over. Assuming that the car was travelling in a horizontal direction
when it left the cliff, estimate the speed with which it went over.
Projectiles
The equation of motion of a projectile subject only to the force of
gravity is
r̈(t) = −gj, (39)
where j is a unit vector pointing vertically upwards.
If the projectile is launched at time t = 0, from the point x = 0,
y = 0, with launch speed u in the (x, y)-plane and launch angle θ
above the horizontal, then the solution of the equation of motion
satisfying these initial conditions is
x = ut cos θ, (40)
1 2
y = ut sin θ − 2 gt . (41)
The maximum height H reached by such a projectile is
u2 sin2 θ
H= . (42)
2g
There are also two results for the range of a projectile launched from a
level surface.
Exercise 28
A shot putter launches a shot at a speed of 13 m s−1 at an angle of π6 above
the horizontal from a height of 1.8 metres above ground level. How far will
the shot travel in the horizontal direction before it hits the ground,
assuming that the ground is horizontal?
Exercise 29
A stone is thrown from a height of 1.5 metres above horizontal ground at
an angle of π4 above the horizontal and lands at a distance of 30 metres
from the point where it was thrown. Estimate the speed with which it was
thrown.
x = ut cos θ, (45)
1 2
y = ut sin θ − 2 gt . (46)
221
Unit 3 Dynamics
Example 11
A golfer wants to play a recovery shot through a copse of trees. There is a
small gap in the foliage at a height of 12 metres and 40 metres in front of
him. He knows that with his usual swing, he hits the ball at about
35 m s−1 . What angle of launch will enable the ball to hit the gap in the
foliage?
Solution
We make the usual choice of axes, with origin at the point of launch. The
golfer wants the trajectory of the ball to pass through the point x = 40,
y = 12 (working in SI units). So from equation (47) with u = 35, we have
9.81
12 = 40 tan θ − 402 (1 + tan2 θ)
2 × 352
( 40 tan θ − 6.407(1 + tan2 θ),
so
6.407 tan2 θ − 40 tan θ + 18.407 = 0.
222
5 Projectiles
In any problem where we need to find a launch angle (given the launch
speed) to hit a specified target, we arrive at a quadratic equation for tan θ,
namely equation (47). So long as this equation has two distinct real roots,
there will be two launch angles that enable the target to be hit. Of course,
the roots may coincide. In either case, we say that the target is
achievable. On the other hand, we may arrive at a quadratic equation
with no real roots. In this case, the target is not achievable – for the given
launch speed, there is no launch angle that enables the target to be hit.
Exercise 30
Suppose that a projectile has launch speed u at an angle θ above the
horizontal, and that it is launched from (0, h) at time t = 0. (That is, the
projectile is launched at a height h above the origin.) By eliminating t
from equations (37) and (38), obtain an equation (relating y to x) for the
trajectory of such a projectile.
We see from Exercise 30 that for a launch at height h above the origin, we
need to add only a term h to the right-hand side of equation (47) for the
trajectory. This result is frequently useful.
Trajectory of a projectile
For a projectile launched at time t = 0 from (0, h) with launch
speed u and launch angle θ above the horizontal, the trajectory has
the equation
g
y = h + x tan θ − x2 2 sec2 θ. (48)
2u
It is often convenient to use the trigonometric identity
sec2 θ = 1 + tan2 θ to give
g
y = h + x tan θ − x2 2 (1 + tan2 θ), (49)
2u
which is a quadratic equation in tan θ.
Exercise 31
A basketball player is 2.6 metres (horizontally) from the basket. The
basket is 3 metres above ground level. The player launches the ball at
7 m s−1 , from a height of 1.8 metres above ground level. What angle of
launch should the player choose?
We now look at an example that requires more work to answer the posed
question.
223
Unit 3 Dynamics
Example 12
This example concerns baseball fielders throwing the ball back to the
catcher. Assume throughout that the point of launch and the point of
impact are at the same horizontal level.
(a) A fielder can just throw a ball a distance of 60 metres. How fast can
the fielder throw the ball?
(b) A fielder needs to throw a ball to the catcher from a distance of
58 metres. Assuming that the fielder throws directly to the catcher at
the speed calculated in part (a), what is the shortest time in which the
fielder can return the ball to the catcher?
(c) Suppose that a second fielder is midway between the first fielder and
the catcher (so that each gap is 29 metres), and that each fielder
throws at the speed calculated in part (a). The first fielder throws to
the second fielder, then the second fielder throws to the catcher. As
well as the time in flight, the second fielder requires 0.3 seconds to
catch and throw the ball. Would this ‘relaying’ result in a quicker
return of the ball to the catcher?
Solution
(a) In the previous subsection we found that the maximum range of a
projectile (for launch speed u) is achieved at a launch angle π4 , and
this maximum range is u2 /g (see equation (44)). So for the fielder, we
have u2 /g = 60, which gives
' √
u = 60g = 60 × 9.81
= 24.26 to 2 d.p.
So the fielder can throw the ball at a speed of approximately
24.3 m s−1 .
(b) Taking the point of launch as origin, and working in metres, the
trajectory of the ball needs to pass through the point (58, 0). So using
equation (47), we need a launch angle θ where
g
It is simpler, as well as
√ more 0 = 58 tan θ − 582 (1 + tan2 θ).
accurate, to use u = 60g here. 2 × 60g
This can be rearranged as
120
tan2 θ − tan θ + 1 = 0.
58
This quadratic equation has solutions tan θ = 0.7696 and
tan θ = 1.299. The corresponding values of θ (between 0 and π2 ) are
0.656 (37.6◦) and 0.915 (52.4◦). Throwing the ball at either of these
angles will return it to the catcher.
To find the time that it takes the ball to reach the catcher, we can use
equation (45). When x = 58, the time t must satisfy
'
58 = ut cos θ = 60g t cos θ.
224
5 Projectiles
225
Unit 3 Dynamics
226
5 Projectiles
the projectile does have a maximum range, and as z = L/R is the only
stationary point, this stationary point must be a maximum. Alternatively,
we could show mathematically that the stationary point given by z = L/R
is a maximum by considering the sign of the second derivative d2 R/dz 2 .
Exercise 32
At a tutorial, one of the students, who happens to be an expert shot
putter, asserts that aiming to launch at an angle π4 has always been good
enough for him. He says that improving launch speed is the key to good
shot putting. Assume that the student launches the shot from a height of
2 metres above ground level.
(a) The student can put a shot 17 metres with a launch angle π4 above the
horizontal. Calculate the speed at which the shot is being launched to
achieve this range.
(b) For launch at the speed calculated in part (a), use equations (51)
and (52) to find the optimum launch angle and the corresponding
range.
(c) If the student achieves a launch speed 1% higher than that calculated
in part (a) and launches at an angle π4 above the horizontal, what
range will he achieve?
227
Unit 3 Dynamics
Exercise 33
As usual in this unit, model the A footballer taking a free kick launches the ball from ground level so that
ball as a particle. This means it just clears a player who is 10 metres away and 2 metres high. The ball
that any swerve that a footballer enters the goal 30 metres away at a height of 2.4 metres.
may achieve by kicking the ball
with spin will be overlooked. (a) Take as the origin the point from which the ball was kicked. Let the
launch speed be u, and let the launch angle above the horizontal be θ.
Use equation (48) (with h = 0) twice to obtain two equations that u
and θ must satisfy.
Multiply one of these equations by a suitable constant, so that the
term (g sec2 θ)/2u2 has the same coefficient in each equation. Then
subtract one equation from the other to eliminate u, and thus obtain
an equation that is satisfied by tan θ. Hence find the launch angle θ.
(b) At what speed was the ball kicked? What period of time elapsed from
the moment the ball was kicked until it entered the goal?
Learning outcomes
After studying this unit, you should be able to:
• understand and use the basic terms for the description of the motion
of particles: position, velocity and acceleration
• understand and use vector functions
• understand the fundamental laws of Newtonian mechanics
• solve mechanics problems in one dimension by drawing a sketch,
choosing a suitable x-axis and origin, making assumptions, drawing a
force diagram, applying Newton’s second law, taking the
x-component, and making suitable substitutions
• solve mechanics problems in one dimension that involve one or more of
the forces of gravity, friction and air resistance
• understand the concept of terminal speed, and use it in solving
mechanics problems in one dimension
• apply Newton’s second law in vector form to problems in more than
one dimension
• solve problems relating to the motion of a projectile in the absence of
air resistance.
228
Solutions to exercises
Solutions to exercises
Solution to Exercise 1
(a) The graph of x against t for the puck is shown below.
x(t)
2
0
0.5 1 1.5 t
−1
(b) (i) The distance between the first player and the back wall is the
distance between where the puck starts and where it first changes
direction. This can be read off the graph as 2 m.
(ii) The second player must be further away than the first player
since the x-coordinate is −1 when the puck changes direction for
a second time, indicating that the second player is 3 m from the
wall.
(iii) The speed at which the puck is travelling is given by the slope of
the distance–time graph. The slope after the second player hits
the puck is greater than the slope after the first player hits the
puck, so the second player gives the puck more speed.
Solution to Exercise 2
) 0 ) 0
dr(t) d 2 d
(a) = t i+ 10t j = 2ti + 10j.
dt dt dt
So the velocity of the particle at t = 1 is
v = 2i + 10j,
and the speed is
√ √
|v| = 4 + 100 = 104 ( 10.20 to 2 d.p.
dr(t)
(b) = (cos t)i − (sin t)j + k.
dt
So the velocity of the particle at t = 1 is
v = (cos 1)i − (sin 1)j + k,
and the speed is
' √
|v| = cos2 1 + sin2 1 + 1 = 2 ( 1.41 to 2 d.p.
229
Unit 3 Dynamics
Solution to Exercise 3
(a) The graph for particle A is its position–time graph. The component of
velocity of the particle along the x-axis, dx/dt, is given by the slope of
this graph: it can be seen that the slope starts at a high value and
decreases as time increases. So particle A is travelling fastest when
t = 0.
(b) The graph for particle B shows the variation with time of the particle’s
component of velocity along the x-axis. For the particle to change
direction, this velocity must change from positive to negative or from
negative to positive (i.e. the graph must cross the t-axis). The graph
shown does not do this, so particle B does not change its direction.
(c) The graph for particle C shows the variation with time of the
particle’s component of acceleration along the x-axis. On the interval
[0, 2.5] the acceleration is positive, so the velocity increases in this
interval. On the interval [2.5, 5] the acceleration is negative, so the
velocity decreases in this interval. To discover whether the particle is
ever stationary for t > 0, we need to know whether its velocity along
the x-axis is ever zero for t > 0. Now, since acceleration is obtained by
differentiating velocity, we can obtain velocity by integrating
acceleration. We also know that the definite integral of a function over
a given interval gives the ‘area’ under the graph of the function, where
‘areas’ below the t-axis are negative. So the ‘area’ under the graph
gives us the velocity.
Now, on the interval [0, 2.5], the ‘area’ under the curve is positive, and
since the particle starts from rest, the velocity after 2.5 seconds is
positive. However, on the interval [2.5, 5], the ‘area’ under the curve is
negative, and furthermore the magnitude of this ‘area’ is greater than
the magnitude of the ‘area’ for the interval [0, 2.5]; therefore the
velocity after 5 seconds is negative. Hence, since the particle has both
positive and negative velocity in the given time interval, it must be
momentarily stationary at some point towards the end of the time
interval.
Solution to Exercise 4
Since
dv
a(t) = = 18t − 20,
dt
we have
*
v = (18t − 20) dt = 9t2 − 20t + A,
where A is a constant.
Using the initial condition v(0) = 3, we obtain A = 3. Hence the
component of the velocity of the particle along the x-axis is given by
v(t) = 9t2 − 20t + 3.
230
Solutions to exercises
where B is a constant.
The initial condition x(0) = 7 gives B = 7. Hence the component of the
position of the particle along the x-axis is given by
x(t) = 3t3 − 10t2 + 3t + 7.
Substituting t = 10 into the expressions for x(t) and v(t) gives
x(10) = 3000 − 1000 + 30 + 7 = 2037,
v(10) = 9 × 102 − 20 × 10 + 3 = 703,
so at time t = 10 the particle is 2037 metres from the origin, with a speed
of 703 m s−1 along the positive x-axis. Thus the position is r(10) = 2037 i
and the velocity is v(10) = 703 i.
Solution to Exercise 5
We have
dv
a(t) = = ge−kt .
dt
Integrating this gives
*
g
v = ge−kt dt = A − e−kt ,
k
where A is a constant.
The initial condition v(0) = 0 gives A = g/k, so the velocity is given by
g g g
v(t) = − e−kt = (1 − e−kt ).
k k k
Then from v(t) = dx/dt we have
* . ,
g g g g
x= − e−kt dt = t + 2 e−kt + B,
k k k k
where B is a constant.
The initial condition x(0) = 0 gives B = −g/k 2 , so the position is given by
g g g g g
x(t) = t + 2 e−kt − 2 = t − 2 (1 − e−kt ).
k k k k k
Therefore the velocity and position of the particle are given by the vector
functions
g$ "
v(t) = 1 − e−kt i,
.kg g ,
r(t) = t − 2 (1 − e−kt ) i.
k k
231
Unit 3 Dynamics
Solution to Exercise 6
Given that a(t) = a0 , the relationship a = v dv/dx gives
dv
v = a0 .
dx
Applying the separation of variables method to this differential equation
gives
* *
v dv = a0 dx,
so
1 2
2v = a0 x + C,
where C is a constant.
Using the initial condition that the velocity is v0 along the x-axis at
x = x0 gives C = 12 v02 − a0 x0 , so
1 2
2v = a0 x + 12 v02 − a0 x0 .
Multiplying through by 2 and rearranging gives
v 2 = v02 + 2a0 (x − x0 ),
as required.
Solution to Exercise 7
(a) To find v in terms of t, substitute dv/dt for a; then v may be found by
direct integration. To find x in terms of t, substitute dx/dt for v and
integrate again.
(b) To find a relationship between v and x, substitute v dv/dx for a. The
result is an equation that can be solved by separation of variables,
whose solution will give the required relationship.
(c) To find x in terms of t, substitute d2 x/dt2 for a, and dx/dt for v. The
result is a linear constant-coefficient second-order differential equation
for the variable x, namely ẍ + 3ẋ + 2x = cos t. This can be solved by
the methods of Unit 1. (The general solution is
1 3
x = Ae−2t + Be−t + 10 cos t + 10 sin t.)
Solution to Exercise 8
(a) (i) The figure below shows the thinking distance in metres against
the speed in miles per hour. The speed of a car before the
thinking phase is the value given in the table. The speed after
the thinking phase is exactly the same, because the driver has
not yet reacted to the hazard. So the acceleration is zero during
this phase, and with x0 = 0, the formula x = x0 + v0 t + 12 a0 t2
reduces to x = v0 t.
232
Solutions to exercises
x
20 Thinking
distances
15
10
0
20 30 40 50 60 70 Speed
233
Unit 3 Dynamics
60
40
20
0
20 30 40 50 60 70 Speed
234
Solutions to exercises
Solution to Exercise 9
Any moving car is subject to resistive forces, namely air resistance, the
internal frictional forces in the car’s engine, transmission and wheel
bearings, and the external frictional forces between the car’s tyres and the
road. In order to maintain a constant velocity, it is necessary to apply a
motive force that balances these resistive forces.
Solution to Exercise 10
The component of the force of gravity in the direction of a unit vector
pointing down the slope is non-zero. If the slope is steep enough, then this
force down the slope will be greater than the resistive force of friction,
causing the toboggan to accelerate.
Solution to Exercise 11
Newton’s second law, F = ma, is a vector equation. Taking the magnitude
of both sides of the equation (and using the fact that mass is always
positive) gives the scalar equation |F| = m|a|, into which the values given
in the question can be substituted (after converting the mass from grams
into the SI unit kilograms) to obtain
10 = 0.2|a|.
This gives |a| = 50, so the force produces an acceleration of magnitude
50 m s−2 .
Solution to Exercise 12
(a) The only forces on the hanging object are the weight W of the object
(a downward force, in the direction of the positive x-axis shown) and
the tension force T due to the string (an upward force, in the T
direction of the negative x-axis). The resultant force on the object is
the sum W + T of these forces. This information is shown in the force m
diagram in the margin. i
W
(b) Applying Newton’s second law to this system gives
ma = W + T.
From the force diagram, we have W = |W| i = mgi, where g is the
magnitude of the acceleration due to gravity. Similarly, the tension
force due to the string is T = −|T| i. The given acceleration of the
hanging object is 1 m s−2 downwards, so a = i. With this information
we can resolve the equation above in the i-direction to obtain
m × 1 = mg − |T|.
Substituting m = 10 and rearranging gives
|T| = 10g − 10 ( 88.1 (using g = 9.81 m s−2 ).
So T = −|T| i = −88.1i, that is, the tension force due to the string is
88.1 newtons in the upward direction.
235
Unit 3 Dynamics
Solution to Exercise 13
Substitute the given force and mass into the equation for Newton’s second
law to obtain
4000 a = 50 000 i + 60 000 j + 100 000 k.
So a = 12.5i + 15j + 25k, which has magnitude (in m s−2 )
√
12.52 + 152 + 252 ( 32. This is well below the threshold of
6g ( 59 m s−2 , so the pilot should remain conscious.
Solution to Exercise 14
! Draw picture # (a) First, we draw a picture.
x
i
! Choose axes # The x-axis is chosen to point vertically downwards, with the origin O
at the top of the well, as shown above.
! State assumptions # The stone is modelled as a particle, and the model assumes that the
! Draw force diagram # only force is the stone’s weight due to gravity. The force diagram is as
follows.
m
i
W
! Apply Newton’s 2nd law # Applying Newton’s second law to the stone gives W = ma. Since
W = mgi, we have ma = mgi, and resolving in the i-direction gives
a = g.
! Solve differential equation # Using a = dv/dt, we obtain
dv
= g.
dt
Integrating this gives
v = gt + A,
where A is a constant.
The initial condition that the stone is dropped from rest (v = 0 when
t = 0) gives A = 0. Hence
v = gt.
236
Solutions to exercises
Solution to Exercise 15
(a) First, we draw a picture. ! Draw picture #
x
i
Choose the x-axis to point vertically upwards, with its origin O at the ! Choose axes #
point from which the ball is thrown, as shown above.
The model assumes that the weight due to gravity is the only force ! State assumptions #
acting on the ball, so the force diagram is as follows. ! Draw force diagram #
m i
W
Applying Newton’s second law to the ball gives ! Apply Newton’s 2nd law #
ma = W.
Since the x-axis points upwards, the weight of the ball is given by
W = −mgi. The acceleration is downwards, so a = ai, where a is
negative. Resolving in the i-direction gives ma = −mg. Dividing by
the mass gives
a = −g.
237
Unit 3 Dynamics
! Solve differential equation # To answer the question, we need an equation relating x to t, and
another relating v to t or to x. Since the acceleration a = −g is
constant, one approach is to use the constant acceleration formulas of
Subsection 1.2. The initial velocity is 10 m s−1 upwards from the
origin. Using the notation of Subsection 1.2, the initial velocity is
v0 = 10i, so v0 = 10 and a0 = −g, and hence
v = v0 + a0 t = 10 − gt,
x = v0 t + 21 a0 t2 = 10t − 21 gt2 .
(The approach of Example 4, using the substitutions a = dv/dt and
v = dx/dt and integrating, leads to the same pair of equations.)
! Interpret solution # The ball reaches its maximum height when v = 0, and from the
equation for v this occurs at time
10
t= ( 1.02 to 2 d.p.
g
Since the motion started at t = 0, the duration of the upward flight of
the ball is about 1 second.
(b) Substituting t = 10/g into the equation for x, we have
100 100
x= − ( 5.10 to 2 d.p.
g 2g
So the maximum height attained by the ball is about 5.1 metres.
The motion up and down has (c) The ball reaches the ground when x = 0, and from the equation for x
the same acceleration −gi, so it this occurs when
can be treated as one motion.
There is no need to consider the 0 = 10t − 12 gt2 = t(10 − 12 gt).
upward and downward motions
separately. Hence t = 0 or t = 20/g ( 2.04 to two decimal places. Now t = 0
corresponds to the time when the ball is thrown, so the ball returns to
the ground after approximately 2 seconds.
(d) Substituting t = 20/g into the equation for v gives
v = 10 − gt = 10 − 20 = −10, so v = 10(−i).
The ball reaches the ground on its return with a speed of 10 m s−1 .
(The (−i) indicates that the ball is now travelling in the direction of
decreasing x, i.e. downwards.)
Solution to Exercise 16
! Draw picture # (a) First, we draw a picture.
x
initial height
i
4.4 m
O
238
Solutions to exercises
The x-axis is chosen to point vertically upwards, with the origin O at ! Choose axes #
ground level, as shown in the diagram. (The other obvious choice for
the origin, which you may well have chosen, is at the point where the
ball is thrown. Both choices of course lead to the same answers below.)
The model assumes that the only force acting on the ball is its weight ! State assumptions #
due to gravity, so the force diagram is as follows. ! Draw force diagram #
m i
W
Applying Newton’s second law to the ball gives ma = W, where ! Apply Newton’s 2nd law #
a = ai and W = −mgi (as the x-axis points upwards). Resolving in
the i-direction gives
ma = −mg, so a = −g.
We want equations relating x to t, and v to x or t. One approach is to ! Solve differential equation #
use the substitution a = dv/dt, to obtain
dv
= −g.
dt
Integrating this, we obtain
v = −gt + A,
where A is a constant.
The initial velocity of the ball is 7.6i, so v = 7.6 when t = 0, which
gives A = 7.6. So
v = −gt + 7.6.
Therefore using the substitution v = dx/dt gives
dx
= −gt + 7.6.
dt
Integrating this gives
x = − 21 gt2 + 7.6t + B,
where B is a constant.
The initial condition that x = 4.4 when t = 0 leads to B = 4.4. So
x = − 21 gt2 + 7.6t + 4.4.
(Since the acceleration is constant, another approach is to use the
constant acceleration formulas to obtain the equations for v and x.
Note that with the choice of origin used here, x0 = 4.4, v0 = 7.6 and
a0 = −9.81. You could use the equation x = x0 + v0 t + 12 a0 t2 to
obtain the equation for x.)
The ball reaches the ground when x = 0. Substituting this into the ! Interpret solution #
equation for x gives a quadratic equation for the time t,
4.905t2 − 7.6t − 4.4 = 0,
239
Unit 3 Dynamics
whose solution is
'
7.6 ± (−7.6)2 − 4 × 4.905 × (−4.4)
t= ,
2 × 4.905
so t ( 1.998 or t ( −0.4489. The negative time is before the ball is
thrown and may therefore be ignored.
So the ball lands about 2 seconds after being thrown.
It may be easier to find the (b) Substituting this time into the equation for v gives
speed v at ground level by
solving v = −gt + 7.6 = −9.81 × 1.998 + 7.6 ( −12.00 to 2 d.p.
v dv/dx = −g So the ball lands with a speed of about 12.0 m s−1 . (The negative sign
to obtain for v confirms that the ball is moving downwards, i.e. the speed is
1 2 1 2
2 v − 2 v0 = −g(x − x0 ), 12.0 m s−1 in the −i direction.)
so when x = 0,
v 2 = v02 + 2gx0 . Solution to Exercise 17
(a) Choose the same x-axis and the same origin as in Example 4, and
proceed in exactly the same way as before until you reach the
equation a = g. Now write a as v dv/dx to obtain
dv
v = g.
dx
Solving this differential equation by the method of separation of
variables, we have
* *
v dv = g dx, so 21 v 2 = gx + A,
where A is a constant.
Now the marble starts from rest, so v = 0 when x = 0, which leads to
A = 0. Hence 21 v 2 = gx, or equivalently,
'
v = 2gx,
where we have taken the positive square root because the velocity is
positive throughout the motion.
(This equation could also have been obtained from the constant
acceleration formula (8).)
(b) Putting v = dx/dt in the equation for v gives
dx '
= 2gx.
dt
Again we use the method of separation of variables to solve this
differential equation. So we have
* * *
1 1 1
1 dt = √ dx = √ x−1/2 dx = √ 2x1/2 + B,
2gx 2g 2g
where B is a constant. So
!
2x
t= + B.
g
240
Solutions to exercises
Solution to Exercise 18
(a) As the question asks for the speed for two different angles of the slope
of the jump, it is sensible (as suggested in Figure 15) to use θ to be an
arbitrary angle of slope and substitute for θ at the interpretation stage.
A diagram of the situation is shown with the question as Figure 15, ! Draw picture #
and is repeated here.
O
x j
65 kg
120 m
i
θ
A
The x-axis is given as pointing down the slope with origin O at the top ! Choose axes #
of the slope, as shown above. The model assumes that the skier can ! State assumptions #
be treated as a particle, and that the only forces acting on the skier
are her weight due to gravity and the normal reaction from the slope. ! Draw force diagram #
N
m j
π
2 −θ i
θ
241
Unit 3 Dynamics
! Apply Newton’s 2nd law # Applying Newton’s second law to the skier gives
ma = W + N.
From the force diagram, N = |N| j. We can resolve W into
components:
W = |W| cos( π2 − θ) i − |W| sin( π2 − θ) j
= mg sin θ i − mg cos θ j.
Now we can resolve in the i-direction to obtain
ma = mg sin θ, so a = g sin θ.
! Solve differential equation # As we want the velocity when the skier has travelled a vertical
distance of 120 metres, it is best to find v as a function of x. Since the
acceleration is constant, we can use equation (8) to obtain
v 2 = v02 + 2ax = 0 + 2(g sin θ)x,
so
'
v= 2gx sin θ.
! Interpret solution # The task is to find the velocity when the skier has travelled a vertical
distance of 120 metres. Now x is the distance travelled down the
slope; so, using trigonometry, x = 120/ sin θ. Substituting for x in the
equation for v gives
!
120 '
v = 2g × × sin θ = 240g ( 48.52 to 2 d.p.
sin θ
So the speed of the skier at the bottom of the slope angled at π4 is
about 48.5 m s−1 .
The substitution of θ at an early (b) In part (a), all mention of θ cancelled from the final expression. So the
stage would have led to a lot of answer remains the same: the speed at the bottom of the π3 slope is
extra work here as v is about 48.5 m s−1 .
independent of θ.
(In the equation for v, note that x sin θ = h is the vertical height thus
the speed at the bottom depends on the height of the slope and not
the angle of the slope.)
Solution to Exercise 19
! Draw picture # (a) Let θ be the angle that the carrier makes with the horizontal, as
shown below, and let m be the mass of the block.
y 3m
j
µ = 0.4 O
µ! = 0.3 x i
θ
242
Solutions to exercises
Choose the axes to be parallel and perpendicular to the carrier, with ! Choose axes #
origin O at the rest position of the concrete block, as shown above.
The block is modelled as a particle, and we assume that the only ! State assumptions #
forces acting on the block are the block’s weight W, the normal
reaction with the carrier N, and the friction force up the slope F.
This gives the following force diagram. ! Draw force diagram #
N j
F
m
i
π
2 −θ
θ
W
243
Unit 3 Dynamics
(b) To solve the dynamics problem when the block is in motion down the
carrier, we start in exactly the same way as for the statics problem. So
we start the analysis of the motion by applying Newton’s second law,
using the same axes, with the origin O at the point at which slipping
first occurs.
! Apply Newton’s 2nd law # Applying Newton’s second law to the block gives
ma = F + N + W.
The acceleration is down the carrier, so a = ai and all the forces are
resolved in exactly the same way as in the statics problem. So we can
resolve in the i-direction to obtain
ma = mg sin θ − |F|.
Now from the moment at which the block begins to slide, |F| = µ" |N|,
so this equation becomes
ma = mg sin θ − µ" |N|.
To find |N|, we resolve the original equation in the j-direction to
obtain
0 = −mg cos θ + |N|.
Therefore |N| = mg cos θ (as before). Thus
ma = mg sin θ − µ" mg cos θ,
so
a = g sin θ − µ" g cos θ.
! Solve differential equation # For a fixed angle θ, the acceleration is constant, so we can use the
constant acceleration formulas from Section 1. Since we want to relate
time to distance travelled, the appropriate formula is equation (7):
x = x0 + v0 t + 21 a0 t2 .
Initially, the block is at rest at x0 = 0, so v0 = 0 and this becomes
x = 12 g(sin θ − µ" cos θ)t2 .
! Interpret solution # The time before the block slides off the back of the truck is calculated
from this equation using the value of θ calculated in part (a).
Substituting the distance travelled (x = 3) and µ" = 0.3, we obtain
Using the exact angle 3 ( 21 g[sin(0.381) − 0.3 cos(0.381)]t2 ( 0.458t2 ,
θ = arctan 0.4 gives t = 2.57 to
two decimal places. so t ( 2.56 to two decimal places.
So once the concrete block begins to slide, it takes about 2.6 seconds
to slide off the back of the truck.
Solution to Exercise 20
! Draw picture # First, we draw a picture.
! Choose axes # The x-axis is given to be parallel to the slope, with origin O at the top of
the slope. Choose the y-axis to be perpendicular to the slope, as shown
below.
244
Solutions to exercises
y
O
x j
65 kg
120 m
µ! = 0.03 i
θ
A
The skier is modelled as a particle, and we assume that the only forces ! State assumptions #
acting on the skier are her weight, the normal reaction with the slope, and
the friction force up the slope. This gives the following force diagram. ! Draw force diagram #
N j
F
m
i
π
2 −θ
θ
W
Applying Newton’s second law to the skier gives ! Apply Newton’s 2nd law #
ma = W + N + F.
From the force diagram, N = |N| j and F = −|F| i. The weight can be
resolved into components as
W = |W| cos( π2 − θ) i + |W| sin( π2 − θ) (−j)
= mg sin θ i − mg cos θ j.
The acceleration is down the slope, so a = ai, and we can resolve in the
i-direction to obtain
ma = mg sin θ + 0 − |F|.
Using |F| = µ" |N| = 0.03|N|, this becomes
ma = mg sin θ − 0.03|N|.
Resolving in the j-direction leads to
0 = −mg cos θ + |N| + 0,
so |N| = mg cos θ. Substituting gives
ma = mg sin θ − 0.03 × mg cos θ,
so
a = g sin θ − 0.03g cos θ.
Using equation (8), we obtain ! Solve differential equation #
2
v = v02 + 2ax = 0 + 2(g sin θ − 0.03g cos θ)x,
so
'
v= 2g(sin θ − 0.03 cos θ)x.
245
Unit 3 Dynamics
Solution to Exercise 21
! Draw picture # First, we draw a picture.
i 100 m
2 kg
x
O
! Choose axes # Choose the x-axis along the direction of motion, with origin O at the point
where the toboggan has speed 2 m s−1 , as shown in the diagram.
! State assumptions # We model the toboggan as a particle. There are three forces acting on the
toboggan: its weight, the normal reaction from the horizontal surface, and
the air resistance force. The vertical forces W and N are in equilibrium
and do not affect the motion. This leaves only the air resistance force R,
! Draw force diagram # which opposes the motion, as shown in the force diagram below.
R i
2 kg
246
Solutions to exercises
dv
mv = −c2 D2 v 2 .
dx
Solving by separation of variables,
* *
1
dv = − k dx,
v
where k = c2 D2 /m, so
ln v = −kx + A,
where A is a constant. Hence
v = Be−kx ,
where B = eA is another constant. The initial condition that the toboggan
is initially moving at 2 m s−1 (i.e. v = 2 when x = 0) gives B = 2, so
v = 2e−kx ,
where k = c2 D2 /m.
Substitute the data given in the question to find the velocity after 100 m: ! Interpret solution #
2 2
c2 D 0.2 × (0.05)
k= = = 2.5 × 10−4 ,
m 2
v(100) = 2 exp(−2.5 × 10−4 × 100) ( 1.95.
So the percentage decrease in velocity is about
v(0) − v(100) 2 − 1.95
100 = 100 = 2.5,
v(0) 2
that is, after 100 metres the toboggan has lost only about 2.5% of its
speed.
Solution to Exercise 22
First suppose that the linear air resistance model applies, so that the
terminal speed is
mg 65 × 9.81
vT = = ( 4 × 105 .
c1 D 1.7 × 10−4 × 10
So DvT ( 4 × 106 , which is greater than 10−5 , so the linear model does
not apply.
Now suppose that the quadratic model applies, so that the terminal speed
is
! !
mg 65 × 9.81
vT = = ( 5.6.
c2 D 2 0.2 × 102
So DvT ( 56, which is greater than 1, so the quadratic model does not
apply either.
The condition for the quadratic model is closer to being satisfied than the
condition for the linear model, so the quadratic model is likely to produce
the better estimate. So the conclusion is that the landing speed of the
parachutist is approximately 6 m s−1 .
247
Unit 3 Dynamics
(Looking at Figure 19, it can be seen that 6 m s−1 (with D = 10, so that
log10 (D|v|) ( 1.8) falls to the right of the range of validity of the
quadratic model. In this region, you can see that the quadratic model lies
just above the experimental curve, so that it gives a slight overestimate of
the air resistance. So it should be expected that the actual landing speed
is slightly greater than 6 m s−1 . Also note that this landing speed would be
achieved by a particle falling under gravity, without air resistance, from a
height of 1.8 m.)
Solution to Exercise 23
You saw in Exercise 22 that the quadratic air resistance model is better
than the linear model for
' problems of this type. Rearrangement of the
quadratic model vT = mg/c2 D2 gives
! !
mg 70 × 9.81
D= 2 = ( 4.5.
c2 vT 0.2 × 132
So the effective diameter needs to be at least 4.5 metres.
As in the solution to Exercise 22, we have
log10 (D|v|) = log10 (4.5 × 13) ( 1.8, so the quadratic model overestimates
the air resistance in this case. So to be safe, a parachute with effective
diameter of 5 or even 6 metres would probably be needed.
Solution to Exercise 24
(a) The equation of motion of the marble is given in the question as
dv
v = g − kv 2 ,
dx
where k = c2 D2 /m is a positive constant.
Proceeding as in Example 8, we solve the differential equation by
separation of variables:
* *
v
dv = 1 dx.
g − kv 2
Now the first difference due to the changing initial conditions occurs,
since if v = 50, then
0.2 × (0.02)2 × (50)2
g − kv 2 = 9.81 − ( −5.6.
0.013
So the denominator of the first integrand is negative (whereas it was
positive before). Rewriting this to make the denominator positive by
taking out the factor −1 gives
* *
v
− dv = 1 dx,
kv 2 − g
so
1
− ln(kv 2 − g) = x + A,
2k
where A is a constant.
248
Solutions to exercises
249
Unit 3 Dynamics
Solution to Exercise 26
We need to find the solution of r̈(t) = −gj satisfying ṙ(0) = u, where
u = (u cos θ)i + (u sin θ)j and r(0) = hj (since the point of launch is (0, h)).
The integral of r̈(t) = −gj is
ṙ(t) = −gtj + c,
where c is a constant vector.
Substituting t = 0 and using the initial condition ṙ(0) = u, we must have
c = u.
Integrating again gives
r(t) = − 12 gt2 j + ut + d,
where d is a constant vector.
Substituting t = 0 and using the initial condition r(0) = hj, we must have
d = hj.
Hence the required solution is
r(t) = − 21 gt2 j + (ut cos θ)i + (ut sin θ)j + hj
= (ut cos θ)i + (h + ut sin θ − 21 gt2 )j.
Solution to Exercise 27
We choose the y-axis vertically upwards and the x-axis horizontal. We
choose the origin at the bottom of the cliff at beach level. Suppose that
the car left the cliff with a speed u, travelling in a horizontal direction (at
right angles to the cliff). Then the initial conditions are r(0) = 18j and
ṙ(0) = ui. We can use the solution of r̈(t) = −gj derived in Exercise 26,
with θ = 0, which is
r(t) = uti + (h − 12 gt2 )j,
or, separated into components,
x = ut,
y = h − 12 gt2 .
We know that x = 8 when y = 0 (assuming that the car hit the ground
exactly 8 metres from the cliff and modelling the car as a particle). So
from the first equation, the car hit the ground at t = 8/u. Substituting
this into the second equation gives
0 = 18 − 12 g(64/u2 ).
Thus, as u is positive,
!
64 × 9.81
u= ( 4.18 to 2 d.p.
2 × 18
So the car left the cliff at a speed of just over 4 m s−1 .
250
Solutions to exercises
Solution to Exercise 28
Taking the origin to be at ground level, and using equations (37) and (38),
the position of the shot at a time t after the launch is given by
x = 13t cos π6 ,
y = 1.8 + 13t sin π6 − 1
2 × 9.81t2 .
To find the time when the shot hits the ground, we substitute y = 0 in the
second equation and solve the resulting quadratic equation for t. The
solutions are t = 1.560 and t = −0.2352. The negative solution represents
a time before the shot is launched and so can be rejected. At t = 1.560, we
have x = 17.57.
So the shot lands at a horizontal distance of 17.57 metres from the point
of launch.
Solution to Exercise 29
Taking the origin to be at ground level, we can use equations (37) and (38)
with h = 1.5 and θ = π4 . Suppose that the launch speed is u. Then
√
equation (37) gives x = ut cos π4 = ut/ 2. If the stone hits the ground
√ √
when t = T , we have 30 = uT / 2, so T = 30 2/u. We know that y = 0
when t = T , so substituting into equation (38) gives
√ - √ #2
30 2 1 g 30 2
0 = 1.5 + u √ −
u 2 2 u
900g
= 31.5 − .
u2
'
This gives u = 30 9.81/31.5 ( 16.74.
So the launch speed was approximately 16.74 m s−1 .
Solution to Exercise 30
Equations (37) and (38) are
x = ut cos θ,
y = h + ut sin θ − 21 gt2 .
From the first equation, t = x/(u cos θ). Substituting this into the second
equation gives
x g . x ,2
y =h+u sin θ −
u cos θ 2 u cos θ
2 g
= h + x tan θ − x 2
sec2 θ.
2u
Alternatively, using sec2 θ = 1 + tan2 θ,
g
y = h + x tan θ − x2 2 (1 + tan2 θ).
2u
251
Unit 3 Dynamics
Solution to Exercise 31
We choose the origin to be at ground level, vertically below the point of
launch. So the equation of the trajectory of the basketball is equation (49)
with h = 1.8 and u = 7 (using SI units). In order for the ball to pass
through the hoop, we want the point x = 2.6, y = 3 to be on this
trajectory. Hence
9.81
3 = 1.8 + 2.6 tan θ − (2.6)2 (1 + tan2 θ).
2 × 72
This simplifies to the quadratic equation
0.6767 tan2 θ − 2.6 tan θ + 1.877 = 0.
(Alternatively, and more efficiently, you may have chosen the origin to be
the point from which the ball was launched. However, this leads to the
same equation for tan θ.)
This equation for tan θ has the two solutions
tan θ = 2.879 and tan θ = 0.963.
Each of these gives a single value for θ in the range 0 ≤ θ ≤ π2 :
θ = 1.236 (70.9◦) and θ = 0.767 (43.9◦).
We see that there are two possible launch angles that enable the target to
be hit. In this example, the choice of a launch angle of approximately 71◦
is more likely to be suitable, since this has the ball descending towards the
net at the steeper angle, so the ball is less likely to catch on the rim of the
basket.
Solution to Exercise 32
(a) We choose the origin to be at ground level, vertically below the point
of launch. So the equation of the trajectory of the shot is
equation (48) with h = 2 and θ = π4 (using SI units). The trajectory
must pass through the point of impact, namely x = 17, y = 0. So the
launch speed u must satisfy the equation
172 × 9.81 × 2
0 = 2 + 17 − .
2u2
(Alternatively, you may have chosen the origin to be the point from
which the shot is launched. Then the equation of the trajectory is
equation (48) with h = 0, and the point of impact is x = 17, y = −2.
However, you should arrive at the same equation for u as above.)
We have u2 = (172 × 9.81)/19, so u ( 12.22.
So the launch speed is about 12.22 m s−1 .
252
Solutions to exercises
253
Unit 3 Dynamics
Solution to Exercise 33
(a) The trajectory must pass through the points (10, 2) and (30, 2.4). So
using equation (48) twice, we have
g
2 = 10 tan θ − 100 2 sec2 θ,
2u
g
2.4 = 30 tan θ − 900 2 sec2 θ.
2u
To eliminate the sec2 θ term, we multiply the first equation by 9 to
obtain
g
18 = 90 tan θ − 900 2 sec2 θ.
2u
Subtracting gives
15.6 = 60 tan θ.
So tan θ = 0.26 and θ = 0.2544 (14.6◦).
(b) Substituting θ = 0.2544 into the first equation in part (a) gives
50g
2 = 10 tan(0.2544) − sec2 (0.2544),
u2
so
50 × 9.81 sec2 (0.2544)
u2 = = 872.28
10 tan(0.2544) − 2
and u ( 29.53.
So the ball was kicked at a speed of approximately 29.5 m s−1 to one
decimal place.
Using equation (45), we have x = ut cos θ. We know that the ball
entered the goal when x = 30, so t = 30/(29.53 cos(0.2544)) ( 1.05 to
two decimal places.
So just over 1 second after having been kicked, the ball entered the
goal.
254
Index
Index
absolute error 29, 30 derivative 6
acceleration 184, 189, 193 of a vector function 183
acceleration due to gravity 125 differentiable function 7
accumulation 4 differential equation 7
achievable target 223 first-order 7
addition of vectors 101, 103 general solution 8
in component form 108 homogeneous 16
air resistance 205, 206 inhomogeneous 16
algebraic rules for scaling and adding vectors 103 linear 16, 34
analytic solution 3 non-homogeneous 16
angle between vectors 115 order 7
approximate solution 24 particular solution 8
arbitrary constant 7 solution 7
area direct integration 11
of a parallelogram 122 direction field 21
of a triangle 122 direction of a vector 97, 99
associated homogeneous equation 46 discriminant 44
associativity 103 displacement 95
auxiliary equation 37 distributivity 103, 112, 120
domain 10
balanced rod 142 dot product 111, 112
birth rate 4 dynamics 96, 177
boundary condition 62
boundary value 62 effective diameter 211
boundary-value problem 62 efficiency 31
equal vectors 99
Cartesian components of a vector 108 equation of a straight line 110
Cartesian coordinates 100, 105 equation of motion 187
Cartesian unit vectors 100, 107 equilibrium 124, 144
centre of mass 141, 182 equilibrium condition
characteristic equation 37 for a particle 124
choosing axes 125 for a rigid body 144
coefficient of dynamical friction 202 Euler, Leonhard 27
coefficient of kinetic friction 202 Euler’s method 27
coefficient of sliding friction 202, 204 exceptional cases in the method of undetermined
coefficient of static friction 130, 132 coefficients 57
commutativity 102, 103, 112 exponential function 51
complementary function 46 extended body 141
component form
first-order differential equation 7
of a cross product 121
force 95, 193, 194
of a dot product 114
mathematical representation 96
of a vector 107, 108
force diagram 124
component of a vector 108
force of gravity 124
in an arbitrary direction 116
friction 202
constant acceleration 189
friction force 129
constant-coefficient equation 34
fundamental laws of Newtonian mechanics 194
cost 31
cross product 118, 120 general solution of a differential equation 8, 43, 46, 47
gradient 21
death rate 4 gravitational force 124
decay constant 15
dependent variable 6 hinge 146
255
Index
257