MAKERERE UNIVERSITY

COLLEGE OF ENGINEERING, DESIGN, ART AND TECHNOLOGY

SCHOOL OF ENGINEERING

DEPARTMENT OF ELECTRICAL AND COMPUTER ENGINEERING

EMT 1101: ENGINEERING MATHEMATICS I LECTURE NOTES 2018/2019

CHAPTER TWO: DIFFERENTIAL CALCULUS

Instructor: Thomas Makumbi

BSc. Eng (MUK, Uganda)
MSc. RET (MUK, Uganda)
MSc. SEE (HIG, Sweden)

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

 DIFFERENTIAL CALCULUS
The Derivative

Given a curve we can always find a deferential quotient as;

y
f 1  x0   lim
x 0 x
f  x0  x   f  x0 
 lim
x 0 x
Which is the differential of f(x) at the point x = x0.
If the limit exists at point x = x0, we say f is differentiable at x = x0. If the limit does not exist
at x = x0, then we say f is non – differentiable at x = x0. It follows that the domain of f1
consists of all points in the domain of f where f is differentiable.
The derivative function
The derivative of a function f(x) on the interval I is a function denoted f1(x) with values
f  x0  x   f x0 
f 1 x   lim , provided that this exists for all values in I. This function is
x 0 x
called the slope function of f(x).
Examples
Find the derivatives of the following functions, from first principles:
1. f ( x)  x
f  x0  x   f x0 
f 1 x   lim
x 0 x
x  x  x x  x  x
 lim *
x 0 x x  x  x

 lim
x  x  x x    x x  x   x 2

x 0
x x  x  x x
x 1 1
 lim  lim 
x 0
x x  x  x x x 0
x  x  x 2 x

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

2. f ( x)  3 x 2  1

f 1  x   lim
f  x 0  x   f  x 0 
 lim
 
3 x  x   1  3 x 2  1
2

x  0 x x  0 x

 lim
 2
 
3 x 2  2 xx  x   1  3 x 2  1
 lim
 3 x 2  6 xx  3x   1  3 x 2  1
2

x  0 x x  0 x
6 xx  3x 
2
 lim  6x
x  0 x
3. f ( x)  x 3  1 (Ans: 3x2)
Differentiation
This is the process of finding f1(x) from f(x) or more explicitly, known as differentiation with
respect to x.
d
f ( x)  f 1 ( x)
dx
Or Df  f 1 , symbol D being called an operator.
d
- the operation of differentiation w.r.t.x.
dx
Fundamental results

1.
d n
dx
 
x  nxn 1 for any rational number.
Examples
(i)
dx
 
d 67
x  67x 66

 29
(ii)
d  1  d  29
 29  
dx  x  dx
x   29 x 30  30
x

  d   1
1
d 1 1
(iii) x   x 2   x 1/ 2  1/ 2  as seen earlier.
dx dx   2 2x 2 x

2.
d
sin x   cos x
dx
Proof from first principles
f ( x  x)  f ( x) sin( x  x)  sin x
f 1 ( x)  lim  lim
x 0 x x 0 x
x  x  x x  x  x
2 cos sin
 lim 2 2
x 0 x

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

  x  x  x  x x
2 cos x   sin cos x   sin sin
 2  2  2  2  x  2
 lim  lim  lim cos x   lim
x 0 x x  0 x x  0
 2  x  0 x
2 2
 cos x
sin x
Recall: lim 0
x 0 x
3.
d
cos x    sin x
dx
Proof
f ( x  x)  f ( x) cos(x  x)  cos x
f 1 ( x)  lim  lim
x 0 x x  0 x
x  x  x x  x  x  x  x
 2 sin sin  2 sin x   sin
 lim 2 2  lim  2  2
x 0 x x  0 x
x
sin
 x  2   sin x
 lim  sin x  
x 0
 2   x
2
Rules of differentiation
We shall express these rules in terms of the functions f, g and also in terms of variables u, v
where and .
1. Addition rules
d
 f ( x)  g ( x)  f 1 ( x)  g 1 ( x)
dx
dy du dv
i.e. if y  u  v , then  
dx dx dx
Example
d
sin x  cos x   cos x  sin x
dx
d  67 1  29
 x  29   67 x  30
66

dx  x  x
2. Chain rule
f  g ( x)   f 1  g ( x) .g 1 ( x)
d
dx
i.e. If y is a function of v and v is a function of x, then y is also a function of x and;
dy dy dv
 .
dx dv dx

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

 Example
dy
Find for the following:
dx
(i) 
y  2x2  3 
2

dy dv
Let v  2 x 2  3 then; y  v 2   2v and  4x
dv dx
Therefore,
dy dy dv
  
 .  2v.4 x  2 2 x 2  3 4 x  8 x 2 x 2  3
dx dv dx

(ii) y  3x 5  5 x 2  1
27
Ans : 135x3x  23x  5x
3 5 2

1
26

(iii) y  sin 27 x
dy dv
Let v  sin x , then; y  v 27   27v 26 and  cos x
dv dx
dy dy dv
Therefore,  .  27v 26 . cos x  27 sin 26 x cos x
dx dv dx
Questions
dy
Find of the following using the chain rule:
dx
1
(i) y
x2 1
 
(ii) y  sin x 27
(iii) y  cos x
(iv) y  cos x  
3. Extended chain rule
We can easily extend the chain rule to show that if y is a function of v1, v1 a function of v2,
and v2 is a function of x, then;
dy dy dv1 dv2
 . .
dx dv1 dv2 dx
And more generally, if y is a function of v1,v2, ……vn a function of x, then;
dy dy dv1 dv
 . .......... .. n
dx dv1 dv2 dx
Example
dy
Use extended chain rule to find of the following:
dx
(i) y  sin x 2  1
Let v1  x 2  1 and v2  x 2  1
Then; y  sin v1 and v1  v21/ 2
dv 1 1/ 2 dy dv
Therefore, 1  v2 ,  cos v1 and 2  2 x
dv2 2 dv1 dx

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

 dy dy dv1 dv2 1 1
 . .  . . cos x 2  1.2 x
dx dv1 dv2 dx 2 x 2  1
x cos x 2  1

x2 1

(ii) y  sinx 2  1

 Ans : dy  x cos x  1
2
  


dx 
sin x 2  1  
1  4 cos(2 x  3) 
(iii) y   Ans :  
sin (2 x  3) 
2
sin 3 (2 x  3) 
4. Product rule
d
 f ( x).g ( x)  f ( x).g 1 ( x)  f 1 ( x).g ( x)
dx
i.e. if y  uv , then;
dy dv du
u v
dx dx dx
Examples
(i) If y  x 2 sin x , then;
dy
 x 2 cos x  2 x sin x
dx
(ii) If y  x cos x , then;
dy 1
  x sin x  cos x
dx 2 x
5. Quotient rule
d  f ( x)  g ( x). f 1 ( x)  f ( x).g 1 ( x)

dx  g ( x)  g ( x)2
u
i.e. if y  , then;
v
du dv
v u
dy
 dx 2 dx
dx v
Examples
x2 1
(i) If y  , then;
x2 1
dy

  
x 2  1 .2 x  x 2  1  4x
dx x 2
 1
2
x 2

1
2

sin x
(ii) If y  , then;
1  cos x

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

 dy 1  cos x  cos x  sin x( sin x)

dx 1  cos x 2
1

1  cos x
The quotient rule may be used to derive another fundamental result:
d
tan x   d  sin x 
dx dx  cos x 
cos x cos x  sin x( sin x) 1
 2
 2
 sec2 x
cos x cos x
d
And similarly; (cot x)   cos ec2 x
dx

6. Inverse rule
If f has an inverse f-1, so that y = f(x) and x = f-1(y), then;
d 1 1
f ( y )  1
dx f ( x)
dx 1
i.e. 
dy dy
dx
Examples
(i) Find
d
dx
 sin 1 x 

Let y  sin 1 x  x  sin y

dx dy 1 1
 cos y ; Hence,  
dy dx cos y  1  x 2
 
For the inequalities,   y  we see that only the positive value of the root should
2 2
be taken. Hence;
dy 1

dx 1 x2
dy
Note: is not defined at x  1 , i.e. is not differentiable at x  1
dx
Question
Prove the following using the inverse rule:
(i)
d
cos1 x   
1
dx 1 x2
(ii)
d
tan 1 x  
1
dx 1 x2

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

 Differentiation of functions involving exponential, logarithmic and hyperbolic functions
The natural logarithm
x
1
ln x   dx for all positive values of x. It follows that if x  1 , lnx is the area under the
1
x
1
curve y  from point 1 to point x on the x – axis. If 0  x  1 , lnx is minus the area under
x
this curve from x to the point 1, and ln1 = 0.
It follows immediately from the above definition that ln x is differentiable for all positive
values of x and that;
d 1
ln x 
dx x
d f 1 ( x)
Generally this leads to the relationship; f ( x) 
dx f ( x)
Examples
Differentiate w.r.t.x the following:
1. lnx 2  5
lnx 2  5  2
d 2x
dx x 5
2. ln3  2 x 
2
ln3  2 x  
d
dx 3  2x
 1 
3. x 2 ln1  
 x
Using the product rule;
2  x 
2
d  2  1   1
 x ln  1     x 
   2 x ln1  
dx   x   11/ x   x
 1  x 
 2 x ln1     
 x   x 1

lnbx 
d b 1
If x  0 and b is a positive constant, then; 
dx bx x

lnbx 
d d
Therefore, (ln x)
dx dx

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

 More examples
Differentiate the following w.r.t.x:
 2x 1 
1. ln 
 2x 1
 2x 1 
Let y  ln   ln(2 x  1)  ln(2 x  1)
 2x 1
dy 2 2 4
Then;    2
dx 2 x  1 2 x  1 4 x  1
2. ln x 2  x  1
Let y  ln x 2  x  1 
1
2

ln x 2  x  1 
dy 1 2 x  1
Then; 
dx 2 x 2  x  1
 x  17   7 1 
3. ln    Ans :  
 2x 1   x 1 2x  1 

4.

x 1 x2  3

3
1  x3

If y 

x 1 x2  3

; then ln y  ln

x 1 x2  3

  1

 ln x  3 ln 1  x 2  ln 1  x 3 
1 x3 3
1 x 3 3 3
1 dy 1
 
6x

x 2


1  x 2 1  x 3   6 x( x)1  x 3   x 2 1  x 2 x
y dx x 1  x 2 1  x 3 x1  x 2 1  x 3 

1  7 x 2  6 x5

   x 1  x 2 1  x3
dy 1  7 x  6 x 1  x 
2 5 2
Hence, 
dx 1  x  3 4/3

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

 The exponential function
From the graph of lnx we can see that the natural logarithm has an inverse function, the graph
of which is shown below:
y

This inverse function is called the exponential function. Thus: y  e x  x  ln y

If x is rational then;
dx 1 dy
x  ln y  y  e x and so  that is  y  ex
dy y dx

Hence,
d x
dx
 
e  ex

This suggests the general expression;

d f ( x)
dx
e  
 f 1 ( x )e f ( x )

Examples
Differentiate the following w.r.t.x:
1. e x  x 1 
2

Let y  e x  , then; dy  2 x  1e x

 x 1  x 1 
2 2

dx
1
2. y  e x

dy 1
  x  2 e1 / x   2 e1 / x
dx x
3. y  e sin x

dy
 cos xesin x
dx

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

 Differentiation of b x
If b is any positive number, we have b  e ln b

Therefore, b x  e ln b   e x ln b . Since e x ln b is defined for all x, this result suggests an obvious

x

way of extending the definition of b x ; we simply redefine b x by saying;

b x  e x ln b for all x. Using this we see that;
d x
dx
 
b  (ln b)e x ln b  b x ln b

Examples
Differentiate w.r.t.x the following:
1. 5 x
Let y  5 x , then; ln y  ln 5 x  x ln 5
1 dy dy
Hence;  ln 5   5 x ln 5
y dx dx
2. 3x  x 1
2


Let y  3x  x 1  ln y  ln 3x  x 1  x 2  x  1 ln 3
2 2

1 dy
y dx
 2 x  1 ln 3 .Therefore;
dy
dx

 2 x  1 3 x  x 1 ln 3
2

3. y  x x

ln y  ln x x  x ln x
1 dy 1
 x x 1 ln x 
dy
Applying the product rule;  x.  ln x . Hence
y dx x dx
 
4. x 2
1 sin x

 Ans : x 2  1 
sin x  2 x sin x
 
 x 2  1  cos x ln x  1  
2

  
Hyperbolic functions
We define hyperbolic sine of x (written as sinhx) and the hyperbolic cosine of x (written as
coshx) by the relations;

sinh x 
1 x
2
  1

e  e  x and cosh x  e x  e  x
2

Note Recall all identities involving hyperbolic sines and cosines.
Derivatives of hyperbolic functions


d
sinh x   cosh x
dx

d
cosh x   sinh x
dx

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

From these results we can deduce that:


d
tanh x   sec h 2 x
dx

d
coth x    cos ech 2 x
dx

d
sec hx   sec hx tanh x
dx

d
cos echx   cos echx coth x
dx
Derivatives of inverse hyperbolic functions
dx 1
These can be derived using the inverse rule of differentiation; 
dy dy
dx
Or alternatively by using, the relations:


sinh 1 x  ln x  x 2  1 ,
cosh1 x  lnx  
x 2  1 and;
1 1 x 
tan nh 1 x  ln 
2 1 x 
The results are:


d

sinh 1 x  1
dx 1 x2

d

cosh1 x   1
dx x2 1

d

tanh 1 x  1
dx 1 x2
Implicit differentiation

From the chain rule it follows that;

d
 
sinh 1 x 
1
dx 1 x2
Examples

1.
d 2
dx
 
y  2y
dy
dx
2.
d
sin y   cos y dy
dx dx
dy
From an implicit equation we can thus find by differentiating both sides of the equation with
dx
respect to x.

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

Examples
1. If x 2  y 2  1 , then;
dy dy x
2x  2 y 0 
dx dx y
2. If x  sin y  x 2 , then
1
2x 
1 dy dy 2 x
 cos y  2x  
2 x dx dx cos y
Special care is needed with terms that contain both x and y, e.g

1.
d
xy  x dy  y
dx dx
2.
dx
x y   x 2 .3 y 2
d 2 3 dy
dx
 2 xy 3  3 x 2 y 2
dy
dx
 2 xy 3

dy 2 dy
sin y.2 x  cos y x 2 x sin y  x 2 cos y
d  x2  dx  dx
3.  
dx  sin y  sin 2 y sin 2 y

4.
d
sinxy
dx
du dy
Let u  xy  x y
dx dx
d
sin u   cosu du
dx dx
d
sin( xy)  cos(xy) x dy  y 
dx  dx 
More examples
Find given that;

1. x 2  xy  y 2
dy dy dy (2 x  y )
2x  x  2 y  
dx dx dx x  2y
x
2. sin   xy
 y
dy
yx
x du dx
Let u   
y dx y2
 du  dy
sin u  xy  cos u    y  x
 dx  dx

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

  x 2
 dy  y  cos   y 
  
 x  y x  dy dy   y 
 cos   dx  x y 
 y  y
2
 dx dx  x 
 x cos   y 2 
 
  y 
Parametric differentiation
A relation between x and y is sometimes given by expressing both x and y in terms of another
variable, say, t called a parameter.
If x = f(t), y = g(t), then by the chain rule, we have;
dy dt dt 1
 g 1 ( x) and since  by the inverse rule, it follows that:
dx dx dx dx dt 
dy
dy g 1 (t ) dy
 i.e.  dt
dx f 1 (t ) dx dx
dt
Examples
1. If x  cos t , y  sin t , then;
dy cos t
   cot t
dx  sin t
1
2. If x  t  , y  3t 1/ 2  t 3 / 2 , then;
t
3 1/ 2 3 1/ 2
t  t
dy 2 2 3t 3 / 2
 
dx 1 2
1 2t  1
t
Significance of sign of derivative
We say that f(x) is increasing in the interval if f(x) increases as x increases from a to
b inclusive. We say that f(x) is increasing at the point x = x0 if there is some neighbourhood of x0
(i.e some interval with point x0 inside it) in which f(x) is increasing.
We say that f(x) is decreasing in the interval if f(x) is decreasing as x increases from a
to b inclusive, and f(x) is decreasing at point x0 if there is some neighbouhood of x0 in which f(x)
is decreasing.
The slope of the curve y = f(x) at any point equals to the value of f1(x) at that point. It follows
that f(x) is increasing at a point where and f(x) is decreasing at a point where
. A point where is called a stationary point and the tangent to the curve is
parallel to the x – axis.

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

Summary
dy
  0  y is increasing
dx
dy
  0  y is decreasing
dx
dy
  0  y is stationary.
dx
Stationary point
If changes sign from positive on the left of the stationary point to negative on its right,
such a point is reffered to as a maximum turning point.
If the sign of f1(x) changes from negative on the left of a stationary point to positive to its right,
then the point is a minimum turning point.
No change of sign from to right of a stationary point normally means a point of inflexion at that
point.
Higher derivatives
The derivative of the function f1 is denoted f11,the derivative of f11 is denoted f11, and so on.
i.e if

and so on.
Example
d2y
Find given that;
dx 2
x2  y2  1

dx 1 d 2x 1
Note: Although  , in general 2
 2
dy dy / dx dy d y / dx2

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

Example 2
d2y
Find when and
dx 2
By chain rule;

dy dx dy  sin t
  sin t and  cos t     tan t
dt dt dx cos t
By chain rule again;

Theorem of Leibnitz
This theorem is an extension of the product rule to higher derivatives. We use the notation:

and
Theorem
If y  uv , then;

dny n n n
 u n v  u v 
C1 n1 1 C2 n2 2
u v  ........  Cr unr vr  .......  uvn
dxn
n n
Where the coefficients C , C ,......... ....... are those which occur in the Binomial expansion of
1 2

.
This theorem can be easily verified for n = 1, 2, 3, ………. Using the product rule. For ,
then;

Question

Find
d6y
dx 6
, when y  x 3 sin x Ans :63x 2
  
 20 cos x  x x 2  90 sin x 

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

Significance of the sign of the second derivative
If:

 the curve is concave upwards

 the curve is convex upwards or concave downwards.
 the curve has a point of inflexion, provided it changes sign.
Examples
Find the points of inflexion on the following curves:
1. y  x 3  6 x 2  7 x  1

dy d2y
 3x  12x  7 and
2
 6 x  12
dx dx 2
For point of inflexion;
d2y
 6 x  12  0  x  2
dx 2
d2y
Clearly  0 and it changes sign at x  2 i.e. the point  2, 3 is a point of inflexion.
dx 2
2. y  x 4  4 x 3  6 x 2  2
dy
 4 x 3  12x 2  12x d2y
 2  12x 2  24x  12  12x  1
2
dx
dx
d2y d2y
The only point where we have  0 is at x = -1, but does not change sign there, hence
dx 2 dx 2
there are no points of inflexion on this curve.
Criteria for maximum and minimum turning points
The curve y =f(x) has a maximum turning point at x = a if f1(a) = 0 and if f1(x) changes sign
from positive to negative as x increases through the value a. Alternatively, there is a maximum
turning point at x = a if f1(a) = 0 and f11(a) < 0
The curve y =f(x) has a minimum turning point at x = a if f1(a) = 0 and f1(a) changes sign from
negative to positive as x increases through the value of a. Alternatively, there is a minimum
turning point at x = a if f1(a) = 0 and f11(a) > 0.
The alternative form does not enable us decide the nature of the turning point if f11(a) = 0, it may
also be inconvenient if f11(x) is difficult to find.

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

Examples
Find the turning points on the following curves and determine their nature.
1. y  2 x 3  15x 2  36x  28

dy
Hence,  0 when x = 2 and x = 3
dx
d2y
When x = 2,  6  0 so that this point is a maximum turning point.
dx2
d2y
When x =3,  6  0 so that this point is a minimum turning point
dx 2
Therefore, the turning points are:
(2, -72), maximum and (3, -109); minimum
2.

Hence only when x =1

d2y
When x =1,  0 , so the test fails to determine the nature of the point. However, it is easy
dx 2
to see that changes sign from – to + as x increases through the value x = 1, so that the
point is a minimum turning point.
3.

d2y
Hence, when x = 2. In this case it is more laborious to find than to check that
dx 2
changes sign from – to + as x increases through value 2.

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

Applications of differential calculus
Introduction
If y = f(x), then the value of at the point x = x0, i.e. f1(x0) measures:
a) The rate of change of y with respect to x at the point x = x0, or, in other words,
b) The slope of the curve y = f(x) at the point x = x0
Recall: y is increasing, decreasing or stationary as x increases through the value x0
accordingly as f1(x0) is positive, negative or zero.
Related rates
If y is a function of x and both x and y are changing with respect to time, then there must be
some connection between their two rates of change. This connection is given by the chain rule,
since we must have:
dy dy dx
 .
dt dx dt
Recommended procedure for solving related rates problems
(i) If the problem can be interpreted geometrically, draw a sketch of the problem.
(ii) Label the important quantities on your sketch, such as variables and constants according to
the statement of the problem.
(iii) From the sketch, together with known relationships; either given in the problem or known
from geometry or trigonometry, write down an equation relating the relevant variables.
(iv) Differentiate both sides of the equation with respect to time, t, using the chain rule where
necessary to obtain an equation relating the rates.
(v) Solve for the desired rate, and after doing so, substitute the given data or information into
this expression to obtain the solution.
(vi) Express the solution in terms of the original problem.
Examples
1. Water pours into a right circular cylindrical tank at the rate of 4 cm3/s. At what rate is the
level rising in the tank?
Solution
The volume (V cm3) of the water in the tank at any time t is given by; V  r 2 h
Where r cm is the radius of the tank and h cm is the depth of the water in the tank.

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

 Hence;
dV dV dh
 .
dt dh dt
dV
 r 2
dh
dV dh dV dh 4
 r 2 . but  4 and so;  2
dt dt dt dt r
4
Hence the level is rising at the rate of cm / s
r 2
2. Solve problem 1. Above for the case in which the tank is conical with vertex angle 600. (Ans:
12
)
r 2
3. An aero plane flying in a straight path at a constant height of 3 km passes directly over a
search light on the ground and is subsequently caught in the beam of the search light with the
beam at an angle of 600 to the level ground. In order to keep the aeroplane in the beam it is
found necessary to start lowering the beam at 40/s. What is the speed of the aeroplane (in
km/h) at that moment?
Solution
Let S be the position of the search light, O the point of the flight path 3 km above S. Let P be
the position of the aero plane at time, t, and suppose the distance OP is x km while SP makes
an angle θ radians with the ground, then:
O P

3 km Beam

θ
S
3 3
tan   x ; x  3 cot
x tan 
dx dx d
 .
dt d dt
dx
 3 cos ec2
d
dx d
  3 cos ec2 .
d dt

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

 4
Given the line SP rotates at 4 degrees/s, then θ is changing at rad/s. since SP is rotating
180
d d 4
clockwise, θ is decreasing so, that is negative. Hence;  .
dt dt 180

Given also   600  , then;
3
dx     4  16
 3 cos ec 2  .  
dt  3   180  180
16
It follows that x is increasing at a rate of km/s i.e 320π km/h, and therefore, the plane is
180
flying at a rate of 320π km/h.

4. A man is standing 30 m from a straight – road track. A train is approaching moving along the
track at a rate of 90 km/h. How fast is the distance between the train and the man decreasing
when the train is 50 m away from the man? (Ans: 72 km/h)
5. A ladder 8 m long leans against a wall 4 m high. The lower end of the ladder is pushed away
from the wall at a rate of 2 m/s. How fast is the angle between the top of the ladder and the
wall changing when the angle between the ladder and the wall is 600? (Ans: 0.5 rad s-1)
Problems concerning maximum and minimum values
Many problems reduce to the fundamental problem of finding the maximum or minimum value
of f(x) in same closed interval, where the function f is continuous on that interval and
differentiable at all points of the interval except (possibly) the end points.
Procedure generally applied
(a) Assign letter names to all variables and constants in the problem and if possible draw a
sketch or geometrical implication of the problem.
(b) Identify the variable for which the extreme is being sought.
(c) Find an equation of this variable in terms of other constants and variables. This is called the
principle equation.
(d) Find any other equations involving the given variables and constants. These are called the
auxiliary equations.
(e) Use the auxiliary equations to substitute for variables in the principle equation until a
function in terms of a single independent variable of interest is obtained.
(f) Determine the closed interval over which the function is defined.
(g) Find the maximum or minimum value of the function over the interval using the
known techniques of differentiation.
(h) Describe the solution found in (g) in the language of the original problem.

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

Examples
1. How should a wire of length 48 cm be bent to produce a rectangle of maximum area?
Suppose the wire is bent at a point to form one side of length x. If the other side of the
rectangle is y cm we require that,
x

2 x  2 y  48
y
y  24  x

The area of the rectangle is therefore, given by;

A  xy  x24  x 
A  24x  x 2
It is clear that A lies between 0 and 24, therefore, we can regard the problem as that of
finding the maximum value of 24 x  x 2 for values of x in the interval; 0  x  24 .
A is continuous and differentiable on this interval and;
dA
 24  2 x
dx
dA
A is maximum, when  0 therefore; x  12 . This means the wire should be bent to form
dx
a square of side 12 cm.
2. A circular cylinder is cut from a cone as shown in the figure below. Given that the radius of
the base of the cone is 2 cm and its height is 12 cm, find the maximum volume that the
cylinder can have.

Solution
Suppose the top of the cylinder is x cm from the vertex of the cone. Then the height of the
cylinder is (12 – x) cm. Let the radius of the cylinder be r cm. From similar triangles, we
have:
r 2 1
 r x
x 12 6
Hence the volume of the cylinder (V cm3) is given by:

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

 2
1 
V  r 2 h  r 2 h    x  12  x 
6 

V
36
12x 2
 x3 
We wish to find the maximum value of V on the interval . Clearly V is
continuous and differentiable on this interval, and;
dV  

dx 36

24x  3 x 2  x8  x 
12

64
Therefore, the stationary values of V occur when x = 0 and x = 8 and are V = 0 and V 
9
respectively.
64
Hence the maximum value of the volume in the interval is Vmax 
9
3. A water trough is to be constructed from 3 metal sheets of length 6 m and width 1 m. It is to
be constructed with end panels in shape of trapezoids. The end panels are made from other
materials. Find the angle at which the metal sheets should be joined so as to provide a trough
of maximum volume. (Ans: 1200)
Approximations and small errors
From the derivative f1(x) it follows immediately that:
f ( x  x)  f ( x)
 f 1 ( x) for small ∆x.
x
Hence;
f ( x  x)  f ( x)  f 1 ( x)x
Therefore, for small ∆x;
 dy 
y   .x
 dx 

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

This approximation is illustrated in the figure below:

P and Q are two points on the curve y = f(x), PT is the tangent to the curve at P, and we are
saying that the length RQ is approximately equal to the length RT provided that Q is close to P.
Consequently, if the values of f(x) and f1(x) are both known at one point, the value of f(x)
anywhere in the neighbourhood of that point can be calculated approximately.
Example
Using the concept of differentiation, calculate 3.987 approximately.

y  x when x  4, y  2
dy 1 1
  when x = 4
dx 2 x 4

Then
 dy 
y   .x
 dx 

 y  . 0.013  0.00327

1
4
Hence; 3.987  2  0.00327  1.99675
Curve sketching
To sufficiently sketch the curve of a rational function, the following points must be noted:
(a) Domain and range of the function.
(b) x and y intercepts.
(c) Establish the interval over which the function f is either increasing or decreasing.
(d) Look for maxima and minima points of the function.
(e) Find the interval over which the function f is concave upwards or concave downwards.
(f) Look for points of inflexion.

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

(g) Look for asymptotes to the curve. An asymptote to a curve is a straight line to which the
shape of the curve approximates at a great distance from the origin i.e. the distance from a
point P on the curve to the line tends to zero as the distance OP from the origin to P tends to
infinity. We distinguish between three types of asymptotes:
1. Horizontal asymptotes (asymptotes parallel to the x – axis)
The line y = b is an asymptote to the curve y = f(x), if f ( x)  b as x   or if
f ( x)  b as x   i.e. lim f ( x)  b or lim f ( x)  b
x  x 
2. Vertical asymptotes (Asymptotes parallel to the y – axis)
The line x = a is an asymptote to the curve y = f(x) if f (x)   as x  a or as
xa
3. Slant asymptotes
The line y  mx  C is an asymptote to the curve y = f(x) if f ( x)  mx  C  0 as x  
or as x  
Examples
Sketch the following curves:
x 1
1. f ( x) 
x2
Solution
Domain and range
x  2  0  x  2
Domain:  ,2U  2,
Range:
y – intercept
1
when x = 0, y 
2

f 1 ( x) 
x  2.1  x  1  1
x  12 x  22
For the interval we use the domain:
Interval f1(x) Increasing/decreasing
+ Increasing
+ Increasing
Maxima or minima
1
f 1 ( x)   0 ; Undefined hence there are no maxima or minima points.
x  22
Concavity
2
f 11 ( x)  2 x  2  
3

x  23

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

Interval f11(x) Concavity
+ Concave upwards
- Concave downwards
Points of inflexion
2
f11(x) = 0; f 11 ( x)   0 ; undefined, hence no points of inflexion.
x  23
Horizontal asymptotes
x 1 1  1/ x
lim f ( x)  b  lim x  2  lim 1  2 / x  1
x  x  x 

Therefore, y = 1 is a horizontal asymptote.

Vertical asymptotes
f(x) is not defined at x = -2,
x 1
lim x  2  
x 2

x 1
lim x  2  
x 2 

hence x = -2 is a vertical asymptote.

Sketch

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

 1 x2
2. f ( x) 
x3
Domain and range
x0
Domain:
Range:
x – intercept
1 x2
When y = 0; 3
 0  1 x2  0
x
x  1 , hence the x – intercepts are; (1, 0) and (-1, 0).
y – intercept
1 0
When x = 0, y  ; undefined, hence no y – intercepts.
0
x 3 (2 x)  1  x 2 3x 2   2 x 4  3x 2  3x 4 x 2  3
f 1 ( x)   
x6 x6 x6
Maxima or minima
x2  3
f 1 ( x)  0   0; x   3
x4
For the interval we use the domain, putting the turning points in consideration:
Interval f1(x) Increasing/decreasing?
+ Increasing
- Decreasing
- Decreasing
+ Increasing

For x   3; y 
 3   0.385   3,  0.385 is a minima point.
1
2

 3 3

1   3 
 0.385   3, 0.385 is a maxima point.
2

For x   3; y 
 3  3

Points of inflexion
 2 x 2  12
f 11 ( x) 
x5
 2 x 2  12
f 11 ( x)   0  2 x 2  12 ; x   6
x5

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

 For x   6; y
1 6   0.34   6 ,0.34
2

 63

1   6 
 0.34   6 ,0.34
2

For x   6; y
 6  3

Horizontal asymptotes
1 x2
lim  0  y  0 is a horizontal asymptote.
x   x3
Vertical asymptote
1 x2
lim   and;
x 0 x3
1 x2
lim  
x 0 x3
Sketch

Functions of two variables

A function of two variables defined on a domain D in a plane is a relationship that associates
x2  y
each point (x, y) in D with a real number f(x, y). e.g. f ( x, y)  2 .
x  y2
The graph of a function of two variables is frequently a height above or below the plane domain.
It is a set of points (x, y, z) in space related by the equation Z = f(x, y).

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

A function of two variables f(x, y) defined over a domain D has got a limit as (x, y) approaches
(a, b).
lim f ( x, y )  l , provided that f(x, y) gets close to l as (x, y) approaches (a, b) along any
( x , y )  a ,b 

curve.
Examples
1. Find lim 4  x  y
x 1,1

lim 4  x  y  lim 3  x  2
( x , y )1,1 x 1

y2  x2
2. lim
 x , y 0, 0  y 2  x 2
 Ans : Limit does not exist
Continuity
If f is a function defined in the neighbourhood of (a, b) at point (x0, y0), f is continuous at (x0, y0)
if the limit as x, y   x0 , y0  or if f x0 , y0  exists i.e.
lim f ( x, y )  f x0 , y0 
 x , y  x0 , y0 

Partial derivatives
f
The partial derivative of f with respect to x is denoted by and is defined by the equation:
x
f f ( x  x, y )  f ( x, y ) f z
 lim i.e. or is the rate of change of f(x, y) with respect to x
x x  0 x x x
when y is held fixed.
Similarly;
f f ( x, y  y )  f ( x, y ) f z
 lim i.e. or is the change of f(x, y) or z with respect to y when
y y 0 y y y
x is held fixed.

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

 f f
The notation; f x  and f y  is commonly used.
x y
f
To find , we regard f(x, y) as a function of the single variable x, treating y as a constant, and
x
we differentiate with respect to x in the usual way.
f
Similarly, we can find by ordinary differentiation with respect to y, treating x as a constant.
y
Examples
1. If z  x 2  y 2  5 xy  xy 2  2 x 2 y , then;
f
 2 x  5 y  y 2  4 xy and
x
f
 2 y  5 x  2 xy  2 x 2
y
y x
2. If z   , then;
x y
f y 1
  2  and
x x y
f 1 x
  2
y x y
3. If z  e x  y
2 2

f
 2 xe x  y and
2 2

x
f
 2 ye x  y
2 2

x
Chain rule
If z  g (u ) where u  h( x, y ) , then z is a function of x and y, and;
z dz u z dz u
 . and  .
x du x y du y
 y
4. If z  tan 1   , then;
x
y
We can let u  and hence z  tan 1 u
x
z dz u 1  y  y
 .  .  2 2
2 
x du x 1  u  x  x  y 2
z dz u 1 1 x
 .  2  
.  2
y du y 1  u  x  x  y 2

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

 1
5. If z 
2x  y
Let u = 2x +y therefore;
z dz u 1 2
 .   2 .2 
x du x u 2 x  y 2
z dz u 1 1
 .   2 .1 
y du y u 2 x  y 2
x
6. If z  y  x ln 
 y
Using product rule;
z    x   x 1 1 x x
 x ln   ln   x. .  ln   1  ln 
x x   y   y x / y  y  y   y
z x  x  x
 1   2   1 
y x/ y  y  y
Note
The idea of partial differentiation extends easily to functions of more than two variables e.g. if z
z z z
is function of three variables x, y, and t then z has three partial derivatives; , and
x y t
z
To obtain we differentiate with respect to x treating both y and t constants, and similarly for
x
z z
and .
y t
Partial derivatives of higher order
z z
If z = f(x, y), then and are themselves functions of x and y, and have partial derivatives
x y
with respect to x and y. we define the second order partial derivatives of z as:
 2 z   z 
    ……………………………………………………………………….(1)
x 2 x  x 
 2 z   z 
    ……………………………………………………………………….(2)
y 2 y  y 
2z   z 
    ………………………………………………………………………(3)
xy x  y 
2z   z 
    ………………………………………………………………………(4)
yx y  x 
The second order derivatives (1) to (4) are also denoted by fxx, fyy, fxy and fyx respectively.

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

It can be shown that;
2 z 2 z

xy yx
This result is known as the commutative property of partial derivatives.
5z
Partial derivatives of any order are defined in a similar manner. For example to find
x 3 y 2
We differentiate with respect to x three times and with respect to y twice, and the result is
independent of the order in which these differentiations are carried out.
Examples
2 z 2 z 2 z 2 z
1. Show that if z  x 3  3xy 2 , then  and 2  2  0
xy yx x y
Solution
z
 3x 2  3 y 2
x
2z
 6x
x 2
2z   z 
    6 y
yx y  x 
z
 6 xy
y
2z
 6 x
y 2
2 z   z 
    6 y
xy x  y 
2z 2z 2z 2z
  and, 2  2  6 x  6 x  0
yx yx x y

2. Show that if z  x 2  y 2  tan 1   , then;

 y
x
2z 2z 1  y  2z x2  y2
  4 tan   and 
x 2 y 2 x yx x 2  y 2
Extended chain rule
Suppose that z =f(x, y) and that x, y are both function of two other variables u, v, then z may be
z  z
expressed in terms of u and v and , may be found by normal methods of partial
u v
z  z
differentiation. Alternatively, , may be found by using a chain rule;
u v

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

z z x z y
 .  . and;
u x u y u
z z x z y
 .  . . These equations can be regarded as an extension of the chain rule.
v x v y v
Example
z z  z z 
Given z = f(x, y) with x  
1 2

u  v 2 and y  uv , show that: u  v
v u
 2 x  y 
2  y x 
Solution
x x y y
 u;  v;  v;  u
u v u v
By extended chain rule;
z z z
u v
u x y
z z z
 v  u . Which gives;
v x y

z z  z z   z z 
u v  u   v  u   v u  v 
v u  x y   x y 
z z z z z z
 uv  u2  uv  v 2  u 2  v 2   2uv
x y x y y x
Total derivative
A simple extension of the chain rule applies when z = f(x, y) and x and y are both functions of
the single variable u. Putting derivatives with respect to v equal to zero in the extended chain rule
we see that the equation reduces to the single equation:
dz z dx z dy
 .  .
du x du y du
dz
Note that in this case z is a function of the single variable u, and (sometimes called the total
du
derivative of z) is not a partial derivative.
Implicit functions
dz
We have previously found using the implicit equation of the form f(x, y) = 0. We can now
du
give a general formula, differentiating both sides of the equation f(x, y) = 0 with respect to x we
get:
dy
fx  fy 0
dx

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101

 dy  f x
 
dx fy
We may extend this idea to the equation f(x, y, z) = 0 which expresses z implicitly as a function
of the two variables x and y. differentiating the equation f(x, y, z) = 0, first with respect to x and
then with respect to y, we get;
z z
fx  fz  0 and f y  f z 0
x y

z  f x z  f y
Hence;  and 
x fz y fz
Examples
dy
1. Find at any point of the curve x 3  y 3  3xy  a 3
dx
Solution
Here we have that;
f ( x, y )  x 3  y 3  3xy  a 3
 
f x  3x 2  3 y  3 x 2  y
f  3 y  3x  3y  x 
y
2 2

dy  f  3x  y  y  x 2 2
  
3y  x 
x
Hence;
dx f y y x 2 2

z z
2. Find and at any point on the surface; x 3  y 3  z 3  3xyz  a 3
x y
f ( x, y, z )  x 3  y 3  z 3  3xyz  a 3
f x  3x 2  yz 

f y  3 y 2  xz 
fz  3z 2
 xy
Therefore;
z f
 x 

 3 x 2  yz yz  x 2
 2

x fz 
3 z 2  xy z  xy and

z

fy


 3 y  xz xz  y 2
2
 2

y fz 
3 z 2  xy z  xy 

END

Thomas Makumbi Department of Electrical &Computer Engineering EMT 1101