uu
mm
mm
K A STROUD
hi ~
ENGINEERING MATHEMATICS
Programmes and Problems
K. A. Stroud
MACMILLAN
K. A. Stroud 1970
Published by
MACMILLAN AND CO LTD
London and Basingstoke
Associated companies in New York, Toronto,
Melbourne, Dublin, Johannesburg and Madras
logical subjects, and for other courses of a comparable level. While formal
proofs are included where necessary to promote understanding, the
emphasis throughout is on providing the student with sound mathematical
skills and with a working knowledge and appreciation of the basic con
cepts involved.The programmed structure ensures that the book is highly
suited for general class use
and for individual selfstudy, and also provides
a ready means for remedial work or subsequent revision.
The book is the outcome of some eight years' work undertaken in the
development of programmed learning techniques in the Department of
Mathematics at the Lanchester College of Technology, Coventry. For the
past four years, the whole of the mathematics of the first year of various
Engineering Degree courses has been presented in programmed form, in
conjunction with seminar and tutorial periods. The results obtained have
proved to be highly satisfactory, and further extension and development
of these learning techniques are being pursued.
Each programme has been extensively validated before being produced
in its final form and has consistently reached a success level above 80/80,
i.e. at least 80% of the students have obtained at least 80% of the possible
marks in carefully structured criterion tests. In a research programme,
carried out against control groups receiving the normal lectures, students
working from programmes have attained significantly higher mean scores
than those in the control groups and the spread of marks has been con
siderably reduced. The general pattern has also been reflected in the results
of the sessional examinations.
The advantages of working at one's own rate, the intensity of the
student involvement, and the immediate assessment of responses, are well
known to those already acquainted with programmed learning activities.
Programmed learning in the first year of a student's course at a college or
university provides the additional advantage of bridging the gap between
the rather highly organised aspect of school life and the freer environment
and which puts greater emphasis on personal responsibility for his own pro
gress which faces every student on entry to the realms of higher education.
Acknowledgement and thanks are due to all those who have assisted
in any way in the development of the work, including those who have
been actively engaged in validation processes. I especially wish to
record my sincere lhanks for the continued encouragement and support
which I received from my present Head of Department at the College,
Mr. J. E. Sellars, M.Sc, A.F.R.Ae.S., F.I.M.A., and also from
Mr. R. Wooldridge, M.C., B.Sc, F.I.M.A., formerly Head of Department,
now Principal of Derby College of Technology. Acknowledgement is also
made of the many sources, too numerous to list, from which the selected
examples quoted in the programmes have been gleaned over the years.
Their inclusion contributes in no small way to the success of the work.
K. A. Stroud
CONTENTS
Preface v
Hints on using the book xii
Useful background information xiii
Introduction 37
Loci problems
Test exercise II
Further problems II
Introduction 73
Graphs of hyperbolic functions
Evaluation of hyperbolic functions
Inverse hyperbolic functions
Log form of the inverse hyperbolic functions
Hyperbolic identities
Trig, identities and hyperbolic identities
Relationship between trigonometric & hyperbolic functions
Test exercise III
Further problems HI
Programme 4: Determinants
Determinants \q\
Determinants of the third order
Evaluation of a third order determinant
Simultaneous equations in three unknowns
Consistency of a set of equations
Properties of determinants
vii
1
Test exercise IV
Further problems IV
Programme 5: Vectors
^/Programme 6: Differentiation
Partial differentiation
25
Small increments
Test exercise IX
Further problems IX
Series
297
Arithmetic and geometric means
Series of powers of natural numbers
Infinite series: limiting values
Convergent and divergent series
Tests for convergence; absolute convergence
Test exercise XI
Further problems XI
Introduction
357
Standard integrals
Functions of a linear function
Integrals of the form
Integration of products 
integration by parts
Integration by partial fractions
Integration of trigonometrical functions .
Introduction 457
Volumes of solids of revolution
Centroid of a plane figure
Centre of gravity of a solid of revolution
Lengths of curves
Lengths of curves  parametric equations
Surfaces of revolution
Surfaces of revolution  parametric equations
Rules of Pappus
Revision summary
Test exerciseXVII
Further problems XVII
t Introduction
517
j. Approximate integration
1 Method 1 by series
1
Introduction 593
Formation of differential equations
Solution of differential equations
Method 1  by direct integration
Method 2  by separating the variables
Method 3 homogeneous equations: by substituting y = vx
Method 4  linear equations: use of integrating factor
Test exercise XXII
Further problems XXII
xi
HINTS ON USING THE BOOK
This book contains twentyfour lessons, each of which has been
written in such a way as to make learning more effective and more
interesting. It is almost like having a personal tutor, for you proceed at
your own rate of learning and any difficulties you may have are cleared
before you have the chance to practise incorrect ideas or techniques.
You will find that each programme is divided into sections called
frames, each of which normally occupies half a page. When you start a
programme, begin at frame 1. Read each frame carefully and carry out
any instructions or exercise which you are asked to do. In almost every
frame, you are required to make a response of some kind, testing your
understanding of the information in the frame, and you can immediately
compare your answer with the correct answer given in the next frame. To
obtain the greatest benefit, you are strongly advised to cover up the
following frame until you have made your response. When a series of dots
occurs, you are expected to supply the missing word, phrase, or number.
At every stage, you will be guided along the right path. There is no need
to hurry: read the frames carefully and follow the directions exactly. In
this way, you must learn.
At the end of each programme, you will find a short Test Exercise.
This is set directly on what you have learned in the lesson: the questions
are straightforward and contain no tricks. To provide you with the
necessary practice, a set of Further Problems is also included: do as many
of these problems as you can. Remember that in mathematics, as in many
other situations, practice makes perfect or more nearly so.
Even if you feel you have done some of the topics before, work
steadily through each programme: it will serve as useful revision and fill
in any gaps in your knowledge that you may have.
USEFUL BACKGROUND
INFORMATION
I. Algebraic Identities
(a + bf = a2 + 2ab+b 2 (a + bf = a 3 + 3a 2 b + 3ab 2 + b 3
(a  bf =a 2  Ixib + b 2
(a b)
3
=a 3 ~ 3a b + 3ab  b
2 2 3
(a + bf = a 4 + 4a 3 b + 6a 2 b 2 + 4ab 3 + 6 4
(a  Z>) = a 4  4a 3 * + 6a 2
4
Z>
2
 4ab 3 + b
4
2
a b 2 = (a b) (a + b). a
3
b 3 ab + b 2 )
= (a b) (a
2
2 2
cosec 2
2
(1) sin + cos = 1 ; sec = 1 + tan 2 0; = 1 + cot 2
*
tan a iDB)^ =
(A + =
1
tan
 tan
A + tan B
A tan B
,. _. tan A  tan B
tan (A  B) _
1 + tan A tan B
_ 2tan0
tan na j^^
,
20 
xin
cos ra
= cos^2  sin 2^
2
= l2sin'f
= 2cos.22 ^ 1
2 tan
tan0=
2
1tan^
sin
n
C
.
sin
=
D
.
2 cos
C+D
 sin CD .
cos n,.
C + cos r.
D= ,
2 cos
C+D
2
cos
CD
2
cos
p.
D  cos C = .
2 sin
.
C+ D
 sin
. CD

xiv
(9) a sin + b cos = A sin (0 + a)
a sin b cos = A sin (0  a)
a cos +b sin = A cos (0  a)
fl cos 6 sin = A cos (0 + a)
[A = vV + 6 2
)
where :
(a=tan 1 (0<a<90)
Slope, m = *! = ZiZZi
dx x2 x l
(2) 0>c/e:
2
Centre at origin, radius r: x2 + y 2 = r
(3) Parabola:
Vertex 2
at origin, focus (a, 0): y = Aax
2
Parametric equations: x = at y = 2at ,
xv
(4) Ellipse:
2 2
(5) Hyperbola:
2 2
Centre at origin, foci ( \/a 2 + b 2 0): ,
x
y
p" = 1
Rectangular hyperbola:
2
Centre at origin, vertex/ y, y") : xy = = c 2 where c = r
2
i.e. xy = c
Parametric equations: xct, y = c/t
xvi
Programme 1
COMPLEX NUMBERS
PART1
Programme 1
2
u* auby *u
obtained
the f
_
formula, x = i
6V(6= 4gc)
9V(8156) _ 9V25 9 + 5
X
4 4 " 4
v _4 14
" *" 4
0r
~T
:. x =1 or 35
neither
It is, of course, neither, since + 8 and  8 are the square roots of 64 and
not of (64). In fact, y/( 64) cannot be represented by an ordinary
number, for there is no real number whose square is a negative quantity.
However, 64 = "1 X 64 and therefore we can write
J5
 tx x = 6V(36100) _ 6 V(~64)
2 =u
5x
5* 6x + 5
i
 To To
:' x= " x
= ' 6
* j0 8
'
^JcT
.'.
x = 06+j08 or x = 06j08
Powers of j
Since stands for _ l), us consider some powers of j.
j V( let
j =V(D j =V(i)
2
1
j =i
r=u;2v)i=i.j
= = j
2
f=j
4
j
=0 2 ) 2 =(i) 2 = i
j
4
i
Note especially the last result: j 4 = 1 . Every time a factor j 4 occurs, it can
be replaced by the factor 1 , so that the power of j is reduced to one of
the four results above.
9 4 2
e.g. j =(j ) j
= (l) 2 j = lj=j
j2o 4
)5 =(1)S =1
=(j
30 4
=(l) 7 (l)=l(l)'
a
j =G )'j
15 4 3 3
and j =(j ) J = lH) = j
5
So, in the same way, j
Programme 1
s 4
since j =(j )j
= 1 j=j
42 =
So (i) j
12 =
00 j
(iii)
U =
J
and (iv) If x
2
 6x + 34 = 0, x =
,xl^3 j 5
Complex numbers 7
The result x = 3 + j5 that we obtained, consists of two separate terms, 3
and j5. These terms cannot be combined any further, since the second is
real part ==
2; imaginary part ==
7 (NOTJ7!)
8
Complex numbers have many applications in engineering. To use them,
we must know how to carry out the usual arithmetical operations.
(5+j7) + (3j4)(6j3)=
Programme 1
2+j6
= 2+j6
10 (i)4+j (ii)jl2
(ii) +j5)(5j4)(2j3)
= 3+J55+J4 + 2+J3 (Take care
with signs!)
= (3 5 + 2) +j(5 +4 + 3)
= 0+J12 = j!2
This is very easy then, so long as you remember that the real and the
imaginary parts must be treated quite separately just like x's andy's in
an algebraic expression.
On to frame 11.
j
Complex numbers 1
These are multiplied together in just the same way as you would deter
mine the product (3x + Ay) (2x + 5y).
(iv)
the two outer terms
the two righthand terms
(3 + J4) (2 + j5)
6+j8 +jl5+j 2 20
.2.
"3" 2
6+J2320 (since =l)
14+J23
Ukewise, (4j5)(3 +]2)
12
22 j7
for: (4j5)(3+j2)=12jl5+j8j 2 10
2
= 12J7 + 10 (j =l)
= 22j7
(3+j4)(2j5)(lj2)
= (6+j8jl5j 2 20)(lj2)
= (6j7 + 20)(lj2)
= (26j7)(lj2)
Finish it off.
Programme 1
13
12J59
for: (26j7)(lj2)
= 26j7j52+j 2 14
= 26 j59 14 = 12J59
Note that when we are dealing with complex numbers, the result of our
calculations is also, in general, a complex number.
(5+j8)(5j8)=
14 89
Here it is:
They are identical except for the middle sign in the brackets,
15
i.e. (5+j8) and (5j8)
16
a complex number
since (7j6) (4 + j3) is a product of two complex numbers which are not
conjugate complex numbers.
17
3+j2
because the conjugate of (3  j2) is identical to it, except for the middle
+ j2), and we know that the product of two conjugate com
sign, i.e. (3
plex numbers is always real.
= 9 + 4=13
Example 2 (2 + j7) (2  j7) = 2
2
 G7) 2 = 4  j 2 49
= 4+49 = 53
. . . and so on.
complex numbers.
18
conjugate
(i) (4j3)(4+j3)
(ii) (4+j7)(4j7)
(hi) (fl+j6)(aj6)
(iv) (xiy)(x+jy)
entirely real.
Complex numbers 1
2
(ii) (4+j7)(4j7) = 4 j 7 2 2
=16+49 = 65
 x2  2 2
x1 + y2
(iv) (x ]y) (x + ]y) = y =
(3j5)(3 +j5) = 3 2 j 2 5 2 =9 + 25 = 34
Revision exercise. 20
12
1. Simplify (i) j (ii) j' (iii) j
2. Simplify:
= 12 _ /;4\3 _ 1 3
1.
10 4 2 2 2 1
GO j = (j ) j = i (D=
4 5 3 3
(m) j" = o ) j =j =[T]
7J24
(iv) (5j4)(5+j4)
= 25 j 2 16 = 25 + 16 = 41
(4j3)(4+j3)=16 + 9: 25
All correct? Right. Now turn on to the next frame to continue the
programme.
11
Complex numbers 1
5 J 4 = 5 4=1.67jl.33
J
3 3 3
7j4 r
But how do we manage with '
4+j3
If we could, somehow, convert the denominator into a real number, we
could divide out as in the example above. So our problem is really, how
can we convert (4 + j3) into a completely real denominator and
this is where our last piece of work comes in.
We know that we can convert (4 + j3) into a completely real number
by multiplying it by its c
23
Conjugate i.e. the same complex number but with the opposite sign
nDDDanDnnnannnnDDnnnanDDDDnnnDnnanaDan
But if we multiply the denominator by (4  j3), we must also multiply
the numerator by the same factor.
ifj=0.64jl48
and the job is done.
To one complex number by another, therefore, we multiply
divide
numerator and denominator by the conjugate of the denominator. This
will convert the denominator into a real number and the final step can
then be completed.
4i5
Thus, to simplify . we shall multiply top and bottom by
12
Programme 1
DnnannnnDannnnDDnnDDDnnaQDDDDDnnDnnnnD
If we do that, we get:
= l2j26
3 +j2
Simplify
1J3
When you have done it, move on to the next frame.
25 Result
03+jll
=
3+jll = _
Q 3+jl . ,
1
10
naDnannDDDnDnnnnnDDDannnnDDDnnnnDDDnnD
Now do these in the same way:
4J5 3+j5
(i) (ii)
v " _/
2j 5J3
(2+j3)(lj2)
(iii)
3+j4
When you have worked these, turn on to frame 26 to check your results.
13
jl Complex numbers 1
13 j6 _
26J12
(2+j3)(lj2)_2j+6_8j
(iii)
(3+J4) 3+J4 3+J4
_ (8j)(3j4)
(3+j4)(3j4)
. _24j354_20j35
9+16 25
= 08J14
And now you know how to apply the four rules to complex numbers.
a + \b and c +]d
Then we have
a +}b = c + \d
ac=](db)
In this last statement, the quantity on the lefthand side is entirely real,
while that on the righthand side is entirely imaginary, i.e. a real quantity
equals an imaginary quantity! This seems contradictory and in general it
justcannot be true. But there is one special case for which the statement
can be true. That is when
14
Programme 1
a c=\{d~b)
can be true only if
ac = 0, i.e. a = c
29
a = 6 and b=~3
aDannnanDnnnnDDnnDnnnDannnDnnnaDnnnDDn
Now what about this one?
Well now, following our rule about two equal complex numbers, what
can we say about (a + b) and (a  b)1
15
Complex numbers 1
+ 6 = 7 and ab = 2 30
since the two real parts are equal and the two imaginary parts are equal.
onaannnnannanaaaaaaaaaaaaaaanaaaaaauaa
This gives you two simultaneous equations, from which you
can deter
mine the values of a and 6.
So what are they?
a = 45; 6 = 25
31
For a + 6=7
2a = 9 .'. a = 45
26 = 5 :. 6 = 25
ab = 2
DDDDDDDaDDDDDnDDDDDDDDDDDDDnnDDDDnnnDD
We see then that an equation involving complex numbers leads to
a
pair of simultaneous equations by putting
16
Programme 1
32 Graphical Representation of a
Although we cannot evaluate a complex number
Complex Number
as a real number, we can
represent it diagrammatically, as we shall now see.
33 direction
DnnDDDnaDnnnnnDnDnnnnnnnDnnDDDDDDDDnnn
If we multiply (+3) by the factor (1), we get (3), i.e. the factor (1)
has the effect of turning the
vector through 1
80
180
+3
3 1
2
Multiplying by (1) is equivalent to multiplying by j ,
i.e. by the factor
j twice. Therefore multiplying by a
single factor j have half the
will
effect and rotate the vector through
j3
only o
x
\ l
3 2 1
17
Complex numbers 1
90 34
DDDnDnnDDDDDDnanDaanDnDDnDDDDnna a a a a a a
The factor j always turns a vector through 90
in the positive direction
of measuring angles, i.e. anticlockwise.
If we now multiply j3 by a
further factor j, we get j 2 3,
(3) and the diagram agrees
i.e.
2 1
Result:
35
+3
*, ~
36 Results:
aaauaaoaauDaanannnuauanaaonannnnunnaaa
we must
we now wish to represent 3 + 2 as the sum of two vectors,
If
one
draw them as a chain, the second vector starting where
the first
finishes. ,,, ,
,
(3) 2j
H *!.
I 3+2=5
single vector
The two vectors, 3 and 2, are together equivalent to a
(giving naturally that
drawn from the origin to the end of the final vector
3 + 2 = 5).
Continue
19
Complex numbers 1
z, = 2+j3
20
Programme 1
40 :=5+2=7 Z>
= 2 +3 = 5
:. OP = z = 7+j5
DnDDDOODDDnnDDDDODDODDODDDDDDDDDDDDDDD
So the sum of two vectors on an Argand diagram
is given by the
of the parallelogram of vectors.
41
diagonal
aaaDaannDDnnDDDnnnDnDnnaDaDaaannnnnnaa
How do we do subtraction by similar means? We do this rather craftily
Z\ 2 2 ~Z\ +(Z2)
21
Complex numbers 1
P (zi+r2 )
42
(* 2 )B
1 X
OA = z =4+j2
1
OB =z 2 =2+j3
OC=z 3 = lj6
r = ^(a 2 +b 2 )
6 = tan"
1
^
a = r cos and 6 = r sin
22
Programme 1
(ii) tan0=^=O75
6 = 3652
z = a + ]b = K cos + j sin 0)
z = 2 + j3 = /(cos +j sin 0)
r
i
=4 + 9=13 r = 3606
DDDnDDnDODnnnDnnnDDDnnDDDDDnDDDDnnnDDn
We have special names for the values of r and .
z = a + )b = r(cos 6 + j sin 6)
23
Complex numbers 1
argz = 6812' 46
z = 2 + j5. Then argz = 6 = tan"
1
1 = 6812'
DDDDDnnnnnanDnnannDnnnDnnnnnnnnaaaDD
Warning. In finding 6, there are of course two angles
between 0 and
360, the tangent of which has the value
 We must be careful to use the
angle in the correct quadrant. Always draw a sketch of the vector to
ensure you have the right one.
e.g. Find argz when z =3 j4.
2148'
24
Programme 1
48
2
v/(4 +3 )
= 2 = 5
r r
= 3652'
tan E = 075 /.
= 360 = 3238'
'
z = 40960 +J38680
z = 16904
49 +J36252
sin210 = sin30
= 2(08660j05)
= l732j
Here you are. What about this one?
Express z = 5(cos 140 + j sin 140) in the form a + }b
What do you make it?
25
Complex numbers 1
z =38300 +J32140 50
Here are the details 
cos 140 = cos 40
sin 140 = sin 40
r
2
=4 2 +5 2 = 16 + 25 = 41
51

r = 6403
tan E = 08 :. E = 3840'

e = l4120'
1500J2598
Turn to frame 52.
26
:
Programme 1
= i b
and tan"
53
Exponential Form of a complex number.
X ,x X
cosx
,
i
1 ~2! 4l 6!
27
>
Complex numbers 1
If we now take the series for e* and write j0 in place of x, we get Jl*
J
2! 3! 4! "
'*!?!?
(& 2! 4! )
4.ua
+j(*i; + _
g.. ..)
3! 5!
= cos +j sin
55
The three ways of expressing a complex number are therefore
(i) z=a+]b
(ii) z = r(cos +j sin 0) . . . . Polar form
(iii) z = r.e) e Exponential form
Remember that the exponential form is obtained from the polar form,
(i) the r value is the same in each case.
(ii) the angle is also the same in each case, but in the
exponential form
the angle must be in radians.
So, knowing that, change the polar form 5(cos 60 + j sin 60) into the
exponential form.
28
Programme 1
56
Exponential form 5c
for we have 5(cos60 +jsin60) r=5
6 = 60 =  radians
n
J
3
.'. Exponential form is 5 e
DnDnnnDnnDnnnnaDDnnDDDDnDDnnDanaDnanDn
And now a word about negative angles
e"J fl
=cos(0)+j sin(0)
= cos0 j sinfl
So we have e = cos 6 +
t'i j sin 6
Make a note of
e~J
e = cos 6  j sin 6 these.
e.g. If
z = 642eJ 1 57 
then
lnz = ln642+jl57
= 18594 +J157
29
Complex numbers 1
lnz = ln38jO236 :
13350J0236 58
DnDDDnDnnDnDDDDDDnDDnDnDanDDnnDDDDDnDn
Finally, here is an example of a rather different
kind. Once you have seen
it done, you will be able to deal with others of this
kind. Here it is.
Express e^ 4
in the form a + ]b
e'J^4 as e^ 4
=e
U _j
>i}
(1

V2 j)
59
This brings us to the end of this programme, except
for the test
exercise. Before you do that, read down the Revision Sheet that follows
in the next frame and revise any points on
which you are not completely
sure
Then turn on and work through the test exercise: you will find the
questions quite straightforward and easy.
30
Programme 1
60 Revision Summary
1 Powers off
2 3 4
i
= V(D, J
=i> 3 =n, i = i
2. Complex numbers
z =a + \b
a = real part
b= imaginary part
2 2
(a+)b)(a]b)=a +b
>"'/IT
lb
= r(cos0 +j sin0)
^e
It X
= r\
a *1
r = V(a 2 +& 2 ); = tairl
{l)
also a=rcos6; b=r sin 9
31
Complex numbers 1
Test Exercise  I Kl
3 12 14
1. Simplify (i)j , (ii)f, (iii)j (iv) .
, j
0) (4j7)(2+j3) (ii)(l+j) 2
(* + >0+j(*.>') = 148+j62
7. Express z = e
1+J7r/2
in the forma +j&.
Now you are ready to start Part 2 of the work on complex numbers.
32
.
Programme 1
Further Problems  1
3. If z = x
2+j3
: +
1
r~
]i
, express z in the form a + ]b.
4. if z = ^4, find the real and imaginary parts of the complex number
z + l
z
2 5
result in the
5. Simplify (2 + j5) + ^jp j(4j6), expressing the
forma + ]b.
6 If z,1 = 2 + i,
J
z, = 2 + i4 and = Z +
23 Z
t 2
, evaluate z 3 in the form
+ b) + )(a b) = (2+ j5) + j(2  j3), find the values of a and
2 b.
9. If (a
)x _ 3x + j4
1 + )y x + 3y
c+]d
33
Complex numbers 1
a _c
real then and (ii) if z is entirely imaginary then = ^.
12. Given that (a + b) + ](a ~b) = (l + j) 2 + j(2 + j), obtain the values of
a and b.
1+ J 3 >*
15. If* is real, show that (2 + j)e< + (2  j) e*
1 "* 3 * is also real.
17. Uzx+iy, where* and y are real, and if the real part of
(z +l)/(z + j) is equal to 1 show that the point z lies on a straight
,
19. tf
R i
+ )jL
]f3
^
""";
*4
^2
"
. where ^i,^2,R3,^4,co,Z.andCarereal,
j
<3c
show that
CR2R3
=
a 2 C 2 Rl + 1
34
Programme 2
COMPLEX NUMBERS
PART 2
Programme 2
Introduction
1
In> Part 1 of this programme on Complex Numbers, we discovered how to
r
2
=a 2 + b
2 .'.
r=y/(a 2 + b 2 )
To be sure that you have taken the correct value of 6 always , DRAW A
SKETCH DIAGRAM to see which quadrant the vector is in.
aDDnnnDannnnDnannaDnDDannQnDDDnDDnDnan
Right. Just by way of revision and as a warming up exercise, do the
following:
Do not forget the sketch diagram. It ensures that you get the correct value
for0.
When you have finished, and not before, turn on to frame 3 to check your
result.
37
Complex numbers 2
Result:
D
13(cos337 23'+jsin33723')
r^
B 12 r
2
= 12 2 + 5
2
= 144+25 = 169
^^
r=13
1
:.
15
j r^V^^1
Y
z UnE=Y2 =04167 '
E = 223T
In this case, 8 = 360 ~E = 360  2237' /. 8 = 33723'
aDDDDDDDDDDnDOnDDDPDDDOnDaDDQDOnaDDDDn
Did you get that right? Here is one more, done in just the same way.
Result:
2 = 6403(cos 21840' + j sin 21840') 4
Here is the working: check yours.
Y
r
2
=5 2 +42 =25 + 16 = 41
 r = V41 =6403
tan E=j= . E = 3840'
In this case, 8 = 180 + = 21840'
So z = 5 j4 = 6403(cos 21840' + sin 21840')
j
DnDanDDGDnDDDDDDDnnannDnDnnnaDDDaDannD
Since every complexnumber in polar form is of the same shape,
i.e.r(cos 8 + j sin 8) and differs from another complex
number simply by
the values of r and 8, we have a shorthand method
of quoting the result
in polar form. Do you remember what
it is? The shorthand way of writing
the result above, i.e. 6403(cos 21840' + sin 21840')
j is
38
Programme 2
6403 l21840'
nnnuuDDononnnnaoaaaaaaaaanaoaaauaaaaaa
Correct. Likewise:
5 5 32215'
5 7 2(cos 322 1 ' +j sin 322 1 ') is written 572
c
5(cosl05+jsinl05) " " 5 105
They arecomplex numbers in polar form. They are all the same
all
r and e
DDDDDnnDnDannnDDnnaDnnnnnnDDDDDnDDDDnD
Now let us consider the following example.
r=5
tan = = 075 /. E = 3652'
J
6 = 360  3652' = 3238'
4j3 = 5(cos3238'+jsin3238')
or in shortened form, z
39
Complex numbers 2
z = 5 3238
DnnanDDnnDnDDnnnnnDannnnnDnDnnDDnDnan
In this last example, we have
z = 5(cos3238'+jsin3238')
But the direction of the vector,
measured from OX, could be given
as 3652', the minus sign show
ing that we are measuring the angle
in the opposite sense from the
usual positive direction.
z = 5(cos3652'jsin3652')
i.e. very much like the polar form but
with a minus sign in the middle.
This comes about whenever we use negative angles.
In the same way z = 4(cos 250 + j sin 250) = 4(cos [110] +
j sin [1 10])
,
= 4( )
Ex. 1 z = 3(cos230+jsin230)
= 3(cos 130 j sin 130).
40
Programme 2
z = 6(cos50jsin50)
DDDDnDDanaDDonDDDnnnDnDaDDnDanDDDDDnna
One moment we agreed that the minus sign comes about by the
ago,
given in this way
use of negative angles. To convert a complex number
'+' in the middle, we simply
back into proper polar form, i.e. with a
negative sign in
work back the way we came. A complex number with a
middle equivalent to the same complex number with a positive
the is
e.g. z = 4(cos 30
 j sin 30)
= 4(cos[30] +jsin[30])
= 4(cos 330 + j sin 330) and we are back in the proper polar form.
You do this one. Convert z = 5(cos 40 j sin 40) into proper polar form.
10
z = 5(cos320+jsin320)
DDnDDDDDDaDnnDDanDDDnananDDDDaannanDan
Here another for you to do.
is
polar form.
Express z = 4(cos 100 j sin 100) in proper
Do not forget, it all depends on the use of negative angles.
41
Complex numbers 2
z = 4(cos260+jsin260) 11
for z = 4(cos 100 j sin 100) = 4(cos [100] + sin [100])
j
nnDDODDDDDDDDODDDDDDDDDDDDDDDDDDDDDDDD
We ought to see how this modified polar form affects our shorthand
notation.
Remember, 5(cos 60 + j sin 60) is written 5 60 1
A 60
using the negative angle we use a
V/60 different symbol i.e. 5 [60
becomes 5 ["60
5
\
5 160"
Similarly, 3(cos 45 j sin 45) = 3 I
45 =
3 [45
12
DnnDnnnnnDaDDDDnnnDDnnnanDnDDDDDnDDnnn
This is easy to remember,
for the sign ... Q resembles the first quadrant and indicates
measuring angles, \ i.e. in the positive direction,
42
Programme 2
13
(0 120
DDDnnDDDnnaaDnananaDDnDDnDannanannnnDn
The polar form at first sight seems to be a complicated way of
complex number. However it is very useful as we shall
representing a see.
+ cos^i sin0 2 )]
Now the brackets (cos 0, cos0 2 sin0 x sin B 2 ) and (sin 0j cos0 2
+ cos0! sin0 2 )
ought to ring a bell. What are they?
DnDaannDnaaDDaDDnnnnaaaanonDDnDnnaDDDD
In that case, ZiZ 2 r x r 2 [cos(0i + 2 )+jsin(0! + 2 )]
we get
43
)
Complex numbers 2
= 8(cosl40+jsin 140)
(ii) a(cos 6 + j sin d) X b(cos +j
) sin
= 15(cos30+jsin30)
Have you got it? No matter what the angles are, all we do is
16
12(cos55+jsin55)
oaaaanaoannaanaannnanannanaanaaanaaana
Now let us see if we can discover a similar set of rules for Division.
44
Programme 2
17  j4
the conjugate of the denominator i.e. 3
anDDDaaoooDQcmDananciQDQnnnnoDnnnonDDDD
Right. Then let us do the same thing with
/iCcosf?! +j sin^i)
r 2 (cos0 2 +j sin0 2 )
DDDDDDDnDDDDDDDDDDDDDDDDnDDaDnnDDDDaDD
That is correct.
eg. ) ..,, . .
l0 . 3(cos31 +jsin31)
2(cos 41 + j sin 41 )
then(i)z!Z 2 = and(ii)~ =
z2
45
Complex numbers 2
19
z x z2 =32(cos88 + jsin88)
^=2(cos42+jsin42)
DDDDDPODDDOaODDDDDDDDDDDDDDDDDDODDDDDD
Of course, we can combine the rules in a single example.
20
Result:
o
24(cos90 +jsin90)
DDOOODODDDOOOaDDDDDDDDDDDDDODDDDDDDDDD
Now what about a few revision examples on the work we have done so
far?
46
.
Programme 2
21 Revision Exercise
Work all these questions and then turn on to frame 22 and check your
results.
10(cosl26+jsinl26)
Cll)
2(cos72 +jsin72)
4. Express in the forma + \b,
(i) 2(cos30+jsin30)
(ii) 5(cos 57 j sin 57)
Solutions are on frame 22. Turn on and see how you have fared.
47
Complex numbers 2
Solutions
22
1. r
2
=2 2 +4 2 =4+ 16 = 20
..
r = 4472
8 tan E = 05 :. = 2634'
x, 4 o~^< '
6 = 15326'
z =4 + j2 = 4472(cos 15326' + j sin 15326')
2(cos72+jsin72)
= 5(cos54+jsin54)
4. (i) 2(cos30+jsin30)
= 2(0866 +j05) = 1732 +j
(ii) 5(cos57jsin57)
= 5(05446 J08387)
= 27230 J41935
48
Programme 2
then z t z2 =r 1 r 2 [cos(6i
+ 2 )+jsin(0i + 2 )]
24
ZiZ 2 z 3 =r 1 r 2 r 3 [cos(e 1
+ 2 +0 3 ) +j sin^! +0 2 + 3 )]
for in multiplication, we multiply the moduli and add the arguments.
DDDDDODDnDDDDDDOOODDDDDDDnDDO^nDDnDDQD
Now suppose that z x z 2) z 3 are , all alike and that each is equal to
z = /(cos 6 + j sin 6). Then the result above becomes
49
.
Complex numbers 2
and so on.
In general, then, we can say
[r(cos0+j sine)]"
26
[/(cos 5 +j sin^)]" = r"(cosn6+j sinnd)
nnnDDDDDDDanDDDDDDDDDnDnnnDnnaDD a a
[2(cos37+jsin37)] 4 =
50
Programme 2
27 16(cosl48+jsinl48)
DnnnnanDDDDDDDDnDDDDnnnnDaDDDnnDnnDDnn
This is where the polar form really comes into its own! For DeMoivre's
theorem also applies when we are raising the complex number to a
fractional power, i.e. when we are finding the roots of a complex number,
e.g. To find the square root of z = 4(cos 70 + j sin 70).
z" = /"(cos
28 z = r(cos 6 + j sin0), then nd + j sin nQ)
DnDQaDDDaDQnDnDnnDDDDannDnnDnanDDnDann
Look again at finding a root of a complex number,. Let us find the cube
Y
rootofz = 8(cos 120 +j sin 120).
Here is the given complex number
shown on an Argand diagram.
z = 8 1
120
Of course, we could say that 9 was
'1 revolution + 120': the vector
would still be in the same position,
or, for that matter, (2 revs. + 120),
(3 revs. + 120), etc.
51
Complex numbers 2
rU 8 4 120 c
j or 8^
i 480
T ci
or 83
840
5 _i
or 83
1200
29
3
DDDDDDDnDnnnnnnDDDanDDDnDDnDDnDnnnanna
If we simplify these, we get
160
we see we have three quite different results for the cube roots of z and
also that the fourth diagram is a repetition of the first. Any subsequent
calculations merely repeat these three positions.
Here they are: The cube roots of z = 8(cos 120 + j sin 120).
30
z, = 2 I
40
z2 = 2 160
?3 = 2 1280
"s
aDDDannDDDDDDnDnaDnnnDaDDnQDnnDnnDDDDn
We see, therefore, that there are 3
cube roots of a complex number.
Also, if you consider the you see that the 3 roots are equally
angles,
spaced round the diagram, any two adjacent vectors being separated
by degrees.
52
Programme 2
31 120
DnDnDnnnaDDnnnDDDDDDDDnnnnDnnDnnnnaDao
32
Example. To find the three cube roots of z = 5(cos 225 + j sin 225)
x i 225 225
The first root is given by z x = z 3 = 5^(cos^5 + j sin ^ )
= l71(cos75+jsin75)
zj = 171 
75
We know that the other cube roots are the same size (modulus), i.e. 1 71
zi = 171 75
I
z 2 = 171 [195
z 3 = l71 315
53
Complex numbers 2
We
number
find any roots of a
in the
complex
same way.
33
(i) Apply DeMoivre's theorem to
find the first of the n roots,
(ii) The other roots will then be
distributed round the diagram
at regular intervals of
n
A complex number, therefore, has
i.e. 180
4 fourth roots,
360
i.e. 90
34
DDDnnnDnnDDDnDDDDDnanannnnnDaDDannnDDD
And now: To find the 5 fifth roots of 1 2 j
300
z=12[300_ .
2, = 12*"
^=12Tl60
z4 = 1 644 1
276 , z5 = 1 644 [348
Sketch them on an Argand diagram, as before.
54
Programme 2
35 ">
Jz 1
X^
z2 i *1 = 1644  60
*2 = 644 
136
X, ^ *3 1644204
*5 *4 = 1644 I
2 76
*3
'5 = 1 644 348
1
>
L
Principal root. Although there are 5 fifth roots of a complex number, we
are sometimes asked to find the principal root. This is always the root
whose vector is nearest to the positive OX axis.
In some cases, it may be the first root. In others, it may be the last
root. The only test is to see which root is nearest to the positive OX axis.
In the example above, the principal root is therefore
DDDDnDnnannDDDDnDnnaDanDnnnDDDDnDDnnaD
lA 20
zx = 1 627 
20
55
Complex numbers 2
Y
37
\ J
Zi = 1627 J 20
2z = 1627 J110
*i 0\ X *3 = 1 627 200
z4 = 1627 [290
aDDDDnnDDDnnDnnannDDDDDaDnnnnnanDDDaDa
And in this example, the principal fourth root is
38
DDDnnDnnnnDnanannnDnnnDnnnnnDnnnnDnnDa
Now you can do one entirely on your own. Here it is.
Find the three cube roots of 6(cos 240 + sin 240).
j Represent them
on an Argand diagram and indicate which is the principal
cube root.
When you have finished it, turn on to frame 39 and check your results.
56
Programme 2
39 Result:
>
4**
z, = 1 817 I
80
j
= 1817 200
K8o zz
z3 = 1817 320
1
N^O"
*2
Principal root : Zj = 817 320
*3
<l
u
DDnnnnnnnQnnannnoanaonQO
Here is the working.
240"
z = 6 I
240 Zi =63 1817 80
5zrr\
The principal root is the root nearest to the positive OX axis. In this case,
then, the principal root is z3 = 1817 ]
320
We have
cos 3d + j sin 30 = (cos 8 + j sin dy
= (c + js)
3
where c = cos 8
s = sin d
Now expand this by the binomial series like (a + b) 3 so that
cos 30 + j sin 30 =
57
Complex numbers 2
C
3 2
+j3c s3cs 2 js 3 41
for: cos 30 + j sin 30 = c 3 + 3c 2 Qs) + 3c(js)
2
+ (js)
3
3 2 2 3
= c +j3c s3cs js sincej
2
=l
3 3
= (c 2
3cs )+j(3c s~s 3 ) 2
j =i
Now, equating real parts and imaginary parts, we get
cos 30 
and sin 30 =
cos 3d = cos 3
sin 30 = 3 cos 2
 3 cos
sin
sin
 sin 3
2
42
If we wish, we can replace sin 2 by (1  2
cos 0)
2
and cos by (1 sin 2 0)
so that we could write the results above as
cos 30 = (all in terms of cos 0)
sin 30 = (all in terms of sin 0)
58
Programme 2
4
Working: cos 40 + j sin 40 = (cos + j sin 0)
4
= (c + js)
4 4
= c + 4c
3
(js) + 6c 2 Cs) 2 + 4c(js) 3 + (js)
4 4
= c + j4c 3
 6c 2 s 2  j4cs 3 + s
s
4  6c s + s4 ) + j(4c 3 s  4cs 3 )
2 2
=
(c
Equating real parts:
4  6c 2 s 2 + 4
cos 46 = c s
c
4
6c (lc 2 ) + (lc2\2
2 2 2
)
4
c  6c
2
+ 6c 4 + 1  2c 2 + c
4
4 2
8c 8c + 1
= 8 cos 4 08cos 2 + 1
x
then =z = cos 6 j sin i
.'.
z + =
z
2 cos 6 and z z
= j 2 sin 6
.'.
z" += = 2 cos n# and z" = = j 2 sin nd
z" z"
z" + t.
z"
= 2 COS H0 z" ^ = j 2 sin0
Mzfe a note of these results in your record book. Then turn on and we
will see how we use them.
59
Complex numbers 2
3
Ex. 1. T5 expand cos
46
From our results, z + = 2 cos i
z
L
/. (2cos0) 3 = (z+) 3
^ 1 ^ 1
3
= z + 3 z + 3I+ I3
z z*
Now here is the trick: we rewrite this, collecting the terms up in pairs
from the two extreme ends, thus 
(2cos0)^3_,3
3
=(z 3 +p)
.
K + ~3(z+)
.U .
z + =1
and z" +
z+ =2 cos0;z 3
+, =2 cos 30 47
3
.'.
(2 cos 0) = 2 cos 30 + 3.2 cos 6
3
8 cos = 2 cos 30 + 6 cos
4cos 3 = cos 30 + 3 cos0
,
3 1
cos 0=(cos30 + 3cos0)
When you have obtained a result, check it with the next frame.
60
Programme 2
48 4  464 cos
sin 6 = [cos 26 + 3]
for, we have:
z  = j 2 sin 6 z n ;
 ^ = 2
i sin nd
4
Q 2 sin 0) = (z ) 4
1 \ 1
"liM?)^!?)*
(2 4 tj)* ! +p) + 6
Now z M +  = 2 cos nd
4
1 6 sin 6 = 2 cos 40  4.2 cos 20 + 6
.'.
sin
4
= [cos 46 4 cos 26 +3]
They are all done the same way: once you know the trick, the rest is
easy.
Loci Problems
49 We
are sometimes required to find the locus of a point which moves
in theArgand diagram according to some stated condition. Before we
work through one or two examples of this kind, let us just revise a
couple of useful points.
You will remember that when we were representing a complex
number in polar form, i.e., z =a + )b =r(cos 6 + j sin d), we said that
(i) r is called the modulus of z and is written 'mod z' or 
z\ and
(ii) 6" " " argument of z " " " 'arg z'
so that 
z 
= \J(a
2
+ b 2 ) and arg z = tan" 1
Similarly, if z = x + 'yy, then \z\ =
and argz=
61
.
Complex numbers 2
2 2
\z\ = V(* + y ) and arg z = tan" 1 fL\
If 2 =x + jy,
.'.
X 4
+ y2 =25
Locus  z ]
= 5
i.e. x 2 +y 2 = 25
62
Programme 2
z+
Jj Ex. 3. Uz =x +jy, find the equation of the locus
1
Since z =x + jy,
z+l=x+]y+l=(x+l)+]y =r \9 l 1
z+ 1
z  1 r2 \
e2 r2
z+ 1
J= Ifil +y 2
y/[(x + l)
2
= ]
r2 \z 2 \
^/[{xlf+y 2 ]
= 2
Wl(x\) 2 +y 2 ]
2
+y 2
(^ + i) =
" (xl) +y 2 2
All that now remains is to multiply across by the denominator and tidy
up the result. So finish it off in its simplest form.
53 We had
(x+ l)
2
+y 2
(xiy^ry
2
So therefore (x+l) +y 2 =4J(xl) 2 +.
2 2 2
x + 2xy + 1 +y = 4(x 2x+ 1 +y 2 )
= 4x 2  8x + 4 + 4y 2
2 2
:. 3x  \0x + 3 + 3y ;
Although this takes longer to write out than either of the first two
examples, the basic principle is the same. The given condition must be
a function of either the modulus or the argument.
63
1
Complex numbers 2
.'
By DeMoivre's theorem, z
2
= r
2
[20
arg(z
2
) = 20=?
".
tan 29= tan (7) =1
4
2 tan 6>
=
1 tan 2
2
.'.
2tan0 = tan 1
But tan d =L
X
^XX
= z! 
,2
2
2xy =y x 2
:.
y =x
2 2
+ 2xy
In that example, the given condition was a function of the argument.
Here is one for you to do:
lfz=x+]y, find the equation of the locus arg(z+ 1)=
Do it carefully; then check with the next frame.
V
=tan^ = V3
x+ ] 3
y=\/3(x+l)
And that is all there is to that.
Now do this one. You will have no trouble with it.
64
Programme 2
This brings us to the end of this programme, except for the final test
exercise. Before you work through it, read down the Revision Sheet
(frame 57), just to refresh your memory of what we have covered in this
programme.
So on now to frame 5 7.
65
Complex numbers 2
Revision Sheet
r  mod z = I
z 
= y/a 2 + b 2
'
= arg z = tan"
,lf*
(f)
2. Negative angles
z=r(cos [0] +j sin [6])
If z i =>"! z2 = r2
jj^; j^2_
then ^ 2
=/,/
2 I
0, + e2
z2 r2 LJ i
4. DeMoivre's theorem
If z = r (cos + sin 0), then z n = /"(cos n0 +
j
j sin nd)
5. Exponential form of a complex number
z = rei e :. In z = In r+jd
7. Loci problems
TTzar 's /Y/ A^o w .yow are razc?y for the Test Exercise on Frame 58.
66
Programme 2
Jj Q Test Exercise II
3
4. If z = 2(cos 25 + j sin 25), find z in polar form.
5. Find the three cube roots of 8 (cos 264 + j sin 264) and state which
of them is the principal cube root. Show all three roots on an Argand
diagram.
4
7. Express cos in terms of cosines of multiples of .
67
Complex numbers 2
Further ProblemsII
1. If z =x + iy, where x and y are real, find the values of x and y when
3z _3z
+ = _4_
J
J J 3j
6. Determine the fourth roots of 16, giving the results in the form
a +)b.
68
. .
Programme 2
12. Solve the equation z 2 +2(1 + j)z + 2 = 0, giving each result in the
form a + \b, with a and b correct to 2 places of decimals.
1
13. Express e ^/ 2 in the form a + jb
6
15. Express sin x as a series of terms which are cosines of angles that
are multiples of x.
17. If z =x + jy, show that the locus are] ^1= is a circle. Find its
z ii 6
\
centre and radius.
2
z+j2 +zj2J 40
z + 2 Z + 2) 77
(i) = 3 (ii) arg
(i) if 
Zj+z 2 ]
=z z 2 1 
,
the difference of the arguments of
Z\ and z 2 is .
69
Complex numbers 2
70
Programme 3
HYPERBOLIC FUNCTIONS
Programme 3
Introduction
When you were first introduced to trigonometry, it is almost certain that
you defined the trig, ratios ' sine, cosine and tangent  as ratios between
the sides of a rightangled triangle. You were then able, with the help of
trig, tables, to apply these new ideas from the start to solve simple right
angled triangle problems and away you went.
You could, however, have started in quite a different way. If a circle
of unit radius is drawn and various constructions made from an external
point, the lengths of the lines so formed can be defined as the sine,
cosine and tangent of one of the angles in the figure. In fact, trig, func
tions are sometimes referred to as 'circular functions'.
This would be a geometrical approach and would lead in due course
to all the results we already know in trigonometry. But, in fact, you did
not start thatway, for it is more convenient to talk about rightangled
triangles and simple practical applications.
Now if the same set of constructions is made with a hyperbola instead
of a circle, the lengths of the lines now formed can similarly be called the
hyperbolic sine, hyperbolic cosine and hyperbolic tangent of a particular
angle in the figure, and, as we might expect, all these hyperbolic functions
behave very much as trig, functions (or circular functions) do.
This parallel quality is an interesting fact and important, as you will
see later for we shall certainly refer to it again. But, having made the
point, we can say this: that just as the trig, ratios were not in practice
defined geometrically from the circle, so the hyperbolic functions are not
in practice defined geometrically from the hyperbola. In fact, the defini
tions we are going to use have apparently no connection with the hyper
bola at all.
So now the scene is set. Turn on to Frame 1 and start the programme.
73
Hyperbolic Functions
You may remember that of the many functions that can be expressed
as a series of powers of x, a common one is e x ,
1
x4
2
1 . ,* *3
2! 3! 4!
If we replace x by x, we get
2! 3! 4!
e q~x
x = hyperbolic sine of x
sinh*
ey  y
So, in the same way,  e
would be written as
sinh y
DDnonDOODDOOODDDDDDOODOOQODODODOQDDODQ
In much the same way, we have two other definitions:
( m ) x + x  hyperbolic tangent of x
e
sinh* =
74
Programme 3
sinh X =
noooanQaoaaannaooDDoaaoonnnQanaoaaaaaQ
Here they are together so that you can compare them.
= e*
e x
sinh x
2
x + x
e e~
cosh x =
2
e
x
e x
tanh x =
e
x + ex
Make a copy of these in your record book for future reference when
necessary.
sinh x '
cosh x = 
tanhx :
DDDnnnanDnDDnDnDnnnnDDDnDDDDDDnDDDDnnn
x x
We started the programme by referring to e and e' as series of
X
2
L
X
3
X4
e* = l
If we subtract, we get
+ +
3 s
2x 2x
2x
3! 5!
Divide by 2
"
x x
=sinh x =x + ^T + 5T
+
2
x we get a similar
(ii) If we add the series for e* and e~ , result.
What is it?
75
:
Hyperbolic Functions
X2 x4 X6
coshx=1+ + + ~6! +
2! 4!
DnDn.nnDnDDnnnannnDQnnapnDDnnDDDDnaDDDn
For we have
'x
=  x+ x +x
e l
jr i i'
x x
cosh* = 1 + ^7+rr+
Move on to Frame 6.
So we have:
sinhx=^
X
+I]+
x^
_+
_ + x^ ...
cosh x=l + ! + 5l +
^+ ...
Note: All terms positive: sinh x has all the odd powers,
coshx has all the even powers.
We cannot easily get a series for tanh x by this process, so we will leave
that one to some other time.
Make a note of these two series in your record book. Then, cover up
what you have done so far and see if you can write down the definitions
of:
76
Programme 3
DnnnnDnannnnnDnnnnnnnDDnanoDDanDnnDaDn
Graphs of Hyperbolic Functions
We shall get to know quite a lot about these hyperbolic functions if we
sketch the graphs of these functions. Since they depend on the values of
x
e* and e we had better just refresh our memories of what these graphs
,
look like.
x x
y= e and y  e cross the >>axis
at the pointy = 1 (e = 1). Each
graph then approaches the xaxis
as an asymptote, getting nearer
and nearer to it as it goes away to
infinity in each direction, without
actually crossing it.
x x
are positive for all values of x
8 e and e
DDDnDDDnaDDnDDDDDDDnnnnnDnnnnnDnDannnD
At any value of x, e.g. x = x iy
x v
e
cosh x = ,
i.e. the value of
77
Hyperbolic Functions
Here it is:
y = cosh x
X1 ~~ x _ n
We see from the graph of y = cosh x that:
(i) cosh = 1
(ii) the value of cosh* is never less than 1
cosh(x) = coshx
(iv) for any given value of coshx, there are two
values of x, equally
spaced about the origin, i.e. x = a.
Now let us see about the graph of y = sinh x in the same sort of
way.
3* D'x
sinha:= 10
x  ex
e
sinh x =
2
On the diagram,
x
CAe
x
CB =e
{ = ex e x
_e* e x
BP
78
Programme 3
y= sinh x
.* _ .x
(i) sinh =
(ii) sinhx can have all values from  to +
sinh(x) = sinh x
(iv) for a given value of sinh x, there is only one real value of x.
y = cosh X
12
y= sinhx
as x increases
i.e. asx^, sinh x  cosh x
79
Hyperbolic Functions
(iii) =
tanh(x) tanh x
(iv) as** 00 , tanhx^1
as x  , tanh x > 1
Finally, let us now sketch all three graphs on one diagram so that we can
compare them and distinguish between them.
80
Programme 3
I J) Revision Exercise
Fill in the following
x
e +e
= coshx
00 = tanhx
(iii) = sinhx
>
<
(iv)
^ y = tcinh x
(v)
y = cosh x
(vi)
y = sinh x
82
Programme 3
x 1
"
275 1
'
275 We now
Now sinh* = \(e  e*) .\ sinh 1275 = Ke  e' ).
1 275
1 27S
Note that when we have done that, e"
' "
is
have to evaluate e .
Let A = e
1 '
275 :. In A= 1 275 and from tables of natural logs we
now find the number whose log is 1275.
So e
1 "
275 =3579 and e"
1 "
275
=^ = 02794
cosh 2156.
In thesame way, you now find the value of
When finished, move on to frame 18.
DDDDnDDDDDDDDDnDDDDDnnnnDDDDDDDDnnD
Here is the working:
2' 1S6 2 "
156
Example 2. cosh 2156 =\(e + e' )
Let A = e 2 156
"
.'.
In A= 2156 :. A= 8637 and  = 01158
83
Hyperbolic Functions
 
e r e
So, evaluating sinh, cosh and tanh is easy enough and depends mainly
on being able to evaluate e*, where k is a given number and that is most
easily done by using natural logs as we have seen.
..
e*  \ = 2950
2
&x = 295V(295 +4) = 295 V(8703 + 4)
2 2
=
295 y/l 2703
=
295+3564
2 2
_ 6514
=
j or
0614 _
7y = 3257 or 0307
But e* is always positive for real values of x. Therefore the only real solu
tion is given by e* = 3257.
:. x= In 3257=11809
..
x= 11809
Exercise 2.
84
Programme 3
1
cosh 2364 = 1507
21
naaannnnDnannaDDnaannDDnDaDnnnnaaaaDDD
1 1
For: To evaluate costf 2364, let x = cosh"" 2364
.'.
cosh* = 2364
+ e'
= 2364 :. e* + Y = 4728
x 2 =0
(e 4728(e*)+l
^
)
e* = 4728V(22364) Vlg .
36 =
22 tanh x = 
DDDDnnDnnaDDDanDDDDannnnanDDnn Dnnnannn
That being so, we can now evaluate tanh" 0623.
Let
1
x = tanh" 0623 .' tanh x = 0623
e c
=06^
x + x
e e~
..
e
* e*=0623(e*+ex )
X
:. (1 0623) e*=(l + 0623) e
x
0377 e* = 1623 e'
1623
e*
02103
( s x\2
1623 T5763
"0377 2) 06340
:. e
x = 2075 03170
:. x= In 2075 = 07299
tanh" 0623 = 0730
1
:.
1
Now one for you to do on your own. Evaluate sinrf 05
85
Hyperbolic Functions
1
sinh" 05 = 04810
23
aQnaaonanaanannaaaaQDqnoQQDQDQDOODODOn
Check your working.
1
Let x = sinh" 05 .'. sinh x = 05
p
e
x _ p
e
x
1
= 05 :. e* = 1
/. (e*) 2  1 = e*
2
(e*) (e*)l=0
a *_ 1V(1+4)_1V5
2
32361 12361
or
2 "* 2
= 16181 or 06181
e* =06181
..
x= ln 16181 =04810
gives no real
1
sinh" 05 = 04810 value of x.
1
And just one more! Evaluate tanh" 075.
24
1
tanhT 075 = 09731
QnQQnnnnonnaonannnnQnnooonoQannaanaQn
Let x= tanh" 075
1
.'. tanh x = 075
= 075
x  x =
e t 075(e* + e"*)
(l075)e*=(l + 075)e"*
025 e* = l75e*
x 175
(e Y = 7
025
e* = V7 = 26458
But remember that e x cannot be negative for real values of x.
Therefore e* = 26458 is the only real solution.
.'.
x = In 26458 = 09731
075 = 09731
1
tanh"
86

Programme 3
,1l 1, fl +x
y = t<mh x=$\n\^^
So that tanlf
1
05=1 In
j^)
= iln3=i(10986) =05493
tanh" (06)
1
=06932
26 nnDDDnnDnnannDnDnnnaDnnnnnDnDDnnnnDnnn
For, tanh"
1
* = jln{ t~z
= 06932
_1
Now, same way, find an expression for sinh x.
in the
1
Start off by saying: Lety = sinrf * x = sinny .'.
y y
&  e^ = x e y  y = 2x
.e y  = 2x
y
(e y )
2
2x(e y )l=0 Now finish it off.
87
Hyperbolic Functions
Result: sinh
1
jc=ln{x+V(^ 2 +l)} 97
nnnnnnnDDnnDDDDDnnDnDDnnnnDnnDnanDDnnD
For (e>) 2 2x(e>)l=0
2
2x V(4x 2 + 4) = 2x 2y/(x + 1)
ey =
^ 2
= xV(* 2 + 1)
ey =x+yj(x 2 + 1) or e y =xsj(x 2 + 1)
y= sinh"
1
* = \n{x + y/(x 2 +1)}
1
Finally, let us find the general expression for cosh x.
Let
1
y = cosh x
,
.'.
x = cosh y =?
ey +
Q y 28
ey L = 2x 2
 2x{t y ) +1=0
e^ W<y
:. +
(e^)
4)
= *V(*'l)
:. ey =x + 2
V(*  1) and ey = xy/(x 2  1)
2
 = 
X+V(* "I)
775
i
.
x\J(x 2
}L>%
2
\)
X\/(x l) ~i
e^W^Oande
c y = x + s/(x
2  1)
^^^
or {x + y/(x
2
 l)}"
1
..
2
y = \n{x+^(x ])} or \n{x+^(x 2 \)}
x = ln{x + V(* 2 ~ 0}
.".
cosh"
1
Notice that the plus and minus signs give two results which are symmetri
cal about thejyaxis (agreeing with the graph of y = coshx).
88
Programme 3
* = ln{* + V(* 2 ~
1
cosh" 1)1
tanh"
iir^
1
*= 5 In I
Add these to your list in your record book. They will be useful.
Compare the first two carefully, for they are very nearly alike. Note
also that sinh"
1
x has only one value,
(j)
1
(ii) cosh" x has two values.
Hyperbolic Identities
30 There is no need to recoil in horror. You will see before long that we
have an easy way of doing these. First of all, let us consider one or two
relationshi ?s based on the basic definitions.
(1) The first set are really definitions themselves. Like the trig, ratios,
we have reciprocal hyperbolic functions:
These remind us, once again, how like trig, functions these hyperbolic
functions are.
89
: : .
Hyperbolic Functions
 x x
sinh* e* e e* + e
(2) Let us consider
cosh* 2 ' 2
31
e* 
= x
e~*
=
x
tanh *
e + e'
cosh* \ sin 6 .
tan 9 = 2 I
cos 9)
x _x
(3) Cosh x = \{e + e"*); sinh x = ^(e*  e )
' cosh * 2
 sinh
2
* = 1
2 2
Jin trig., we have cos + sin = 1 , so there is a difference in

( sign here. )
On to frame 32.
Divide by cosh
2
* : 1 r= = r^* 32
cosrr* cosh
.'.
1  tanh 2 * = sech *
2
.'. sech
2
*= 1  tanh 2 *
.*.
cosech 2 * = coth 2 *  1
2 2
I In trig., we have cosec = 1 + cot , so there is a sign difference I
I here too. J
90
Programme 3

cosh x + sinh * = e* and cosh x sinh x= e~
(0
(ii)
2*
cosh * 2 sinh x cosh x + sinh * = e"
2 2
g
2X _ e
2X
2 sinh x cosh x = = sinh 2x
.'.
sinh 2* = 2 sinh x cosh x
If however we add the two lines together, we get ....
2 2 2*
35 2(cosh * + sinh *) = e + e
2 2 = cosh 2x
.'.
cosh * + sinh *
2 2
.'.
cosh 2x = cosh * + sinh *
.'.
cosh 2
* = 1 + sinh 2 *
Or we could say
2
cosh *  1 = sinh 2 *
.'.
2 2
cosh 2* = cosh * + (cosh *  1)
.'.
cosh 2* = 2 cosh *
2 1
Now we will collect all these hyperbolic identities together and com
pare them with the corresponding trig, identities.
91
Hyperbolic Functions
(1)
Trig. Identities
cot* = 1/tan*
Hyperbolic Identities
coth* = 1/tanh*
36
sec* = 1/cos* sech* = 1/cosh*
cosec* = 1/sin* cosech* = 1/sinh*
2 2
(2) cos * + sin *= 1 cosh 2 *  sinh 2 * = 1
2
sec * = 1 + tan 2 * sech 2 * =  tanh 2 *
1
2
cosec * = 1 + cot 2 * 2
cosech * = coth 2 *  1
(3) sin 2* = 2 sin * cos * sinh 2* = 2 sinh * cosh *
cos 2* = cos 2*  sin 2 * cosh 2* = cosh 2 * + sinh 2*
= 1 2 sin 2* = 1+2 sinh
2
*
= 2 cos 2*  1 = 2 cosh 2 *  1
tan 2x = J^
1 tan * 2 becomes tanh 2* =
1
2tanh *
+ tanh 2 *
So providing you know your trig, identities, you can apply the rule
to form the corresponding hyperbolic identities.
92
y
Programme 3
je
e =cos0 +jsin0
,e  j sin 6
and e~ = cos 6
2 cos 6
38
e g
ei + eJ
So that. cos 8
39 cosh j0
cos 9 = cosh j0
je = cos0 j sin0
e~
41 2j sin 6
So that, j sin 6 =
e>
e  e>
e
93
Hyperbolic Functions
42
sinh j0
Now do exactly the same with the second result above, i.e. put 8  jx
in the relationship j sin 8
= sinh j0 and simplify the result. What do you get?
44
j sinhx = sinjx
94
Programme 3
45 Now let us collect together the results we have established. They are
so nearly alike, that we must distinguish between them.
Copy the complete table into your record book for future use.
cos \y by
and sin \y by
So that
sin(x + ]y) = sin x cos \y + cos x sin \y
becomes sin(jc + \y) = sin x cosh y + j cos x sinh y
Note: sin(x + \y)is a function of the angle (x + j y), which is, of course,
95
Hyperbolic Functions
.'.
cos(x  jj>) = cos x cosh ^ + j sin x sinh >>
49
All that now remains is the test exercise, but before working through
look through your notes, or revise any parts of the programme on
it,
96
.
Programme 3
2
2. If v = 18 L tanh^, find v when d = 40 and L = 315.
3. On the same axes, draw sketch graphs of (i)y = sinh x, (ii)j' = coshx,
(iii)^ = tanhx
2x
6. If tanh x =r, find e and hence evaluate x.
y=C coshpr
97
. )
Hyperbolic Functions
Further Problems  HI
2 cosh 2x  sinh 2x = 2
(fl + 6) 2 e 2x =a 2 6 2
1
5. Evaluate (i) tanh" '075, (ii) cosh 2.
7. Express
and & to 4
(i) cosh Ll and (ii) sinh
significant figures.
^ in the form a + ]b, giving a
3 cosh 2x = 3 + sinh 2x
1+tanh * 2*
11. Proveth a t = e
1  tanh x
98
'
Programme 3
13. If x = x
ln tan find e and e * and hence show that
sinh x = tan 6
[54 ,
sin 2A sinh 2B
tan x =
1 + cos 2A cosh 2B
cosh a
18 ifx =
a? '
s' n ^ at + s' n fl ^
calculate \ when a = 0215 and =
t 5.
2 I cosh a/  cos at
,
1
a
19. Prove that rantr !
*, \ = \n
[x +0*2 ) a
2
20. Given that sinh
1
x = ln{x + \/(x + 1)}, show that, for small values
ofx,
*3
sinh
t 1
x
*
x
6
+
JX 5
40
.
99
Programme 4
DETERMINANTS
Programme 4
1 Determinants
You are quite familiar with the method of solving a pair of simultaneous
equations by elimination.
e.g. To solve 2x + 3y + 2 = ... (i)
3x + Ay + 6 = ... (ii)
So &c + 12>> + 8 =
9jc + 1 2y + 18 =
Then, by subtraction, we get x + 10 = 0, i.e. x = 10. By substituting back
in either equation,we then obtain y = 6.
So finally, x=10, y = 6
That was trivial. You have done similar ones many times before. In just
the same way, if
a xx +b y +di
x
= ... (i)
a xx +b^y + di =
a 2 x + b 2y + d 2 =
become a x b 2 x + b^b^y + b 2 d\ 
Subtracting, we get
(a l b2 a 2 bi)x + b 2 di ~bid 2 =0
so that {a l b2 a 2 bi)x = bid 2 b 2 di
Then x =
101
Determinants
b 1 d2 b 2 di
X =
alb2 a 2 b l
In practice, this result can give a finite value for x only if the
denominator is not zero. That is, the equations
tfi* + b xy + dx =
a 2 x + b^y + d2 =
give a finite value for x provided that {a x b 2 ~a 2 bi) f 0.
4x + 3y  7 =
0i6 2
~ a ibi  3.3 4.2
= 98=1
This is not zero, so there (will  be a finite value of x.
will not
will
ai bi
a \b 2 a 2 b\ 
ax bx
For to represent a x b 2 a 2 b\ then we must multiply the terms
3 7
eg
= 3.2 5.7 = 6 35 =29
5 2
6 5 5
So
1 2 1
102
Programme 4
6 5
= 125 =
1 2
DDnaDDnnnnnDnnDnnDDnnnonDDDDnDDnnnDDD
rows and two columns) and represents a y b 2 a 2 b\. You can easily
4 ?. 7 4 2 1
(i) ,
(ii) ,
Oii)
b 3 6 3 4 3
4 ?.
7 4
(ii) = 7.36.4 = 2124 = 3
6 i
2 1
we found that x= ij2 11 and the numerator and the denominator
aib 2 2 b\
b 1 d2 b 2 di = ; axb2 a 2 b =
x
103

Determinants
bi di i bx
i
b2 d2 a2 b2
.b y d 2 b 2 di _a\d 2 a 2 dx
we have and
a\b 2 a 2 b\ """ ^ fli6 2 a 2^i
So, x= and y =
bi d, ai di 8
b2 d2 a2 d2
X = and y =
bi i 61
a2 b2 2 b2
X 1 y 1
and
6, dt ai 61 "i di flj 61
Z>
2 d2 a2 b2 a2 d2 a2 b2
X ~y 1
bi dx ai d\ 1
i 61
b2 d2 a2 d2 
a2 62
Afafce a note 0/ fese results and then turn on to the next frame.
104
Programme 4
So if a yx +bty +di =0
a 2 x + b^y + d2 =
X y 1
Then
bi dt i dx 1\ by
b2 d2 a2 d2 a2 b2
Each variable is divided by a determinant. Let us see how we can get
them from the original equations.
(i) Consider .
. Let us denote the determinant in the denominator
bx dx
b2 d2
bi d,
by Aj , i.e. A i
t
b2 d2
To form Ai from the given equations, omit the xterms and write down
the coefficients and constant terms in the order in which they stand.
<tix +biy +d t = 6, d.
gives
a 2x + bjy + d 2 = b2 d2
y ai di
(ii) Similarly for , let A2
di "2 d2
a2 d2
To form A 2 from the given equations, omit the j;terms and write down
the coefficients and constant terms in the order in which they stand.
aix + biy +d t
= di
gives A2 =
a 2 x + b^y + d2 = 2 d2
1
(iii) For the expression denote the determinant by Aq.
a\ by
a2 b2
To form Aq from the given equations, omit the constant terms and write
down the coefficients in the order in which they stand
aix + b x y + d x =0 ay b x
gives
a 2 x + b2y + d2 = a2 b2
105
Determinants
 ~y  *
A, A2 Aq
To find A , omit the constant terms
5 2
Ao = 5.43.2 = 206= 14
3 4
..
Ao = 14 ... (i)
A, =
Ai =42
11
2 19
for Ai = 34 76 = 42 (iO
4 17
Similarly, to find A2 , omit the j>terms
5 19
A,= 8557 = 28 (iii)
3 17
Substituting the values of A t , A2 A , in the key, we get
x _y _ J_
42 ~ 28 ~ 14
42 28
X = "'i; y" = ~2
14 14
,y 12
Now for another example.
Example 2. Solve by determinants 2x+3>'14 =
\3x2y+ 5 =
First of all, write down the key:
x _^y
_ _ J_
Ai A2 "'Aq
(Note that the terms are alternately positive and negative.)
2 3
Then A, 4 9 =13 0)
3 2
Now you find Ai and A2 in the same way.
106
.
Programme 4
13 Ai =13; A 2 =52
3 14 3 14
A, = =
2 5 5 2
= 15 28 = 13. :. Aj =13
?, 14 ?. 14
A, = =
i 5 t> 3
= 10 (42) =52 :. A2 = 52
x _ y _ 1
So that " "
A! A2 A
_Ai_13_, . _,
A2 52 = "
' _
= _
at^3 4
"zn
Do not forget the key
x _y _ 1
A^ "A2 ~A^
with alternate plus and minus signs.
107
Determinants
x = 2; y = 4
15
Here is the working in detail:
4x  3y + 20 =
A ~ A2 Ao
3x + 2y 2 = t
4 3
Ao = 8(9) = 8 + 9 = 17
3 2
3 20
Ai = 6  40 = 34
2 2
4 20
A2 = = 8  60 = 68
3 2
Ai _34
2 :. x = 2
Ao 17
A2 68 .
:. j=4
nDODnnDDDnaDnnnnDDnaDDDnDDDnnannnDnDnn
Now, by way of revision, complete the following:
5 6
(0
7 4
5 2
00 3 4
(iii)
b c
P q
(iv)
r s
Here are the results. You must have got them correct.
=
16
(0 2042 22
(ii) 20 6 = 26
(iii) ac bd
(iv) ps rq
108
.
Programme 4
17 A determinant of the third order will contain 3 rows and 3 columns, thus:
0i bi Cx
a2 b2 c2
3 b3 c3
Each element in the determinant is associated with its MINOR, which
is found by omitting the row and column containing the element concerned.
b2 ?2 :
i !
bi Cll
e.g. the minor of a, is obtained
b3 c3 \ a2 \
b2 c2
!
3 !
b3 c3
a2 \a l \ bi i C1
the minor of b\
i
is obtained
a3 c3 a 2 \b 2 c2
\
a3 :
b3
c3
a2 b2 01
the minor of c t is obtained
a3 b3 2 ~b 2 c2 \
by ct
18 Minor of a 2 is
b3 c3
since, to find the minor of a 2 we simply ignore the row and column con
,
taining a 2 , i.e.
! fli !
bj. Cl
1 a2 j
fc 2 Cl
:
a 3i ^>3 c3 Similarly, the minor of b 3 is
ax Cy
19 Minor of b 3 
a2 c2
01 61 i Cl
02 62 '
Ci
% b3 \
c3 \
Now on to frame 20.
J
109
Determinants
i bt Ci
~
a\ ^7 b t
a2 c2 + c, a2 b2
a3 c3 a3 b3
a3 b3 c3
Example 1.
1 3 2
1 5 7 3 4 7 + 2 4 5
4 5 7 =
4 8 2 8 2 4
2 4 8
= 1(5.8  4.7)  3(4.8  2.7) + 2(4.4  2.5)
= l(4028)3(3214) + 2(1610)
= 1(12) 3(18) + 2(6)
= 12 54 + 12 =30
Here is another.
Example 2. 3 2 5
21
3 6 7 2 4 7 + 5 4 6
4 6 7
9 2 ?. 2 ?, 9
2 9 2
Example 3. Evaluate 2 7 5
4 6 3
8 9 1
Expand along the top row, multiply each element by its minor, and
+ and signs to the products.
assign alternate
no
Programme 4
22 Result 38
For 2 7 5
6 3 7 4 3 + 5 4 6
4 6 3
9 1 8 1 8 9
9 1
We obtained the result above by expanding along the top row of the
given determinant. If we expand down the first column in the same way,
still assigning alternate + and  signs to the products, we get
2 7 5
2 6 3 4 7 5 +8 7 5
4 6 3
9 1 9 1 6 3
8 9 1
y^ We can, if we wish, expand along any row or column in the same way,
multiplying each element by its minor, so long as we assign to each
product the appropriate + or sign. The appropriate 'place signs' are
given by +_ + _ +
 +  +  .
+  +  + .
 + + 
etc., etc
The key element (in the top lefthand corner) is always + . The others are
then alternately + or  as you proceed along any row or
, down any column.
So in the determinant 13
7
5 6 9
4 2 8
Ill
Determinants
24
since in a third order determinant, the 'place signs' are
+  +
Remember that the top lefthand element
 +  always has a + place sign. The others
+  + follow from it.
6 8 4
1 9 5
3 7 2
7 6 4 + 8 3 2 9 3 2
6 8 4
1 5 1 5 6 4
1 9 5
Result
78
25
for 6 4 + 8 3 2 9 3 2
1 5 1 5 6 4
= 7(30 4) + 8(1 2) 5
9(12 12)
= 7(26) + 8(13) 9(0)
= 182 + 104 = 78
5 7 2
112
Programme 4
26 Answer 119
6 1 3  + 
5 7 2 +  +
= 5 3 4 7 2 4 + 2 2 3
1 3 6 3 6 1
4 6 9
27
Result 143
For 12 87 2 8 + 3 1 8  1 1 2
7 3 1 6 9 4 9 4 6
4 6 9
= 7(1 8 48) + 3(9 32) 1(6 8)
= 7(30) + 3(23)l(2)
= 21069 + 2= 143
DDDDnDDDDnDnDnDnnnDnDnDDDnnDnDDDnDnnDn
We have seen how we can use second order determinants to solve
simultaneous equations in 2 unknowns.
We can now extend the method to solve simultaneous equations in
3 unknowns.
113
Determinants
b2 c2 d2 ci d2 a2 b2 d2 a2 b2 c2
b3 c3 di 3 b3 d3 "3 b3 c3
x ZL z ^i_
Ai A2 aT A
where Ai = the det. of the coefficients omitting the xterms
" "
A2 = " " " " " jterms
" "
A3 = " " " " " zterms
" "
Ao = " " " " " constant terms.
3y z 4 =
'2x +
3x+ y + 2z 13 =
x + 2y  Sz + 1 1 =
2L=Z2. z ^1
First the key: :
A, A2 a! Ao
2 3 1
?. 1 7. 3 3 7 1 3 1
A = 3 1 2 =
2 5 1 5 1 2
1 2 5
= 18 + 515 = 28
(ii) Now you find Ai , in the same way.
114
Programme 4
29 A =56
t
.
1 . x _l
But
A "A i
" 56 28
Note that by this method we can evaluate any one of the variables,
without necessarily finding the others. Let us do another example.
z
30 Aj A2 a; Ao
To find j, we use
A2 A
Therefore, we must find A 2 and Ao
4.x + 2j>  3z + 8 =
To find A2 , omit the ^terms.
2 5 11
1 6 + 5 + 11 1
.".
A2 1
3
3 8
3
2(8  18) + 5(8 + 24) + 1 1(3  4)
20 + 16077 = 63
To find A , omit the constant terms
Ao =
115
Determinants
Ao=21
31
for 2 1 5
2 1 1
 1 1 1 5 1 1
Ao = 1 1 1 =
2 3 4 3 4 2
4 23
= 2(3  2)  l(3  4)  5(2 + 4)
= 2 + 730 = 21
So we have A
~y
2
=
i
.
A
. .
. A2
y =ir z,
Aq 21
63
:.y=3
The important things to remember are
X ~y _ z
(i) The key "<5
A, A2 A; ^0
(ii) To find Ai , which is associated with x in this case, omit the x terms
and form a determinant with the remaining coefficients and con
stant terms. Similarly for A2 A 3 A, , .
Next frame.
l. x+2y3z 3 =
2x y i 11 =0 Find y.
3x + 2y + z + 5=0
 Ay
3x + 2z + 8 = 
5y3z+
x + 5y 2 = \ Find* andz.
5x + 3y  z+ 6 J
2x2y z 3 =
4x + 5y  2z + 3=0 f
Find x,y and z.
3x + Ay  3z + 7 =
When you have finished them all, check your answers with those given in
the next frame.
116
Programme 4
1. y=4
2. x=2; z = 5
3. x = 2; y = _ 1; z = 3
34
The answer to question No. 1 in the revision test was y = 4
Did you get that one right? If so, move on straight away to frame 41.
If you did not manage to get it right, let us work through it in detail.
A!

To fill in the missing terms, take each variable in turn, divide it by the
associated determinant, and include the appropriate sign.
So what do we get?
On to frame 35.
117
.
Determinants
x _~y_ z _ H_ 35
a7 ~a7~a^"a
16

y = l
AA 2
A
.
A2
y = ir
A
To find A 2 we
,
36
So 1 3 3
A2 = 21 11
3 1 5
^and ^^
So we get A2 =
118
, Programme 4
37 A, = 120
A A2 = 120
So
A =
38
Ao
39 1 1 2 2 1 3 2 1
Ao =
2 1 3 1 3 2
Now, evaluating the second order determinants in the usual way gives
that
Ao=
119
Determinants
Ao =30
40
for A = 1(1 + 2) 2(2 + 3) 3(4 + 3)
= 1(1) 2(5) 3(7)
= 11021 =30
So Aq=30.
So we have
_A 2 _120_
y A 30
:.y = A
in detail.
x = 2 41
The answers to No. 2 in the revision exercise were
z = 5
Did you get those correct? If so, turn on right away to frame 51 . If not,
follow through the working. Here it is:
x _
...
_ ...  ...
120
Programme 4
42
A ~ A2
t
A3 A
X 1 Ai
i.e. x =
Ao Ao
1
and
z
i.e. z = A 3
A, Ao Ao
/. Ax
43
4 2 8
A t = 5 3 2
3 1 6
3 2 2 5 2 + 8 5 3
A, =
1 6 3 6 3 1
121
Determinants
A, =48 44
for A =4(18
t + 2)2(306)+8(5 + 9)
= 4(16) 2(24) + 8(4)
= 6448 + 32 = 9648 = 48
:. Aj = 48
A,
34 8
45
A3 = 1 5 2
5 3 6
A3 =
46
3 5 2 + 4 1 2 +8 1 5
A3 =
3 6 5 6 5 3
Now evaluate the second order determinants and finish it off. So that
A3 =
On to frame 47.
122
Programme 4
47 120
:.A 3 =120
Ao =
48
3 4 2
Ao = 1 5 3
5 3 1
Now expand this along the top row as we have done before. Then
evaluate the second order determinants which will appear and so
find the
value of A .
Ao =
123
Determinants
Ao=24 49
for
5 3 +4 1 3 + 2 1 5
Ao =
3 1 5 1 5 3
:. A =24
48
x=2 50
(120) _
z 24 ~ 5 z = 5
Well, there you are. The method is the same ev(:ry time but take
care not to make a slip with the signs.
Now what about question No. 3 in the revision exercise. Did you get
that right? If so, move on straight away to frame 52.
have another go at it. Here are the equations again: copy them
If not,
down and then find*,.}> andz.
2x2y z3 =
Ax + Sy  2z + 3 =
1
3x + Ay  3z + 7 =
When you have finished this one, turn on to the next frame and check
your results.
124
Programme 4
51 Answers to No. 3
2, y=~\, z = 3
Here are the main steps, so that you can check your own working.
_^L = IZ = JL =Zi_
Ax A2 A 3 A
2 1 3
A, = 5 2 3 = 54
4 3 7
2 1 3
A2 = 4 2 3 = 27
3 3 7
2 2 3
A3 = 4 5 3 = 81
3 4 7
2 2 1
Ao = 4 5 2 = 27
3 4 3
X 1
=ii=2
x =
A, Ao Ao 27
x=2
y 1 A2
A2 A y= A 27
^=1
z 1
z =_A 3 = li = 3
a; A A 27
z =3
All correct now?
125
Determinants
3x y4 = (0
2x + 3y~8 = (ii)
x y4=Q (iii)
If we solve equations (ii) and (iii) in the usual way, we find that x = 1 and
y = 2.
If we now substitute these values in the lefthand side of (i), we obtain
3x j4 = 324= 3 (and not as the equation states).
The solutions of (ii) and (iii) do not satisfy (i) and the three given
equations do not have a common solution. They are thus not consistent.
There are no values ofx and y which satisfy all three equations.
If equations are consistent, they have a
53
common solution
3x + y 5 = (i)
2x + 3y  8 = (ii)
x  2y + 3 = (iii)
3x +y 5 =3 + 25 =
i.e. all three equations have the common solution x= l,y = 2 and the
equations are said to be c
126
Programme 4
54 consistent
a x x + by +di = (i)
a 2x + b 2 y + d2  00
a 3 x + b 3y +d 3 = (iii)
i.e. ( a 2 x + biy + d2 =
a 3 x + 63J + d3 =
we get "
Ai A2 " A
b2 d2 a2 d2 a2 b2
where A, = , A2 = , A =
b3 d3 a3 d3 3 b3
A _ A2
and y  ~
t
so that x  V" 7
Ao Aq
i.e. fliAi ~b y A 2 +d v A =
d2 a2 <2 2 fl
2 > 2
i.e. 61 + di
d3 a3 d3 fl
3 Z> 3
fll 61 C?i
i.e. a2 i2 c? 2
a3 63 c? 3
which is therefore the condition that the three given equations are
consistent.
127
Determinants
55
Example 1. Test for consistency
1
5
= 2 4 1  1 1 1 _5 1 4
1
4 1
1 10 3 10 13 1
1 10
= 2(40+l)l(103)5(l12)
= 2(39)(13)5(13)
= 78 + 13 +65 =78 + 78 =
The given equations therefore consistent.
(are/are not)
are
J Ax + 2y  k = For consistency,
4 2 k =
[2x y + 3k =
2 1 3&
:. 3 2 k 1 4 k +2 4 2
=
1 3k 2 3k 2 1
{ 3x+ ly + 1 =
128
y
Programme 4
57 k = 2 or

1
2k 5 3 =
3 7 1
5 3 (*+0 2k 3 + 1 2k 5
=
7 1 3 1 3 7
58 k = 1 or x
For 1 1 k
fc 3 11 =
2 48
1 3 11  1 k 11 k k 3
=
4 8 2 8 2 4
.'.
fc = 1 or k =
129
Determinants
Properties of determinants 59
Expanding a determinant which the elements are large numbers can be
in
a very tedious affair. It is possible, however, by knowing something of the
properties of determinants, to simplify the working. So here are some of
the main properties. Make a note of them in your record book for future
reference.
ax bx
h b. a2 b2
2. If two rows (or two columns) are interchanged, the sign of the
determinant is changed.
a2 b2 at b\
at bi a2 b2
3. If two rows (or two columns) are identical, the value of the deter
minant is zero.
=
a2 a2
4. If the elements of any one row (or column) are all multiplied by a
common factor, the determinant is multiplied by that factor.
ka x kbt a\ bx
a2 b2 a2 b2
5. If the elements of any one row (or column) are increased (or decreased)
by equal multiples of the corresponding elements of any other row (or
column), the value of the determinant is unchanged.
a x +kbi bx fii bi
a 2 + kb 2 b2 a2 b2
DannnaDnnnnnDnDDnnDDDnDDDnnDnDDDDDDDDD
NOTE: The properties stated above are general and apply not only to
second order determinants, but to determinants of any order.
130
Programme 4
369 2
58
(Rule 5)
369 2
= (58)(2)(0) = 116
Example 2. Evaluate 1 2 2
61 4 3 5
column 2 minus column 3 will
give us one zero
4 2 7
1 0. 2
2 3
We could take a factor (1) from
1
the top row and another factor
5
(1) from the bottom row.
(1X1) 2 3
5 1
131
Determinants
Example 3. Evaluate
62
You do that one, but by way of practice, apply as many of the listed
properties as possible. It is quite fun.
4 2 2
We can take out a factor 2 from
2 4 2 each row, giving a factor 2 3 i.e. ,
2 1 1
2 1
column 1 minus twice (column
1 1 1 3) will give another zero in the
same row.
1 1 2
1
Expanding along the top row will
1 1 1 now reduce this to a second order
determinant.
3 1 2
11 1
11 1
4
4
1
= 8 (4) = 32
4
132
Programme 4
34 x  2
In this type of question, we try to establish common factors wherever
possible. For example, if we add row 2 and row 3 to row 1 we get ,
(x + 2) (x + 2) (x + 2)
5 x+ 1 1
3 4 x 2
Taking out the common factor (pc + 2) gives
(*+2) 1 1
5 jc + 1
34 x2
Now if we take column 1 from column 2 and also from column 3, what
do we get?
When you have done it, move on to the next frame.
We now have (x + 2) 1
65 5 jc4 4 =
3 1  x + 1
Expanding along the top row, reduces this to a second order determinant.
(x + 2) x4 4
1 x+ l
If we now multiply out the determinant, we get
(x + 2) [(x4)(x+ l)4] =0
(x + 2) (x  3x  8) =
2
:.
2
x+2= or jc 3x8 =
133
Determinants
Result: 66
x = 4 or 1 \/6
5 x 3
3 2 x
x+4 x +4 x +4
Take out the common
x+ 2 2 1
factor (x + 4)
3 2
(x+4) 1 1
Take column 3 from
x + 2 2 = column 1 and from
column 2
3 2
(*+4)
=
This now reduces to
x+ 1 1
second order
jc 3 2 x
(x + 4) x + l 1
Subtract column 2 from
=
x 3 2 x column 1
(x + 4) x 1
We now finish it off
5 2x
:. (x+4)(2xx,22 + 5) =
x  2x 
2
;. x + 4= or 5 =
onaannnnoDDDQaoaaaaonnaDnaDaananonaoaa
You have now reached the end of this programme on determinants
except for the Test Exercise which follows in frame 67. Before you work
through brush up any parts of the work about which you are at all
it,
134
Programme 4
Qg Test Exercise IV
Answer all the questions. Take your time and work carefully. There is
DnnannnDnnnDnDDaannnnnannnnaDDDnnDnDnn
1. Evaluate (a) 1 1 2 (b) 1 2 3
2 1 1 3 1 2
1 2 1 2 3 1
'2x+3y~ z 13 =
*3.y+4z5 =
2x+ y+ z3 =
[4x + 3y + 5zl=Q
4. Find the values of k for which the following equations are consistent
3x + 5y+k =
2x + y 5 =
QnnonaDnaQaQQnnaaoannaciaQLjnannaQanna
135
Determinants
Further Problems IV
11 9 13 2 535 984
15 17 19 3 642 1107
16 10 18 77 112 39
34 6 38 74 HI 37
3. Solve by determinants
4x5v + 72 =14
9x + 2v + 3z = 47
x y Sz  11
4x  3y + 2z = 7
2x  Ay  z = 3
5. Solve by determinants
3x + 2y  2z = 16
4x + 3y + 3z = 2
2x .y + z=l
6. Find the values of X for which the following equations are consistent
5x+(\+l)v5 =
(\1)jc + 7^ + 5 =
3x + 5v + 1 =
136
Programme 4
k 1
1 k 1 =
1 k
(b) Factorise 1 1 1
a b c
3 3 3
a b c
2 x+3 6 =
3 4 x +6
10. Find the values of x that satisfy the equation
x 3 +x 2+ x
3 3 1 =
2 2 2
11. Express 1
2 2 2
a' b c
(b+cf2 < j. \2
{c+af ( M2
(a+bf j.
137
Determinants
14. Find values of fc for which the following equations are consistent.
x+ (1 + k)y + 1=0
(2 + k)x + 5y  10 =
x + 7y + 9 =
15. Express 1 + x2 yz 1 as a product of four linear factors.
1+y 2 zx 1
1+z 2 xy 1
x+3 x+l
2 =
x+3 x+2 1
+M >rM iy = W
(iM, 2
M 2x + 2M 2 + (Mi  M )z = >>
2
~M 2 y+QU + M )z = l 2
6ii  i
2 + 2/ 3 = 4
3r\  \2i 2 +8i 3 =
By the use of determinants, find the value of it and hence solve com
pletely the three equations.
20. Find the angles between = and =n that satisfy the equation
2 2
1 + sin cos 4 sin 20
2 2
sin 1 + cos 4 sin 20 =
2 2
sin cos 1 +4 sin 20
138
Programme 5
VECTORS
Programme 5
(b) A vector quantity is defined completely when we know not only its
magnitude (with units) but also the direction in which it operates, e.g.
force, velocity, acceleration. A vect< quantity necessarily involves
direction as well as magnitude.
vector
(ii) its
direction
So that:
(i) A temperature of 100C is a quantity.
2
(ii) An acceleration of 98 m/s vertically downwards is a
quantity.
141
Vectors
(i) scalar, (ii) vector, (iii) vector, (iv) scalar, (v) vector
Since, in (ii), (iii) and (v), the complete description of the quantity
includes not only its magnitude, but also its
direction
Vector representation
A vector quantity can be represented graphically by a line, drawn so that:
(i) the length of the line denotes the magnitude of the quantity,
according to some stated vector scale,
(ii) the direction of the line denotes the direction in which the vector
quantity acts. The sense of the direction is indicated by an arrow
head.
35
AB=
On to frame 7.
142
Programme 5
Similarly, if two vectors a and b are such that b = a, what can we say
about (i) their magnitudes,
N
8 (i) Magnitudes are equal,
(ii) The vectors are parallel but opposite in sense.
Types of vectors
(i) A position vector AB occurs when the point A is fixed.
(ii) A line vector is such that it can slide along its line of action, e.g. a
by its magnitude and direction and can be drawn as any one of a set
of equallength parallel lines.
So on now to frame 1 0.
M3
Vectors
Addition of vectors 10
The sum of two vectors, AB and BC, is defined as the single or equivalent
or resultant vector AC
i.e. AB + BC = AC
or a + b = c
A "s B
To find the sum of two vectors a and b, then, we draw them as a chain,
starting the second where the first ends: the sum
then given by the
c is
single vector joining the start of the first to the end of the second.
I
then the
ttt'then 1 magnitude of the vector sum r of these two forces will be
r=50N 11
for
*'' r
2
=p 2 +q 2
? = 1600 + 900 = 2500
r = V2500=50N
the start of the first to the end of the last  in this case, AE. This follows
directly from our previous definition of the sum of two vectors.
R
Q *" A t Similarly,
PQ + QR + RS + ST =
144
Programme 5
12 PT
Now suppose that in another case, we draw the vector diagram to find the
sum of a", b,c,d,e, and discover that the resulting diagram is, in fact, a
closed figure.
13
Sum of the vectors =
For we said in the previous case, that the vector sum was given by the
single equivalent vector joining the beginning of the first vector to the
end of the last.
But, if the vector diagram is a closed figure, the end of the last vector
coincides with the beginning of the first, so that the resultant sum is a
vector with no magnitude.
145
Vectors
AQ 14
Right. Now what about this one?
Also ED = DE
ABCB + CDED AB + BC + CD + DE
AE.
>Iow you do this one
15
For:
AB + BCDCAD = AB + BC + CD + DA
and the lettering indicates that the end of the last vector coincides with
the beginning of the first. The vector diagram is thus a closed figure and
therefore the sum of the vectors is 0.
(i) PQ + QR + RS + ST =
(ii) AC + CLML =
(iii) GH + HJ + JK + KL+LG =
(iv) AB + BC + CD + DB =
When you have finished all four, check with the results in the next frame.
146
Programme 5
(i) PQ + QR + RS + ST = PT
(iii) GH + HJ + JK + KL+LG =
[Since the end of the last vector coincides with the
beginning of the first.]
e.g. i^"''"^\.c
5
\ >.
\ J_ VY = a + b+c+d
Example 1.
AB = AC + GH+HB
DC =
say
147
Vectors
DC = DG + GH + HC
18
So we have
A AB = AG + GH + HB
DC = DG + GH + HC
..
AB + DC = AG + GH + HB + DG + GH + HC
= 2GH + (AG + DG) + (HB + HC)
Now, G is the mid point of AD. Therefore, vectors AG and DG are equal
in length but opposite in sense.
..
DG = AG
Similarly HC=HB
:. AB + DC = 2GH + (AGAG) + (HBHB)
= 2GH
Next frame.
19
Example 2.
Points L, M, N are mid points of the sides AB, BC, CA, of the triangle
ABC. Show that
(i) AB + BC + CA =
(ii) 2AB + 3BC + CA = 2LC
(hi) AM BN CL + + = 0.
148
r
Programme 5
To show that 2 AB + 3 BC + CA = 2 LC
B M
From the figure
2AB + 3BC + CA =
21 2LC
For 2 AB + 3 BC + CA = 4 AL + 3 BL + 3 LC + CL + LA
= 4AL3AL + 3LCLCAL
= 4AL4AL+3LCLC
= 2LC
Now part (iii)
To prove that AM + BN + CL =
From the figure in frame 20, we can say
AM =+ BM
AB
BN=BC +CN
Similarly CL=
22 CL = CA + AL
So AM + BN + CL = AB + BM + BC + CN + CA + AL
= (AB + BT + CA) + (BM + CN"+AL)
= (AB + BC + CA) + i(BC + CA + AB)
= Finish it off.
149
Vectors
23
AM + BN + CL =
Since AM + BN + CL = (AB + BC + CA) + KBC + CA + AB)
Now AB + BC + CA is a closed figure .". Vector sum =
and BC + CA + AB is a closed figure .'. Vector sum =
:. AM + BN + CL =
Here is another.
Example 3.
24
A D
To prove that AB + AD + CB + CD = 4PQ
Taking the vectors on the lefthand side, one at a time, we can write
AB = AP + PQ + QB
AD = AP + PQ + QD
CB =
CD=
t
CB = CP + PQ + QB CD = CP ; + PQ + QD 25
Adding all foilr lines together, we have
AB + AD + CB + CD = 4PQ + 2AP + 2CP+ 2QB + 2QD
= 4 PQ + 2 (AP + CP) + 2(QB + QD)
150
Programme 5
26 AP+CP =
Since P is the mid point of AC .'.
AP = PC
:. CP = PC = AP
.'.
AP + CP = APAP=0.
In the same way, (QB + QD) =
27
QB + QD =
Since Q is the mid point of BD .'. QD = QB
.'.
QB + QD = QBQB =
.'.
AB + AD + CB + CD = 4PQ + +
= 4PQ
Prove by vectors that the line joining the midpoints of two sides of a
triangle is parallel to the third side and half its length.
A
Let D and E be the midpoints of AB
and AC respectively.
DE = DA + AE
Now express DA and AE in terms of BA and AC respectively and see if
you can get the required results.
151
)
Vectors
Now for the next section of the work: turn on to frame 30.
then a=ai
Similarly, if we define / to be a unit vector in the OY direction,
then b = bj
r =a i +b j
152
Programme 5
31 Let Zj = 2i + 4/ and z 2 = 5/ + 2/
Y
5 H
fi + z2 = OB
= (2 + 5)z+(4 + 2)/
= 7/ + 6/
If Zi = 3/ + 2/ and z2 = 4/ + 3/
Zi +z 2 =3z+2/+4i + 3/
= li + 5/
32 Z2 Z! = 1/ + 1/
= 4/ + 3/3z2;
= 1/ + 1/
Similarly, if z"i
= 5 i  2/; z2 = 3/ + 3/; z3 = 4/  1/,
153
Vectors
0)12/ ;
(ii)2/4/
33
Here is the working:
(i) zj + z 2 + z 3 = 5/  2/ + 3/ + 3/+ 4/  1/
= (5 +3 + 4)/ + (3 21)/
= 12/
(ii) Z!  z2  f3 = (5/ 2/). (3/ + 3/)  (4/ 1/)
3 4)/
= (5 + (2 3 + 1)/

= 2i 4/
Now this one.
AB =
AB = 2/7/
34
for we have OA + AB = OB (from the diagram)
.'.
AB = OBOA
= (5/ 2/) (3/ + 5/) = 2i 7/
On to frame 35.
Vectors in space z
The axes of reference are defined by 35
the 'righthand' rule.
154
Programme 5
36 ox
Vector OP is defined by its
components
a along OX
J* b " OY
c " OZ
OP 2 =a 2 +ft 2 +c 2
37 PQ = V29 == 5385
i
For, if PQ =4i + 3j+2k
2
pq= V(4 +3 2 +2 2 )
= V(16 + 9 + 4) = V29=5385
155
Vectors
Direction cosines JQ
The direction of a vector in three dimensions is determined by the angles
which the vector makes with the three axes of reference.
z
Let 0? = r = ai + bj + ck
Then
a
 = cos a .. a = r cos a
7=cos0 b = r cos (3
Also a
1
+b z +<r =r
2 2
.'.
r cos a + r 2 cos 2 (3 + r 2 cos 2 7 =r 2
2 2 2
.".
cos a + cos 3 + cos 7 = 1
If / = cos a
w = cosj3 then /
2
+m +2 2
= l
n = cos 7
Note: [l,m,n]_ written in square brackets are called the direction cosines
of the vector OP and are the values of the cosines of the angles which the
vector makes with the three axes of reference.
So for the vector r=ai + bj + ck
So, with that in mind, find the direction cosines [l,m,n] of the vector
T=3i2j+6k
Then to frame 39.
7=3i2j + 6k 39
a = 3,b=2,c = 6 r = V(9 + 4 + 36)
/. r = V49 =7
:. /=; m=\; n =
\
Just as easy as that! On to the next frame.
156
Programme 5
For example
OA.OB =
41 OA.OB = ^
For, we have:
OA.OB = OA.OB.cos
= 5.7.cos45
= 1 35V2
V2 2
.90
157
Vectors
42
since, in this case, a.b = a. b. cos 90 = a.b.O =
So, the scalar product of any two vectors at right angles to each other is
always zero.
And in this case now, with two vectors in the same direction, 6 = 0,
* so a.b =
a.b
43
since a.b. = a. b.cos0 = a.b. I = a.b
Now suppose our two vectors are expressed in terms of the unit vectors.
Then A.B = (a xi
+ bij+ c y k).(a 2 i + b2]' + c 2 k)
= a l a 2 i.i+ aib 2 i.j + aiC 2 i.k + b 1 a 2 /.i + bib 2 j.j
+ bic 2 j.k + c x a 2 k.i + c 1 b 2 k.j + cc 2 k.k
So, using the results (i) and (ii), we can simplify the expression for A.B
above to give
AJ =
158
a ,
Programme 5
44 A.B = a a 2 + b x b 2 + c t c 2
x
i.e. we just sum the products of coefficients of the unit vectors alonj;
corresponding axes.
e.g. If A= 2/ + 3/ + 5k and B = 4/ + lj + 6k
45
for P.Q = 3.2 + (2).3 + l(4)
= 6  64 :. P.Q =4
Now we come to:
(Axe) 
(AX B)= ABsin6
Note that BXA reverses the direction
of rotation and the product vector
would now act downward, i.e.
(B X A) = (A X B)
(BxA)
If = 0,then(AX B) =
159
Vectors
9= 0,(AX B)=0
46
0=9O,(AX B)=AB
A=a l
i + bij+c l k and B = a 2 i + b 2 j + c 2 k
:. i X / = k
}Xk= i \ (ii)
kX i = j
iXj=(jXi)
since the sense of
j X k= (k X j)
rotation is reversed.
kXi = (i X k)
Now with the results of (i) and (ii), and this last reminder, you can
simplify the expression for AX B.
Remove the zero terms and tidy up what is left.
160
Programme 5
47 AXB = {bic 2 b 2 c t )i + (a 2 c t
 iC
2 )/ + {a\b 2 ~a 2 bi)k
Now we could rearrange the middle term slightly and rewrite it thus:
then AX B / / k
0i &1 Cl
a2 b2 c2
and that is the easiest way to write out the vector product of two vectors.
Note: (i) the top row consists of the unit vectors in order, /, /, k
(ii) the second row consists of the coefficients of A
(iii) the third row consists of the coefficients of B.
48
PXQ = i j k
2 4 3
1 5 2
FXQ=
161
Vectors
PX Q=23/ + +
7/ 6fc
49
PXQ / / k
2 4 3
1 5 2
4 3 2 3 + /c 2 4
5 2 1 2 1 5
Also AX B = z
i k
ii bx Cx
a2 b2 c2
PX Q = 6z" + 7/ + 23fc
=
50
for PX Q / ;'
k
34 2
2 51
4 2 / 3 2 +k 3 4
5 1 2 1 2 5
162
Programme 5
{mm')
2  ri) 2
Then (PP') = (/ I'f + (m m'f + (n
= I
2  2.1.1' + l'
2
+m 2  l.m.rri + m 2
+ n 2  Inn + n'
2
= (l
2
+m +n 2 2
) + (I'
2
+ m' 2
+ ri
2
)
 2{ll' + mm + nri)
But (/ 2 +m 2
+n 2 )= 1 and (/'
2
+m' 2 +n' 2 )= 1 as was proved earlier.
2
:. (PP') = 22(//' +mm' + nri) (i)
2
(PP') = OP2 + OP' 2  2.0P.0P! cos e
= 1 + 1 2.1.1.cos0 [ OP and OP' are )
= 22cos0 (ii)
\unitvectors j
(PP')
2
= 22(//' +mm +nri)
and (PP')
2
= 2  2 cos 6
:. cos 9 =
i.e. just sum the products of the corresponding direction cosines of the
two given vectors
So, if [/, m,n] = [05, 03, 04]
and [l',m',ri] = [025, 06, 02]
the angle between the vectors is 6 =
163
Vectors
6= IT 53
for, we have
cos 6 = 11' + mm + nri
= (05) (025) + (03) (06) + (04) (02)
= 0125 + 018  008
= 0308  008 = 0225
0=77
54
'vfe'
m
*<k'
n=
^
2 2
for P =V(2 + 3 + 42 ) = V(4 + 9 + 16) =
r V29
fe 3
/
V29
_c_ 4
r \729
"
"2 3 4
' U,m,n] =
V29 '
%/29 '
\/29
Now find the direction cosines [/', m, ri] of Q in just the same way.
164
Programme 5
55
since r'=Ql = V(4
2
+3 2 +2 2 ) = V(16 + 9 + 4) = V29
3
.. [1 ,m,n] =
V29 '
V29 '
V29
We already know that, for P,
[l,m,n] = _?_ _1 J_
V29 '
V29 '
V29
So, using cos d =11' + mm + nri', you can finish it off and find the angle
6. Off you go.
= 762'
56
a = 2 4 ^ 3 (3) 4 2
for COS P +   + 
29 29 29
02414
=
1T
2
Now on to frame 5 7.
p^ Direction ratios
\0P\ = ^a 2 +b 2 +c 2
r =
I =
r
, ?n =
r
, n = r
We can see that the components, a,b,c, are proportional to the direction
cosines, /, m, n, respectively and they are sometimes referred to as the
Note that the direction ratios can be converted into the direction cosines
by dividing each of them by r (the magnitude of the vector).
Now turn on to frame 58.
165
.
Vectors
Here is a short summary of the work we have covered. Read through it. 58
Summary
1 A scalar quantity has magnitude only ; a vector quantity has both
magnitude and direction.
2. The axes of reference, OX, OY, OZ, are chosen so that they form a
righthanded set. The symbols /',/, k denote unit vectors in the direc
tions OX, OY, OZ, respectively.
2 2 2
If OP = ai + bj +ck, then OP =  r = V(tf + b +c )
3. The direction cosines [I, m, n] are the cosines of the angles between
the vector and the axes OX, OY, OZ respectively.
i _b n
_c
i
For any vector
^ i
l ~ji m,
and l
2
+m 2 +n 2 =\
a2 b2 c2
166
Programme 5
59 Now you are ready for the Test Exercise below. Work through all the
questions. Take your time over the exercise: the problems are all straight
forward so avoid careless slips. Diagrams often help where appropriate.
So off you go.
Text Exercise V
3. Find the direction cosines of the vector joining the two points
(4, 2, 2)and (7, 6, 14).
(i) the value of A.B and the angle between the vectors A and B.
(ii) the magnitude and the direction cosines of the product vector
(AXB) and also the angle which this product vector makes
with the vector C.
167
Vectors
Further Problems V
2. Find the direction cosines of the vectors whose direction ratios are
and (1,2, 3). Hence find the acute angle between the two
(3, 4, 5)
vectors.
7. Find the scalar product (A.B) and the vector product (A X B), when
(i) A = + 2/ k, B = 2/ + 3; + k
i
168
Programme 5
11. A, B, C are vectors defined by A= 8/ + 2] 3k, B = 3/ 6/+ 4/fc, and
C= 2/  2/ fc, where z, /', k are mutually perpendicular unit vectors.
(i) Calculate A.B and show that A and B are perpendicular to each
other
(ii) Find the magnitude and the direction cosines of the product
vector(A XTT)
13. If position vectors, OA, OB, OC, are defined by OA = 2i /+ 3k,
OB = 3/ + 2/ 4/fc, OC = /' + 3/  2k, determine
169
Programme 6
DIFFERENTIATION
1
Programme 6
dy
y=f(x)
dx
1. xn nx"
1
2. ex tx
3. e
kx ke !cx
4. ax a x .\na
1
5. \nx
X
1
\og a x
x. In a
I sinx cosx
cosx  sin*
2
9. tan* sec x
10. cot X cosec
2
x
11. secx sec x.tanx
171
Differentiation
r ie
l
x
}=e x and T{e x } = e x
'
dx dx
(i) y = sinh x y= ~
^_e*(e)_e* +e^
cosh;c
"dx 2 2
x x
e + e'
(ii) y = cosh x y
2
x
dy _tx + (~e ) _ e
x  x
. e~ _
" dx sinhx
2 2
Move on to frame 3.
1. xs 11. cosx
2. sinx 12. sinhx
3x
3. e 13. cosec x
3
4. lnx 14. a
5. tan x 15. cot X
X 16. a
x
6. 2
4
7. secx 17. x'
8. coshjc 18. loga x
9. logio* 19.
x/2
10. e* 20. e
When you have finished them all, turn on to the next frame to check your
results.
172
Programme 6
Here are the results. Check yours carefully and make a special note of
any where you may have slipped up.
1 5.x
4
11. sin x
2. cosx 12. coshjc
3x
3. 3e 13. cosecx.cotx
4. 1/x 14.
5. sec
2
* 15. 
cosec x
2
6. 2* In 2 x
16. a lna
7. secx.tanx 17. 4x' s
8. sinhx 18. l/(xlna)
\<? h
x
10. e 20.
If by chance you have not got them all correct, it is well worth while
returning to frame 1 , or to the list you copied, and brushing up where
necessary. These are the tools for all that follows.
When you are sure you know the basic results, move on.
Functions of a function
Sin x is of x since the value of sin x depends on the value of
a function
the angle x. Similarly, sin(2x + 5) is a function of the angle (2x + 5) since
173
Differentiation
DDDnDnDnnnDDnnnnnDnnnnDDnnannannnnDDnn
We very often need to find the differential coefficients of such func
tions of a function. We could do them from first principles:
gives us
dy
du
not
dy
dx
,
. To convert our result into the required
n coefficient
we use = dy
dy
du .
,
.
i.e.
1i
we multiply
. . dy ,
,
~ (which
, .
we have) by ,
, . du
=
,
to obtain
.
du r
i.e. ^ = 5.
'
j {cos(5x  4)}= sin(5x  4) X 5 = 5 sin(5x  4)
_ L dy dy du du
But r = T  and r = cos x
,
dx du dx dx
d sinx
{e }=e sinx .cosx
dx
This is quite general.
If J = /(") andw = F(x), then ^ = ^ 
, i.e. if y = In F, where F
r \. c ^ dx du dx
is a function of x, then
dy = dy_ dF^ \_dF
dx dF dx F dx
So,if>' = lnsinx & = _L .
C os x = cot x
dx sinx
dF
It is of utmost important not to forget this factor , so beware!
174
Programme 6
;. i^ = 2
sec (5x  4) X the diff. of the function (5x  4)
dx
= 2
sec (5.x  4) X 5 = 5 sec (5x
2  4)
(ii) y = (4x 3)
s
Basic standard form isy = X s ,
 = 5x
A
dx
Here, (4x  3) replaces the single x
:. %=
dx
5(4x  3)
4
X the diff. of the function (4x  3)
4 4
= 5(4*  3) X 4 = 20(4*  3)
y = cos(7x + 2)
;
7 sin(7x + 2)
DnnnDnaoDDDDnDDDLinnnnDnnnannDDnDnnDana
Right, now you differentiate these:
6
1. y = (4x5) 
3 ~x
2. y= e
3. j> = sin 2x
2
4. y = cos(x )
5. y= ln(3 4 cosjc)
The results are on frame 10. Check to see that yours are correct.
175
\
Differentiation
Results: 10
1.  <n 6
j = (4x5) ^
~y.=
= (s(Av
6(4x5) A = 1A(A
<n 5 .4 V  ^\$
24(4x5)
3 x
2. .y = e
^= e
3
^(l)=e 3
>;
3. v = sin 2x f
= cos 2x.2 = 2 cos 2x
ax:
2 2 2
4. ^ = cos(x ) ^ = sin(x ).2x=2x sin(x )
9. j = cos 3 (3jc).
10. j/=log 10 (2xl)
Results: 11
sin2x
6 . ^=e ^=e sin2x 2cos2x = 2cos2x.e sin2 *
dx
? dy
7. y= sin x f= 2 sin x cos x = sin 2x
dx
8. y= In cos 3x / = ^ (3
dx cos 3x v
sin 3x) = 3 tan 3x
3
9. y= cos (3jc) ^ = 3 cos 2 (3x).(3 sin 3x) = 9 sin 3* cos 2 3x
1 2
10. j,=log i0 (2xl) ^ = (2c _ 1 )lnl0
2=
(2x _ 1 )lnl0
176
Programme 6
1. Products
Ify = uv, where u and v are functions of x, then you already know
that
dy _ dv ,
du
ax ax ax
= 3
e.g. If y a: , sin 3x
dy
then = x 3 3 cos 3x + 3x sin 3x
.
dx
= 3x 2 (x cos 3a: + sin 3x)
2*
So what is the differential coefficient of e In 5jc?
13  = (i t 21 5 ,)
2x 2x =
for y= e In 5x, i.e. u = t , v In 5x
^=e
dx
2 *L.5
5x
+ 2e 2 *ln5;t
2
= e *(i + 2 In Sx)
Now here is a short set for you to do. Find f when
dx
2
1. y =x tanx
sx
2. y= e (3x + 1)
3. _y = x cos 2x
4. _y = x3 sin 5x
2
5. ^ =x In sinh x
Ill
Differentiation
Results:
14
1. j=x 2 tanx .'. ? = x 2 sec 2 * + 2x tan x
dx
= x(x sec 2 x + 2 tanx)
2. >> = e
5
*(3x +1) :.^=e sx .3 + 5e sx (3x + l)
= s s
e *(3 + 15x + 5) = e *(8 + 15x)
*
5. j> = x2 In sinhx .".
f
= x2 . cosh x + 2x In sinh x
dx sinh x
= x(x coth x + 2 In sinh x)
2. Quotients
Example 1 If v =
sin 3x < (*+ 1)3 cos 3s sin 3s.l
x+1 '
dx (jc + 1>2
2
*!lnx.2e 2x
Example 2. Ifj/ = ^
,
dy =
dx
.
e
?
e
4x
p
(I21nx)
21nx
X
=
You do this one. If y
x2
j >
'
~T
dx
=
178
Programme 6
*
i sin
_
~
2(* sin 2x + cos 2a:)
v.3
dy dv du
So: For_y=wv, ,
..(0
a* dx dx
du dv
dy_ _ dx dx
forv=^ (ii)
dx v
Be sure that you know these.
1__
n SGC
2
X
cos x
In the same way we can obtain the diff. coefft. of tanh x
" dx cosh 2 *
x
cosh 2 *  sinh
2
*
2
cosh *
1
2
2 seen *
cosh *
2
.. j (tanh *) = sech *
Add this last result to your list of differential coefficients in your notebook.
So what is the diff. coefft. of tanh(5* + 2)?
179
Differentiation
d_
(tanh(5x + 2)]= 5 sech (5x
2
+ 2) 18
dx
d
for we have : If i
tanh x
i

= sech x
7
= 5 sech 2 (5x + 2)
Fine. Now move on to frame 1 9 for the next part of the programme.
19
Logarithmic differentiation
are used when there are just twofactor functions, i.e. uv or. When there
v
are more than two functions in any arrangement top or bottom, the diff.
coefft. is best found by what is known as 'logarithmic differentiation'.
In y= In u + In v  In w
Now differentiate each side with respect to x, remembering that u, v, w
and y are all functions of x. What do we get?
180
Programme 6
20 1 dy 1 du 1 dv 1 dw
y dx u dx V dx w dx
dy _uv t\ du 1 dv 1 dw \
dx w \u dv v dx w dx I
If ~
find T
dx
y= .". In y= In (x )
2
+ In (sin x)  In (cos 2x)
y dx
dy
=
1
x
.
2x A
+
sinx
1
cosx
cos 2x
1
(2 sin 2x)
= + cot x + 2 tan 2x
x
.
..
dy = x sin x
r 
2 1 ,
1 cot
.
a:
,
+ 2 tan 2x
.
dx cos 2x \x )
This is a pretty complicated result, but the original function was also
somewhat involved!
You do this one on your own:
If y =x 4 e 3x tanx, then
dx
181
x
Differentiation
dy 3x 4 secx 2 22
~x 4 e 3X tanx 
j. \
+ 3 +
dx tanx
1 dy _ 1 j
1
3X 3e +
_44x + ir
.y dx x tanx
sec2jc
= 1+3 +
x tanx
2
4y 3x 4 sec x
.
..f=x
4
e
JJC j.
tanx
(
+ 3 +
dx { tan x
There it is.
Always use the log. diff. method where there are more than two func
tions involved in a product or quotient (or both).
Here is just one more for you to do. Find^ given that ,
=
y z? cosh 2x
x
J=^r
dx x cosh 2x
(4^2tanh2x
\ x 23
Working. Check yours.
e
4
y= 5 In 7 = ln(e *) ln(x 3 ) In (cosh 2x)
x cosh 2x
"
I *1= 4^ 4e 13 *
3x
2 
1
2 sinh 2x
^ dx e x cosh 2x
= 4 2tafth2x
x
"
^_.
.. ^ rr
I
42tanh2x
3 1
dx x cosh 2x I x I
Well now, before continuing with the rest of the programme, here is a
revision exercise for you to deal with.
182
Programme 6
3x
2. e sin 4x
,, sin2x
2x + 5
(3x + 1) cos 2x
5. xs sin 2x cos 4x
When you have finished them all (and not before) turn on to frame 25
to check your results.
183
Differentiation
Solutions
25
r\ . dy \ \
,ln4* i . .
1. (!)  ^%4=^_
(ii) v = In sin 3x .'.
i = 
dx
^
3x
sin
3 cos 3x
= 3 cot 3x
3 3x
2. v = e * sin 4jc .'. p = e A cos 4x + 3e 3 * sin 4x
dx
3
= e *(4 cos Ax + 3 sin 4x)
(3x + l)cos2x
4. ^=i 2*
e
dy
7 dx
1
3.x
1
+ 1
3+
cos 2x
* (2 sin 2x)  ?e 2x
3
 2 tan 2x  2
3jc+ 1
(3x + ) cos 2x 3
2 tan 2x  2
1 (
2*
dx e bx + 1 I
5. _y = x5 sin 2x cos Ax
.'.
In jy = ln(x 5 ) + ln(sin 2x) + In (cos Ax)
1 dy^\. 5x^ 2cOS
j ox x sin
?
2x
C
+
cos
*
4.x
( 4 sin4x)
= + 2 cot 2x4tan4x
x
So far so good. Now on to the next part of the programme on frame 26.
184
Programme 6
26 Implicit functions
routine.
2
+y 2 = 25, as stands, an example of an function.
x it is
DDDnDaDDnnnDannnDDDDDDnnDDnnnDDDnnnDDn
Once again, all we have to remember is that y is a function of x. So, if
2 2
x +j; = 25,letusfind^.
ax
If we differentiate as it stands with respect to x, we get
2x + 2y Q
dx
=
2x + 2y^
dx
=
dy . dy = x
:. yf=x
dx
..
f
dx y
As you will have noticed, with an implicit function the differential coef
ficient may contain (and usually does) both* and
185
Differen tiation
y
DnnananoDnDnDDDnnDDDDDDDnnnDnnnDDanDaD 28
Let us look at one or two examples.
If x
2
+ y2  2x~ 6y + 5 = find^ and^p at x= 3,j> = 2.
Example 1. 0,
dx dx
:.(3y6)g = 22*
dy _ 22x _ 1 x
dx 2y  6 _>>  3
(y3)(l)(lx)^
~,
Then
^ rf,ir>
v _ dJ' x l

dx
2
dx dx\y3) iy~3)
2
Oy)(lx)^
dx
_
(y3) 2
d 2 y. (32)(l3)2 _ l(4) .
dx 2 (23) 2 1
At (3,2) f
Jx
= 2, ^
dx
= :
2
Now this one. If x + 2xy + 3y 2 = 4, find^
Away you go, but beware of the product term. When you come to 2xy
treat this as (2x)(y).
x 2 +2xy + 3y 2 =4 _
2x + 2xfx + 2y + 6y dx
dy __(2x + 2y)__(x+y)
dx (2x + 6y) (x + 3y)
And now, just one more:
If x3 +y 3 + 3xy 2 = 8, find ^ Tum fQ frame 2Q fof fhe solution
186
:
Programme 6
30 Solution in detail
x z + y i + 3xy 2
3x 2 +3y 2 ^+3x.2y^+3y 2 =
. dy _ (x
2
+y 2 )
" ~dx (y
2
+ 2xy)
That is really all there is to it. All examples are tackled the same way.
l
The key to it is simply that y is a function of x' and then apply the
'function of a function' routine.
31
Parametric equations
In some cases, it is more convenient to represent a function by expressing
x and y separately in terms of a third independent variable, e.g. y = cos It,
x = sin t. In this case, any value we give to t will produce a pair of values
for x and j% which could if necessary be plotted and provide one point of
the curve of y = f(x).
The third variable, e.g. t, is called a parameter, and the two expressions
for x and y parametric equations. We may still need to find the differen
tial coefficients of the function with respect to x, so how do we go
about it?
Let us take the case already quoted above. The parametric equations
of a function are given as y = cos 2t, x = sin t. We are required to find
2
dy d y
expressions ciorf and ry
,
187
Differentiation
could only give us something called pf which has no meaning and is
very meaning of fd
y
2
^ 33
2
d y d tdy\ dl , . \
Therefore we say
J ^(4
dx\
sin t)
'
=(4
dt\
sin t).
I
dx
.
2
d y A 1
rn = 4 COS t. = 4
dx cos t
gi
Let us work through another one. What about this?
The parametric equations of a function are given as
y= 3 sin#  sin 0,
3
x = cos 3
Find ~ and 2
dx dx Turn on to frame 34.
Programme 6
dy
~ = 3 cos 6  2
34 3 sin cos
.'.
c?x
X = cos !
3cos 2 0(sin6l)
d0
= 3 cos sin(
3
dx d6
dy _dy d6 _
^"Tdx
3 cos d (1
V
 sin
.
d)
^ 3 C os 2
i

sin 5
3
3 cos
j = cot
3 cos 2 fl sin d
l
OX
Also
d^y_d
dx
1
( cosec 2 )
3 cos 2 6 sin
1
dx 2 3 cos
2
sir?0 V
Now here is one for you to do in just the same way
2  3t 3 + It
If x y
J !
find^
l + ? '
l + t dx
W/?e >>ow /zflve done it, move on to frame 35.
dy = }_
35 dx 5
.
23? dx (1+Q (3) (2 3Q
For
1 +t It (T+Tp
3 + 2? dy (1+0 (2) (3 +20
1+? dt
dx 33/ 2 + 3? 5
* (i +o 2
^(i+0 2
dy. 2 + 2?32? :
dt (Ho 2 (TT77
2
dy_ <^ <# _ 1 (1+ ?) _ 1
2 ~5
dx dt' dx (1+?) 5 dx
And now here is one more for you to do to finish up this part of the work.
It is done in just the same way as the others.
f and
find
ax ft
ax
189
Differentiation
dx
.'.
jT = a( sin 8 + 8 cos 8 + sin 8) = a 8 cos <
do
dy dy dd . . 1
~=^7 = a8 smd
.
.. = tan 6
dx dd dx ad cos 6
$
dx
= tan0
d y_ d u .. du .. dd
rr= 7 (tan0) = ^(tan0).
_ !
1
sec
0 COS 8
. d*y_ 1
dx 2 3
'
00 cos
190
Programme 6
37 Test Exercise VI
Do all the questions. Write out the solutions carefully. They are all quite
straightforward.
(vii) e
2x
sin 3x (viii)
(x + lf j
^ e
xcos2x
sin x
dy d 2y
2./ If x
2
+y 2  2x + 2y = 23, find and at the point where
/ f
dx dxf2
V x =2, y = 3.
x 3 +y 3 +4xy 2 = 5
simplest forms.
191
Differentiation
Further Problems  VI
\
,.. N . ,
(li)ln(secx + tanx)
,
. >. : ,....
(ni>
\/
.
4
suvx 3
cos x
f
e
3. If v is a function of x, andx = 7
e' + 1
sln 5*
. 5. Differentiate: (i) y= e (u )y = n  ^jjjl
2
<i v
9. If x
2
+ 2xy + 3y 2 = l, prove that (x + tyf "di +2 = 0.
192
Programme 6
12. The parametric equations of a curve are* = cos 2B,y=\ + sin 20.
2
c ,dy an d y at =
. .
Show 2 OTX
16. that y= e" sin 4mx is a solution of the equation
f^ + 2*f + y + *> = (
fcf
19. If> = e" (A cosh ?f + B sinh qf) where A, B, q and fc are constants,
show that
X 5
20. If sinh y= {T* :\ show that
4 + 3 sinh x'
= ,
u
dx 4 + 3 sinh*
193
Programme 7
DIFFERENTIATION APPLICATIONS
PART1
Programme 7
Y
u fy = dy
where m = slope =r
i
ox dx
f
y = mx + c
y^j, i.e. 5y = x + 7
DnanDDnnDaDDnnaDnnnnannnDDnDaDDnDnnDnD
Sometimes we are given the slope, m, of a straight line passing through
a given point {x x y\ ) and we are required to find
, its equation. In that
case, it is more convenient to use the form
y~y\ =m(xx l )
For example, the equation of the line passing through the point (5,3)
with slope 2 is simply which simplifies to
195
Differentiation Applications 1
y~3 = 2(x~5)
\.e.y3 = 2x\0 y = 2x7
DnnnnDDDnDDnDDnDDDaDaDDDDDnDnnDDDDDnDD
Similarly, the equation of the line through the point (2,1) and
having a slope  is
y(i)={x(Z)
:. y + l=^(x + 2)
2y + 2 =x + 2
y= x
2
So, in the same way, the line passing through (2,3) and having
slope (2) is
y = i2x
For y(S) = 2(x2)
.'
y+ 3 = 2x + 4 :. y= 1  2x
DDDDanaDnDnDaaanDDnanoDDnDDnnnaanDnnnn
Right. So in general terms, the equation of the line passing through the
point {x l ,y l ) withslope m is
Turn on to frame 5.
196
Programme 7
DnDDnnnDnnnDDaDDDDDDDnnDaDnnDnnnDanDnD
So for one last time:
P has coordinates (4,3) and the slope m of
If a point a straight line
through P is 2, then the equation of the line is thus
y  3 = 2(x  4)
= 2x8
.'.
y=2x5
The equation of the line through P, perpendicular to the line we have
just considered, will have a slope mi , such that m rri\ = 1
j,3 = I(*4)
= x/2 + 2
2y = Ax  5 and 6y = 2 3x
If we convert each of these to the form y= m x + c, we get
5
(i) y = 2x  and (ii) j> =
1
^+
1
197
J1
Differentiation Applications 1
Result:
DDDnDnDnnDanQanannnnDnapaDDDnDQDDDDDaD
For if we convert each to the fornix = mx + c, we get
x 2
(i) y= 3x~5 and 7
=3+
y= =
()
2~J and y x + 6
_ 1
m;m l mmifl
l :. Not perpendicular.
(iii) j = 3x + 2 and _y = ~ 3
m3i 1
;m =1 :. mm 1
=\ Perpendicular.
m =j ; m, = 5 :. mm l
= 1 Perpendicular
Remember
to each other, then
that if y =mx + c and y = m x x+ d are perpendicular 8
m wj =1, i.e.m, =_
Here is one further example:
A line AB passes through the point P (3,2) with slope ~\ Find its .
198
:
Programme 7
y+ 2 2 + 1
1
.y = "2
2
2y+x+l=
1 1
_
Equation of CD: slop 2 WIl = = 2
m
7 (2) =2(jc 3)
^ +2 == 2x6
^ =
2x8
So we have:
D
r /" 2x8
10 m rrii =1
naDnnDanDnnDnnnnnDDDonDDDnnaDDnDDDDnDD
And now, just one more to do on your own.
The point P(3, 4) is a point on the line y = 5x\l.
Find the equation of the line through P which is perpendicular to the
given line.
That should not take long. When you have finished it, turn on to the
next frame.
199
Differentiation Applications 1
5y + x = 23
^4=I(jc3)
Sy  20 = x + 3 :. 5y + x = 23
nooaaaQOQOonQananaannaanQQonanana
Tangents and normals to a curve at a given point.
The slope of a curve, y =/(*), at a point P on the curve is given by the
slope of the tangent at P.
*lf
7
It is also given by the value of ^ at the point P,
' X
DDnDODDDDnDDDDDDDDDDDDDnDDaDDDDnnnDDDD 12
Correct. This is sufficient information for us to find the equation of the
tangent. Let us do an example.
e.g. Find the equation of the tangent to the curves =
2x 3 + 3x 2 ~2x 3
at the point P,jc = l,y = 0.
h,
2
~r = 6x + 6x 2
ax
Slope of tangent (^1 :
6 + 62= 10, i.e.m= 10
\dx( x= 1
200
Programme 7
13 1
Slope of normal =
1_
Slope of tangent 10
DDDDDDaDDDDnDnnnnanDDnDDDDDDnDnDDnaaD
The normal also passes through P, i.e. when x = 1 ,y = 0.
.'.
Equation of normal is: y0=  (x  1)
10y = x + 1 \0y+x=l
That was very easy. Do this one just to get your hand in:
Find the equations of the tangent and normal to the curve
y = x 3  2x 2 + 3x  1 at the point (2,5).
Off you go. Do it in just the same way.
14 Tangent: y = Ix  9 Normal: ly + x = 37
y = x  2x 2
3
+ 3x  1
dy dy
.
y 5 = l{x  2) Tangent is y
~
= Ix  9
1
For normal, slope =  =I
slope of tangent 7
:.y5={x2)
7y35=x + 2
Normal is ly + x = 37
201
Differentiation Applications 1
2x + 2y^ +
dx
3x^+3y
dx
=
dy _ 2x + 3y
dx 2y + 3x
Therefore, at x = l,y = 2,
dy..
dx
dy _ 2 + 6 dy = J&_
16
,
'
dx 4+ 3 dx 1
7y14 = 8x + 8
.'.
Tangent is ly + 8x = 22
1 7
Slope =
Slope of tangent 8
7,
Normal passes through (1,2) :.y 2 Jxl)
8j16 = 7x7
Normal is 8y = Ix + 9 That's that!
Now try this one:
Find the equations of the tangent and normal to the curve
x3 +x 2 y+y 3 7 = at the point x = 2,j = 3.
202
2
Programme 7
Results:
17
Tangent: 31j + 24x = 141 Normal: 24y =31*+ 10
3x
2
+x 2
^+2xy
dx
+ 3y 2
^
dx
= Q
2 ^L = (3x d = _ 3x 2 +2xy
(x +3y 2 )$ 2
^ 3X +lXy)
+2xy)
+2xv)  dx 2
dx x +3y 2
A t(2 '^
AU/ 3) ^i = i2i2 = _24
" dx 4 + 27 31
Find the equations of the tangent and normal at the point for which
r= 2.
First find the value of
d
 when t = 3.
dx
3r . dx_{\ +r)33f _3 + 3t  3t
x 2
1 +f " dt (1 + f) (1+0 (1 +0 :
r . dy _(1 +t)2tt 2 _ It + 2f 2  1
2
_2t + t
y 14/ " n A2 2
1 +r
rlt
'"
dr + r)
(1 J (1 + r) (1 + ff
dy ^dy dt _ 2t + t
2
(1 + t)
2
_ 2t + t
2
;. Att = 2&= ' <2x
dx dt '
dx (1 + tf '
3 3 3
To get the equation of the tangent, we must know the x and_y values of a
point through which it passes. At P
2
3f _ 6 _6_ _ ? _4
jc
:
1 +r 1+2 3 '
J 1 +r 3
203
Differentiation Applications 1
77
normal to the curve at t = .
Results:
Tangent: 2y + Ax = 3 Normal: 4y = 2x + 1
20
Working:
dy
y  cos 2f .'. ~=2 sin 2t = 4 sin t cos t
dt
dx
x = sin t .'.
7T = cos t
dt
dy dy dt = 4 sm t cos f
_
=
j
:J7
1 =4 sin
.
dx dt dx cos ?
Tangent is y ^ = 2(jc
 ~) .'.
2y  \ = Ax + 2
.".
2y + Ax = 3 (Tangent)
Equation is y \
2
= \_
2
(*4)
' 4^2 = 2^1
'
4y = 2x 1 (Normal)
204
6
Programme 7
21 Before we leave this part of the programme, let us revise the fact that
we can easily find the angle between two intersecting curves.
that point, and ~= tan 6, where is the angle of slope, then we can
use these facts to determine the angle between the curves at their point
of intersection. One example will be sufficient.
e.g. Find the angle between^
2
= 8x and x 2 + y 2 = 16 at their point of
intersection for which y is positive.
=
8 11314
~ =
3314
or
19314

Now we have to find f for each of the two curves. Do that
ax
2 dy. . dy _ 4 _ 1
(i) y = Sx :. 2y " dx 1099
22 =1099
dx
= 4742'
y 3641 0910
tan 0! /. di
2x + 2.y^ =
dx
^
dx
=
_x = _F65_7 = 04551
y 3641
205
Differentiation Applications 1
Curvature
The value of at any point on a curve denotes the slope (or direction)
of the curve at that point. Curvature is concerned with
how quickly the
curve ischanging direction in the neighbourhood of that point.
Let us see in the next few frames what it is all
about.
24
Let us first consider the change in direction of a
curves =/(*) between
the pomts P and Q as shown. The direction of a curve is measured by the
slope of the tangent.
fix)
206
.
Programme 7
the arc PQ
25
i.e. we must know the change of direction, but also how far along the
curve we must go to obtain this change in direction.
Now let us consider the two points, P and Q, near to each other, so
that PQ is a small arc (= 6 s). The change in direction will not be great,
so that if 8 is the slope at P,
Y '
 then the angle of slope at Q can
be put as 8 + 88.
26 between 8 and s, and usually all we have is the equation of the curve,
y =/(x) and the coordinates of P. So we must find some other way
round it.
arc PQ = 8s =
207
Differentiation Applications 1
arcPQ = Ss=R50 27
5s = R60 :. ^4
8s R
If 5 s > 0, this becomes = which is the curvature at P.
nDDDDnnDDQDDDDDDDnanDOODDDDDODDDDnDDD
This is not so. For the curvature = and we have just shown that
j
g K is the denominator,
. so that a large value for R gives a small
value for the fraction and hence a small value for the
curvature.
You can see it this way. If you walk round a circle with a large radius
R
then the curve is relatively a gentle one, i.e. small value
of curvature, but
if R is small, the curve is more abrupt.
So once again, if R is large, the curvature is
208
Programme 7
R is large, the
29 If curvature is small
Let us consider our two points P and Q again. Since 8 s is very small,
there is little difference between
the arc PQ and the chord PQ, or
between the direction of the chord
and that of the tangent.
ds
ds dx
3/2
3/2 3/2
Now sec
3 2
= (sec 0) = (1 + tan 2 0) = {
1 + (^ ) }
2
d y
3/2
dd 17 l
1+ ()
3/2 R =
ds R"
(
1+ >
2
1 2
dx
209
Differentiation Applications 1
R
2
dx
..
1
..*= Ax 2 =s
and
S =8* 3=
^
At (2,2)
v
'
*: =
dx
! = !.
4 '
^
dx5
= 8=i
8
, R= {ut!H'Ui!r =(2) .
ft=2V2
:. R = 2\/2 = 2828 units.
Example 2. If y = x + 3x 2  3
, find R at x = 0.
rf>\
2,3/2 31
R=
2
dx
i
*: =
dx
.**,A^n^i
i+^_^2 Ao.. dy
,($.
i
dx" dx 2
Find the radius of curvature of the curve y 2 =^ at the point (l,4)
When you have finished, check with the solution oh frame 32.
210
Programme 7
R= 521 units
32
Here is the solution in full.
2
dy_ ^_ dy _ 3x
=
y dx = 4 " dx Sy
.
y "
W
(dy\ f. {251 3/2
1 + } + _9J
> _ r 16) _ \16J _ 8 125 _ 1 25 _5_
R =
64 24 24
2
dx '
R= 521 units
 =
e.g. \fx = sin0 andy = 1 cos 0, find Rwhen0 = 60
>
* = 0sin0 =lcosfl l
dy = dy_dB_
dY dx dd dx '
y  1  cos 8 /. ^ = sin
I
I
. dv . 1 _ sing
dx 1  cos 1  cos
dx dx\lcos0) dd \lcosd\dx
(1  cos 8)
2
1  cos 8
0 cos 2 8 2
sin 6 _ cos  1
_ cos 1
=
(1  cos 0)
3 "
(i  cos 0)
3
(i  cos ey
l l
:. At = 60,$ = = 4
dx* (T^l? i
.
R _03p/ 2 _2 _8 3
_ R = 2 units
211
,
Differentiation Applications 1
You notice in this last example that the value of R is negative. This
merely indicates which way the curve is bending. Since R is a physical
34
length, then for all practical purposes, R is taken as 2 units long.
If the value of R is to be used in further calculations however, it is
usually necessary to maintain the negative sign. You will see an example
of this later.
Result:
R= 3 units 35
For x=2 3
cos 6 :. dx
do
= 6 cos 2 (sin
v 6)' = 6 sin 8 cos
2
(
M U ${*.}. U^,} dd
dx
_
6
sec
2
sin 9 cos
2
8
1
6 sin0 cos 4
A tg = 45^=
'dx
.*
6(i)(l)
=^ = ^2
6
R=
h&r ()'
2
3/2
^_Z
2
V2 2V2
dx 3
.3.2V2 =
3
2V2
' R= 3 units
212
j
Programme 7
\
k=y l + LC =yi + R cos 6
1
1 c \
That is, h=X! Rsinfl
{
rv /
\
\
r* \ R 1 k =yi + R cos0
\ //
\
\
'Li /\P{x,.y,) where Xi andj'i are the co
I 1
ordinates of P, R is the radius of
yf
l
V i
curvature at P, 6 is the angle of
< h
" =1
slope at P, i.e. tan
of the centre
Example. Find the radius of curvature and the coordinates
37 of curvature of the curve y= _
3x
at the point (2,3).
dx 1 W
2
d2 y_d
=^H3^}=2(3x)(l)=^3
dx
:. Atx = 2,
^.=^=2
2
dx 1
(3/2
213
Differentiation Applications 1
NOTE: If, by chance, the calculated value of R is negative, the minus sign
must be included when we substitute for R in the expressions for and k. ft
Example. Find the radius of curvature and the centre of curvature for
given by
h =Xi Rsin0 \ , . ,
_ .I and in these expressions
r
k=y
,
l
,
+ R cos0 J
i.e. tan0
\dx) ]
Now, in the problem stated above, 8 is a parameter and not the angle
214
Programme 7
y= sin
2
t,x = 2 cos f , at the point for which t = j
Start off by finding the radius of curvature only. Then check your
result so far with the solution given in the next frame
before setting out
2
7y
= sin f .'. p = 2 sin t cos t
dt
_ dx
x = 2cosf .. =2sinf
.
dt
dy _dy dt _2 sin f cos t _ _^ c
f
~dx dt' dx 2 sin f
A..".fMi..f4
A*gi~')n~'}r.i,s
. d 2 y_ 1
""
dx 2 2
R= Wi
dV " i
2 2
dx
_25n/5_5\/5_ 5 (22361)
8 4 4
=
lM805 = _ 27951
4
R =2795
215
Differentiation Applications 1
Also Xi = 2 cos 60 = 2. = 1
7i = sin'
Calculate them and when you have finished, move on to the next frame.
43
Results: /z=025; fc=l75
00969
h = 025
216
.
Programme 7
Find the equation of the normal at the point for which 8 = = 45.
i d2 y
6. Given that x = 1 + sin 8, y= sin 8 5 cos 28, show that 75 = 2. Find
the radius of curvature and the centre of curvature for the point on
this curve where 8 = 30.
217
. .
Differentiation Applications 1
2 2
4. If x y + xy x*y 3 +16 = 0, find^ in its simplest form. Hence
find the equation of the normal to the
curve at the point (1,3).
2 2
6. If 2x +y  6y  9x = 0, determine the equati6n of the normal to
the curve at the point (1,7).
8. Find the equation of the normal at the point x = a cos d,y = b sin 0,
of the ellipse ~ +^ = 1 . The normal at P on the ellipse meets the
major axis of the ellipse at N. Show that the locus
of the midpoint
of PN is an ellipse and state the lengths of its
principal axes.
218
Programme 7
X ~X
9. For the point where the curve y =~r j passes through the origin,
determine:
10. In each of the following cases, find the radius of curvature and the
coordinates of the centre of curvature for the point stated.
(1)^4=1 at (0,4)
y = Ax x  3 at x = 25
2 2
(ii)
12. If 3ay
2
= x(x a) 2 with a > 0, prove that the radius of curvature at
the point (3a, 2d) is
13. If x = 26 sin 20 and y = 1  cos 20, show that/ = cot and that
ax
Hr2 = i. z~z
4 If P is the radius of curvature at any point on the
dx 4sin
2 = 8y.
curve, show that p
2
1 5 Prove that the centre of curvature {h, k) at the point ?(at 2at) on ,
2 2 3
the parabola y = 4ax has coordinates h = 2a + 3at =
k 2at , .
219
Differentiation Applications 1
18. Find the radius of curvature and the coordinates of the centre of
curvature of the curves = 3 lnx, at the point where it meets the
xaxis.
19. Show that the numerical value of the radius of curvature at the point
. If C is the centre
of curvature at the origin and S is the point (a, O), show that
OC = 2 (OS).
2
20. The equation of a curve is 4y =x 2 (2~x 2 ).
(i) Determine the equations of the tangents at the origin,
(ii) Show that the angle between these tangents is
1
tan" (2\/2).
(iii) Find the radius of curvature at the point (1,1/2).
220
Programme 8
DIFFERENTIATION APPLICATIONS
PART 2
Programme ,
indicates the principal value of the angle whose sine is the value x;
to see what we mean by that, move on to frame 2.
05 = 30
1
sin"
to 180
and no other angle!
223
Differentiation Applications 2
Principal value of 9 = 45
sin"
1
07071 =45
DnnnnDDannnDDnDnnDnnnnnDnnnDnanDDnDnnD
1
In the same way, we can find the value of tan" y/3.
Iftan 6 = V3 = 1 7321 then 9 = 60 or 240. Quoted in the range 0 to
,
x, X
\ /
The principal value of the angle is the one nearer to the positive OX
direction, i.e. in this case, tan" \J3
1
=
tan"V3 = 60
" Y,
Y,
224
Programme 8
tan" (1)
1 = 45
.135
\
Work through it carefully and then check your result with that on frame 6.
1
cos" (0866) = 150
For we have
cos"
1
(0866)= 150
1 1 1
So to sum up, the inverse trig, functions, sin *, cos" *, tan" *
indicate thep v of the angles having the value
of the trig, ratio stated.
225
Differentiation Applications 2
principal value
y= sin"
1
* .".
x = siny
dx dy
j= cosy .. j =
dy dx
dy._ 1
8
dx cosy
2  2  x2
.'.
cos y = 1 sin ^= 1 (since x= sin y)
2
.".
cos y = \/( 1 ~x )
"'
dx VO x 2 )
d \
{
.
sin
_,
x\ }=z 1_ ^ x
dx\ j V(l^ 2 )
Go through the same steps and finally check your result with that on
frame 9.
226
Programme 8
d_ { '
cos lx
dx [ ] Vo^
Here is the working:
Let y= cos x .'.
x = cos j
dx dy 1
dy dx smy
cos
2
y+ sin
2
y= 1 .'. sin
2
y = 1  cos 2 _y = 1  x2
2
siny = V(l x )
=
dy
" dx~sj{\x 2 )" "'cos
d _, 1 . i 1
x]=j^, \
!
dx )
1
(i)
iH sin"*rv(i* 2
)
Different only in sign.
(n) cos x = 7 7;
1
Now you find the differential coefficient of tan x. The working is
slightly different, but the general method the same. See what you get and
then move to frame 10 where the detailed working is set out.
10 d
dx
t*
{ tan
1
x)= }
1
l+x 25
l
Working: Let y= tan x .".
x = tan_y.
dx 2
r = sec
2
y= 1 + tan y = 1 + x2
dy 
^=i
dy
+JC 2
dx 1
l
+ x2
dx{ {
tan
1
x
j
}
=
1 + x2
*
iH
cos
~
v vfe }
=
2
)
(ii)
d U i 1 1
v tan x )
=. n (iii)
dx\ + x2 I 1
Copy these results into your record book. You will need to remember them.
On to the next frame.
227
. 1
Differentiation Applications 2
Example 1. Find ^ ,
given thatj> = (1 ~x 2 ) sin"
1
*
* I
.2 =
2
~
dx 1 + (2x  l)
2
1 + 4x 2  4x + 1
2 + 4x
2
 Ax 2x
2  2x + 1
and so on.
12
Here you are. Here is a short exercise. Do them all: then check your
results with those on the next frame.
Revision Exercise
Differentiate with respect to x:
1
1 y= sin" 5x
1
2. y= cos" 3*
3. y = tan" 2x 1
4. y = sin' 1 (x 2 )
5. j=x 2 .sin 1
()
228
.
Programme 8
Results:
13
1 v = sin" Sx
1
:. ^ =
dx V{l(5^) 2 }
_ s = _
V{1  25x }
2
V d 1 2
3 'v = tan"
tan
1
2v
2x ^
1+(2jc)2
? 
2 j^^
4. ^ = sm
1
(x
2
) ^=
^ 7r
V{l(x 2 2
)
1
.2x =
f~ V(lx
2x
4
)
1 a
5
'' ta
(f)S* 7p^)i^^(f)
2V{iV
=
vcfe)
+ ^ sin
"1
(!)
Right, now on to the next frame.
In just the same way that we have inverse trig, functions, so we have
inverse hyperbolic functions and we would not be unduly surprised if
their differential coefficients bore some resemblance to those of the
inverse trig, functions.
Anyway, let us see what we get. The method is very much as before.
(i) y= sinh"
1
x To find ^
ax
First express the inverse statement as a direct statement.
y = sinh"
1
x .'.
x = sinh y .".
dy
= cosh y ' f =
dx cosh j'
We now need to express coshj in terms of x
2
cosh^ =y/(x + 1)
dy _ I d_( ,i 1 = 1 .
&mh X
dx V(* 2 + 1)" dx \ j V(*
2
+ 1)
Let us obtain similar results for cosh" 1 * and tanh" 1 * and then we will
take a look at them.
So on to the next frame.
229
Differentiation Applications 2
(ii) y= cosh"
1
x .'. x = coshy
dx dy = 1
 = sinhy ^ .
,
.. rT
dy dx sinhy
Now cosh
2
y  smb. 2 y = 1 smh 2 y = cosh 2 ^ 
.'. 1 = x2  1
2
.'. sinh y =\J{x ~ 1)
" dx
dy_
2
.'.
l =
" 1 )" d* cosh x V(x 2  { \
VC* 1)
dy _ 1
y = tanh *x
dx 1 x 2 16
for:
1
>' = tanh x .'. x = tanhj>
dy
>,= ltanh2 ,= lx 2 :.%^
tanhx U^
Now here are the results, all together, so that we can compare them.
1
Ijsinhxj = (iv)
V(x2 + i}
d_ 1
COSh_lx 2
(v)
dx { jV(x l)
sK'+ri? (v
Make a note of these in your record book. You will need to remember
these results.
Now on to frame 1 7.
230
Programme 8
1
Example 1. y= cosh j32x
. dy = 1 (2) = 2
" dx 2
V((3   7(9  1 2x + Ax 
2
2x) 1} 1)
2 2 1
= =
V(8  \2x + Ax )
2
V(*  3x +
2
2
2y/(x  3x + 2) 2)
Example 2. y= tanrf
1
l^)
, dy_ 1
2 = _! 1
" ^ ,_/3*\
1
2
4 _9^ 4
i
U / 16
= 16 1= 12
16 9x 2 '
4 169jc 2
1
Example 3. y= sinrT {tan jc }
2
ty  l
2 _ sec *
dx ^(tan 2 * + 1)' VWc * 2
= secx
I
O Here are a few for you to do.
Exercise
Differentiate:
1
1. y= sinh" 3x
2. y= cosh '
(
2 J
1
3. 7 = tanh" (tanx)
1 2
A.y = sinh V(* ~ 1)
1 2 *)
5. j= cosh" (e
231
Differentiation Applications 2
Results:
19
dy _ 1
v 3=
1 .
1. y= sinh" 3x
dx V{(3^) 2 + 1}' 2
V(9* +l)
'(?) t^ 1
._
5
2
5
(?) ir M ^_ l)
// 25x
2
4 \ V(25x 2 4)
2
1 2 S6CX
jtanh" (tan*) ^= t
_ tan 2 ^ . sec x=
1 tan"*
1 2
^=
/
sinh { v (x l)}
dy _ 1
J_, , ,ci
^V(x 2
1+1)2 (X 1}
^)=V(* 2 1)
2X
1 2 2
2e
^ = cosh (e *) :. j =
2 2
,2e 4JC
dx V{(e *) 1} V(e D
All correct?
1 1 1
1
sin" x 2
sinh" x 2
V0* ) V(* + 1)
1
1 1
1
cos" * cosh" *
V(l* 2 ) 2
V(*  1)
1
1 1
tan" x tanh"
1
*
1+* 2 l* 2
Itwould be a good idea to copy down this combined table, so that you
compare and use the results. Do that: it will help you to remember them
and to distinguish clearly between them.
232
Programme 8
21 Before you do a revision exercise, cover up the table you have just copied
and see if you can complete the following correctly.
1.
_ i dy ^
If^sin*, ;
2. If, = co,',,g =
3. If, = tan*,g =
_1
4. y
If J = sinh A:,r=
dx
5.
1
If, = cosh" *, 7 =
1 =
6. If, = tanh *, r
Now check your results with your table and make a special point of
brushing up any of which you are not really sure.
22
Revision Exercise
Differentiate the following with respect to x:
1
1. tan (sinh;>c)
1
2. sinlT (tan x)
1
3. cosh" (sec x)
1
4. tanh" (sin x)
5 sin'
.
If)
Take care with these; we have mixed them up to some extent.
When you have finished them all  and you are sure you have done
what was required check your results with those on frame 23.
233
Differentiation Applications 2
Solutions 23
1. j^tanCsinh*) jLjtan" 1 *} =
^
'
dy
r =
dx
,
1 + sinn x
1
. , 7 .
,
cosh x =
coshx
rj = seen x
cosh x
,
2. j,
l
= rinlf (tan*)
fj^ x =
} ^Ti)
.'.
dy
f
= ;
.
2
1
dx ^(tan * + 1)
.. sec^x =
i
, *
yseCjc
2
sec 5
= sec x
1
cosh =
3. y= cosh" (sec x)
^( "^J ^^l)
dy
= , ,
1
dx V(sec 2 x 
, r; sec x. tan x = ~r 5z
Vtan x
x
se c x tan
1)
secx
y=
1
tanlT (sin x)  
tanlT
1
x =
y^j
J
:.
dy
r =
dx
,
1
rr"
 sin x
1
cos x= 5xx
cos
cos
= sec x
5 =sin
1 x dT,\
i^" .,}., 7 ^ I
5

^ ii
a dx" * rvo )
If you have got those all correct or nearly all correct you now
know quite a lot about the differential coefficients of Inverse Trig, and
Hyperbolic Functions.
You are now ready to move on to the next topic of this programme, so
off you go to frame 24.
234
Programme ,
y = fix)
25
y = fix)
On to frame 26.
235
Differentiation Applications 2
y = f(x)
Y (max)
A
Point of S
^^^^/+
^ inflexion
y ^/S~ \ ^S. c

t + r '
\
X (min)
y^*""
"^^
]
o
X^ B
!
i

^>f I
; o ]
dec
^iX S^Z ^3 X
Therefore, we obtain the first rule, which is that for turning points,
dx
Turn on to frame 27.
236
Programme
% against jr.
dx Y A
y = fix)
y = f'(oc)
y = f"(x)
From the first derived curve, we see that for turning points,
dx
From the second derived curve, we see that
y d2
for maximum y, r^ is negative
f
for minimum j',
*y
75
.
is
.,.
positive
d2 y
for PofL 2 is zero
dx
Copy the diagram into your record book. It summarizes all the facts on
max. and min. values so far.
237
Differentiation Applications 2
We shall need both the first and second differential coefficients, so find
x x
them ready. If y =   2x + 5, then f =
dv
and
3 2 dx
d2 y
dx 2
238
Programme ,
30 dx dx*
DaDnDnDDnnDDDnDDnnnDnnDDDnnDnDDnDnaaaD
(i) Turning points occur at =
/. x2 x2 = :. (x2)(x+l) = :. jc = 2andx=l
i.e. turning points occur at x = 2 and x=l.
(ii) To determine the type of each turning point, substitute x = 2 and
cfy
then x = 1 in the expression for j
At x = 1 , ^
dx 2
= 2 1, i.e. negative /. x = 1 gives y mstx .
Also, we can see at a glance from the function, that when x = 0,y = 5.
A
6V6
31 We know that (i) at x = 1 7 max = 65
,
B
2
1 /3
 
5 x
(ii) atx = 2, ^m in = l
(iii) at x= 0, y = 5
Joining up with a smooth curve gives:
Y
X, ( \ 2 3 4 5 X
There is no point of inflexion like the point C on this graph. Move on.
239
Differentiation Applications 2
All that was just by way of refreshing your memory on work you have 32
done before. Now let us take a wider look at these
Points of Inflexion
('positive
"J
negative 
Which?
J
zero
Correct. The slope can of course be positive, negative or zero in any one
case, but there is no restriction on its sign.
DnDDnDDDnnDanDDDDnQDnnnnDaDnDDDDDnDDaD
A point of inflexion, then, is simply a point on a curve at which there is a
change in the d of b
240
Programme 8
direction of bending
DODDDDQDDDDDnDDDDDDDDDDDDnnnDODDDnDDDD
If the slope at a PofI is not zero, it will not appear in our usual max.
dy

and min. routine, for will not be zero. How, then, are we going to
find where such points of inflexion occur? Let us sketch the graphs of the
slopes as we did before.
L
% N\
LH
always negative.
dy_
dx In curve 1 ,
r reaches a minimi
dx
value but not zero.
x
In curve 2, r reaches a maxinr!
dx
value but not zero.
For both points of inflexion, i.J
x = X4 and x = x s J
d2 v
dx
So, is this the clue we have been seeking? If so, it simply means that to
find the points of inflexion we differentiate the function of the curve
d2 y
twice and solve the equation 72 ~ 0
That sounds easy enough! But turn on to the next frame to see what is
involved.
241
Differentiation Applications 2
This is perfectly true. Unfortunately, this is not the whole of the story,
cPy
for it is also possible for 7^ to be zero at points other than points of
inflexion!
d2 y
So if we solve 75 = 0, we cannot as yet be sure whether the solution
dx2
242
Programme 8
d2 y
37 In the case of the real PofI the graph of
^ crosses the xaxis.
2
d y
In the case of no PofI, the graph of
^ only touches the x axis
A*y
and k adoe s not change sign.
dx*
DDDnDDnnDDDnnDDnnnnDnnnnnDDDDnnDDnnnDn
This is the clue we have been after, and gives us our final rule.
y d 2
d2 y
For a point of inflexion, pi  and there is a change of sign of pi
as we go through the point.
try =
For a with change of sign
38 PofI,
dP
This last phrase is allimportant.
DGnnnnDnDnnnanDnnDnnnnnDDDnnDDnnnnDDDD
Example 1. Find the points of inflexion, if any, on the graph of the function I
y^jTx + 5.
d
(i) Diff. twice. =x 2
x2, 4$= 2x  1
d2 y_
For PofI, = 0, with change of sign. .'. 2x  1 = .'. x=
dx
1
If there is a PofI, it occurs at x=
(ii) Test for change
and a point
of sign. We take
just after x=
11
x, i.e.
a point just before
x = + a, where a is
x=
1
z, i.e.
a small positive
x = r~ a,
dry
quantity, and investigate the sign of pi at these two values of x.
Turn on.
243
Differentiation Applications 2
dx 2 ^ 1 39
Atx = a, ^A = 2(
(i)
dP~ 2( a)l = l2al 2
= 2a (negative)
:
2a (positive)
2
^ d
change in sign of H, as we go through x =
r
There is a
.'.
There is a point of inflexion at x=
If you look at the sketch graph of this function which you have
already drawn, you will see the point of inflexion where the righthand
curve changes to the lefthand curve.
Example 2. Find the points of inflexion on the graph of the function ZL]
y = 3x s  5xA + x + 4
d2 v
First, differentiate twice and solve the equation j^ = 0. This will give the
sure until we have then tested for a change of sign in ^ We will do that
.
in due course.
d2 v
So start off by finding an expression for Hj and solving the equation
dx
d2 y _
n
244
Programme 8
41 We have: y = 3x s  5x 4 +x + 4
/. ^
dx
=15x 4 20* 3 + 1
d 2 y_.
For Pof1, rj = 0, with change of sign.
comes the test for a change of sign. For each of the two values of x we
have found, i.e. x = and x = 1 , take points on either side of it, differing
from it by a very small amount.
(i) Forx =
(ii) For x = 1
At x =  a,  2 a
1
^ = 60(1 a) (l 1)
= (+)(+)() = negative
Change in sign.
Therefore, the only point of inflexion occurs when x 1 , i.e. at the point
x=\,y = Z
That is just about all to it. The functions with which we have
there is
Now turn on to the next frame and complete the Test Exercise awaiting
you. The questions are all very straightforward and should not cause you
any anxiety.
245
Differentiation Applications 2
y=
1
(i) sin" (3* + 2)
1
.... cos" *
(")/=
(j
(iv) y= cosh" (1
1
 3x)
1
(v) y= sinlf (cos x)
(vi) y= tanh" 5x
1
3. Find the stationary values ofy and the points of inflexion on the
graph of each of the following functions, and in each case, draw a
sketch graph of the function.
y = x  6x 2 + 9x
3
(i) + 6
(ii) y = x+
x
(iii) y =xe'
Well done. You are now ready for the next programme.
246
Programme 8
1 1 + tan x
1. Differentiate (i) tan {
tanxj
2
(ii) jcV(1 a: ) sin" 1 y/(lx 2 )
1
sin" *
2. If y =j(l _ x 2y prove that
d
(i) {\x*) =xy+\
3. Find when w
(i) y
J = tan1 '
f
dx \\\x
(ii) ^ = tanh M (
2x
2;
x + Jy = 1 and
d y
dx
2
=f
= x t
&5y
4 . Hence find the maximum and
minimum values of y.
9. If y
3
= 6xy x 3  1 ,
prove that f
= % _ ~ and that the maximum
3
value of y occurs where x = 8 + 2\/l4 and the minimum value
where x 3 =82Vl4.
247
.
Differentiation Applications 2
x
10. For the curve y = e sin x, express j in the form Ae x cos(x + a)
and show that the points of inflexion occur at x = j + kit for any
integral value of k.
(i) y = 2x 3  5x 2 + Ax  1
,... x(xl)
248
.
Programme 8
18. The velocity (v) of a piston is related to the angular velocity (oj) of
19. A right circular cone of base radius r, has a total surface area S
prove that the vertical angle (0) of the cone for maximum volume
1
is given by =2 sin (jj.
d x dx
20. Show that the equation 4tj + 4jjt + y?x = is satisfied by
2
e" M A
''
x = (At + B) , where and B are arbitrary constants. If
dx
x= and
at
t = C when t = 0, find A and B and show that the
2
maximum value of x is and that
2C
this occurs when t =
lie li
249
Programme 9
PARTIAL DIFFERENTIATION
PART1
Programme 9
Partial differentiation
1
By dy 9V 3V
= and j Now we have a new one, ris called the ^partial differential
.
J
r.
Bx dx dh dh
coefficient of V with respect to h and implies that for our present
purpose, the value of r is considered as being kept
constant
DDDaDnaDDnnDannunDDnaDnDnnnnDDDDDnnnnn
V = nr2,,h. To find
'
.3V
ov
we differentiate the given expression, taking all
3V
K
251
Partial Differentiation 1
DDnnDDDDDDDDDDDDQDnODnDQaDDDDDDDDDDnDO
Another Example
r Let us consider the area of the curved
surface of the cylinder.
7 A = 2nrh
3A
To find^ we differentiate the expression for A with respect to r, keep
ing all other symbols constant.
To find we
9A
differentiate the expression for A with respect to h, keep
JjA = 9A =
A= 2irrh 2irh and 2irr
dr dh
ODDDDnUDDDDDDDDDDDDDDDDDUDDDDDDDDnnDDO
Of course, we are not restricted tothe mensuration of the cylinder.
The same happen with any function which is a function of two
will
independent variables. For example, consider z = x 2 y 3 .
'i =2^ ^l
3=
dz
(ii) To find^, differentiate w.r.t. y, regarding x as a constant.
~=x
dy
2 2 2 2
3y = 3x y
252
Programme 9
constant
DnonnnnnnnnooQannDDnnonQaaoaQnonoon
Here are one or two examples:
2
Partial diff. w.r.t. x oix = 2x
" " " " xy = a constant factor)
''
y (y is
2
" " " y 2 =0 (y is a constant term)
= 2x+y
du ,
2
Partial diff. w.r.t. y of x
2
=0 (x is a constant term)
" " " " " xy = x (x is a constant factor)
=x + 2y
by
Another example on frame 6.
= + 3v 2  2x 2 = 3y 2  2x 2
by
And it is all just as easy as that.
i
This is a product, and the usual product rule applies except that we
253
Partial Differentiation 1
Results:
 = 24x + 14y
bx
^
^
= 14jc20j
9z
.
= (4x  2y) (3 + 0) + (3x + Sy) (4  0)
bx
= 1 2x  6y + \2x + 20y = 24a: + 14y
9z
= (4*  2y) (0 + 5) + (3x + Sy) (0  2)
9>>
= 20x  lQy  6x  lOy = 14x20y
There we are. Now what about this one?
Example* U z=^LZZ t
find ^ and ^
x +y bx by
Applying the quotient rule, we have
az_ fr+jO(20)(2xjQ(l+0) .
3y
bx (x+y) 2 (x+yf
aM ~(x+y) 2
by~ (x+yf
That was not difficult. Now you do this one:
T

=
5x +y _ 9z 9z
If z find 5 and *
x  jf,
ly ox dy
When you have finished, on to the next frame.
bz lly bz _ llx
bx (x  2y) 2 3y (x2^) 2 8
Here is the working:
9z ,
(x2y)(S+0)(Sx+y)(l0)
bx
5s 10y5sj;_ 1 !>
(x2y?  Oi,^
(*2y) :
9z
(ii) To find^ we , regard x as being constant.
y
. 9z (s2y)(0 + l)(5s+>Q(02)
" by (x  2yf
(s2>>) 2 (s2y) a
In practice, we do not write down the zeros that occur in the working,
but that is how we think.
Let us do one more example, so turn on to the next frame.
254
Programme 9
dz
(i) ^ , we treaty as constant, and
10 constant
DaanDnnnnnDDDnDnnDnannDannaaDDDnaDnann
Fine. Now here is a short exercise for you to do by way of revision
Exercise
1 2 = 4x 2 + 3xy + Sy 2
2. z = (3jc + 2y) (4x  5y)
3. z = tan(3* + 4y)
4. z =
sin(3x + 2y)

xy
Finish them all, then turn to frame 11 for the results.
255
..
Partial Differentiation 1
^~=8x
dx
+ 3y
L.
^=3x
by
+ lOy
2. z = (3x + 2y) (Ax  5y)
^=24x7y ^=7x20y
Ox ay
3. z = tan(3x + Ay)
dx
= 3 sec 2
(3x + Ay) ^
ay
= 4 sec 2
(3x + Ay)
_ sin(3x +
=
4. z
xy
bz 3x cos(3x + 2v) sin(3x + 2v)
bx x 2y
bz 2y_ cos(3x + 2v> sin (3.x + 2y)
by xy 2
DDnDnDDnnnaanQaDDDnnaDDnnnnDaDaDnDnana
If you have got all the answers correct, turn straight on to frame 15.
If you have not got all these answers, or are at all uncertain, move to
frame 12.
= Ax 2 + 3xy + 5y 2
1 z
2
To find
bz
, regard y as a constant.
r2 = 3x +
Y= + 3x + 10y, i.e. 3x + lOy .*. 10)>
256
Programme 9
13 3. z = tan(3x + Ay)
2 2
Y = sec (3x + Ay) (3) = 3 sec (3x + Ay)
2
+ Ay) (4) = 4 sec 2 (3x + Ay)
Y= sec (3x
sin(3x + 2y )
4 2 =
xy
dz _ xy cos(3x + 2y) (3)  sin(3x + 2y) (y)
bx x 2y 2
_ 3x cos(3x + 2y)  sin(3x + 2y)
x 2y
dz
Now have another go at finding r in the same way.
14 Here it is:
_ sin(3x + 2y)
=
z
xy
that for the time being, all the independent variables except the one you
are using are kept, constant and behave like constant factors or constant
terms according to their positions.
257
,
Partial Differentiation 1
The expression = 6x + Ay
bz
is itself a function of x and y. We could
Note that the operation now being performed is given by the lefthand
of the two symbols in the denominator.
2
b z 9 (3z\ i)L A , 1 .
So we have this:
= 3x 2 + Axy  Sy 2
z
16
dx
= 6x + Ay = Ax10y
by
2=6
bx
^
by.bx
A
Of course, we could carry out similar steps with the expression for ^ on
the right. This would give us:
2   10
dy
bx.by
2
KT
Note !, Z
that ^ ;
ay. ax
oy\ox)
means ^ r so  \
0Z \
means }
Z
ox. by
258
Programme 9
2
a z
rr means
a Ldz
r{
17
DDDnDDDnDnnnDnDDnnnDnDDnnnnnDDnnDDnnDD
Collecting our previous results together then, we have
z = 3x\ + 4xy  5y
2
, = 6x + 4y 4 * 1 *
; dx SfTy'
'^ 2 z
f
9.T
*>= 4 a =4
dy.dx a.x.a.y
We as, that .
see, in this case, ^ .
ay. ax
= .
dx.dy
2
a z
.
/
,'f^T = \5x
; dx
2
+6xy ^=3x
dy
'.'
2
+12y 2
2
i, a z
= 30x + 6y = 24y
^2
a*
2 ti
oy
2
* J!* = 6x a z
= 6x
dy.dx dx.dy
2 2
3 z 3 z
Again in this example also, we see that . Now do this one.
dy.dx dx.dy
It looks more complicated, but is done in just the same way. Do not rush
your time and all will be well. Here it is. Find all the first
at it; take
and second partial differential coefficients of z =x.cosyy.cosx.
Then to frame 19.
259
Partial Differentiation 1
So we get
dx =
bz
cos v + y.smx
. 3z
r =
by
x.sin y cos
2 2
b z d z
^^x.cosy
2 2
z
 a r
z
by. ox
= sin v A
+
. .
sin x b
bx.dy
.  sin
.
v +
+
.
sin a:
A A
And again, =r
2Z
by.bx
2Z
bx.by

In fact this will always be so for the functions you are likely to meet, so
that there are really three different second partial diff. coeffts. (and not
~2
four). In practice, if you have found r ^ it is a useful check to find
\2
'
V = \n(x 2 +y 2 )
3V 1 2x
2x = 
bx (jc
2
+ y2 ) x 2 +y 2
3
2
V = {x 2 +y 2 )22x.2x
2
bx
2
(x +y 2 ) 2
= 2x
2
+2y 2 4x 2 = 2y 2  2x 2 (i)
2 2 2 2 2 2
(x +y ) (x +y )
a v
2
Now you find7y in the same way and hence prove the given identity.
by
260
Programme 9
01
I We had found that
9
rrz =
2
V 2y
;*V~;
2
~2x 2
5^
9x (x + .y )
2
So making a fresh start from V = ln(x +y 2
), we get
9y = _i_, 2v= 2y
9j
2
x +j
2
^ x 2 +/
9
2
V^(* 2 +y 2 )22y.2y
2 2
by (x +y 2 ) 2
_ 2x
2
+ 2y 2  4y 2 _ 2x 2  2y 2 (ii)
2 2 2
(x
2
+J ) (x +/) 2
Substituting now the two results in the identity, gives
2
JlY 3
2
V= 2y 2x 2 2x
2
 2y 2
2 2
9x by* (x +y 2 )2 (x
2
+y 2 ) 2
z
2y  2x + 2x  2y
2 z 2
_
"^
(x
2
+j 2 ) 2
Afew o to frame 22.
2 2
Here we are told that Visa function of (x + y ) but the precise nature
of the function is not given. However, we
Much more than documents.
Discover everything Scribd has to offer, including books and audiobooks from major publishers.
Cancel anytime.